Context Navigation

Tests for https://bugs.webkit.org/show_bug.cgi?id=27099
Make hidden elements not focusable. This is important becaue we should
not allow tabbing to a hidden element.

The patch also makes a element without href behave like generic elements
when it comes to focus. Added tests that asserts that this works as
expected.

fast/events/click-focus-anchor-expected.txt: Added.
fast/events/click-focus-anchor.html: Added.
fast/events/resources/tabindex-focus-blur-all.js: (test): (testProgrammaticFocus):
fast/events/tab-focus-anchor-expected.txt: Added.
fast/events/tab-focus-anchor.html: Added.
fast/events/tab-focus-hidden-expected.txt: Added.
fast/events/tab-focus-hidden.html: Added.
fast/events/tabindex-focus-blur-all-expected.txt:

2009-09-06 Erik Arvidsson <arv@chromium.org>

Reviewed by Eric Seidel.

Make hidden elements not focusable.

Clean up supportsFocus and isFocusable.
supportsFocus now strictly means that the node can be focused by default.
This is true for links, form controls and content editable elements for example.
isFocusable means that the user can actually focus it.
All (2) calls to supportsFocus have been updated to take this into account.

Make a elements without href work correctly as well.

https://bugs.webkit.org/show_bug.cgi?id=27099

Tests: fast/events/click-focus-anchor.html

fast/events/tab-focus-anchor.html
fast/events/tab-focus-hidden.html

dom/Element.cpp: (WebCore::Element::focus):
dom/Node.cpp: (WebCore::Node::supportsFocus): (WebCore::Node::isFocusable):
dom/Node.h:
html/HTMLAnchorElement.cpp: (WebCore::HTMLAnchorElement::supportsFocus): (WebCore::HTMLAnchorElement::isMouseFocusable): (WebCore::HTMLAnchorElement::isKeyboardFocusable):
html/HTMLAnchorElement.h:
html/HTMLAreaElement.cpp: (WebCore::HTMLAreaElement::supportsFocus):
html/HTMLAreaElement.h:
html/HTMLElement.cpp: (WebCore::HTMLElement::supportsFocus):
html/HTMLElement.h:
html/HTMLFieldSetElement.cpp: (WebCore::HTMLFieldSetElement::supportsFocus):
html/HTMLFieldSetElement.h:
html/HTMLFormControlElement.cpp: (WebCore::HTMLFormControlElement::supportsFocus): (WebCore::HTMLFormControlElement::isFocusable):
html/HTMLFrameElementBase.cpp: (WebCore::HTMLFrameElementBase::supportsFocus):
html/HTMLFrameElementBase.h:
html/HTMLLegendElement.cpp: (WebCore::HTMLLegendElement::supportsFocus):
html/HTMLLegendElement.h:
html/HTMLOptGroupElement.cpp: (WebCore::HTMLOptGroupElement::supportsFocus): (WebCore::HTMLOptGroupElement::isFocusable):
html/HTMLOptGroupElement.h:
html/HTMLOptionElement.cpp: (WebCore::HTMLOptionElement::supportsFocus): (WebCore::HTMLOptionElement::isFocusable):
html/HTMLOptionElement.h:
page/EventHandler.cpp: (WebCore::EventHandler::dispatchMouseEvent):
page/FocusController.cpp: (WebCore::FocusController::advanceFocus):
svg/SVGAElement.cpp: (WebCore::SVGAElement::supportsFocus): (WebCore::SVGAElement::isFocusable):
wml/WMLAElement.cpp: (WebCore::WMLAElement::supportsFocus):
wml/WMLAElement.h:
wml/WMLFormControlElement.cpp: (WebCore::WMLFormControlElement::supportsFocus): (WebCore::WMLFormControlElement::isFocusable):
wml/WMLFormControlElement.h:

9:15 PM Changeset in webkit [48105] by eric@webkit.org

3 edits in trunk/WebCore

2009-09-06 Joel Stanley <joel@jms.id.au>

Reviewed by Dimitri Glazkov.

Expose functions to set the caret blink interval for Linux Chromium.
Default value is set to that of RenderTheme::caretBlankInterval().
Uses caretBlinkIntervalInternal to avoid overriding
RenderThemeChromiumSkia::caretBlankInterval to preserve the disabling
of blinking caret required for layout tests.
https://bugs.webkit.org/show_bug.cgi?id=28931

rendering/RenderThemeChromiumLinux.cpp: (WebCore::RenderThemeChromiumLinux::RenderThemeChromiumLinux): (WebCore::RenderThemeChromiumLinux::setCaretBlinkInterval): (WebCore::RenderThemeChromiumLinux::caretBlinkIntervalInternal):
rendering/RenderThemeChromiumLinux.h:

7:27 PM Changeset in webkit [48104] by adachan@apple.com

2 edits in trunk/WebCore

Remove an extra </File> in WebCore.vcproj.

WebCore.vcproj/WebCore.vcproj:

3:44 PM Changeset in webkit [48103] by xan@webkit.org

2 edits in trunk/WebKit/gtk

2009-09-06 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

Update the unitests to latest changes wrt scrollbar policy.

tests/testwindow.c: (test_webkit_window_scrollbar_policy):

1:43 PM Changeset in webkit [48102] by timothy@apple.com

6 edits
1 add in trunk/WebCore

Adds a persistent setting for color format in the Web Inspector.

https://bugs.webkit.org/show_bug.cgi?id=28889

Reviewed by Dan Bernstein.

English.lproj/localizedStrings.js: New strings.
inspector/front-end/Color.js: Minor style fixes.

(WebInspector.Color.prototype.toString): Added. Replaces the individual
"toFoo" functions. Takes a optional arguments for the color format to return.

inspector/front-end/Images/paneSettingsButtons.png: Added.
inspector/front-end/StylesSidebarPane.js:

(WebInspector.StylesSidebarPane.prototype._changeColorFormat): Update the sections to show
the new color format.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.processColor.changeColorDisplay):
Use the new toString function on Color.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.processColor): Use the persistent setting
to format colors by default. Add a tooltip for the color swatch.

inspector/front-end/inspector.css: New styles for the settings menu.
inspector/front-end/inspector.js:

(WebInspector.loaded): Initilize the colorFormat preference.

7:31 AM Changeset in webkit [48101] by xan@webkit.org

15 edits
4 adds in trunk

2009-09-06 Martin Robinson <martin.james.robinson@gmail.com>

Reviewed by Gustavo Noronha, Jan Alonzo and Xan Lopez.

[Gtk] Expose a database API
https://bugs.webkit.org/show_bug.cgi?id=27899

Expose an HTML5 database API for GTK+.

GNUmakefile.am:

LayoutTests:

2009-09-06 Martin Robinson <martin.james.robinson@gmail.com>

Reviewed by Gustavo Noronha, Jan Alonzo and Xan Lopez.

[Gtk] Expose a database API
https://bugs.webkit.org/show_bug.cgi?id=27899

Expose an HTML5 database API for GTK+.

platform/gtk/Skipped:

WebKit/gtk:

2009-09-06 Martin Robinson <martin.james.robinson@gmail.com>

Reviewed by Gustavo Noronha, Jan Alonzo and Xan Lopez.

[Gtk] Expose a database API
https://bugs.webkit.org/show_bug.cgi?id=27899

Expose an HTML5 database API for GTK+.

WebCoreSupport/ChromeClientGtk.cpp: (WebKit::ChromeClient::print):
webkit/webkit.h:
webkit/webkitdefines.h:
webkit/webkitprivate.cpp: (webkit_init):
webkit/webkitprivate.h:
webkit/webkitsecurityorigin.cpp: Added. (webkit_security_origin_finalize): (webkit_security_origin_dispose): (webkit_security_origin_set_property): (webkit_security_origin_get_property): (webkit_security_origins): (webkit_security_origin_class_init): (webkit_security_origin_init): (webkit_security_origin_get_protocol): (webkit_security_origin_get_host): (webkit_security_origin_get_port): (webkit_security_origin_get_web_database_usage): (webkit_security_origin_get_web_database_quota): (webkit_security_origin_set_web_database_quota): (webkit_security_origin_get_all_web_databases): (WebKit::kit): (WebKit::core): (webkit_security_origin_get_web_database):
webkit/webkitsecurityorigin.h: Added.
webkit/webkitwebdatabase.cpp: Added. (webkit_web_database_finalize): (webkit_web_database_dispose): (webkit_web_database_set_property): (webkit_web_database_get_property): (webkit_web_database_class_init): (webkit_web_database_init): (webkit_web_database_set_security_origin): (webkit_web_database_set_name): (webkit_web_database_get_security_origin): (webkit_web_database_get_name): (webkit_web_database_get_display_name): (webkit_web_database_get_expected_size): (webkit_web_database_get_size): (webkit_web_database_get_filename): (webkit_web_database_remove): (webkit_remove_all_web_databases): (webkit_get_web_database_directory_path): (webkit_set_web_database_directory_path): (webkit_get_default_web_database_quota):
webkit/webkitwebdatabase.h: Added.
webkit/webkitwebframe.cpp: (webkit_web_frame_new): (webkit_web_frame_get_focused_accessible_element): (webkit_web_frame_get_horizontal_scrollbar_policy):
webkit/webkitwebframe.h:
webkit/webkitwebview.cpp: (webkit_web_view_class_init):

5:38 AM Changeset in webkit [48100] by Laszlo Gombos

3 edits in trunk/WebCore

2009-09-06 Laszlo Gombos <Laszlo Gombos>

Reviewed by Eric Seidel.

[Qt] Remove Qt v4.6 check from the prefetchDNS implementation
https://bugs.webkit.org/show_bug.cgi?id=28993

platform/network/qt/DnsPrefetchHelper.cpp:
platform/network/qt/DnsPrefetchHelper.h:

12:24 AM Changeset in webkit [48099] by eric@webkit.org

24 edits
3 deletes in trunk

2009-09-06 Cameron McCormack <cam@mcc.id.au>

Reviewed by Eric Seidel.

Drop <definition-src> support
https://bugs.webkit.org/show_bug.cgi?id=28991

Update test not to expect a <definition-src> element to be of type
SVGDefinitionSrcElement.

svg/dom/font-face-elements-expected.txt:
svg/dom/font-face-elements.html:

2009-09-06 Cameron McCormack <cam@mcc.id.au>

Reviewed by Eric Seidel.

Drop <definition-src> support
https://bugs.webkit.org/show_bug.cgi?id=28991

http://dev.w3.org/SVG/profiles/1.1F2/publish/changes.html#FontsChapter

DerivedSources.cpp: Remove references to definition-src.
DerivedSources.make: Ditto.
GNUmakefile.am: Ditto.
WebCore.gypi: Ditto.
WebCore.pro: Ditto.
WebCore.vcproj/WebCore.vcproj: Ditto.
WebCore.xcodeproj/project.pbxproj: Ditto.
bindings/objc/DOM.mm: (WebCore::createElementClassMap): Ditto.
bindings/objc/DOMSVG.h: Ditto.
bindings/v8/DerivedSourcesAllInOne.cpp: Ditto.
bindings/v8/V8DOMWrapper.cpp: Ditto.
bindings/v8/V8Index.cpp: Ditto.
bindings/v8/V8Index.h: Ditto.
svg/SVGAllInOne.cpp: Ditto.
svg/SVGDefinitionSrcElement.cpp: Removed.
svg/SVGDefinitionSrcElement.h: Removed.
svg/SVGDefinitionSrcElement.idl: Removed.
svg/SVGFontFaceElement.cpp: (WebCore::SVGFontFaceElement::rebuildFontFace): Don't search for definition-src child.
svg/svgtags.in: Remove definition-src.

2009-09-06 Cameron McCormack <cam@mcc.id.au>

Reviewed by Eric Seidel.

Drop <definition-src> support
https://bugs.webkit.org/show_bug.cgi?id=28991

MigrateHeaders.make: Remove reference to ObjC definition-src binding class.

2009-09-06 Cameron McCormack <cam@mcc.id.au>

Reviewed by Eric Seidel.

Drop <definition-src> support
https://bugs.webkit.org/show_bug.cgi?id=28991

When SVG 1.1 Second Edition is a Recommendation we can remove the
entry altogether.

projects/svg/status.xml: Update page to state that <definition-src> is not supported.

Sep 5, 2009:

9:09 PM Changeset in webkit [48098] by Dimitri Glazkov

3 edits in trunk/WebCore

2009-09-05 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

[Chromium] Roll out http://trac.webkit.org/changeset/48094, because the
logic of retrieving default caret blink interval was incorrect (non-existent)
and caused layout test flakiness.

rendering/RenderThemeChromiumLinux.cpp:
rendering/RenderThemeChromiumLinux.h:

8:57 PM MathML Goals edited by alex@milowski.com

(diff)

4:47 PM Changeset in webkit [48097] by eric@webkit.org

2 edits in trunk/WebKit/qt

2009-09-05 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Eric Seidel.

[Qt] r47424 broke binary compatibility
https://bugs.webkit.org/show_bug.cgi?id=28996

Append the new DnsPrefetchEnabled attribute at the end of the enum.

Api/qwebsettings.h:

10:42 AM Changeset in webkit [48096] by jianli@chromium.org

2 edits in trunk/WebCore

Back up partial change in r48072 in order to fix chromium reliability build break.

Reviewed by NOBODY (Chromium reliability build break).

bindings/v8/V8AbstractEventListener.cpp:

(WebCore::V8AbstractEventListener::invokeEventHandler):

2:13 AM Changeset in webkit [48095] by eric@webkit.org

6 edits in trunk/WebCore

2009-09-05 Nicolas Weber <thakis@chromium.org>

Reviewed by Darin Fisher.

https://bugs.webkit.org/show_bug.cgi?id=28975
Fix file upload on chrome/mac. Images that have "Hide Extension" set,
such as screenshots by default, can now be uploaded to sites such as
imgur.com.

platform/FileSystem.h: Add pathGetDisplayFileName().
platform/chromium/FileChooserChromium.cpp: (WebCore::FileChooser::basenameForWidth): Call pathGetDisplayFileName() instead of pathGetFileName().
platform/chromium/FileSystemChromiumLinux.cpp: (WebCore::pathGetDisplayFileName): Implement pathGetDisplayFileName() by delegating to pathGetFileName().
platform/chromium/FileSystemChromiumMac.mm: (WebCore::pathGetFileName): Now returns a real filename instead of a presentational one, which fixes the problem addressed by this change. (WebCore::pathGetDisplayFileName): New function that returns a presentational filename; does what pathGetDisplayFileName() did prior to this patch.
platform/chromium/FileSystemChromiumWin.cpp: (WebCore::pathGetDisplayFileName): Implement pathGetDisplayFileName() by delegating to pathGetFileName().

2:05 AM Changeset in webkit [48094] by eric@webkit.org

3 edits in trunk/WebCore

2009-09-05 Joel Stanley <joel@jms.id.au>

Reviewed by David Levin.

Expose functions to set the caret blink interval for Linux Chromium.
https://bugs.webkit.org/show_bug.cgi?id=28931

rendering/RenderThemeChromiumLinux.cpp: (WebCore::RenderThemeChromiumLinux::setCaretBlinkInterval): (WebCore::RenderThemeChromiumLinux::caretBlinkInterval):
rendering/RenderThemeChromiumLinux.h:

1:43 AM Changeset in webkit [48093] by eric@webkit.org

2 edits in trunk/LayoutTests

2009-09-05 Eric Seidel <eric@webkit.org>

No review, results update only.

Update results after r48043.

fast/js/JSON-stringify-expected.txt:

Sep 4, 2009:

9:10 PM Changeset in webkit [48092] by mrowe@apple.com

4 edits in branches/safari-4-branch

Versioning.

9:09 PM Changeset in webkit [48091] by mrowe@apple.com

1 copy in tags/Safari-6531.17

New tag.

9:08 PM Changeset in webkit [48090] by mrowe@apple.com

2 edits in branches/safari-4-branch/WebCore

Merge r48085.

9:08 PM Changeset in webkit [48089] by mrowe@apple.com

4 edits in branches/safari-4-branch/WebCore

Merge r48070.

9:07 PM Changeset in webkit [48088] by mrowe@apple.com

2 edits in branches/safari-4-branch/WebCore

2009-09-04 Anders Carlsson <andersca@apple.com>

Reviewed by Adele Petersen.

Fix <rdar://problem/7192902>

Never throw away the first frame in an animation.

platform/graphics/BitmapImage.cpp: (WebCore::BitmapImage::destroyDecodedData):

9:07 PM Changeset in webkit [48087] by mrowe@apple.com

4 edits in branches/safari-4-branch/WebCore

Merge r48062.

8:54 PM Changeset in webkit [48086] by dimich@chromium.org

4 edits
1 add in trunk/WebCore

Prevent UI freeze in case when too many timers are in the timer queue.
https://bugs.webkit.org/show_bug.cgi?id=23865

Reviewed by David Levin.

The fix measures the elapsed time while executing timers. If we have too many
timers and it takes significant time to fire, quit the loop and reschedule.
This lets the run loop process user input (close the window for example).

platform/ThreadTimers.h:
platform/ThreadTimers.cpp:

(WebCore::ThreadTimers::sharedTimerFiredInternal):
(WebCore::ThreadTimers::fireTimersInNestedEventLoop):
Removed functions and members used before to grab a list of all expired timers.
Now we fetch one by one, measuring time and quitting the loop if it takes too long.

platform/Timer.cpp:

(WebCore::TimerBase::isActive):
(WebCore::TimerBase::setNextFireTime):
Since timers are now fired one by one, there is no need to keep track of updated timers.

manual-tests/input-starved-by-timers.html: Manual test that attempts to freeze browser by

creating of enough timers. It then keeps UI frozen and after 10 seconds starts to drain the timers
so the browser un-freezes. This is instead of a regular layout test because we don't have a way
to inject an input events via regular run loop in DRT - instead they are directly dispatched by
eventSender, which does not reproduce the UI freeze that happens with real input.

8:47 PM Changeset in webkit [48085] by andersca@apple.com

2 edits in trunk/WebCore

<rdar://problem/7201063>

Reviewed by Adele Peterson.

Detach any custom scrollbars before detaching the document.

page/Frame.cpp:

(WebCore::Frame::setView):

7:19 PM Changeset in webkit [48084] by oliver@apple.com

3 edits in trunk/JavaScriptCore

Windows build fix

7:07 PM Changeset in webkit [48083] by oliver@apple.com

6 edits in trunk

ToString? conversion should use the actual toString function for String objects.

Reviewed by Gavin Barraclough.

Remove incorrect specialisations of toString conversions on StringObject.

6:13 PM Changeset in webkit [48082] by Laszlo Gombos

5 edits in trunk/WebCore

Unreviewed, build fix.

Remove remaining references to UserStyleSheetLoader after r48075.

WebCore.gypi:
WebCore.xcodeproj/project.pbxproj:
page/qt/FrameQt.cpp:
page/wince/FrameWince.cpp:

5:35 PM Changeset in webkit [48081] by eric@webkit.org

7 edits in trunk

2009-09-04 Michelangelo De Simone <micdesim@gmail.com>

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=28988
Minor fix: "novalidate" attribute renamed in "noValidate", according to
HTML5 specs:
http://www.whatwg.org/specs/web-apps/current-work/#dom-fs-novalidate

fast/forms/novalidate-attribute-expected.txt:
fast/forms/resources/novalidate-attribute.js:

2009-09-04 Michelangelo De Simone <micdesim@gmail.com>

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=28988
Minor fix: "novalidate" attribute renamed in "noValidate", according to
HTML5 specs:
http://www.whatwg.org/specs/web-apps/current-work/#dom-fs-novalidate

html/HTMLFormElement.cpp: (WebCore::HTMLFormElement::noValidate): (WebCore::HTMLFormElement::setNoValidate):
html/HTMLFormElement.h:
html/HTMLFormElement.idl:

5:18 PM Changeset in webkit [48080] by dimich@chromium.org

2 edits in trunk/WebCore

Remove unused line of code from WorkerContext
https://bugs.webkit.org/show_bug.cgi?id=28990

Reviewed by David Levin.

workers/WorkerContext.cpp:

(WebCore::WorkerContext::importScripts):

5:03 PM Changeset in webkit [48079] by Darin Adler

2 edits in trunk/WebCore

Fix assertion seen when running buildbot.

Patch by Darin Adler <Darin Adler> on 2009-09-04
Reviewed by Oliver Hunt.

bindings/js/JSDOMBinding.h:

(WebCore::DOMObjectWithGlobalPointer::createStructure): Create the structure without
the HasDefaultMark bit since this has a custom mark function.

3:27 PM Changeset in webkit [48078] by ap@apple.com

2 edits in trunk/LayoutTests

Fix expected test output.

fast/dom/DOMImplementation/detached-doctype-expected.txt:

2:54 PM Changeset in webkit [48077] by sfalken@apple.com

3 edits in trunk/JavaScriptCore

Windows build fix.

JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Add new export.
JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def: Add new export.

2:45 PM Changeset in webkit [48076] by sfalken@apple.com

3 edits in trunk/JavaScriptCore

Windows build fix.

JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Remove unneeded export.
JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def: Remove unneeded export.

2:33 PM Changeset in webkit [48075] by hyatt@apple.com

11 edits
2 deletes in trunk/WebCore

https://bugs.webkit.org/show_bug.cgi?id=28984, remove the remote stylesheet loader on Mac/Qt.
Make sure the other code path (now used by everybody) can handle data URLs.

Reviewed by Adam Roben.

As per the post on webkit-dev, the ability to work remotely isn't working properly anyway (it doesn't
hold up the pending sheet count, so you'd just FOUC on every Web page), and it's extremely unlikely
that anybody is relying on this support.

Tor Arne agreed it was ok to disable (he enabled it on the Qt side), and only Mac/Qt were using this
code.

WebCore.pro:
WebCore.xcodeproj/project.pbxproj:
dom/Document.cpp:

(WebCore::Document::userStyleSheet):

dom/Document.h:
loader/FrameLoader.cpp:

(WebCore::FrameLoader::begin):

loader/UserStyleSheetLoader.cpp: Removed.
loader/UserStyleSheetLoader.h: Removed.
page/Frame.cpp:

(WebCore::Frame::reapplyStyles):

page/Frame.h:
page/Page.cpp:

(WebCore::Page::userStyleSheetLocationChanged):
(WebCore::Page::userStyleSheet):

page/mac/FrameMac.mm:
page/qt/FrameQt.cpp:

2:11 PM Changeset in webkit [48074] by pfeldman@chromium.org

4 edits in trunk/WebCore

2009-09-04 Mikhail Naganov <mnaganov@chromium.org>

Reviewed by Timothy Hatcher.

Add "sampling profiler" preference, simplify data display when it's set.

https://bugs.webkit.org/show_bug.cgi?id=28969

inspector/front-end/ProfileDataGridTree.js: (WebInspector.ProfileDataGridNode.prototype.get data.formatMilliseconds): (WebInspector.ProfileDataGridNode.prototype.get data):
inspector/front-end/ProfileView.js: (WebInspector.ProfileView):
inspector/front-end/inspector.js:

1:55 PM Changeset in webkit [48073] by hyatt@apple.com

1 edit in trunk/WebCore/page/Frame.cpp

Remove bogus assert that is affecting layout tests.

1:40 PM Changeset in webkit [48072] by jianli@chromium.org

10 edits in trunk/WebCore

[V8] Run-time exception in onmessage handler is not forwarded to the worker object.
https://bugs.webkit.org/show_bug.cgi?id=28980

Reviewed by Adam Barth.

Tested by worker-close.html.

bindings/v8/ScriptFunctionCall.cpp:
bindings/v8/V8AbstractEventListener.cpp:

(WebCore::V8AbstractEventListener::invokeEventHandler):

bindings/v8/V8Utilities.cpp:

(WebCore::getScriptExecutionContext):
(WebCore::reportException):

bindings/v8/V8Utilities.h:

(WebCore::getScriptExecutionContext):

bindings/v8/custom/V8MessageChannelConstructor.cpp:

(WebCore::CALLBACK_FUNC_DECL):

bindings/v8/custom/V8SharedWorkerCustom.cpp:

(WebCore::CALLBACK_FUNC_DECL):

bindings/v8/custom/V8WorkerCustom.cpp:

(WebCore::CALLBACK_FUNC_DECL):

bindings/v8/custom/V8XMLHttpRequestConstructor.cpp:

(WebCore::CALLBACK_FUNC_DECL):

bindings/v8/custom/V8XMLHttpRequestCustom.cpp:

(WebCore::CALLBACK_FUNC_DECL):

12:53 PM Changeset in webkit [48071] by hyatt@apple.com

3 edits in trunk/WebCore

Remove unused m_sheet variable and setCSSStyleSheet methods.

Reviewed by Anders Carlsson.

dom/Document.cpp:
dom/Document.h:

12:39 PM MathML Goals edited by alex@milowski.com

(diff)

12:31 PM Changeset in webkit [48070] by andersca@apple.com

4 edits in trunk/WebCore

(Based on a patch from Adele).

Reviewed by David Hyatt and Adele Peterson.

Fix <rdar://problem/7185875>.

page/Frame.cpp:

(WebCore::Frame::setView):
Detach any custom scroll bars from the old view.

page/FrameView.cpp:

(WebCore::FrameView::detachCustomScrollbars):
Simplify the check (and make it work when body is null) by just checking that the scroll bar owner is
not a RenderPart object.

rendering/RenderObject.cpp:

(WebCore::RenderObject::destroy):
ASSERT that frame() is not null here and also add a null check so that release builds won't crash.

12:09 PM Changeset in webkit [48069] by hyatt@apple.com

4 edits in trunk/WebCore

Make m_userStyleSheetLoader in Frame an OwnPtr.

Reviewed by Adam Roben.

page/Frame.cpp:

(WebCore::Frame::Frame):
(WebCore::Frame::~Frame):

page/Frame.h:
page/mac/FrameMac.mm:

(WebCore::Frame::setUserStyleSheetLocation):
(WebCore::Frame::setUserStyleSheet):

12:05 PM MathML Goals created by alex@milowski.com

12:03 PM Changeset in webkit [48068] by Darin Adler

10 edits in trunk/JavaScriptCore

DateInstance object collected on ARM JIT (JSValue: WTF_USE_JSVALUE32)
https://bugs.webkit.org/show_bug.cgi?id=28909

Patch by Darin Adler <Darin Adler> on 2009-09-04
Reviewed by Geoff Garen.

Part two.

Make some improvements to garbage collection code:

1) Create a runtime assertion that catches any classes that

override markChildren but have the HasDefaultMark bit set.

2) Remove checks of the mark bit outside the MarkStack::append

function; they are redundant.

3) Improve the efficiency of the asObject and asArray functions

when called on JSCell* to avoid a round trip to JSValue.

4) Make more callers use the checked asCell and asObject

casting functions rather than unchecked casts.

5) Removed the JSCell::marked function and other GC-related

functions because these operations are no longer things that
code other than the core GC code needs to do directly. Fixed
callers that were calling them.

runtime/Collector.cpp:

(JSC::Heap::markConservatively): Removed unneeded call to MarkStack::drain.
(JSC::Heap::markProtectedObjects): Removed unneeded check of the mark
bit and call to MarkStack::drain.
(JSC::Heap::collect): Removed unneeded checks of the mark bit and also
changed call to SmallStrings::mark to call markChildren instead to match
the rest of the objects.
(JSC::typeName): Removed unneeded cast to JSObject*.

runtime/JSArray.h:

(JSC::asArray): Added an overload for JSCell* and changed the JSValue
version to call it. Removed some unneeded casts.
(JSC::JSArray::markChildrenDirect): Marked this function inline. It's in
a header, and if not marked inline this could lead to linking problems.
(JSC::MarkStack::markChildren): Added. This helper function is used by
the drain function to avoid repating code. Also added the code here to
check fro default mark violations in debug code. If a markChildren
function adds something to the mark stack, but the type info claimed
hasDefaultMark was true, then we will get an assertion now. Also fixed
the assertion about the mark bit to use the Heap function directly
because we don't have a JSCell::marked function any more.
(JSC::MarkStack::drain): Changed a local variable from "v" to "value",
and from "currentCell" to "cell". Changed to call markChildren in two
places instead of repeating a chain of if statements twice. Changed
code that reads and writes the mark bit to use Heap::isCellMarked and
Heap::markCell so we can eliminate the JSCell::marked and
JSCell::markCellDirect functions.

runtime/JSCell.h: Removed JSCell's markCellDirect and marked member

functions. Added a comment explaining that asCell should be deprecated
in favor of the JSValue asCell member function.
(JSC::MarkStack::append): Added the assertion that catches callers
that have set the HasDefaultMark bit incorrectly. Changed
code that reads and writes the mark bit to use Heap::isCellMarked and
Heap::markCell so we can eliminate the JSCell::marked and
JSCell::markCellDirect functions. Moved the overload of
MarkStack::append for JSValue here so it can call through to the cell
version. The old version had a copy of all the code instead, but that
repeated the conversion from JSValue to JSCell* and the check for
whether a value is a cell multiple times.
(JSC::Structure::markAggregate): Moved this function here to avoid
dependencies for Structure.h, since this calls MarkStack::append.

runtime/JSObject.cpp:

(JSC::JSObject::markChildren): Added code to clear
m_isCheckingForDefaultMarkViolation so the marking done by JSObject
doesn't trigger the assertion.

runtime/JSValue.h: Moved some stray includes that were outside the

header guard inside it. Not sure how that happened! Removed the
GC-related member functions markChildren, hasChildren, marked, and
markDirect.

runtime/JSWrapperObject.h: Made markChildren private.

(JSC::JSWrapperObject::createStructure): Added. Fixes a bug where the
HasDefaultMark bit was set.

runtime/MarkStack.h: Added m_isCheckingForDefaultMarkViolation and

initialized it to false. Moved the append function body from here to
JSCell.h. Added a declaration of a private markChildren function used
inside the drain function.

runtime/SmallStrings.cpp:

(JSC::SmallStrings::markChildren): Changed the name and style of this
function to match other functions. This allows us to share the normal
mark stack code path.

runtime/SmallStrings.h: Changed the name and interface of mark to

the more-normal markChildren style.

runtime/Structure.h: Moved the body of markAggregate into the

JSCell.h to avoid a circular dependency with JSCell.h.

11:53 AM Changeset in webkit [48067] by Darin Adler

13 edits in trunk

JavaScriptCore: DateInstance object collected on ARM JIT (JSValue: WTF_USE_JSVALUE32)
https://bugs.webkit.org/show_bug.cgi?id=28909

Patch by Darin Adler <Darin Adler> on 2009-09-04
Reviewed by Geoff Garen.

Part one.

Make some improvements to garbage collection code:

1) Fix the two classes that had the default mark bit set but

should not.

2) Remove checks of the mark bit outside the MarkStack::append

function; they are redundant.

3) Make more callers use the checked asCell and asObject

casting functions rather than unchecked casts.

4) Removed some GC-related functions because these operations are

no longer things that code other than the core GC code needs
to do directly. Fixed callers that were calling them.

bytecode/CodeBlock.cpp:

(JSC::CodeBlock::markAggregate): Removed unneeded check of the mark
bit before calling MarkStack::append.

interpreter/Register.h: Removed unneeded marked and markChildren

functions.

jit/JITStubs.cpp:

(op_eq): Removed unneeded assertions, instead using checked casting
functions such as asObject.

runtime/ArgList.h: Added now-needed forward declaration of MarkStack.

runtime/GetterSetter.cpp:

(JSC::GetterSetter::markChildren): Remmoved unneeded check of the mark bit.

runtime/GlobalEvalFunction.h:

(JSC::GlobalEvalFunction::createStructure): Added. Fixes a bug where the
HasDefaultMark bit was set.

runtime/JSCell.cpp:

(JSC::JSCell::getObject): Use asObject to avoid a direct static_cast.

runtime/JSObject.h:

(JSC::asObject): Added an overload for JSCell* and changed the JSValue
version to call it.
(JSC::JSValue::get): Use asObject to avoid a direct static_cast.

runtime/JSValue.h: Moved some stray includes that were outside the

header guard inside it. Not sure how that happened! Removed the
GC-related member functions markChildren, hasChildren, marked, and
markDirect.

runtime/JSWrapperObject.h: Made markChildren private.

(JSC::JSWrapperObject::createStructure): Added. Fixes a bug where the
HasDefaultMark bit was set. Later we may want to optimize this for
wrapper types that never have cells in their internal values, but there
is no measured performance regression in SunSpider or V8 doing this
all the time.

runtime/MarkStack.cpp: Tweaked formatting.

JavaScriptGlue: * JSValueWrapper.cpp:
(JSValueWrapper::JSObjectMark): Removed a check of the mark
bit. It's OK to do more work in this case, and there is no
longer a public function to access the mark bit.

Reviewed by Geoff Garen.

11:35 AM WikiStart edited by alex@milowski.com

(diff)

11:34 AM WikiStart edited by alex@milowski.com

added MathML section (diff)

11:22 AM Changeset in webkit [48066] by kevino@webkit.org

4 edits in trunk

wx build fix. Switch USE_ defines over to the compiler so that they can be
checked by files not including config.h (like WebCorePrefix.h).

11:07 AM Changeset in webkit [48065] by timothy@apple.com

3 edits in trunk/WebCore

Show color swatches in the Web Inspector before the color text so clicking them will not
shift the swatch. Also makes multiple swatches in the single property toggle the color format.

https://bugs.webkit.org/show_bug.cgi?id=28978

Reviewed by Darin Adler.

inspector/front-end/StylesSidebarPane.js:

(WebInspector.StylePropertyTreeElement.prototype.updateTitle): Some refactoring to consolidate
the processing we do on the property value. This eliminated the old nickname code, since the new
WebInspector.Color class handles this. We could also simplify the color regex since more
is handled by the Color class. Also no longer uses innerHTML for the linkify code.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.processValue): Helper function to
process a value given a regex and processor functions.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.linkifyURL): Make an anchor for the
URL with "url()" syntax surrounding it.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.processColor): Makes a color, if
there was an exception just return a text node.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.processColor.changeColorDisplay.changeTo):
Moved from later in the file.
(WebInspector.StylePropertyTreeElement.prototype.updateTitle.processColor.changeColorDisplay): Ditto.

inspector/front-end/inspector.css:

10:59 AM Changeset in webkit [48064] by Dimitri Glazkov

13 edits in trunk

WebCore:

2009-09-04 Mark Mentovai <mark@chromium.org>

Reviewed by Dave Hyatt.

https://bugs.webkit.org/show_bug.cgi?id=28614

Account for scrollbar state changes that occur during layout.

platform/ScrollView.cpp: (WebCore::ScrollView::updateScrollbars):

Perform a layout prior to checking whether the scrollbar modes are
off, on, or automatic. The modes may change during layout.

WebCore.base.exp:
WebCore.order:
page/Frame.cpp: (WebCore::Frame::createView):
page/FrameView.cpp: (WebCore::FrameView::FrameView): (WebCore::FrameView::resetScrollbars): (WebCore::FrameView::layout):
page/FrameView.h:
platform/ScrollView.h:

Eliminate duplicated (and incorrect) scrollbar mode tracking between
FrameView and ScrollView.

WebKit/mac:

2009-09-04 Mark Mentovai <mark@chromium.org>

Reviewed by Dave Hyatt.

https://bugs.webkit.org/show_bug.cgi?id=28614

Account for scrollbar state changes that occur during layout.

WebView/WebDynamicScrollBarsView.mm: (-[WebDynamicScrollBarsView updateScrollers]):

Perform a layout prior to checking whether the scrollbar modes are
off, on, or automatic. The modes may change during layout.

WebView/WebFrameView.mm: (-[WebFrameView _install]):

Eliminate duplicated (and incorrect) scrollbar mode tracking between
FrameView and ScrollView.

WebKit/qt:

2009-09-04 Mark Mentovai <mark@chromium.org>

Reviewed by Dave Hyatt.

https://bugs.webkit.org/show_bug.cgi?id=28614

Account for scrollbar state changes that occur during layout.

Api/qwebframe.cpp: (QWebFrame::setScrollBarPolicy):

Eliminate duplicated (and incorrect) scrollbar mode tracking between
FrameView and ScrollView.

10:53 AM Changeset in webkit [48063] by yael.aharon@nokia.com

5 edits
2 adds in trunk

WebCore: hostname and host are mixed up when manipulating anchor elements.
https://bugs.webkit.org/show_bug.cgi?id=28954

Patch by Yael Aharon <yael.aharon@nokia.com> on 2009-09-04
Reviewed by Darin Adler.

Swapped the implementation of host and hostname, and made sure not to return
the port number if it is default for the given protocol.
FireFox also avoids returning the protocol number if it is default.

Test: fast/dom/Element/hostname-host.html

10:19 AM Changeset in webkit [48062] by andersca@apple.com

4 edits in trunk/WebCore

Fix <rdar://problem/7192902>.

Reviewed by Darin Adler and Steve Falkenburg.

Add the necessary plumbing for having QTMovieWin use WebCore timers.

platform/graphics/win/MediaPlayerPrivateQuickTimeWin.cpp:

(WebCore::TaskTimer::initialize):
Create the timer and call QTMovieWin::setTaskTimerFuncs.

(WebCore::TaskTimer::setTaskTimerDelay):
Start the timer.

(WebCore::TaskTimer::stopTaskTimer):
Stop the timer.

(WebCore::TaskTimer::fired):
Call QTMovieWin::taskTimerFired().

(WebCore::MediaPlayerPrivate::load):
Initialize the Task timer.

platform/graphics/win/QTMovieWin.cpp:

(updateTaskTimer):
Call the global stopTaskTimer function and setTaskTimerDelay.

(QTMovieWin::taskTimerFired):
Make this a member function.

(QTMovieWin::setTaskTimerFuncs):
Set the global function pointers.

(QTMovieWin::initializeQuickTime):
(DllMain):
No need to use the QT shared timer.

platform/graphics/win/QTMovieWin.h:

10:08 AM Changeset in webkit [48061] by levin@chromium.org

2 edits in trunk/WebCore

Fix Chromium Win compile. Missing return statement in
WebCore::RenderThemeChromiumSkia::paintSliderThumb.
https://bugs.webkit.org/show_bug.cgi?id=28974

Patch by Tony Chang <tony@chromium.org> on 2009-09-04
Reviewed by David Levin.

No new tests, build fix.

rendering/RenderThemeChromiumSkia.cpp:

(WebCore::RenderThemeChromiumSkia::paintSliderThumb):

10:07 AM Changeset in webkit [48060] by Adam Roben

2 edits in trunk/WebKit/win

Make WebView function without a parent window

When WebView has no parent window, it becomes a message-only window,
which hides it from the screen and prevents it from appearing on the
taskbar.

Fixes <http://webkit.org/b/28971> <rdar://problem/4976998>
WebView::initWithFrame fails if there's no host window

Reviewed by Dave Hyatt.

WebView.cpp:

(WebView::initWithFrame): m_hostWindow is null, use HWND_MESSAGE as
our parent window, which will make us a message-only window.
(WebView::setHostWindow): If he new host window is null, use
HWND_MESSAGE as our parent window, which will make us a message-only
window.

10:06 AM Changeset in webkit [48059] by Dimitri Glazkov

7 edits
3 deletes in trunk

2009-09-04 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

[Chromium] Rolling out http://trac.webkit.org/changeset/48035, because it broke
the build.
https://bugs.webkit.org/show_bug.cgi?id=28696

10:02 AM Changeset in webkit [48058] by andersca@apple.com

2 edits in trunk/WebCore

Fix vcproj file.

WebCore.vcproj/WebCore.vcproj:

10:01 AM Changeset in webkit [48057] by hyatt@apple.com

29 edits
6 adds in trunk

WebCore: https://bugs.webkit.org/show_bug.cgi?id=28890, make simple user script injection work.

Reviewed by Adam Roben.

This patch adds new API for adding and removing user scripts from page groups. User scripts
are bundled together in isolated worlds (you can have multiple scripts together in the same
world).

Added userscripts/ directory for holding new tests (along with a simple test of script injection).

WebCore.base.exp:
WebCore.vcproj/WebCore.vcproj:
WebCore.xcodeproj/project.pbxproj:
bindings/js/ScriptController.cpp:

(WebCore::ScriptController::evaluateInIsolatedWorld):

bindings/js/ScriptController.h:
bindings/v8/ScriptController.cpp:

(WebCore::ScriptController::evaluateInIsolatedWorld):

bindings/v8/ScriptController.h:
loader/FrameLoader.cpp:

(WebCore::FrameLoader::finishedParsing):
(WebCore::FrameLoader::dispatchDocumentElementAvailable):

page/Frame.cpp:

(WebCore::Frame::injectUserScripts):
(WebCore::Frame::injectUserScriptsForWorld):

page/Frame.h:
page/PageGroup.cpp:

(WebCore::PageGroup::~PageGroup):
(WebCore::PageGroup::addUserScript):
(WebCore::PageGroup::removeUserContentForWorld):
(WebCore::PageGroup::removeAllUserContent):

page/PageGroup.h:

(WebCore::PageGroup::userScripts):

page/UserScript.h: Added.

(WebCore::UserScript::UserScript):
(WebCore::UserScript::source):
(WebCore::UserScript::url):
(WebCore::UserScript::patterns):
(WebCore::UserScript::worldID):
(WebCore::UserScript::injectionTime):

page/UserScriptTypes.h: Added.

(WebCore::):

WebKit/mac: https://bugs.webkit.org/show_bug.cgi?id=28890, make simple user script injection work.

Reviewed by Adam Roben.

This patch adds new API for adding and removing user scripts from page groups. User scripts
are bundled together in isolated worlds (you can have multiple scripts together in the same
world).

Added userscripts/ directory for holding new tests (along with a simple test of script injection).

WebView/WebView.mm:

(+[WebView _addUserScriptToGroup:source:url:worldID:patterns:injectionTime:]):
(+[WebView _removeUserContentFromGroup:worldID:]):
(+[WebView _removeAllUserContentFromGroup:]):

WebView/WebViewPrivate.h:

LayoutTests: https://bugs.webkit.org/show_bug.cgi?id=28890, make simple user script injection work.

Reviewed by Adam Roben.

This patch adds new API for adding and removing user scripts from page groups. User scripts
are bundled together in isolated worlds (you can have multiple scripts together in the same
world).

Added userscripts/ directory for holding new tests (along with a simple test of script injection).

platform/mac/userscripts: Added.
platform/mac/userscripts/script-run-at-end-expected.txt: Added.
userscripts: Added.
userscripts/script-run-at-end.html: Added.

9:42 AM Changeset in webkit [48056] by abarth@webkit.org

1 edit in trunk/WebKit/win/WebCoreSupport/WebFrameLoaderClient.cpp

Added missing ;

9:34 AM Changeset in webkit [48055] by ap@apple.com

4 edits
2 adds in trunk

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=26402
Crashes when using a detached DocumentType node

Test: fast/dom/DOMImplementation/detached-doctype.html

DocumentType nodes are the only one that can have null document(), and they can never have
children, so they are a degenerate case for NodeLists.

dom/Node.cpp: (WebCore::Node::childNodes): Don't try register node list with document if there's no document. Since there are no changes possible for a DocumentType's node list, this is not necessary. (WebCore::Node::registerDynamicNodeList): Ditto. (WebCore::Node::unregisterDynamicNodeList): Ditto. I couldn't find a scenario where this was invoked for a DocumentType node, but I also couldn't prove that it can't be (the de-registration code path that I saw taken was via Node destructor, and that already has a null check for document).

bindings/js/JSNodeCustom.cpp: (WebCore::JSNode::addEventListener): Bail out if there is no document (just as we already do if there is no window). (WebCore::JSNode::removeEventListener): Ditto.

9:34 AM Changeset in webkit [48054] by abarth@webkit.org

2 edits in trunk/WebKit/win

2009-09-04 Adam Barth <abarth@webkit.org>

Unreviewed attempt at a build fix.

Added missing include.

WebCoreSupport/WebFrameLoaderClient.cpp:

9:23 AM Changeset in webkit [48053] by abarth@webkit.org

5 edits in trunk

2009-09-04 Adam Barth <abarth@webkit.org>

Unreviewed attempt at a build fix.

Fix typo of IWebFrameLoadDelegatePrivate2 and make
IWebFrameLoadDelegatePrivate inherit from IWebFrameLoadDelegatePrivate
as requestd by Adam Roben.

Interfaces/IWebFrameLoadDelegatePrivate2.idl:
WebCoreSupport/WebFrameLoaderClient.cpp: (WebFrameLoaderClient::didDisplayInsecureContent): (WebFrameLoaderClient::didRunInsecureContent):

2009-09-04 Adam Barth <abarth@webkit.org>

Unreviewed build fix.

Update declaration of FrameLoadDelegate to reflect that
IWebFrameLoadDelegatePrivate2 inherits from
IWebFrameLoadDelegatePrivate.

DumpRenderTree/win/FrameLoadDelegate.h:

8:53 AM Changeset in webkit [48052] by vestbo@webkit.org

2 edits in trunk/LayoutTests

[Qt] Fix brainfart in r48050, tests should have paths

Rubber-stamped by Simon Hausmann.

platform/qt/Skipped:

8:49 AM Changeset in webkit [48051] by abarth@webkit.org

13 edits
2 adds in trunk

2009-09-04 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=24696

Plumb mixed content notifications to WebFrameLoadDelegatePrivate.

WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::didDisplayInsecureContent): (WebFrameLoaderClient::didRunInsecureContent):
WebView/WebDelegateImplementationCaching.h:
WebView/WebFrameLoadDelegatePrivate.h: Added.

2009-09-04 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=24696

Plumb mixed content notifications to IWebFrameLoadDelegatePrivate.

Interfaces/IWebFrameLoadDelegatePrivate2.idl: Added.
Interfaces/WebKit.idl:
WebCoreSupport/WebFrameLoaderClient.cpp: (WebFrameLoaderClient::didDisplayInsecureContent): (WebFrameLoaderClient::didRunInsecureContent):
WebKit.vcproj/Interfaces.vcproj:

2009-09-04 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=24696

Add testing instrumentation for mixed content.

DumpRenderTree/mac/FrameLoadDelegate.mm: (-[FrameLoadDelegate webView:]): (-[FrameLoadDelegate webView:didRunInsecureContent:]):
DumpRenderTree/win/FrameLoadDelegate.cpp: (descriptionSuitableForTestResult): (FrameLoadDelegate::QueryInterface): (FrameLoadDelegate::didDisplayInsecureContent): (FrameLoadDelegate::didRunInsecureContent):
DumpRenderTree/win/FrameLoadDelegate.h:

8:34 AM Changeset in webkit [48050] by vestbo@webkit.org

2 edits in trunk/LayoutTests

[Qt] Skip tests failing due to missing DRT features

Rubber-stamped by Simon Hausmann.

platform/qt/Skipped:

7:04 AM Changeset in webkit [48049] by xan@webkit.org

2 edits in trunk/WebCore

2009-09-04 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

[GTK] loadFinished never emitted
https://bugs.webkit.org/show_bug.cgi?id=28935

Do not report errors right away from
ResourceHandle::start. Instead return true and schedule an error
to be reported in the handle later. Otherwise WebCore gets
confused and thinks the resource never finished loading, making it
not emit the finished status.

platform/network/soup/ResourceHandleSoup.cpp: (WebCore::startHttp): (WebCore::):

7:00 AM Changeset in webkit [48048] by yong.li@torchmobile.com

2 edits in trunk/JavaScriptCore

2009-09-03 Yong Li <yong.li@torchmobile.com>

Reviewed by David Levin.

Remove unnecessary dependency on unistd.h
https://bugs.webkit.org/show_bug.cgi?id=28962

runtime/Completion.cpp:

3:28 AM Changeset in webkit [48047] by eric@webkit.org

3 edits in trunk/WebCore

2009-09-04 Ben Murdoch <benm@google.com>

Reviewed by Eric Seidel.

[V8] DOM Storage bindings: Event Handler should create StorageEvents
https://bugs.webkit.org/show_bug.cgi?id=28942

Tests:
Causes the following tests that were failing with V8 to pass:

storage/domstorage/localstorage/onstorage-attribute-setwindow.html
storage/domstorage/localstorage/complex-keys.html
storage/domstorage/localstorage/onstorage-attribute-markup.html
storage/domstorage/localstorage/simple-events.html
storage/domstorage/localstorage/onstorage-attribute-set-attribute.html
storage/domstorage/localstorage/index-get-and-set.html
storage/domstorage/sessionstorage/onstorage-attribute-setwindow.html
storage/domstorage/sessionstorage/onstorage-attribute-markup.html
storage/domstorage/sessionstorage/simple-events.html
storage/domstorage/sessionstorage/onstorage-attribute-set-attribute.html
storage/domstorage/sessionstorage/index-get-and-set.html

Code changes:

bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::convertEventToV8Object): Create StorageEvents.
bindings/v8/custom/V8StorageCustom.cpp: (WebCore::storageGetter): Check if we are retrieving the length property and return the number of items in the object rather than a property named length when invoking the getItem() member of local/sessionStorage.

2:04 AM Changeset in webkit [48046] by pfeldman@chromium.org

19 edits
1 add in trunk/WebCore

2009-09-03 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

https://bugs.webkit.org/show_bug.cgi?id=28911

WebCore.gypi:
WebCore.vcproj/WebCore.vcproj:
inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::dispatchOnInjectedScript):
inspector/InspectorBackend.h:
inspector/InspectorBackend.idl:
inspector/front-end/ConsoleView.js: (WebInspector.ConsoleView.prototype.doEvalInWindow): (WebInspector.ConsoleView.prototype._formatarray): (WebInspector.ConsoleView.prototype._formatnode):
inspector/front-end/DOMAgent.js:
inspector/front-end/ElementsPanel.js: (WebInspector.ElementsPanel.this.treeOutline.focusedNodeChanged): (WebInspector.ElementsPanel.prototype.searchCanceled): (WebInspector.ElementsPanel.prototype.performSearch):
inspector/front-end/InjectedScript.js: (InjectedScript.dispatch):
inspector/front-end/InjectedScriptAccess.js: Added. (InjectedScriptAccess._installHandler.InjectedScriptAccess.methodName): (InjectedScriptAccess._installHandler):
inspector/front-end/MetricsSidebarPane.js: (WebInspector.MetricsSidebarPane.prototype.update): (WebInspector.MetricsSidebarPane.prototype.editingCommitted):
inspector/front-end/ObjectPropertiesSection.js: (WebInspector.ObjectPropertiesSection.prototype.update): (WebInspector.ObjectPropertyTreeElement.prototype.onpopulate): (WebInspector.ObjectPropertyTreeElement.prototype.applyExpression):
inspector/front-end/PropertiesSidebarPane.js: (WebInspector.PropertiesSidebarPane.prototype.update):
inspector/front-end/ResourcesPanel.js: (WebInspector.ResourceSidebarTreeElement.prototype.ondblclick):
inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.doEvalInCallFrame):
inspector/front-end/StylesSidebarPane.js: (WebInspector.StylesSidebarPane.prototype.update): (WebInspector.StylePropertiesSection.prototype.editingSelectorCommitted): (WebInspector.BlankStylePropertiesSection.prototype.editingCommitted): (WebInspector.StylePropertyTreeElement.prototype):
inspector/front-end/inspector.html:

1:47 AM Changeset in webkit [48045] by eric@webkit.org

5 edits in trunk/WebCore

2009-09-04 Tony Chang <tony@chromium.org>

Reviewed by Eric Seidel.

Implement <input type="range"> for Chromium Linux by painting
the slider and the thumb using skia.
https://bugs.webkit.org/show_bug.cgi?id=28928

Covered by existing tests (we're currently hitting an assert for them).

rendering/RenderThemeChromiumLinux.cpp: (WebCore::RenderThemeChromiumLinux::adjustSliderThumbSize):
rendering/RenderThemeChromiumLinux.h:
rendering/RenderThemeChromiumSkia.cpp: (WebCore::drawVertLine): (WebCore::drawHorizLine): (WebCore::drawBox): (WebCore::RenderThemeChromiumSkia::paintSliderTrack): (WebCore::RenderThemeChromiumSkia::paintSliderThumb):
rendering/RenderThemeChromiumSkia.h:

1:40 AM Changeset in webkit [48044] by eric@webkit.org

3 edits in trunk

2009-09-04 Albert Bachand <albertb@google.com>

Reviewed by Eric Seidel.

Modify the condition regarding the key modifier for opening combo
boxes under GTK from just PLATFORM(GTK) to also look for
(PLATFORM(CHROMIUM) && PLATFORM(LINUX)).

WebCore/dom/SelectElement.cpp:
WebCore/manual-tests/select-popup-on-spacebar.html:

1:32 AM Changeset in webkit [48043] by eric@webkit.org

6 edits
6 adds in trunk

2009-09-04 Kent Tamura <tkent@chromium.org>

Reviewed by Eric Seidel.

Tests for <input type=color> validation.
https://bugs.webkit.org/show_bug.cgi?id=28966

fast/forms/ValidityState-patternMismatch-unsupported-expected.txt: Added.
fast/forms/ValidityState-patternMismatch-unsupported.html: Added.
fast/forms/ValidityState-typeMismatch-color-expected.txt: Added.
fast/forms/ValidityState-typeMismatch-color.html: Added.
fast/forms/resources/ValidityState-patternMismatch-unsupported.js: Added.
fast/forms/resources/ValidityState-typeMismatch-color.js: Added.

2009-09-04 Kent Tamura <tkent@chromium.org>

Reviewed by Eric Seidel.

Implement <input type=color> as a text field.
ValidityState::typeMismatch supports for type=color. https://bugs.webkit.org/show_bug.cgi?id=28966

Tests: fast/forms/ValidityState-patternMismatch-unsupported.html

fast/forms/ValidityState-typeMismatch-color.html

html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::valueMissing): (WebCore::HTMLInputElement::patternMismatch): (WebCore::HTMLInputElement::setInputType): (WebCore::HTMLInputElement::formControlType): (WebCore::HTMLInputElement::saveFormControlState): (WebCore::HTMLInputElement::restoreFormControlState): (WebCore::HTMLInputElement::accessKeyAction): (WebCore::HTMLInputElement::rendererIsNeeded): (WebCore::HTMLInputElement::createRenderer): (WebCore::HTMLInputElement::appendFormData): (WebCore::HTMLInputElement::valueWithDefault): (WebCore::HTMLInputElement::storesValueSeparateFromAttribute): (WebCore::HTMLInputElement::defaultEventHandler): (WebCore::HTMLInputElement::isRequiredFormControl): (WebCore::HTMLInputElement::dataList):
html/HTMLInputElement.h: (WebCore::HTMLInputElement::): (WebCore::HTMLInputElement::isTextField):
html/ValidityState.cpp: (WebCore::ValidityState::typeMismatch): (WebCore::ValidityState::isValidColorString):
html/ValidityState.h:

1:07 AM Changeset in webkit [48042] by levin@chromium.org

2 edits in trunk/WebCore

[V8] Crash when updating document cache
https://bugs.webkit.org/show_bug.cgi?id=28965

Patch by Mads Ager <ager@chromium.org> on 2009-09-04
Reviewed by David Levin.

Initialization of a V8 JavaScript context can fail. When that
happens, do not attempt to update the document wrapper cache.

This code path is exercised by http/tests/security/javascriptURL/javascriptURL-in-new-iframe.html.

bindings/v8/V8Proxy.cpp:

(WebCore::V8Proxy::updateDocument):

12:50 AM Changeset in webkit [48041] by eric@webkit.org

2 edits in trunk/WebKit/qt

2009-09-04 Laszlo Gombos <Laszlo Gombos>

Reviewed by Eric Seidel.

[Qt] Fix tst_QWebPage::database autotest failure
https://bugs.webkit.org/show_bug.cgi?id=28961

Make sure that the test case enables the feature before
the feature gets tested

tests/qwebpage/tst_qwebpage.cpp: (tst_QWebPage::database):

12:15 AM Changeset in webkit [48040] by eric@webkit.org

5 edits in trunk

2009-09-04 Yaar Schnitman <yaar@chromium.org>

Reviewed by Eric Seidel.

Two bugs:

https://bugs.webkit.org/show_bug.cgi?id=25489
CSS property background-position is now serialized in shorthand string as
background-position-x background-position-y, as the specification requires
(See http://www.w3.org/TR/2008/WD-css3-background-20080910/#the-background-position).

https://bugs.webkit.org/show_bug.cgi?id=26541
CSS background-color is now be serialized before other background properties,
as CSS specification requires (See http://dev.w3.org/csswg/cssom/#css-value).

An existing layout test that accepted wrong order and skipped position test
had to be modified.

fast/dom/background-shorthand-csstext-expected.txt:
fast/dom/background-shorthand-csstext.html:

2009-09-04 Yaar Schnitman <yaar@chromium.org>

Reviewed by Eric Seidel.

Two bugs:

https://bugs.webkit.org/show_bug.cgi?id=25489
CSS property background-position is now serialized in shorthand string as
background-position-x background-position-y, as the specification requires
(See http://www.w3.org/TR/2008/WD-css3-background-20080910/#the-background-position).

https://bugs.webkit.org/show_bug.cgi?id=26541
CSS background-color is now be serialized before other background properties,
as CSS specification requires (See http://dev.w3.org/csswg/cssom/#css-value).

An existing layout test that accepted wrong order and skipped position test
had to be modified.

css/CSSMutableStyleDeclaration.cpp: (WebCore::CSSMutableStyleDeclaration::getPropertyValue):

12:07 AM Changeset in webkit [48039] by eric@webkit.org

4 edits in trunk/WebKit/qt

2009-09-04 Benjamin C Meyer <benjamin.meyer@torchmobile.com>

Reviewed by Eric Seidel.

Any QtWebKit application crashes on exit if the inspector is
used during the session rather then exiting.

If a QWebView has a QWebPage that is destroyed the QWebView
does not update its pointer and will crash the next time
it access the page pointers (such as in its destructor).

InspectorClientView should not call deleteLater when the page
is destroyed because it does not have a parent and is a top
level widget. close() needs to be called so that QApplication
can exit the application if quitOnLastWindowClosed is set
and a InspectorClientView is the last window, otherwise
the application will never exit.

Api/qwebview.cpp: (QWebViewPrivate::_q_pageDestroyed): (QWebView::setPage):
Api/qwebview.h:
WebCoreSupport/InspectorClientQt.cpp: (WebCore::InspectorClientView::InspectorClientView):

Sep 3, 2009:

10:24 PM Changeset in webkit [48038] by eric@webkit.org

4 edits
3 adds in trunk

2009-09-03 Kent Tamura <tkent@chromium.org>

Reviewed by David Levin.

Test for HTMLTextAreaElement.textLength
https://bugs.webkit.org/show_bug.cgi?id=28929

fast/forms/resources/textarea-textlength.js: Added.
fast/forms/textarea-textlength-expected.txt: Added.
fast/forms/textarea-textlength.html: Added.

2009-09-03 Kent Tamura <tkent@chromium.org>

Reviewed by David Levin.

Add support for HTMLTextAreaElement.textLength
https://bugs.webkit.org/show_bug.cgi?id=28929

Test: fast/forms/textarea-textlength.html

html/HTMLTextAreaElement.h: (WebCore::HTMLTextAreaElement::textLength):
html/HTMLTextAreaElement.idl:

5:27 PM Changeset in webkit [48037] by ap@apple.com

3 edits
3 adds in trunk

Reviewed by Darin Adler.

<rdar://problem/7180197>, https://bugs.webkit.org/show_bug.cgi?id=28822
REGRESSION(r31231): Creating document with current document's DOCTYPE causes crashes

Test: fast/dom/DOMImplementation/createDocument-with-used-doctype.html

dom/DOMImplementation.cpp: (WebCore::DOMImplementation::createDocument): Even though we cannot raise an exception immediately out of fear of breaking Acid3, we shouldn't use a DocumentType node in two documents.

4:57 PM Changeset in webkit [48036] by beidson@apple.com

3 edits
9 adds in trunk

WebCore: Page Cache should support pages with Frames.
https://bugs.webkit.org/show_bug.cgi?id=13631

Reviewed by Sam Weinig.

Tests: fast/events/pageshow-pagehide-on-back-cached-with-frames.html

fast/loader/frames-with-unload-handlers-in-page-cache.html

loader/FrameLoader.cpp:

(WebCore::FrameLoader::canCachePageContainingThisFrame):
(WebCore::FrameLoader::logCanCacheFrameDecision):

LayoutTests: Page Cache should support pages with Frames.
https://bugs.webkit.org/show_bug.cgi?id=13631

Reviewed by Sam Weinig.

fast/events/pageshow-pagehide-on-back-cached-with-frames-expected.txt: Added.
fast/events/pageshow-pagehide-on-back-cached-with-frames.html: Added.
fast/events/resources/pageshow-pagehide-subframe-cachable.html: Added.
fast/events/resources/pageshow-pagehide-subsubframe-cachable.html: Added.
fast/loader/frames-with-unload-handlers-in-page-cache-expected.txt: Added.
fast/loader/frames-with-unload-handlers-in-page-cache.html: Added.
fast/loader/resources/notify-done.html: Added.
fast/loader/resources/subframe-with-subsubframe-with-unload-handler.html: Added.
fast/loader/resources/subframe-with-unload-handler.html: Added.

4:23 PM Changeset in webkit [48035] by eric@webkit.org

7 edits
5 adds in trunk

2009-09-03 Steve VanDeBogart <vandebo@chromium.org>

Reviewed by Eric Seidel.

Check that middle click triggers an onpaste event in Linux.
https://bugs.webkit.org/show_bug.cgi?id=28696

platform/gtk/editing/pasteboard: Added.
platform/gtk/editing/pasteboard/middle-click-onpaste-expected.txt: Added.
platform/gtk/editing/pasteboard/resources: Added.
platform/gtk/editing/pasteboard/resources/TEMPLATE.html: Added.
platform/gtk/editing/pasteboard/resources/middle-click-onpaste.js: Added. (handlePaste):

2009-09-03 Steve VanDeBogart <vandebo@chromium.org>

Reviewed by Eric Seidel.

Handle middle click in Chromium like QT
https://bugs.webkit.org/show_bug.cgi?id=28696

Tested by middle-click-onpaste.html.

platform/Pasteboard.h:
platform/chromium/ChromiumBridge.h:
platform/chromium/ClipboardChromium.cpp: (WebCore::ClipboardChromium::getData):
platform/chromium/PasteboardChromium.cpp: (WebCore::Pasteboard::Pasteboard): (WebCore::Pasteboard::isSelectionMode): (WebCore::Pasteboard::setSelectionMode): (WebCore::Pasteboard::canSmartReplace): (WebCore::Pasteboard::plainText): (WebCore::Pasteboard::documentFragment):
platform/chromium/PasteboardPrivate.h: (WebCore::PasteboardPrivate::):

4:10 PM Changeset in webkit [48034] by beidson@apple.com

2 edits in trunk/WebCore

More partial work towards "Page Cache should support pages with Frames"
https://bugs.webkit.org/show_bug.cgi?id=13631

Reviewed by Sam Weinig.

Fix a bug where subframes weren't asked if they were cacheable.

No new tests, especially since page caching for frames isn't turned on yet.
A layout test will come in my next patch, which enables frames in the page cache.

loader/FrameLoader.cpp:

(WebCore::FrameLoader::canCachePageContainingThisFrame):

4:08 PM Changeset in webkit [48033] by abarth@webkit.org

2 edits in trunk/WebKit/mac

2009-09-03 Adam Barth <abarth@webkit.org>

Unreviewed build fix.

Change notImplemented() to a FIXME.

WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::didDisplayInsecureContent): (WebFrameLoaderClient::didRunInsecureContent):

3:59 PM Changeset in webkit [48032] by abarth@webkit.org

18 edits in trunk

2009-09-03 Adam Barth <abarth@webkit.org>

Reviewed by eric@webkit.org.

https://bugs.webkit.org/show_bug.cgi?id=24696

Added mixed content methods to FrameLoaderClient.

loader/EmptyClients.h: (WebCore::EmptyFrameLoaderClient::didDisplayInsecureContent): (WebCore::EmptyFrameLoaderClient::didRunInsecureContent):
loader/FrameLoaderClient.h:

2009-09-03 Adam Barth <abarth@webkit.org>

Reviewed by eric@webkit.org.

https://bugs.webkit.org/show_bug.cgi?id=24696

Stub implementations of mixed content methods of FrameLoaderClient.

WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::didDisplayInsecureContent): (WebKit::FrameLoaderClient::didRunInsecureContent):
WebCoreSupport/FrameLoaderClientGtk.h:

2009-09-03 Adam Barth <abarth@webkit.org>

Reviewed by eric@webkit.org.

https://bugs.webkit.org/show_bug.cgi?id=24696

Stub implementations of mixed content methods of FrameLoaderClient.

WebCoreSupport/WebFrameLoaderClient.h:
WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::didDisplayInsecureContent): (WebFrameLoaderClient::didRunInsecureContent):

2009-09-03 Adam Barth <abarth@webkit.org>

Reviewed by eric@webkit.org.

https://bugs.webkit.org/show_bug.cgi?id=24696

Stub implementations of mixed content methods of FrameLoaderClient.

WebCoreSupport/FrameLoaderClientQt.cpp: (WebCore::FrameLoaderClientQt::didDisplayInsecureContent): (WebCore::FrameLoaderClientQt::didRunInsecureContent):
WebCoreSupport/FrameLoaderClientQt.h:

2009-09-03 Adam Barth <abarth@webkit.org>

Reviewed by eric@webkit.org.

https://bugs.webkit.org/show_bug.cgi?id=24696

Stub implementations of mixed content methods of FrameLoaderClient.

WebCoreSupport/WebFrameLoaderClient.cpp: (WebFrameLoaderClient::didDisplayInsecureContent): (WebFrameLoaderClient::didRunInsecureContent):
WebCoreSupport/WebFrameLoaderClient.h:

2009-09-03 Adam Barth <abarth@webkit.org>

Reviewed by eric@webkit.org.

https://bugs.webkit.org/show_bug.cgi?id=24696

Stub implementations of mixed content methods of FrameLoaderClient.

WebKitSupport/FrameLoaderClientWx.cpp: (WebCore::FrameLoaderClientWx::didDisplayInsecureContent): (WebCore::FrameLoaderClientWx::didRunInsecureContent):
WebKitSupport/FrameLoaderClientWx.h:

2:37 PM Changeset in webkit [48031] by xan@webkit.org

2 edits in trunk/WebKit/gtk

2009-09-03 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

Access private member variables directly instead of using the
accessor.

WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::createFrame): (WebKit::FrameLoaderClient::windowObjectCleared):

1:25 PM Changeset in webkit [48030] by atwilson@chromium.org

2 edits in trunk/LayoutTests

Exception.toString() on Qt seems to print out a different value than other platforms.
https://bugs.webkit.org/show_bug.cgi?id=28952

Reviewed by Dave Levin.

platform/qt/Skipped:

Disabled new worker tests that rely on the string format of DOM exceptions until bug #28952 is resolved.

12:38 PM Changeset in webkit [48029] by kevino@webkit.org

19 edits
6 adds in trunk

Reviewed by Kevin Ollivier.

[wx] Frames support
https://bugs.webkit.org/show_bug.cgi?id=19041

12:35 PM Changeset in webkit [48028] by yael.aharon@nokia.com

3 edits in trunk/WebCore

[Qt] Page content is not displayed in case of HTTP status error
https://bugs.webkit.org/show_bug.cgi?id=28949

Patch by Yael Aharon <yael.aharon@nokia.com> on 2009-09-03
Reviewed by Simon Hausmann.

If the server sent content, show it to the user instead of an error,
even if the HTTP status code is not success.

platform/network/qt/QNetworkReplyHandler.cpp:
platform/network/qt/QNetworkReplyHandler.h:

12:07 PM Changeset in webkit [48027] by ddkilzer@apple.com

4 edits in trunk/WebKitTools

<http://webkit.org/b/28880> svn-apply --force doesn't actually work

Reviewed by Eric Seidel.

This fixes "svn-apply --force" and adds unit tests for the
scm.apply_patch() method which uses this script.

Scripts/svn-apply: Created $globalExitCode variable that defaults to 0. Exit with a value of $globalExitCode when the script is finished. (applyPatch): Ignore a non-zero $exitCode if $force is true, but set $globalExitCode to $exitCode so that svn-apply exits with a non-zero status if any patches did not apply cleanly. Also print out the actual patch command if $force was not true.

Scripts/modules/scm.py: (scripts_directory): Added. Extracted from script_path(). (script_path): Extracted scripts_directory().
Scripts/modules/scm_unittest.py: Import urllib. (SVNTestRepository.setup): Save the original working directory in test_object since this represents the WebKit repository from where the unit tests are run. (SCMTest): Created new super class to hold utility methods. (SCMTest._create_patch): Creates a patch file on disk and a dictionary for use with scm.svn_apply(). (SCMTest._setup_webkittools_scripts_symlink): Sets up a symlink back to WebKitTools/Scripts in the test repository so that scm.apply_patch() is able to find the svn-apply script. (SVNTest): Inherit from SCMTest instead of unittest.TestCase. (SVNTest.tearDown): Make sure to change directories back to the original_path before the next test. (SVNTest.test_apply_svn_patch): New test case for applying an svn patch with scm.apply_patch(). (SVNTest.test_apply_svn_patch_force): New test case for applying an svn patch with scm.apply_patch() that conflicts. (GitTest): Inherit from SCMTest instead of unittest.TestCase. (GitTest.tearDown): Make sure to change directories back to the original_path before the next test. (GitTest.test_apply_git_patch): New test case for applying a git patch with scm.apply_patch(). (GitTest.test_apply_git_patch_force): New test case for applying a git patch with scm.apply_patch() that conflicts.

11:36 AM Changeset in webkit [48026] by atwilson@chromium.org

7 edits
2 adds in trunk/WebCore

Need to update v8 bindings to support passing multiple ports to postMessage()
https://bugs.webkit.org/show_bug.cgi?id=28839

Reviewed by Dimitri Glazkov.

Added support for passing MessagePortArray to v8 bindings.

New tests that now pass with V8:

fast/events/message-port-multi.html
fast/workers/worker-context-multi-port.html
fast/workers/worker-multi-port.html

WebCore.gypi:

Added V8MessageEventCustom.cpp and V8MessagePortCustom.h

bindings/v8/custom/V8CustomBinding.h:

Added custom postMessage() handlers for various classes.

bindings/v8/custom/V8DOMWindowCustom.cpp:

(WebCore::DOMWindowPostMessage):
Added custom handler that supports MessagePortArray.

bindings/v8/custom/V8DedicatedWorkerContextCustom.cpp:

(WebCore::DedicatedWorkerContextPostMessage):
Added custom handler that supports MessagePortArray.

bindings/v8/custom/V8MessageEventCustom.cpp: Added.

(WebCore::MessageEventPorts):
Added getter for ports attribute.
(WebCore::MessageEventInitMessageEvent):
Added custom handler that supports MessagePortArray.

bindings/v8/custom/V8MessagePortCustom.cpp:

(WebCore::MessagePortPostMessage):
Added custom handler that supports MessagePortArray.
(WebCore::getMessagePortArray):
Added helper function that supports converting from a sequence-like object to a MessagePortArray.

bindings/v8/custom/V8MessagePortCustom.h: Added.

Added declaration for getMessagePortArray().

bindings/v8/custom/V8WorkerCustom.cpp:

(WebCore::WorkerPostMessage):
Added custom handler that supports MessagePortArray.

11:36 AM Changeset in webkit [48025] by atwilson@chromium.org

47 edits
1 copy
12 adds in trunk

WebCore: Need to update JS bindings and IDL files to support multiple message ports in postMessage()
https://bugs.webkit.org/show_bug.cgi?id=28460

Reviewed by Sam Weinig.

Added new toJSSequence() API which validates that a JSValue meets the WebIDL criteria for a sequence.

Tests: fast/events/message-port-multi.html

fast/workers/worker-context-multi-port.html
fast/workers/worker-multi-port.html

GNUmakefile.am:
WebCore.gypi:
WebCore.pro:
WebCore.vcproj/WebCore.vcproj:
WebCore.xcodeproj/project.pbxproj:
WebCoreSources.bkl:

Added JSMessagePortCustom.h and JSMessageEventCustom.cpp.

bindings/js/JSDOMBinding.cpp:

(WebCore::toJSSequence):
Added toJSSequence() API to do validation on sequence-like objects per WebIDL.

bindings/js/JSDOMBinding.h:
bindings/js/JSDOMWindowCustom.cpp:

(WebCore::JSDOMWindow::postMessage):
Added custom handler for postMessage() that handles being passed MessagePortArrays.

bindings/js/JSDedicatedWorkerContextCustom.cpp:

(WebCore::JSDedicatedWorkerContext::postMessage):
Added custom handler for postMessage() that handles being passed MessagePortArrays.

bindings/js/JSMessageEventCustom.cpp: Added.

(WebCore::JSMessageEvent::ports):
Added custom ports() getter that converts from MessagePortArray to JSArray.
(WebCore::JSMessageEvent::initMessageEvent):
Added support for passing a MessagePortArray.

bindings/js/JSMessagePortCustom.cpp:

(WebCore::JSMessagePort::postMessage):
Added custom handler for postMessage() that handles being passed MessagePortArray.
(WebCore::fillMessagePortArray):
New helper routine that validates/converts from a JS sequence o a WebCore::MessagePortArray.

bindings/js/JSMessagePortCustom.h: Added.
bindings/js/JSWorkerCustom.cpp:

(WebCore::JSWorker::postMessage):
Added custom handler for postMessage() that handles being passed MessagePortArray.

dom/MessageEvent.cpp:
dom/MessageEvent.h:

(WebCore::MessageEvent::ports):
Changed ports() to return a MessagePortArray* since this value can be null.

dom/MessageEvent.idl:

Updated IDL to match HTML5 spec (now accepts MessagePortArrays instead of a solitary MessagePort).

dom/MessagePort.cpp:
dom/MessagePort.h:
dom/MessagePort.idl:

Updated IDL to match HTML5 spec (postMessage() now accepts MessagePortArrays instead of a solitary MessagePort).

page/DOMWindow.h:
page/DOMWindow.idl:

Updated IDL to match HTML5 spec (postMessage() now accepts MessagePortArrays instead of a solitary MessagePort).

workers/DedicatedWorkerContext.cpp:
workers/DedicatedWorkerContext.h:
workers/DedicatedWorkerContext.idl:

Updated IDL to match HTML5 spec (postMessage() now accepts MessagePortArrays instead of a solitary MessagePort).

workers/Worker.cpp:
workers/Worker.h:
workers/Worker.idl:

Updated IDL to match HTML5 spec (postMessage() now accepts MessagePortArrays instead of a solitary MessagePort).

LayoutTests: Need to update JS bindings and IDL files to support multiple message ports in postMessage()
https://bugs.webkit.org/show_bug.cgi?id=28460

Reviewed by Sam Weinig.

Updated layout tests to match new postMessage() and MessageEvent APIs.

Added new layout tests to test sending/receiving multiple ports.

fast/dom/Window/window-postmessage-args-expected.txt:
fast/dom/Window/window-postmessage-args.html: