WebKit-https.git
10 years ago2009-08-27 Priit Laes <plaes@plaes.org>
xan@webkit.org [Thu, 27 Aug 2009 07:26:02 +0000 (07:26 +0000)]
2009-08-27  Priit Laes  <plaes@plaes.org>

        Reviewed by Xan Lopez.

        [GTK] Out-of-srcdir build problem
        https://bugs.webkit.org/show_bug.cgi?id=28741

        * GNUmakefile.am, autotools/webkit.m4:
        Fix problem with out-of-srcdir builds. Also bump required glib version.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47811 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Cameron McCormack <cam@mcc.id.au>
eric@webkit.org [Thu, 27 Aug 2009 06:38:22 +0000 (06:38 +0000)]
2009-08-26  Cameron McCormack  <cam@mcc.id.au>

        Reviewed by David Kilzer.

        Make prepare-ChangeLog notice property changes
        https://bugs.webkit.org/show_bug.cgi?id=28675

        Make the generated ChangeLog entry include a short description of
        property changes if there were such changes.  Also make
        prepare-ChangeLog not bail if the only changes are property changes.

        * Scripts/prepare-ChangeLog:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Yaar Schnitman <yaar@chromium.org>
abarth@webkit.org [Thu, 27 Aug 2009 06:33:52 +0000 (06:33 +0000)]
2009-08-26  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Adam Barth.

        Re-baselined and created new isolatedWorld tests to play nice with localStorage & sessionStorage.
        https://bugs.webkit.org/show_bug.cgi?id=27992

        * http/tests/security/isolatedWorld/all-window-properties.html: Re-baselined.
        * http/tests/security/isolatedWorld/storage-properties-expected.txt: Added.
        * http/tests/security/isolatedWorld/storage-properties.html: Added.
        * http/tests/security/isolatedWorld/storage-prototype-expected.txt: Added.
        * http/tests/security/isolatedWorld/storage-prototype.html: Added.
        * platform/qtk/Skipped: Added new tests.
        * platform/mac/Skipped: Added new tests.
        * platform/qt/Skipped: Added new tests.
        * platform/win/Skipped: Added new tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Dumitru Daniliuc <dumi@chromium.org>
abarth@webkit.org [Thu, 27 Aug 2009 05:22:54 +0000 (05:22 +0000)]
2009-08-26  Dumitru Daniliuc  <dumi@chromium.org>

        Reviewed by Adam Barth.

        Removing trailing whitespaces from all storage classes.

        https://bugs.webkit.org/show_bug.cgi?id=28075

        * storage/ChangeVersionWrapper.cpp:
        * storage/Database.cpp:
        * storage/Database.h:
        * storage/DatabaseDetails.h:
        * storage/DatabaseThread.cpp:
        * storage/DatabaseTracker.cpp:
        * storage/DatabaseTracker.h:
        * storage/LocalStorageTask.cpp:
        * storage/LocalStorageTask.h:
        * storage/LocalStorageThread.cpp:
        * storage/LocalStorageThread.h:
        * storage/OriginQuotaManager.cpp:
        * storage/OriginUsageRecord.cpp:
        * storage/OriginUsageRecord.h:
        * storage/SQLError.h:
        * storage/SQLResultSetRowList.h:
        * storage/SQLStatement.cpp:
        * storage/SQLStatement.h:
        * storage/SQLStatementCallback.h:
        * storage/SQLStatementErrorCallback.h:
        * storage/SQLTransaction.cpp:
        * storage/SQLTransaction.h:
        * storage/SQLTransactionCallback.h:
        * storage/SQLTransactionErrorCallback.h:
        * storage/Storage.cpp:
        * storage/Storage.h:
        * storage/StorageArea.h:
        * storage/StorageAreaImpl.cpp:
        * storage/StorageAreaImpl.h:
        * storage/StorageAreaSync.cpp:
        * storage/StorageAreaSync.h:
        * storage/StorageEvent.cpp:
        * storage/StorageEvent.h:
        * storage/StorageMap.cpp:
        * storage/StorageMap.h:
        * storage/StorageNamespaceImpl.cpp:
        * storage/StorageSyncManager.cpp:
        * storage/StorageSyncManager.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47808 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Thu, 27 Aug 2009 05:21:29 +0000 (05:21 +0000)]
2009-08-26  Adam Barth  <abarth@webkit.org>

        Reviewed by Oliver Hunt.

        Don't let local files access web URLs
        https://bugs.webkit.org/show_bug.cgi?id=28480

        * Api/qwebsettings.cpp:
        (QWebSettings::QWebSettings):
2009-08-26  Adam Barth  <abarth@webkit.org>

        Reviewed by Oliver Hunt.

        Don't let local files access web URLs
        https://bugs.webkit.org/show_bug.cgi?id=28480

        * WebPreferences.cpp:
        (WebPreferences::initializeDefaultSettings):
2009-08-26  Adam Barth  <abarth@webkit.org>

        Reviewed by Oliver Hunt.

        Don't let local files access web URLs
        https://bugs.webkit.org/show_bug.cgi?id=28480

        A bunch of our LayoutTests rely on our old behavior, so we explicitly
        grant local files universal access during testing.  Mainly, these tests
        involve making XMLHttpRequests for data URLs.

        * DumpRenderTree/mac/DumpRenderTree.mm:
        (resetDefaultsToConsistentValues):
        * DumpRenderTree/qt/DumpRenderTree.cpp:
        (WebCore::WebPage::WebPage):
        * DumpRenderTree/win/DumpRenderTree.cpp:
        (resetDefaultsToConsistentValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: added some exported symbols
alice.liu@apple.com [Thu, 27 Aug 2009 02:45:53 +0000 (02:45 +0000)]
Windows build fix: added some exported symbols

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=28763, flickering when opacity changes.
hyatt@apple.com [Thu, 27 Aug 2009 02:26:31 +0000 (02:26 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=28763, flickering when opacity changes.

Reviewed by Anders Carlsson.

Make opacity changes from < 1 to 1 and from 1 to < 1 cause a relayout to happen.  We need
to do this in order for overflow information to be properly recomputed.

* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::diff):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: Removed some exported symbols that no longer exist.
ggaren@apple.com [Thu, 27 Aug 2009 02:07:24 +0000 (02:07 +0000)]
Windows build fix: Removed some exported symbols that no longer exist.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-08-26
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Thu, 27 Aug 2009 02:02:25 +0000 (02:02 +0000)]
2009-08-26  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by NOBODY (Build fix).

        Fixup test after r47801: skip the first 'waiting' event because it is fired
        when we call play() and there isn't enough media to begin playback immediately,
        but we are interested in the 'waiting' event fired when the movie runs
        out of loaded media.

        * http/tests/media/video-play-stall-seek.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47803 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agox86-64 GTK broken due to code offsets changing, pointers sometimes packed into immedi...
barraclough@apple.com [Thu, 27 Aug 2009 01:16:53 +0000 (01:16 +0000)]
x86-64 GTK broken due to code offsets changing, pointers sometimes packed into immediates.
https://bugs.webkit.org/show_bug.cgi?id=28317

Reviewed by Olliejver Hunt.

We rely on a slightly OS X specific behaviour, that x86-64 applications have a 4Gb zero page,
so pointers are never representable as a 32-bit integer, and always have to be represented by
a separate immediate load instruction, rather than within the immediate field of an arithmetic
or memory operation.

We explicitly check for a couple of cases where a value might be representable in 32-bit, but
these probably never kick in on Mac OS, and only kick in to hose GTK.  Deleting these does not
show a performance degradation on SunSpider.  Remove.

* assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::storePtr):
(JSC::MacroAssemblerX86_64::branchPtr):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Thu, 27 Aug 2009 00:47:41 +0000 (00:47 +0000)]
2009-08-26  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        HTMLMediaElement sometimes loses events
        https://bugs.webkit.org/show_bug.cgi?id=28762
        <rdar://problem/7172437>

        A media element triggers the resource selection algorithm if 
        play() is called when the networkState attribute is NETWORK_EMPTY, but
        it also immediately queues 'play' and either 'waiting' or 'playing' events.
        One of the steps in preparing to load is to flush pending events, so those 
        events were lost if the load timer fired quickly enough. Fixed by deleting
        pending events before scheduling the load timer.

        Test: media/audio-play-event.html

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::scheduleLoad):
            Do nothing if the load timer is already scheduled. Call prepareForLoad so queue of
            pending events is flushed immediately.
        (WebCore::HTMLMediaElement::scheduleNextSourceChild):
            New, start the load timer without flushing pending events.
        (WebCore::HTMLMediaElement::load):
            Call prepareForLoad();
        (WebCore::HTMLMediaElement::prepareForLoad):
            New, load setup code moved from loadInternal.
        (WebCore::HTMLMediaElement::loadInternal):
            Moved some setup code to prepareForLoad so it can be invoked immediately before
            arming the load timer.
        (WebCore::HTMLMediaElement::setNetworkState):
            Call scheduleNextSourceChild instead of scheduleLoad as the later now clears
            pending events.
        * html/HTMLMediaElement.h:
            Declare scheduleNextSourceChild.

2009-08-26  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        HTMLMediaElement sometimes loses events
        https://bugs.webkit.org/show_bug.cgi?id=28762
        <rdar://problem/7172437>

        * media/audio-play-event.html:
        * media/audio-play-event-expected.txt:
             Added.

        * media/video-play-empty-events-expected.txt:
            Updated results for previously missed events.

        * media/video-src-remove.html:
        * media/video-timeupdate-during-playback-expected.txt:
            Restructured to make results less timing dependent.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=28751
pkasting@chromium.org [Wed, 26 Aug 2009 23:40:08 +0000 (23:40 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=28751
Switch ICO decoder from a Vector of raw pointers to a Vector of
OwnPtrs now that that is possible.  Makes code slightly simpler.

Reviewed by Eric Seidel.

* platform/image-decoders/ico/ICOImageDecoder.cpp:
(WebCore::ICOImageDecoder::decodeAtIndex):
(WebCore::ICOImageDecoder::processDirectory):
* platform/image-decoders/ico/ICOImageDecoder.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoA bit of Collector refatoring.
ggaren@apple.com [Wed, 26 Aug 2009 23:00:39 +0000 (23:00 +0000)]
A bit of Collector refatoring.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-08-26
Reviewed by Oliver Hunt.

SunSpider says no change. v8 says 1.003x faster (1.02x faster on splay).

* JavaScriptCore.exp:

* runtime/JSCell.cpp:
(JSC::JSCell::toPrimitive):
(JSC::JSCell::getPrimitiveNumber):
(JSC::JSCell::toBoolean):
(JSC::JSCell::toNumber):
(JSC::JSCell::toString):
(JSC::JSCell::toObject): Removed pure virtual functions from
JSCell, so the collector can construct one. This allowed
me to remove a bunch of ASSERT_NOT_REACHED throughout the
code, too.

* runtime/JSCell.h:
(JSC::JSCell::JSCell): ditto
(JSC::Heap::heap): Inlined this function because it's trivial.

* JavaScriptCore.exp:

* runtime/Collector.cpp:
(JSC::Heap::destroy):
(JSC::Heap::allocateBlock):
(JSC::Heap::freeBlock):
(JSC::Heap::freeBlocks): Renamed freeHeap to freeBlocks, since
it doesn't actually free the Heap object.
(JSC::Heap::heapAllocate):
(JSC::Heap::sweep):
* runtime/Collector.h: Refactored block allocation and destruction
into helper functions.

* runtime/GetterSetter.cpp:
* runtime/JSAPIValueWrapper.cpp:
* runtime/JSPropertyNameIterator.cpp: Removed dummy implementations
of pure virtual functions. (See above.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: https://bugs.webkit.org/show_bug.cgi?id=28718, image inside relpositioned...
hyatt@apple.com [Wed, 26 Aug 2009 22:30:11 +0000 (22:30 +0000)]
WebCore: https://bugs.webkit.org/show_bug.cgi?id=28718, image inside relpositioned inline doesn't paint.
Make sure vertical overflow is still computed across self-painting layers, even though it isn't
propagated to parents.

Reviewed by Dan Bernstein.

Added fast/repaint/rel-positioned-inline-with-overflow.html

* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::computeVerticalOverflow):

LayoutTests: Add layout test for relpositioned inline regression.

Reviewed by Dan Bernstein.

* fast/repaint/rel-positioned-inline-with-overflow.html: Added.
* platform/mac/fast/repaint/rel-positioned-inline-with-overflow-expected.checksum: Added.
* platform/mac/fast/repaint/rel-positioned-inline-with-overflow-expected.png: Added.
* platform/mac/fast/repaint/rel-positioned-inline-with-overflow-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: ResourceRequestBase::*[Aa]llowHTTPCookies shouldn't mention http.
levin@chromium.org [Wed, 26 Aug 2009 21:57:22 +0000 (21:57 +0000)]
WebCore: ResourceRequestBase::*[Aa]llowHTTPCookies shouldn't mention http.
https://bugs.webkit.org/show_bug.cgi?id=28735

Patch by David Levin <levin@chromium.org> on 2009-08-26
Reviewed by Darin Fisher.

Renaming the method to imply its wider implications of affecting cookies
over any protocol that supports them.

No change in functionality so no new tests.

* WebCore.order:
* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::DocumentThreadableLoader):
(WebCore::DocumentThreadableLoader::makeSimpleCrossOriginAccessRequest):
(WebCore::DocumentThreadableLoader::makeCrossOriginAccessRequestWithPreflight):
* platform/network/ResourceRequestBase.cpp:
(WebCore::ResourceRequestBase::adopt):
(WebCore::ResourceRequestBase::copyData):
(WebCore::ResourceRequestBase::allowCookies):
(WebCore::ResourceRequestBase::setAllowCookies):
(WebCore::equalIgnoringHeaderFields):
* platform/network/ResourceRequestBase.h:
(WebCore::ResourceRequestBase::ResourceRequestBase):
* platform/network/cf/ResourceRequestCFNet.cpp:
(WebCore::ResourceRequest::doUpdatePlatformRequest):
(WebCore::ResourceRequest::doUpdateResourceRequest):
* platform/network/mac/ResourceRequestMac.mm:
(WebCore::ResourceRequest::doUpdateResourceRequest):
(WebCore::ResourceRequest::doUpdatePlatformRequest):
* platform/network/soup/ResourceRequestSoup.cpp:
(WebCore::ResourceRequest::updateFromSoupMessage):

WebKit/win: ResourceRequestBase::*[Aa]llowHTTPCookies shouldn't mention http.
https://bugs.webkit.org/show_bug.cgi?id=28735

Patch by David Levin <levin@chromium.org> on 2009-08-26
Reviewed by Darin Fisher.

* WebMutableURLRequest.cpp:
(WebMutableURLRequest::setHTTPShouldHandleCookies):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore:
mitz@apple.com [Wed, 26 Aug 2009 21:54:11 +0000 (21:54 +0000)]
WebCore:
Made RenderInline::clippedOverflowRectForRepaint() actually include
overflow.

Reviewed by Dave Hyatt.

Test: fast/repaint/inline-overflow.html

* rendering/RenderInline.cpp:
(WebCore::RenderInline::linesVisibleOverflowBoundingBox): Added. Returns
    the bounding box of all lines’ visible overflow boxes (under the
    assumption that the first and last lines determine the top and
    bottom, respectively).
(WebCore::RenderInline::clippedOverflowRectForRepaint): Changed to use
    linesVisibleOverflowBoundingBox() instead of linesBoundingBox().
* rendering/RenderInline.h:

LayoutTests:
Test that RenderInline::clippedOverflowRectForRepaint() actually
includes overflow.

Reviewed by Dave Hyatt.

* fast/repaint/inline-overflow-expected.checksum: Added.
* fast/repaint/inline-overflow-expected.png: Added.
* fast/repaint/inline-overflow-expected.txt: Added.
* fast/repaint/inline-overflow.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago=== End re-roll-in of r47738:47740 with Windows crash fixed ===
ggaren@apple.com [Wed, 26 Aug 2009 21:33:30 +0000 (21:33 +0000)]
=== End re-roll-in of r47738:47740 with Windows crash fixed ===

2009-08-26  Geoffrey Garen  <ggaren@apple.com>

        Build fix: start out with a 32-bit value to avoid a shortening warning.

        * runtime/Collector.cpp:
        (JSC::Heap::sweep):

2009-08-24  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Substantially reduced VM thrash in the GC heap.

        1.08x faster on v8 (1.60x faster on v8-splay).

        1.40x faster on bench-alloc-nonretained.

        1.90x faster on bench-alloc-retained.

        SunSpider says no change.

        * runtime/Collector.cpp:
        (JSC::Heap::heapAllocate): Fixed a long-standing bug: update a few local
        variables unconditionally after calling collect(), since they may be used
        even if we don't "goto scan". (In the bug I saw, usedBlocks got out of
        sync with heap.usedBlocks).
        (JSC::Heap::sweep): Keep enough free heap space to accomodate
        the number of objects we'll allocate before the next GC, plus 25%, for
        good measure.
        * runtime/Collector.h: Bumped the block size to 256k. This seems to give
        the best cache performance, and it prevents us from initiating lots of
        VM traffic to recover very small chunks of memory.

=== Begin re-roll-in of r47738:47740 with Windows crash fixed ===

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFollow-up to http://trac.webkit.org/changeset/47790. Add windows
jparent@chromium.org [Wed, 26 Aug 2009 21:26:25 +0000 (21:26 +0000)]
Follow-up to trac.webkit.org/changeset/47790.  Add windows
specific results that are the same as the results next to the test,
since windows falls back on platform/mac.

Patch by Julie Parent <jparent@chromium.org> on 2009-08-26
Reviewed by Maciej Stachowiak.

* platform/win/http/tests/xmlhttprequest/xmlhttprequest-missing-file-exception-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47794 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: minor style correction and include fix for notifications
levin@chromium.org [Wed, 26 Aug 2009 20:47:45 +0000 (20:47 +0000)]
WebCore: minor style correction and include fix for notifications
https://bugs.webkit.org/show_bug.cgi?id=28745

Patch by John Gregg <johnnyg@google.com> on 2009-08-26
Reviewed by David Levin.

No change in behavior, so no tests.

* bindings/js/JSDesktopNotificationsCustom.cpp: add include of Error.h
* notifications/Notification.cpp: change enum style to InterCaps
(WebCore::Notification::Notification):
* notifications/NotificationPresenter.h:
(WebCore::NotificationPresenter::):

WebKitTools: Minor style correction and include fix for notifications
https://bugs.webkit.org/show_bug.cgi?id=28745

Patch by John Gregg <johnnyg@google.com> on 2009-08-26
Reviewed by David Levin.

* DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
  change to correct EnumStyle
(DRTDesktopNotificationPresenter::checkNotificationPermission):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47793 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebKitTools: XMLHttpRequest.withCredentials=false shouldn't save cookies.
levin@chromium.org [Wed, 26 Aug 2009 20:45:39 +0000 (20:45 +0000)]
WebKitTools: XMLHttpRequest.withCredentials=false shouldn't save cookies.
https://bugs.webkit.org/show_bug.cgi?id=28743

Patch by David Levin <levin@chromium.org> on 2009-08-26
Reviewed by Alexey Proskuryakov.

Added the support to the layout test controller on OSX and
Windows (for CFNETWORK) to allow for changing the accept cookie
policy.

* DumpRenderTree/LayoutTestController.cpp:
(LayoutTestController::LayoutTestController): Added a bool
to track the state of accepting cookies.
(setAlwaysAcceptCookiesCallback): Standard wrapper method
to go from js to a C++ method.
(LayoutTestController::staticFunctions): Added the
setAlwaysAcceptCookies method to the js layoutTestController.
* DumpRenderTree/LayoutTestController.h:
(LayoutTestController::alwaysAcceptCookies): Returns the value.
* DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
(LayoutTestController::setAlwaysAcceptCookies): Stub out method.
* DumpRenderTree/mac/DumpRenderTree.mm:
(resetDefaultsToConsistentValues): Reset the accept cookie to
its default.
* DumpRenderTree/mac/LayoutTestControllerMac.mm:
(LayoutTestController::setAlwaysAcceptCookies): Does the work for
OSX to change the cookie accept policy.
* DumpRenderTree/win/DumpRenderTree.cpp:
(setAlwaysAcceptCookies): Method to handle all the calls necessary
to change the accept cookie policy on Windows.
(resetDefaultsToConsistentValues): Reset the accept cookie to
its default.
* DumpRenderTree/win/DumpRenderTreeWin.h:
* DumpRenderTree/win/LayoutTestControllerWin.cpp:
(LayoutTestController::setAlwaysAcceptCookies): Stub out method.
* DumpRenderTree/wx/LayoutTestControllerWx.cpp:
(LayoutTestController::setAlwaysAcceptCookies): Stub out method.

LayoutTests: XMLHttpRequest.withCredentials=false shouldn't save cookies.
https://bugs.webkit.org/show_bug.cgi?id=28743

Patch by David Levin <levin@chromium.org> on 2009-08-26
Reviewed by Alexey Proskuryakov.

Added a test to verify that XMLHttpRequest.withCredentials=false doesn't
save cookies.

* http/tests/xmlhttprequest/cross-origin-cookie-storage-expected.txt: Added.
* http/tests/xmlhttprequest/cross-origin-cookie-storage.html: Added.
* http/tests/xmlhttprequest/resources/cross-origin-set-cookies.php: Added.
* http/tests/xmlhttprequest/resources/cross-origin-check-cookies.php: Added the ability
to check for a cookie by name.
* platform/gtk/Skipped: Skip the new test.
* platform/qt/Skipped: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47792 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: postMessage() spec now supports sending arrays of ports
atwilson@chromium.org [Wed, 26 Aug 2009 20:45:24 +0000 (20:45 +0000)]
JavaScriptCore: postMessage() spec now supports sending arrays of ports
https://bugs.webkit.org/show_bug.cgi?id=26902

Reviewed by David Levin.

Added OwnPtr to VectorTraits so we can store OwnPtrs in Vectors.

* wtf/VectorTraits.h:

WebCore: postMessage() spec now supports sending arrays of ports
https://bugs.webkit.org/show_bug.cgi?id=26902

Reviewed by David Levin.

Added support for sending multiple ports at once via postMessage(). JS bindings will be updated in an upcoming patch.

No new tests, since the new functionality is not exposed via new JS bindings yet, so the old tests suffice.

* WebCore.xcodeproj/project.pbxproj:
Mark MessagePort.h and MessagePortChannel.h as private since they are used by WebKit now.
* dom/MessageEvent.cpp:
Updated APIs to support passing MessagePortArrays.
(WebCore::MessageEvent::MessageEvent):
(WebCore::MessageEvent::initMessageEvent):
(WebCore::MessageEvent::messagePort):
* dom/MessageEvent.h:
(WebCore::MessageEvent::create):
Now accepts a MessagePortArray parameter instead of a single MessagePort.
(WebCore::MessageEvent::ports):
* dom/MessagePort.cpp:
(WebCore::MessagePort::postMessage):
(WebCore::MessagePort::dispatchMessages):
(WebCore::MessagePort::disentanglePorts):
Added new static function to extract a MessagePortChannelArray from a MessagePortArray.
(WebCore::MessagePort::entanglePorts):
Added new static function to generate a MessagePortArray (entangled ports) from a MessagePortChannelArray.
* dom/MessagePort.h:
* dom/MessagePortChannel.cpp:
EventData now contains a MessagePortChannelArray field instead of a single MessagePortChannel.
(WebCore::MessagePortChannel::EventData::create):
(WebCore::MessagePortChannel::EventData::EventData):
* dom/MessagePortChannel.h:
(WebCore::MessagePortChannel::EventData::channels):
* page/DOMWindow.cpp:
(WebCore::PostMessageTimer::PostMessageTimer):
(WebCore::PostMessageTimer::event):
(WebCore::DOMWindow::postMessage):
Now accepts a MessagePortArray parameter instead of a single MessagePort.
* page/DOMWindow.h:
* workers/DedicatedWorkerContext.cpp:
(WebCore::DedicatedWorkerContext::postMessage):
Now accepts a MessagePortArray parameter instead of a single MessagePort.
(WebCore::DedicatedWorkerContext::dispatchMessage):
* workers/DedicatedWorkerContext.h:
* workers/Worker.cpp:
(WebCore::Worker::postMessage):
Now accepts a MessagePortArray parameter instead of a single MessagePort.
(WebCore::Worker::dispatchMessage):
* workers/Worker.h:
* workers/WorkerContextProxy.h:
* workers/WorkerMessagingProxy.cpp:
Updated WorkerObjectProxy and WorkerContextProxy API implementations to support sending multiple MessagePorts.
(WebCore::MessageWorkerContextTask::create):
(WebCore::MessageWorkerContextTask::MessageWorkerContextTask):
(WebCore::MessageWorkerContextTask::performTask):
(WebCore::MessageWorkerTask::create):
(WebCore::MessageWorkerTask::MessageWorkerTask):
(WebCore::MessageWorkerTask::performTask):
(WebCore::WorkerMessagingProxy::postMessageToWorkerObject):
(WebCore::WorkerMessagingProxy::postMessageToWorkerContext):
* workers/WorkerMessagingProxy.h:
* workers/WorkerObjectProxy.h:
Updated WorkerObjectProxy and WorkerContextProxy APIs to support sending multiple MessagePorts.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd Mac specific results for xhr-missing-file-expection test since it
jparent@chromium.org [Wed, 26 Aug 2009 20:26:27 +0000 (20:26 +0000)]
Add Mac specific results for xhr-missing-file-expection test since it
incorrectly always has statusText of "OK" due to
https://bugs.webkit.org/show_bug.cgi?id=24572 and update the results
that are next to the file to use "Not Found" since that is more correct
and matches IE, FF, Chrome.

Patch by Julie Parent <jparent@chromium.org> on 2009-08-26
Reviewed by Alex Proskuryakov.

https://bugs.webkit.org/show_bug.cgi?id=28723

* http/tests/xmlhttprequest/xmlhttprequest-missing-file-exception-expected.txt:
* platform/mac/http/tests/xmlhttprequest/xmlhttprequest-missing-file-exception-expected.txt: Added.
* platform/win/Skipped: Stop skipping the test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47790 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 26 Aug 2009 20:25:54 +0000 (20:25 +0000)]
2009-08-26  Dimitri Glazkov  <dglazkov@chromium.org>

        Unreviewed.

        Remove accidentally left in clobber option.
        https://bugs.webkit.org/show_bug.cgi?id=28400

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Fumitoshi Ukai <ukai@chromium.org>
eric@webkit.org [Wed, 26 Aug 2009 20:05:43 +0000 (20:05 +0000)]
2009-08-26  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        SocketStreamHandle interface for WebSocket API
        https://bugs.webkit.org/show_bug.cgi?id=28037

        Add build systems only for GNUmakefile.am and WebCore.xcodeproj.now.
        Other build systems will be updated once the code is functional.
        Tests will be landed once this code is complete and functional.

        * GNUmakefile.am:
        * WebCore.xcodeproj/project.pbxproj:
        * platform/network/SocketStreamErrorBase.cpp: Added.
        * platform/network/SocketStreamErrorBase.h: Added.
        * platform/network/SocketStreamHandleBase.cpp: Added.
        * platform/network/SocketStreamHandleBase.h: Added.
        * platform/network/SocketStreamHandleClient.h: Added.
        * platform/network/cf/SocketStreamError.h: Added.
        * platform/network/cf/SocketStreamHandle.h: Added.
        * platform/network/cf/SocketStreamHandleCFNet.cpp: Added.
        * platform/network/soup/SocketStreamError.h: Added.
        * platform/network/soup/SocketStreamHandle.h: Added.
        * platform/network/soup/SocketStreamHandleSoup.cpp: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47788 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-17 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 26 Aug 2009 20:04:20 +0000 (20:04 +0000)]
2009-08-17  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Mark Rowe.

        Add canary-style Chromium WebKit build slave to the waterfall.
        https://bugs.webkit.org/show_bug.cgi?id=28400

        * BuildSlaveSupport/build.webkit.org-config/config.json: Added one Chromium/Windows slave
        * BuildSlaveSupport/build.webkit.org-config/master.cfg: Added support for Chromium slave commands.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Mads Ager <ager@chromium.org>
eric@webkit.org [Wed, 26 Aug 2009 19:55:10 +0000 (19:55 +0000)]
2009-08-26  Mads Ager  <ager@chromium.org>

        Reviewed by Dimitri Glazkov.

        [V8] setTimeout does not convert non-string argument to string
        https://bugs.webkit.org/show_bug.cgi?id=28739

        Convert non-string argument to setTimeout to string in the V8
        bindings.

        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::V8Custom::WindowSetTimeoutImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUpgrade WordPress to the latest version.
mrowe@apple.com [Wed, 26 Aug 2009 18:48:42 +0000 (18:48 +0000)]
Upgrade WordPress to the latest version.

* blog/wp-includes/version.php:
* blog/wp-login.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Wed, 26 Aug 2009 18:08:26 +0000 (18:08 +0000)]
2009-08-26  Xan Lopez  <xlopez@igalia.com>

        Rubber-stamped by Gustavo Noronha.

        Remove duplicated files from file list.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMore export fixes.
oliver@apple.com [Wed, 26 Aug 2009 17:40:37 +0000 (17:40 +0000)]
More export fixes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoHopefully fix all the exports from JSC on windows
oliver@apple.com [Wed, 26 Aug 2009 17:33:46 +0000 (17:33 +0000)]
Hopefully fix all the exports from JSC on windows

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoBuild fix
oliver@apple.com [Wed, 26 Aug 2009 17:12:36 +0000 (17:12 +0000)]
Build fix

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47781 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[ES5] Implement getOwnPropertyDescriptor
oliver@apple.com [Wed, 26 Aug 2009 16:52:15 +0000 (16:52 +0000)]
[ES5] Implement getOwnPropertyDescriptor
https://bugs.webkit.org/show_bug.cgi?id=28724

Reviewed by Gavin Barraclough.

JavaScriptCore:
Implement the core runtime support for getOwnPropertyDescriptor.
This adds a virtual getOwnPropertyDescriptor method to every class
that implements getOwnPropertySlot that shadows the behaviour of
getOwnPropertySlot.  The alternative would be to make getOwnPropertySlot
(or PropertySlots in general) provide property attribute information,
but quick testing showed this to be a regression.

WebCore:
Implement the WebCore side of getOwnPropertyDescriptor.  This
requires a custom implementation of getOwnPropertyDescriptor
for every class with a custom implementation of getOwnPropertySlot.

The bindings generator has been updated to generate appropriate
versions of getOwnPropertyDescriptor for the general case where
a custom getOwnPropertyDescriptor is not needed.  ES5 is vague
about how getOwnPropertyDescriptor should work in the context of
"host" functions with polymorphic [[GetOwnProperty]], so it seems
okay that occasionally we "guess" what attributes -- eg. determining
whether a property is writable.

Test: fast/js/getOwnPropertyDescriptor.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Vincent Untz <vuntz@gnome.org>
xan@webkit.org [Wed, 26 Aug 2009 14:08:32 +0000 (14:08 +0000)]
2009-08-26  Vincent Untz <vuntz@gnome.org>

        Reviewed by Xan Lopez.

        GObject checks for NULL (and not 0) in g_object_get, so use that
        as we were already doing elsewhere to make the compiler happy.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Alpha Lam <hclam@chromium.org>
eric@webkit.org [Wed, 26 Aug 2009 14:06:08 +0000 (14:06 +0000)]
2009-08-26  Alpha Lam  <hclam@chromium.org>

        Reviewed by David Levin.

        RenderThemeMac.mm has incorrect casting for media control element.
        https://bugs.webkit.org/show_bug.cgi?id=28730

        MediaControlMuteButtonElement was incorrectly casted to
        MediaControlPlayButtonElement. This is corrected in this patch.

        No new tests because this is code cleanup and is already covered by existing
        media layout tests.

        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::paintMediaMuteButton):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Robert Sesek <rsesek@chromium.org>
eric@webkit.org [Wed, 26 Aug 2009 13:56:09 +0000 (13:56 +0000)]
2009-08-26  Robert Sesek  <rsesek@chromium.org>

        Reviewed by Eric Seidel.

        [Chromium] Multi-value select boxes cannot do non-contiguous selection on Mac
        https://bugs.webkit.org/show_bug.cgi?id=28670

        Change the condition regarding the key modifier check for non-contiguous
        selection on a multi-value list from just PLATFORM(MAC) to
        PLATFORM(MAC) || (PLATFORM(CHROMIUM) && PLATFORM(DARWIN)).

        No new tests. This patch only changes the condition on which a certain mouse
        event modifier key is used.

        * dom/SelectElement.cpp:
        (WebCore::SelectElement::listBoxDefaultEventHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-26 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Wed, 26 Aug 2009 13:31:23 +0000 (13:31 +0000)]
2009-08-26  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        https://bugs.webkit.org/show_bug.cgi?id=28738
        Expose undo/redo functionality in WebKitWebView

        Expose undo/redo functionality in WebKitWebView, keeping the
        default bindings as they were.

        * WebCoreSupport/EditorClientGtk.cpp:
        (WebKit::):
        * webkit/webkitwebview.cpp:
        (webkit_web_view_real_undo):
        (webkit_web_view_real_redo):
        (webkit_web_view_class_init):
        (webkit_web_view_can_undo):
        (webkit_web_view_undo):
        (webkit_web_view_can_redo):
        (webkit_web_view_redo):
        * webkit/webkitwebview.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoHow many copies of the parameters do you need?
barraclough@apple.com [Wed, 26 Aug 2009 07:46:47 +0000 (07:46 +0000)]
How many copies of the parameters do you need?
https://bugs.webkit.org/show_bug.cgi?id=28701

Reviewed by Darin Adler.

The function parameters in JSC get copied a lot - and unnecessarily so.

Originally this happened due to duplicating FunctionBodyNodes on recompilation,
though the problem has been exacerbated by copying the parameters from the
original function body onto the executable, then back onto the real body that
will be generated (this happens on every function).  And this is all made worse
since the data structures in question are a little ugly - C style arrays of C++
objects containing ref counts, so they need a full copy-construct (rather than
a simple memcpy).

This can all be greatly simplified by just punting the parameters off into
their own ref-counted object, and forgoing all the copying.

~no performance change, possible slight progression.

* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::BytecodeGenerator):
* bytecompiler/BytecodeGenerator.h:
(JSC::BytecodeGenerator::makeFunction):
* parser/Nodes.cpp:
(JSC::FunctionParameters::FunctionParameters):
(JSC::FunctionBodyNode::FunctionBodyNode):
(JSC::FunctionBodyNode::finishParsing):
* parser/Nodes.h:
(JSC::FunctionBodyNode::parameters):
(JSC::FunctionBodyNode::parameterCount):
* runtime/Executable.cpp:
(JSC::FunctionExecutable::~FunctionExecutable):
(JSC::FunctionExecutable::compile):
(JSC::FunctionExecutable::reparseExceptionInfo):
(JSC::FunctionExecutable::fromGlobalCode):
(JSC::FunctionExecutable::paramString):
* runtime/Executable.h:
(JSC::FunctionExecutable::FunctionExecutable):
(JSC::FunctionExecutable::parameterCount):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Wed, 26 Aug 2009 05:49:03 +0000 (05:49 +0000)]
2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Dan Bernstein

        r47744 causes media/video-controls-with-mutation-event-handler.html to crash
        https://bugs.webkit.org/show_bug.cgi?id=28732

        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlVolumeSliderElement::MediaControlVolumeSliderElement):
            Don't call setAttribute from the construtor, any registered mutation event
            listener will cause a crash.

        * rendering/RenderMedia.cpp:
        (WebCore::RenderMedia::createVolumeSlider):
            Set the maxAttr here instead of in the constructor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed. Skip the new test fast/dom/Window/timer-resume-on-navigation-back.html
dimich@chromium.org [Wed, 26 Aug 2009 05:46:44 +0000 (05:46 +0000)]
Unreviewed. Skip the new test fast/dom/Window/timer-resume-on-navigation-back.html
on gtk and qt - their DRT does not yet implement overridePreference.

* platform/gtk/Skipped:
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: Timers from cached pages fire instantly rather than the after the specified...
dimich@chromium.org [Wed, 26 Aug 2009 05:17:15 +0000 (05:17 +0000)]
WebCore: Timers from cached pages fire instantly rather than the after the specified timeout delay
https://bugs.webkit.org/show_bug.cgi?id=28683

Reviewed by Brady Eidson.

Test: fast/dom/Window/timer-resume-on-navigation-back.html

* history/CachedPage.cpp:
(WebCore::CachedPage::restore): Remove duplicated Frame::restore() call, since
    it should be done only once in FrameLoader::open(cachedFrame)
* page/DOMTimer.cpp: Added a debug-only flag and ASSERT to catch out-of-order suspense/restore.
(WebCore::DOMTimer::DOMTimer): Ditto.
(WebCore::DOMTimer::suspend): Ditto.
(WebCore::DOMTimer::resume): Ditto.
* page/DOMTimer.h: Ditto.

LayoutTests: Timers from cached pages fire instantly rather than the specified timeout delay
https://bugs.webkit.org/show_bug.cgi?id=28683

Reviewed by Brady Eidson.

Added test to verify the fix - it uses page cache and measures the timer callback fire time.

* fast/dom/Window/timer-resume-on-navigation-back-expected.txt: Added.
* fast/dom/Window/timer-resume-on-navigation-back.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Wed, 26 Aug 2009 05:10:38 +0000 (05:10 +0000)]
2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        Setting the volume of a media element has no effect if called too early
        https://bugs.webkit.org/show_bug.cgi?id=28731

        * manual-tests/audio-volume.html:
            Added.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivate::setVolume):
        (WebCore::MediaPlayerPrivate::setRate):
            Don't bail if metadata hasn't been loaded, it is fine to set rate and/or
            volume as soon as the movie has been created.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoDocumentThreadableLoader::getShouldUseCredentialStorage should allow for synchronous...
levin@chromium.org [Wed, 26 Aug 2009 03:02:16 +0000 (03:02 +0000)]
DocumentThreadableLoader::getShouldUseCredentialStorage should allow for synchronous callbacks from SubresourceLoader::create.
https://bugs.webkit.org/show_bug.cgi?id=28728

Patch by David Levin <levin@chromium.org> on 2009-08-25
Reviewed by Maciej Stachowiak.

Test: This code path is exercised while running xhr layout tests in chromium.

* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::getShouldUseCredentialStorage): When called
back synchronously, the loader is 0, so the assert is changed to allow for that.
(WebCore::DocumentThreadableLoader::loadRequest): Synchronous callbacks may be
done before SubresourceLoader::create finishes, so ensure that any previous loader
-- from a preflight request -- is cleared before calling SubresourceLoader::create.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRubber stamped by Steve Falkenburg.
bfulgham@webkit.org [Wed, 26 Aug 2009 01:06:11 +0000 (01:06 +0000)]
Rubber stamped by Steve Falkenburg.

debug_wincairo.vsprops was linking against wrong C runtime.

* win/tools/vsprops/debug_wincairo.vsprops: Link to correct
  C runtime (as in the standard 'debug.vsprops' file.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoStop installing IDL files in to the framework.
mrowe@apple.com [Wed, 26 Aug 2009 00:57:58 +0000 (00:57 +0000)]
Stop installing IDL files in to the framework.

* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Kent Tamura <tkent@chromium.org>
eric@webkit.org [Wed, 26 Aug 2009 00:51:02 +0000 (00:51 +0000)]
2009-08-25  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Support for HTMLInputElement::list and HTMLInputElement::selectedOption.
        https://bugs.webkit.org/show_bug.cgi?id=27756

        Tests: fast/forms/input-list.html
               fast/forms/input-selectedoption.html

        * html/HTMLAttributeNames.in:
        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::parseMappedAttribute):
        (WebCore::HTMLInputElement::list):
        (WebCore::HTMLInputElement::selectedOption):
        * html/HTMLInputElement.h:
        * html/HTMLInputElement.idl:
2009-08-25  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Support for HTMLInputElement::list and HTMLInputElement::selectedOption.
        https://bugs.webkit.org/show_bug.cgi?id=27756

        * fast/forms/input-list-expected.txt: Added.
        * fast/forms/input-list.html: Added.
        * fast/forms/input-selectedoption-expected.txt: Added.
        * fast/forms/input-selectedoption.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Cameron McCormack <cam@mcc.id.au>
eric@webkit.org [Tue, 25 Aug 2009 23:24:02 +0000 (23:24 +0000)]
2009-08-25  Cameron McCormack  <cam@mcc.id.au>

        Reviewed by Darin Adler.

        make-script-test-wrappers should be executable
        https://bugs.webkit.org/show_bug.cgi?id=28669

        Make make-script-test-wrappers and update-sources-list.py both be
        executable.

        * Scripts/update-sources-list.py:
        * Scripts/make-script-test-wrappers:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Tue, 25 Aug 2009 23:07:19 +0000 (23:07 +0000)]
2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Oliver Hunt.

        <video> and <audio> controller should be accessible (edit)
        https://bugs.webkit.org/show_bug.cgi?id=28081

        Test: accessibility/media-element.html

        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * WebCoreSources.bkl:
            Add AccessibilityMediaControls.cpp/.h

        * accessibility/AXObjectCache.cpp:
        * accessibility/AXObjectCache.h:
        (WebCore::AXObjectCache::getOrCreate):
            Create AccessibilityMediaControl. Update Copyright

        * accessibility/AccessibilityMediaControls.cpp:
        * accessibility/AccessibilityMediaControls.h:
            Added.

        * accessibility/AccessibilityObject.cpp:
        * accessibility/AccessibilityObject.h:
            Add isMediaTimeline. Update Copyright.

        * accessibility/AccessibilitySlider.h:
            Make constructor protected instead of private so AccessibilityMediaTimeline can
            call it.

        * accessibility/mac/AccessibilityObjectWrapper.mm:
        (-[AccessibilityObjectWrapper accessibilityAttributeNames]):
            Add NSAccessibilityValueDescriptionAttribute to range attributes.
        (-[AccessibilityObjectWrapper subrole]):
            Return NSAccessibilityTimelineSubrole for AccessibilityMediaTimeline.

        * dom/Node.h:
            add isMediaControlElement.

        * page/mac/WebCoreViewFactory.h:
        * platform/LocalizedStrings.h:
        * platform/mac/LocalizedStringsMac.mm:
        (WebCore::localizedMediaControlElementString):
        (WebCore::localizedMediaControlElementHelpText):
        (WebCore::localizedMediaTimeDescription):
            New, return localized media controller strings.

        * platform/gtk/LocalizedStringsGtk.cpp:
        (WebCore::mediaElementLoadingStateText):
        (WebCore::mediaElementLiveBroadcastStateText):
            Move inside of "#if ENABLE(VIDEO)"
        (WebCore::localizedMediaControlElementString):
        (WebCore::localizedMediaControlElementHelpText):
        (WebCore::localizedMediaTimeDescription):

        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlElement::MediaControlElement):
            Initialize m_displayType.
        (WebCore::MediaControlElement::MediaControlElement):
        (WebCore::MediaControlStatusDisplayElement::MediaControlStatusDisplayElement):
            Fix style.
        (WebCore::MediaControlInputElement::MediaControlInputElement):
            Set m_displayType based on pseudo style ID.
        (WebCore::MediaControlInputElement::MediaControlInputElement):
            rename local "o" to "object".
        (WebCore::MediaControlMuteButtonElement::MediaControlMuteButtonElement):
        (WebCore::MediaControlPlayButtonElement::MediaControlPlayButtonElement):
        (WebCore::MediaControlSeekButtonElement::MediaControlSeekButtonElement):
        (WebCore::MediaControlRewindButtonElement::MediaControlRewindButtonElement):
        (WebCore::MediaControlReturnToRealtimeButtonElement::MediaControlReturnToRealtimeButtonElement):
        (WebCore::MediaControlTimelineElement::MediaControlTimelineElement):
        (WebCore::MediaControlFullscreenButtonElement::MediaControlFullscreenButtonElement):
        (WebCore::MediaControlTimelineElement::MediaControlTimelineElement):
        (WebCore::MediaControlTimelineElement::MediaControlTimelineElement):
        (WebCore::MediaControlTimelineElement::MediaControlTimelineElement):
            Update for MediaControlInputElement constructor change.
        (WebCore::MediaControlTimeDisplayElement::formatTime):
            New, moved here from RenderMedia so AccessibilityMediaTimeDisplay can use it.
        (WebCore::MediaControlTimeDisplayElement::setCurrentValue):
            New, set text to current time.
        * rendering/MediaControlElements.h:

        * rendering/RenderMedia.cpp:
        (WebCore::RenderMedia::updateControls):
            Create controls in tab order.
        (WebCore::RenderMedia::formatTime):
            Moved to MediaControlTimeDisplayElement::formatTime.
        (WebCore::RenderMedia::updateTimeDisplay):
            Call MediaControlTimeDisplayElement::setCurrentValue.
        * rendering/RenderMedia.h:

2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Oliver Hunt.

        <video> and <audio> controller should be accessible
        https://bugs.webkit.org/show_bug.cgi?id=28081

        Add localized strings for media controller accessiblility.

        * English.lproj/Localizable.strings:

2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Oliver Hunt.

        <video> and <audio> controller should be accessible
        https://bugs.webkit.org/show_bug.cgi?id=28081

        * WebCoreSupport/WebViewFactory.mm:
        (-[WebViewFactory localizedMediaControlElementString:]):
        (-[WebViewFactory localizedMediaControlElementHelpText:]):
        (-[WebViewFactory localizedMediaTimeDescription:]):
            New.

2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Oliver Hunt.

        <video> and <audio> controller should be accessible
        https://bugs.webkit.org/show_bug.cgi?id=28081

        * WebCoreLocalizedStrings.cpp:
        (WebCore::localizedMediaControlElementString):
        (WebCore::localizedMediaControlElementHelpText):
        (WebCore::localizedMediaTimeDescription):
            New.

2009-08-25  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Oliver Hunt.

        <video> and <audio> controller should be accessible
        https://bugs.webkit.org/show_bug.cgi?id=28081

        * accessibility/media-element.html: Added.
        * platform/mac/accessibility/media-element-expected.txt: Added.
        * media/video-controls-visible-audio-only.html:
        * platform/mac-leopard/accessibility/media-element-expected.txt: Added.
        * platform/mac-leopard/media/controls-drag-timebar-expected.txt: Added.
        * platform/mac-leopard/media/controls-strict-expected.txt: Added.
        * platform/mac-leopard/media/video-controls-rendering-expected.txt: Added.
        * platform/mac-leopard/media/video-display-toggle-expected.txt: Added.
        * platform/win/media/controls-strict-expected.txt: Added.
        * platform/win/media/video-controls-rendering-expected.txt: Added.
        * platform/win/media/video-display-toggle-expected.txt: Added.
            Added.

        * platform/mac-leopard/media/audio-controls-rendering-expected.txt:
        * platform/mac-leopard/media/controls-after-reload-expected.txt:
        * platform/mac-leopard/media/controls-styling-expected.txt:
        * platform/mac-leopard/media/video-zoom-controls-expected.txt:
        * platform/mac/media/audio-controls-rendering-expected.txt:
        * platform/mac/media/controls-after-reload-expected.txt:
        * platform/mac/media/controls-styling-expected.txt:
        * platform/mac/media/video-zoom-controls-expected.txt:
        * platform/win/media/audio-controls-rendering-expected.txt:
        * platform/win/media/controls-after-reload-expected.txt:
        * platform/win/media/controls-drag-timebar-expected.txt:
        * platform/win/media/controls-styling-expected.txt:
            Update for controller element tab order change.

        * platform/gtk/Skipped:
        * platform/win/Skipped:
            Skip new test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebKit/win: Build fix for Windows.
bfulgham@webkit.org [Tue, 25 Aug 2009 23:03:04 +0000 (23:03 +0000)]
WebKit/win: Build fix for Windows.

Revise Debug_Cairo build target to build testapi, jsc, WinLauncher,
DumpRenderTree, and TestNetscapeAPI using WinCairo debug build
settings so they link against the proper libraries.

* WebKit.vcproj/WebKit.sln:

WebKitTools: Build fix

Revise Debug_Cairo targets to point inherit from the
debug_wincairo.vsprops property sheet so that they link
against the proper libraries in Debug build.

* DumpRenderTree/win/DumpRenderTree.vcproj:
* DumpRenderTree/win/TestNetscapePlugin/TestNetscapePlugin.vcproj:
* WinLauncher/WinLauncher.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoVersioning.
mrowe@apple.com [Tue, 25 Aug 2009 22:59:55 +0000 (22:59 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago* JavaScriptCore.vcproj/jsc/jsc.vcproj: Add Debug_CFLite target
bfulgham@webkit.org [Tue, 25 Aug 2009 22:54:30 +0000 (22:54 +0000)]
* JavaScriptCore.vcproj/jsc/jsc.vcproj: Add Debug_CFLite target
  that inherits from the debug_wincairo property sheet and therefore
  links to the proper debug library.
* JavaScriptCore.vcproj/testapi/testapi.vcproj: Add Debug_CFLite target
  that inherits from the debug_wincairo property sheet and therefore
  links to the proper debug library.

Reviewed by NOBODY (Buildfix).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRolled out r44739. Causes crash on Windows.
sfalken@apple.com [Tue, 25 Aug 2009 22:43:20 +0000 (22:43 +0000)]
Rolled out r44739. Causes crash on Windows.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[V8] Fix Canvas layout test failures in Chromium build.
jianli@chromium.org [Tue, 25 Aug 2009 22:38:31 +0000 (22:38 +0000)]
[V8] Fix Canvas layout test failures in Chromium build.
https://bugs.webkit.org/show_bug.cgi?id=28720

Reviewed by David Levin.

The previous fix r47757 does not handle the V8 object conversion correctly.

* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::CALLBACK_FUNC_DECL):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::CALLBACK_FUNC_DECL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[V8] Fix compiling errors in Chromium build caused by Canvas3D changes from r47752.
jianli@chromium.org [Tue, 25 Aug 2009 21:04:58 +0000 (21:04 +0000)]
[V8] Fix compiling errors in Chromium build caused by Canvas3D changes from r47752.
https://bugs.webkit.org/show_bug.cgi?id=28717

Reviewed by David Levin.

* bindings/v8/DOMObjectsInclude.h:
* bindings/v8/DerivedSourcesAllInOne.cpp:
* bindings/v8/V8Index.cpp:
* bindings/v8/V8Index.h:
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::CALLBACK_FUNC_DECL):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::CALLBACK_FUNC_DECL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Alpha Lam <hclam@chromium.org>
eric@webkit.org [Tue, 25 Aug 2009 20:40:16 +0000 (20:40 +0000)]
2009-08-25  Alpha Lam  <hclam@chromium.org>

        Reviewed by David Levin.

        [chromium] Implement media volume slider for chromium
        https://bugs.webkit.org/show_bug.cgi?id=28715

        Provided the implementation of the volume slider and its container for
        chromium port. With this change there will be a usable volume control
        slider for chromium theme.

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

        * css/mediaControlsChromium.css: CSS style for the volume slider.
        * rendering/RenderThemeChromiumSkia.cpp:
        (WebCore::mediaSliderThumbImage): Returns the image for slider thumb.
        (WebCore::mediaVolumeSliderThumbImage): Returns the image for volume slider thumb.
        (WebCore::RenderThemeChromiumSkia::paintMediaVolumeSliderTrack): Paints the track with one vertical white line.
        (WebCore::RenderThemeChromiumSkia::adjustSliderThumbSize): Adjusts thumb sizes according to the images.
        (WebCore::RenderThemeChromiumSkia::paintMediaSliderThumb): Paints slider thumb image.
        (WebCore::RenderThemeChromiumSkia::paintMediaVolumeSliderThumb): Paints volume slider thumb image.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFixed a build problem on Tiger
cmarrin@apple.com [Tue, 25 Aug 2009 19:56:36 +0000 (19:56 +0000)]
Fixed a build problem on Tiger

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Missed checking in 2 Canvas3D files
cmarrin@apple.com [Tue, 25 Aug 2009 19:45:34 +0000 (19:45 +0000)]
    Missed checking in 2 Canvas3D files
        https://bugs.webkit.org/show_bug.cgi?id=28018

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Mads Ager <ager@chromium.org>
abarth@webkit.org [Tue, 25 Aug 2009 19:31:51 +0000 (19:31 +0000)]
2009-08-25  Mads Ager  <ager@chromium.org>

        Reviewed by Adam Barth.

        [V8] Exception from JavaScript propagates to main script
        https://bugs.webkit.org/show_bug.cgi?id=26433

        Test that both compile time and runtime exceptions thrown in
        javascript URLs are isolated from the main script.

        * fast/dom/javascript-url-exception-isolation-expected.txt: Added.
        * fast/dom/javascript-url-exception-isolation.html: Added.
        * fast/dom/resources/javascript-url-exception-isolation.js: Added.
2009-08-25  Mads Ager  <ager@chromium.org>

        Reviewed by Adam Barth.

        [V8] Exception from JavaScript propagates to main script
        https://bugs.webkit.org/show_bug.cgi?id=26433

        Make sure that exceptions thrown both at compile time and at
        runtime in javascript URLs are isolated from the main script.

        Test: fast/dom/javascript-url-exception-isolation.html

        * bindings/v8/V8Proxy.cpp:
        (WebCore::V8Proxy::evaluate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Added all new Canvas3D files to the build
cmarrin@apple.com [Tue, 25 Aug 2009 18:45:56 +0000 (18:45 +0000)]
    Added all new Canvas3D files to the build
        https://bugs.webkit.org/show_bug.cgi?id=28018

        This also hooks up CanvasRenderingContext as the common base class for
        CanvasRenderingContext2D and CanvasRenderingContext3D. And it adds a bit
        of logic throughout to support the new CanvasRenderingContext3D object.
        But nothing is hooked up yet.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoPLATFORM(CFNETWORK) should be USE(CFNETWORK).
levin@chromium.org [Tue, 25 Aug 2009 18:30:00 +0000 (18:30 +0000)]
PLATFORM(CFNETWORK) should be USE(CFNETWORK).
https://bugs.webkit.org/show_bug.cgi?id=28713

Patch by David Levin <levin@chromium.org> on 2009-08-25
Reviewed by Adam Roben.

JavaScriptCore:

* wtf/Platform.h: Added a #define to catch this issue in the
future. The define would generate an error on gcc without the
space in the expansion, but Visual C++ needs the space to cause an error.

WebKit/win:

Also, moved the #if header section after the other #includes.

* WebError.cpp:
(WebError::localizedDescription):

WebKitTools:

* DumpRenderTree/win/DumpRenderTree.cpp:
(main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: Revise CFLite Debug build to emit DLL's with _debug label.
bfulgham@webkit.org [Tue, 25 Aug 2009 17:14:06 +0000 (17:14 +0000)]
JavaScriptCore: Revise CFLite Debug build to emit DLL's with _debug label.
https://bugs.webkit.org/show_bug.cgi?id=28695.

Reviewed by Steve Falkenburg.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Modify
  Cairo debug build to inherit from new debug_cairo property sheet.
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCFLite.vsprops:
  Modify to look for debug CFLite when in debug build.

WebKit/win: Revise CFLite Debug build to emit DLL's with _debug label.
https://bugs.webkit.org/show_bug.cgi?id=28695.

Reviewed by Steve Falkenburg.

* WebKit.vcproj/WebKit.vcproj: Modify Debug_Cairo target to
  inherit from new debug_wincairo property sheet, instead of
  the current debug sheet.

WebKitLibraries: Revise CFLite Debug build to emit DLL's with _debug label.
https://bugs.webkit.org/show_bug.cgi?id=28695.

Reviewed by Steve Falkenburg.

* win/tools/vsprops/debug_wincairo.vsprops: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoREGRESSION (r47630): fast/repaint/background-generated.html is failing
mitz@apple.com [Tue, 25 Aug 2009 17:09:31 +0000 (17:09 +0000)]
REGRESSION (r47630): fast/repaint/background-generated.html is failing
https://bugs.webkit.org/show_bug.cgi?id=28704

Reviewed by Anders Carlsson.

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::calculateFillTileSize): Handle SizeNone.
* rendering/style/FillLayer.cpp:
(WebCore::FillLayer::FillLayer): Initialize the size type to SizeNone.
* rendering/style/FillLayer.h:
(WebCore::FillSize::FillSize): Changed assignment to initialization and
    removed redundant assignment to size.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Yusuke Sato <yusukes@chromium.org>
eric@webkit.org [Tue, 25 Aug 2009 16:28:54 +0000 (16:28 +0000)]
2009-08-25  Yusuke Sato  <yusukes@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] Complex text doesn't show up with text stroking
        https://bugs.webkit.org/show_bug.cgi?id=28707

        Skip the NULL check of m_hdc if Windows GDI is not in use.

        * platform/graphics/chromium/FontChromiumWin.cpp:
        (WebCore::Font::drawComplexText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Update outdated metrics in css1/font_properties. All tests pass.
kenneth@webkit.org [Tue, 25 Aug 2009 13:14:41 +0000 (13:14 +0000)]
[Qt] Update outdated metrics in css1/font_properties. All tests pass.

Patch by Renata Hodovan <hodovan.renata@stud.u-szeged.hu> on 2009-08-25
Reviewed by Ariya Hidayat.

* platform/qt/Skipped:
* platform/qt/css1/font_properties/font-expected.txt:
* platform/qt/css1/font_properties/font_family-expected.txt:
* platform/qt/css1/font_properties/font_size-expected.txt:
* platform/qt/css1/font_properties/font_style-expected.txt:
* platform/qt/css1/font_properties/font_variant-expected.txt:
* platform/qt/css1/font_properties/font_weight-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Antonio Gomes <antonio.gomes@openbossa.org>
kenneth@webkit.org [Tue, 25 Aug 2009 13:12:07 +0000 (13:12 +0000)]
2009-08-25  Antonio Gomes   <antonio.gomes@openbossa.org>

        Reviewed by Ariya Hidayat.

        Bug 28708 - Make possible to better use ResourceError in FrameLoaderClientQt class.

        * Api/qwebframe.cpp:
        (QWebFrame::requestedUrl):
        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::FrameLoaderClientQt):
        (WebCore::FrameLoaderClientQt::dispatchDidFinishLoad):
        (WebCore::FrameLoaderClientQt::postProgressFinishedNotification):
        (WebCore::FrameLoaderClientQt::dispatchDidFailProvisionalLoad):
        (WebCore::FrameLoaderClientQt::dispatchDidFailLoad):
        * WebCoreSupport/FrameLoaderClientQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Update outdated font metrics and add expected files for new tests.
kenneth@webkit.org [Tue, 25 Aug 2009 13:08:29 +0000 (13:08 +0000)]
[Qt] Update outdated font metrics and add expected files for new tests.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-08-25
Reviewed by Ariya Hidayat.

* platform/qt/Skipped:
* platform/qt/fast/backgrounds/001-expected.txt:
* platform/qt/fast/backgrounds/animated-gif-as-background-expected.txt: Added.
* platform/qt/fast/backgrounds/animated-svg-as-background-expected.txt: Added.
* platform/qt/fast/backgrounds/animated-svg-as-mask-expected.txt: Added.
* platform/qt/fast/backgrounds/background-clip-text-expected.txt: Added.
* platform/qt/fast/backgrounds/background-inherit-color-bug-expected.txt:
* platform/qt/fast/backgrounds/background-origin-root-element-expected.txt: Added.
* platform/qt/fast/backgrounds/background-position-1-expected.txt: Added.
* platform/qt/fast/backgrounds/background-position-rounding-expected.txt: Added.
* platform/qt/fast/backgrounds/mask-composite-expected.txt: Added.
* platform/qt/fast/backgrounds/opacity-on-document-element-expected.txt: Added.
* platform/qt/fast/backgrounds/repeat/mask-negative-offset-repeat-expected.txt: Added.
* platform/qt/fast/backgrounds/repeat/negative-offset-repeat-expected.txt:
* platform/qt/fast/backgrounds/repeat/noRepeatCorrectClip-expected.txt:
* platform/qt/fast/backgrounds/size/backgroundSize15-expected.txt:
* platform/qt/fast/backgrounds/size/zero-expected.txt: Added.
* platform/qt/fast/backgrounds/solid-color-context-restore-expected.txt:
* platform/qt/fast/backgrounds/svg-as-mask-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Alpha Lam <hclam@chromium.org>
eric@webkit.org [Tue, 25 Aug 2009 12:04:25 +0000 (12:04 +0000)]
2009-08-25  Alpha Lam  <hclam@chromium.org>

        Reviewed by David Levin.

        Media controls panel does not have a volume control slider
        https://bugs.webkit.org/show_bug.cgi?id=28241

        Introduced "-webkit-media-controls-volume-slider",
        "-webkit-media-controls-volume-slider-container" and the supporting
        implementation. Also added appearances of "media-volume-slider" and
        "media-volume-slider-container".

        The implementation of the controls are in WebCore::MediaControlElements.
        Logic to trigger the display of the volume controls are in WebCore::RenderMedia.

        No new tests since this change doesn't have theme implementation and the
        volume slider is disabled by default.

        * accessibility/AccessibilitySlider.cpp:
        (WebCore::AccessibilitySlider::orientation): Marks that volume slider is vertical.
        * css/CSSPrimitiveValueMappings.h:
        (WebCore::CSSPrimitiveValue::CSSPrimitiveValue): Added corresponding CSS values.
        * css/CSSSelector.cpp:
        (WebCore::CSSSelector::extractPseudoType): ditto.
        * css/CSSSelector.h:
        (WebCore::CSSSelector::): ditto.
        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::SelectorChecker::checkOneSelector): ditto.
        * css/CSSValueKeywords.in: ditto.
        * css/mediaControls.css: Makes the volume slider disabled by default.
        * platform/ThemeTypes.h: Defines theme parts for volume slider and its container.
        (WebCore::):
        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlVolumeSliderContainerElement::MediaControlVolumeSliderContainerElement):
        (WebCore::MediaControlVolumeSliderContainerElement::styleForElement):
        Adjust the CSS values for "display", "postion", "left" and "top" according to member variables.
        (WebCore::MediaControlVolumeSliderContainerElement::setVisible): Affects the "display" CSS value.
        (WebCore::MediaControlVolumeSliderContainerElement::setPosition): Affects the "left" and "top" CSS values.
        (WebCore::MediaControlVolumeSliderContainerElement::hitTest): Hit testing this element.
        (WebCore::MediaControlVolumeSliderElement::MediaControlVolumeSliderElement):
        (WebCore::MediaControlVolumeSliderElement::defaultEventHandler): Modifies the volume using the slider value, allows dragging.
        * rendering/MediaControlElements.h:
        (WebCore::): Added definition for MediaControlVolumeSliderElement and MediaControlVolumeSliderContainerElement.
        (WebCore::MediaControlVolumeSliderContainerElement::isVisible): Inline getter.
        * rendering/RenderMedia.cpp:
        (WebCore::RenderMedia::styleDidChange): Updates styles for volume container and volume slider.
        (WebCore::RenderMedia::createVolumeSliderContainer): Creates volume slider container and attach to panel.
        (WebCore::RenderMedia::createVolumeSlider): Creates volume slider and attach to volume slider container.
        (WebCore::RenderMedia::updateControls): Updates volume container and volume slider.
        (WebCore::RenderMedia::updateVolumeSliderContainer): Sets the visibility and position of volume slider container and its child.
        (WebCore::RenderMedia::forwardEvent):
        Hit testing the volume slider container and volume slider to toggle the visibility and forward events to slider.
        * rendering/RenderMedia.h:
        Added definition for RenderMedia::updateVolumeSliderContainer().
        * rendering/RenderMediaControls.cpp:
        (WebCore::RenderMediaControls::paintMediaControlsPart): Not implemented.
        * rendering/RenderSlider.cpp:
        (WebCore::RenderSlider::createThumbStyle): Makes volume slider vertical.
        (WebCore::RenderSlider::thumbRect): ditto.
        (WebCore::RenderSlider::mouseEventIsInThumb): ditto.
        (WebCore::RenderSlider::setValueForPosition): ditto.
        (WebCore::RenderSlider::positionForOffset): ditto.
        (WebCore::RenderSlider::currentPosition): ditto.
        (WebCore::RenderSlider::trackSize): ditto.
        * rendering/RenderTheme.cpp:
        (WebCore::RenderTheme::adjustStyle): Adjust style for volume slider thumb.
        (WebCore::RenderTheme::paint): Paints volume control parts using theme.
        * rendering/RenderTheme.h:
        (WebCore::RenderTheme::paintMediaVolumeSliderContainer): Naive implementation to be overridden.
        (WebCore::RenderTheme::paintMediaVolumeSliderTrack): ditto.
        (WebCore::RenderTheme::paintMediaVolumeSliderThumb): ditto.
        * rendering/style/RenderStyleConstants.h:
        (WebCore::): Constants for volume control parts.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt/Mac] Use CONFIG+=build_all only when building libraries
vestbo@webkit.org [Tue, 25 Aug 2009 11:36:12 +0000 (11:36 +0000)]
[Qt/Mac] Use CONFIG+=build_all only when building libraries

Reviewed by Ariya Hidayat.

If no configuration is specified when building WebKit we pass the
debug_and_release option to QMake which results in Makefiles for
both configurations being generated.

Previously we built both of these configurations by default, for
all targets (both the QtWebKit framework/dyldlib and the various
executables such as QtLauncher and tests). This makes sense for
the libraries, which get the _debug suffix and can be loaded on
demand by setting the DYLD_IMAGE_SUFFIX, but for executables we
ended up building the same executable twice.

We now only build one instance of each executable, and since this
is a developer build we build the debug-version. Passing either
--debug or --release to build-webkit will override this, and
even in the default case the release version can still be built
by running 'make release' in the the build directory of each
target.

* WebCore.pro:
* DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
* Scripts/webkitdirs.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-25 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 25 Aug 2009 07:18:44 +0000 (07:18 +0000)]
2009-08-25  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Jan Alonzo.

        https://bugs.webkit.org/show_bug.cgi?id=28690
        Build failure in webkitgtk+-1.1.3

        Do not use deprecated function to unref a pixbuf.

        * platform/gtk/CursorGtk.cpp:
        (WebCore::Cursor::Cursor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Hironori Bono <hbono@chromium.org>
abarth@webkit.org [Tue, 25 Aug 2009 06:53:30 +0000 (06:53 +0000)]
2009-08-24  Hironori Bono  <hbono@chromium.org>

        Reviewed by Adam Barth.

        Fix Bug 27827 "[Chromium] Functions Keys don't work in google spreadsheet".
        <https://bugs.webkit.org/show_bug.cgi?id=27827>.

        Because of the lack of mappings from GDK key-codes to WebKit key-codes,
        Chromium cannot send valid key-codes to JavaScript when a user types
        function keys. This change just copies the mappings from 'KeyEventGtk.cpp'.

        To write layout tests for this issue, added mappings from function-key
        names to platform-specific key-codes to EventSendingController objects
        so that eventSender.keyDown() can send function-key events without using
        platform-specific key codes. (Unfortunately, this eventSender.keyDown() change
        is only for Mac. So this change adds this new test to Skipped tests for other
        platforms to prevent this change from crashing the build trees.)

        * fast/events/keydown-function-keys-expected.txt: Added.
        * fast/events/keydown-function-keys.html: Added.
        * platform/gtk/Skipped:
        * platform/qt/Skipped:
        * platform/win/Skipped:
2009-08-24  Hironori Bono  <hbono@chromium.org>

        Reviewed by Adam Barth.

        Fix Bug 27827 "[Chromium] Functions Keys don't work in google spreadsheet".
        <https://bugs.webkit.org/show_bug.cgi?id=27827>.

        Because of the lack of mappings from GDK key-codes to WebKit key-codes,
        Chromium cannot send valid key-codes to JavaScript when a user types
        function keys. This change just copies the mappings from 'KeyEventGtk.cpp'.

        To write layout tests for this issue, added mappings from function-key
        names to platform-specific key-codes to EventSendingController objects
        so that eventSender.keyDown() can send function-key events without using
        platform-specific key codes. (Unfortunately, this eventSender.keyDown() change
        is only for Mac. So this change adds this new test to Skipped tests for other
        platforms to prevent this change from crashing the build trees.)

        Test: fast/events/keydown-function-keys.html

       * platform/chromium/KeyCodeConversionGtk.cpp: Add mappings from GDK key-codes
        to WebKit key-code for function keys.
        (WebCore::windowsKeyCodeForKeyEvent):
2009-08-24  Hironori Bono  <hbono@chromium.org>

        Reviewed by Adam Barth.

        Fix Bug 27827 "[Chromium] Functions Keys don't work in google spreadsheet".
        <https://bugs.webkit.org/show_bug.cgi?id=27827>.

        Because of the lack of mappings from GDK key-codes to WebKit key-codes,
        Chromium cannot send valid key-codes to JavaScript when a user types
        function keys. This change just copies the mappings from 'KeyEventGtk.cpp'.

        To write layout tests for this issue, added mappings from function-key
        names to platform-specific key-codes to EventSendingController objects
        so that eventSender.keyDown() can send function-key events without using
        platform-specific key codes. (Unfortunately, this eventSender.keyDown() change
        is only for Mac. So this change adds this new test to Skipped tests for other
        platforms to prevent this change from crashing the build trees.)

        * DumpRenderTree/mac/EventSendingController.mm:
        (-[EventSendingController keyDown:withModifiers:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoBuild fix: start out with a 32-bit value to avoid a shortening warning.
ggaren@apple.com [Tue, 25 Aug 2009 04:23:59 +0000 (04:23 +0000)]
Build fix: start out with a 32-bit value to avoid a shortening warning.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-08-24
* runtime/Collector.cpp:
(JSC::Heap::sweep):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSubstantially reduced VM thrash in the GC heap.
ggaren@apple.com [Tue, 25 Aug 2009 04:09:29 +0000 (04:09 +0000)]
Substantially reduced VM thrash in the GC heap.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-08-24
Reviewed by Oliver Hunt.

1.08x faster on v8 (1.60x faster on v8-splay).

1.40x faster on bench-alloc-nonretained.

1.90x faster on bench-alloc-retained.

SunSpider says no change.

* runtime/Collector.cpp:
(JSC::Heap::heapAllocate): Fixed a long-standing bug: update a few local
variables unconditionally after calling collect(), since they may be used
even if we don't "goto scan". (In the bug I saw, usedBlocks got out of
sync with heap.usedBlocks).
(JSC::Heap::sweep): Keep enough free heap space to accomodate
the number of objects we'll allocate before the next GC, plus 25%, for
good measure.
* runtime/Collector.h: Bumped the block size to 256k. This seems to give
the best cache performance, and it prevents us from initiating lots of
VM traffic to recover very small chunks of memory.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=28691
barraclough@apple.com [Tue, 25 Aug 2009 02:53:51 +0000 (02:53 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=28691
Do not retain ScopeNodes outside of parsing

Reviewed by Oliver Adler & Darin Hunt.

There is now no need for these to exist outside of parsing - their use in the runtime is replaced by Executable types.

* bytecode/EvalCodeCache.h:
(JSC::EvalCodeCache::get):
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::emitNewFunction):
(JSC::BytecodeGenerator::emitNewFunctionExpression):
* bytecompiler/BytecodeGenerator.h:
(JSC::BytecodeGenerator::makeFunction):
* debugger/Debugger.cpp:
(JSC::Debugger::recompileAllJSFunctions):
(JSC::evaluateInGlobalCallFrame):
* debugger/DebuggerCallFrame.cpp:
(JSC::DebuggerCallFrame::evaluate):
* interpreter/Interpreter.cpp:
(JSC::Interpreter::execute):
(JSC::Interpreter::prepareForRepeatCall):
(JSC::Interpreter::privateExecute):
* jit/JITStubs.cpp:
(JSC::DEFINE_STUB_FUNCTION):
* parser/Nodes.cpp:
(JSC::ScopeNodeData::ScopeNodeData):
(JSC::ProgramNode::create):
(JSC::EvalNode::create):
(JSC::FunctionBodyNode::create):
* parser/Nodes.h:
(JSC::ScopeNode::adoptData):
(JSC::FunctionBodyNode::parameterCount):
* parser/Parser.cpp:
* parser/Parser.h:
(JSC::Parser::arena):
(JSC::Parser::Parser):
(JSC::Parser::parse):
* runtime/ArrayPrototype.cpp:
(JSC::isNumericCompareFunction):
(JSC::arrayProtoFuncSort):
* runtime/Completion.cpp:
(JSC::checkSyntax):
(JSC::evaluate):
* runtime/Executable.cpp:
(JSC::FunctionExecutable::~FunctionExecutable):
(JSC::EvalExecutable::compile):
(JSC::ProgramExecutable::checkSyntax):
(JSC::ProgramExecutable::compile):
(JSC::FunctionExecutable::compile):
(JSC::EvalExecutable::generateJITCode):
(JSC::ProgramExecutable::generateJITCode):
(JSC::FunctionExecutable::generateJITCode):
(JSC::FunctionExecutable::reparseExceptionInfo):
(JSC::EvalExecutable::reparseExceptionInfo):
(JSC::FunctionExecutable::recompile):
(JSC::FunctionExecutable::fromGlobalCode):
(JSC::FunctionExecutable::copyParameters):
(JSC::FunctionExecutable::paramString):
* runtime/Executable.h:
(JSC::ScriptExecutable::ScriptExecutable):
(JSC::ScriptExecutable::sourceID):
(JSC::ScriptExecutable::sourceURL):
(JSC::ScriptExecutable::lineNo):
(JSC::ScriptExecutable::lastLine):
(JSC::ScriptExecutable::usesEval):
(JSC::ScriptExecutable::usesArguments):
(JSC::ScriptExecutable::needsActivation):
(JSC::ScriptExecutable::recordParse):
(JSC::EvalExecutable::bytecode):
(JSC::EvalExecutable::jitCode):
(JSC::ProgramExecutable::bytecode):
(JSC::ProgramExecutable::reparseExceptionInfo):
(JSC::ProgramExecutable::jitCode):
(JSC::FunctionExecutable::FunctionExecutable):
(JSC::FunctionExecutable::make):
(JSC::FunctionExecutable::bytecode):
(JSC::FunctionExecutable::isGenerated):
(JSC::FunctionExecutable::name):
(JSC::FunctionExecutable::parameterCount):
(JSC::FunctionExecutable::jitCode):
* runtime/FunctionConstructor.cpp:
(JSC::constructFunction):
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::numericCompareFunction):
* runtime/JSGlobalObjectFunctions.cpp:
(JSC::globalFuncEval):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=28698
antti@apple.com [Tue, 25 Aug 2009 00:50:26 +0000 (00:50 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=28698
Avoid unnecessary transformations for the window resizer rect.

Reviewed by Simon Fraser.

* platform/Scrollbar.cpp:
(WebCore::Scrollbar::setFrameRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Anders Carlsson.
ap@apple.com [Tue, 25 Aug 2009 00:42:52 +0000 (00:42 +0000)]
    Reviewed by Anders Carlsson.

        <rdar://problem/7161900> REGRESSION: 100 MB membuster leak on windows due to 47391
        (DNS prefetching support)

        * platform/network/cf/DNSCFNet.cpp: (WebCore::prefetchDNS): Use a run loop from secondary
        thread on Windows, because there isn't one on main one.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Darin Adler.
ap@apple.com [Tue, 25 Aug 2009 00:09:39 +0000 (00:09 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=28374
        Cross-scheme requests should not be blocked by appcache rules

        Test: http/tests/appcache/different-scheme.html

        * loader/appcache/ApplicationCacheHost.cpp:
        (WebCore::ApplicationCacheHost::shouldLoadResourceFromApplicationCache):
        Resources with different schemes can never be loaded from appcache.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Darin Adler and Sam Weinig.
ap@apple.com [Tue, 25 Aug 2009 00:03:19 +0000 (00:03 +0000)]
    Reviewed by Darin Adler and Sam Weinig.

        https://bugs.webkit.org/show_bug.cgi?id=28688
        REGRESSION(r24994): Cannot create a frame with a javascript URL

        Test: http/tests/security/javascriptURL/javascriptURL-in-new-iframe.html

        * bindings/js/JSAttrCustom.cpp: (WebCore::JSAttr::setValue):
        * bindings/js/JSElementCustom.cpp: (WebCore::allowSettingSrcToJavascriptURL):
        * bindings/js/JSHTMLFrameElementCustom.cpp: (WebCore::allowSettingJavascriptURL):
        * bindings/js/JSHTMLIFrameElementCustom.cpp: (WebCore::JSHTMLIFrameElement::setSrc):
        Omit the checks if there is no document in frame yet.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 24 Aug 2009 22:58:39 +0000 (22:58 +0000)]
2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Skip the correct test (media/video-currentTime-set.html, not
        media/video-currentTime-set2.html).

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 24 Aug 2009 22:47:53 +0000 (22:47 +0000)]
2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Skip 3 new tests, and reenable some of the media tests, now that
        the bot it fixed to handle the fact that it doesn't have a sound
        card.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 24 Aug 2009 21:56:40 +0000 (21:56 +0000)]
2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed. Skip again tests that we enabled, since they are
        failing on the bot. The tests do pass, but there is an environment
        configuration problem with GStreamer on the bot (possibly related
        to the fact that it has no sound card).

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago* runtime/ObjectPrototype.cpp:
darin@apple.com [Mon, 24 Aug 2009 21:37:29 +0000 (21:37 +0000)]
* runtime/ObjectPrototype.cpp:
(JSC::ObjectPrototype::put): Landed revised version I had tested but forgot
to land. Leave out the branch, since we don't need one.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 24 Aug 2009 21:21:03 +0000 (21:21 +0000)]
2009-08-24  Nate Chapin  <japhet@chromium.org>

        Reviewed by Eric Seidel.

        Fix crash in fast/css/rem-dynamic-scaling.html due to a freed
        RenderStyle being accessed.

        https://bugs.webkit.org/show_bug.cgi?id=28681

        * dom/Element.cpp:
        (WebCore::Element::recalcStyle): Make currentStyle a RefPtr.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Mark Rowe.
adachan@apple.com [Mon, 24 Aug 2009 21:08:11 +0000 (21:08 +0000)]
    Reviewed by Mark Rowe.

        Skip loader/go-back-to-different-window-size.html for qt and gtk.

        * platform/gtk/Skipped:
        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoArray index miss case creates a string every time
darin@apple.com [Mon, 24 Aug 2009 21:05:13 +0000 (21:05 +0000)]
Array index miss case creates a string every time
https://bugs.webkit.org/show_bug.cgi?id=28664

Patch by Darin Adler <darin@apple.com> on 2009-08-24
Reviewed by Geoff Garen.

SunSpider test results I saw:

    0.5% faster overall
    1% faster on crypto-aes
    20% faster on crypto-md5
    13% faster on crypto-sha1

* runtime/ObjectPrototype.cpp:
(JSC::ObjectPrototype::ObjectPrototype): Initialize m_hasNoPropertiesWithUInt32Names
to true.
(JSC::ObjectPrototype::put): Clearly m_hasNoPropertiesWithUInt32Names if the new
property has a name that is the string form of a UInt32.
(JSC::ObjectPrototype::getOwnPropertySlot): Don't call JSObject::getOwnPropertySlot
if m_hasNoPropertiesWithUInt32Names is true, and it is highly likely to be true.

* runtime/ObjectPrototype.h: Added declarations for the above.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/5743105> Long freezing during loading and selecting text
mitz@apple.com [Mon, 24 Aug 2009 20:45:26 +0000 (20:45 +0000)]
<rdar://problem/5743105> Long freezing during loading and selecting text
of a large table
https://bugs.webkit.org/show_bug.cgi?id=14918

Reviewed by Simon Fraser.

* rendering/RenderView.cpp:
(WebCore::RenderView::setSelection): Added calls to
    beginDeferredRepaints() and endDeferredRepaints() around the
    invalidation of the selection, in order to better coalesce multiple
    invalid rectangles.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 24 Aug 2009 20:26:51 +0000 (20:26 +0000)]
2009-08-24  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        WebInspector: Simplify findTreeElement code, get rid of potential infinite loop.

        https://bugs.webkit.org/show_bug.cgi?id=28680

        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel.prototype.updateBreadcrumb):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeOutline.prototype.findTreeElement):
        * inspector/front-end/treeoutline.js:
        (TreeOutline.prototype.getCachedTreeElement):
        (TreeOutline.prototype.findTreeElement):
        * inspector/front-end/utilities.js:
        (isAncestorNode):
        ():

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 24 Aug 2009 20:20:44 +0000 (20:20 +0000)]
2009-08-24  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        WebInspector: Do expand local scope when hitting the break

        https://bugs.webkit.org/show_bug.cgi?id=28678

        * inspector/front-end/ScopeChainSidebarPane.js:
        (WebInspector.ScopeChainSidebarPane.prototype.update):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Brady Eidson.
adachan@apple.com [Mon, 24 Aug 2009 20:16:00 +0000 (20:16 +0000)]
    Reviewed by Brady Eidson.

        Add test for https://bugs.webkit.org/show_bug.cgi?id=28659.
        Cached page should be formatted according to the current window size, not the cached frame's size.

        * loader: Added.
        * loader/go-back-to-different-window-size-expected.txt: Added.
        * loader/go-back-to-different-window-size.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Steve Falkenburg.
adachan@apple.com [Mon, 24 Aug 2009 19:55:23 +0000 (19:55 +0000)]
    Reviewed by Steve Falkenburg.

        Make sure we use the current FrameView's frame rect when
        going back to a cached page.

        https://bugs.webkit.org/show_bug.cgi?id=28659

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::open):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Eric Seidel <eric@webkit.org>
eric@webkit.org [Mon, 24 Aug 2009 19:10:17 +0000 (19:10 +0000)]
2009-08-24  Eric Seidel  <eric@webkit.org>

        No review, just updating results.

        Support placeholder on textarea
        https://bugs.webkit.org/show_bug.cgi?id=21248

        Update expected test results form windows bot.

        * platform/win/fast/forms/basic-textareas-expected.txt:
        * platform/win/fast/forms/textarea-placeholder-pseudo-style-expected.txt: Added.
        * platform/win/fast/forms/textarea-placeholder-set-attribute-expected.txt: Added.
        * platform/win/fast/forms/textarea-placeholder-set-value-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Cameron McCormack <cam@mcc.id.au>
eric@webkit.org [Mon, 24 Aug 2009 18:35:22 +0000 (18:35 +0000)]
2009-08-24  Cameron McCormack  <cam@mcc.id.au>

        Reviewed by Darin Adler.

        Modifying <text rotate=""> doesn't clear the corresponding SVGAnimatedNumberList
        https://bugs.webkit.org/show_bug.cgi?id=28673

        * svg/dom/resources/text-rotate-live.js: Added.
        (getRotate):
        (getAndSetRotate):
        * svg/dom/text-rotate-live-expected.txt: Added.
        * svg/dom/text-rotate-live.html: Added.
2009-08-24  Cameron McCormack  <cam@mcc.id.au>

        Reviewed by Darin Adler.

        Modifying <text rotate=""> doesn't clear the corresponding SVGAnimatedNumberList
        https://bugs.webkit.org/show_bug.cgi?id=28673

        Test: svg/dom/text-rotate-live.html

        * svg/SVGNumberList.cpp:
        (WebCore::SVGNumberList::parse): Clear the list before adding the parsed
        numbers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 24 Aug 2009 18:07:31 +0000 (18:07 +0000)]
2009-08-24  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by NOBODY (build fix)

        Turn off ENABLE_3D_CANVAS in the xconfig files.

        * Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 24 Aug 2009 16:40:34 +0000 (16:40 +0000)]
2009-08-24  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        WebInspector: Only expand local scope when hitting the break

        https://bugs.webkit.org/show_bug.cgi?id=28678

        * inspector/front-end/ScopeChainSidebarPane.js:
        (WebInspector.ScopeChainSidebarPane.prototype.update):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore:
mitz@apple.com [Mon, 24 Aug 2009 16:26:47 +0000 (16:26 +0000)]
WebCore:
background-size fails to parse if a single length/percentage/auto is followed by a comma
https://bugs.webkit.org/show_bug.cgi?id=28674

Reviewed by Simon Fraser.

* css/CSSParser.cpp:
(WebCore::CSSParser::parseFillSize): Added an allowComma parameter,
    which is set to false when this method consumes a comma. Moved
    handling of 'contain' and 'cover' from parseFillProperty into this
    method.
(WebCore::CSSParser::parseFillProperty): Set allowComma to true before
    processing the next value, and pass allowComma to parseFillSize.
* css/CSSParser.h:

LayoutTests:
background-size fails to parse if a single length/percentage/auto is followed by a comma
https://bugs.webkit.org/show_bug.cgi?id=28674

Reviewed by Simon Fraser.

* fast/backgrounds/size/parsing-background-size-values-expected.txt:
* fast/backgrounds/size/resources/parsing-background-size-values.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 24 Aug 2009 16:02:47 +0000 (16:02 +0000)]
2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed. Fix a typo in my distcheck build fix.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 24 Aug 2009 15:41:58 +0000 (15:41 +0000)]
2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Xan Lopez.

        News about the 1.1.13 release.

        * NEWS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-24 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 24 Aug 2009 15:28:24 +0000 (15:28 +0000)]
2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Xan Lopez.

        Bump package and so versions for 1.1.13 release.

        * configure.ac:

WebKit/gtk

2009-08-24  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Xan Lopez.

        Add new symbols, and an index of new symbols for the 1.1.13
        release.

        * docs/webkitgtk-docs.sgml:
        * docs/webkitgtk-sections.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-08-17 Martin Robinson <martin.james.robinson@gmail.com>
kov@webkit.org [Mon, 24 Aug 2009 14:55:07 +0000 (14:55 +0000)]
2009-08-17  Martin Robinson  <martin.james.robinson@gmail.com>

        Reviewed by Gustavo Noronha.

        [GTK] Support for cursors that are images
        https://bugs.webkit.org/show_bug.cgi?id=28346

        Added support for cursors which are images for CursorGtk. Added
        a manual test for this.

        * manual-tests/gtk/cursor-image.html: Added.
        * manual-tests/gtk/resources/redcursor.cur: Added.
        * platform/gtk/CursorGtk.cpp:
        (WebCore::Cursor::Cursor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed. Add missing files from former commit 47709.
kenneth@webkit.org [Mon, 24 Aug 2009 14:48:01 +0000 (14:48 +0000)]
Unreviewed. Add missing files from former commit 47709.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-08-24
* platform/qt/fast/backgrounds/svg-as-background-1-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-2-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-3-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-4-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-5-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-6-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore:
kov@webkit.org [Mon, 24 Aug 2009 14:30:38 +0000 (14:30 +0000)]
WebCore:

2009-08-24  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        [GTK] Integrate GStreamer video with the graphics backend
        https://bugs.webkit.org/show_bug.cgi?id=16356

        Improves GStreamer backend implementation of MediaPlayerPrivate class.

        * platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp:
        (WebCore::mediaPlayerPrivateErrorCallback): Send proper MediaPlayer
        error for different GStreamer errors to MediaPlayerPrivate.
        (WebCore::MediaPlayerPrivate::MediaPlayerPrivate): Move static
        initialization boolean to a better place.
        (WebCore::MediaPlayerPrivate::play):
        (WebCore::MediaPlayerPrivate::pause):
        (WebCore::MediaPlayerPrivate::duration): Return if error occured, do not
        consider the source as stream if duration query failed.
        (WebCore::MediaPlayerPrivate::currentTime): Return if error occured, use
        an early return.
        (WebCore::MediaPlayerPrivate::seek): Return if error occured.
        (WebCore::MediaPlayerPrivate::setEndTime): Remove implementation since this
        function doesn't seem to be used anywhere in WebCore.
        (WebCore::MediaPlayerPrivate::paused):
        (WebCore::MediaPlayerPrivate::seeking):
        (WebCore::MediaPlayerPrivate::setRate): Seek to current time when rate is set.
        (WebCore::MediaPlayerPrivate::maxTimeBuffered): Return if error occured.
        (WebCore::MediaPlayerPrivate::maxTimeSeekable): Ditto.
        (WebCore::MediaPlayerPrivate::maxTimeLoaded): Ditto.
        (WebCore::MediaPlayerPrivate::bytesLoaded):
        (WebCore::MediaPlayerPrivate::totalBytesKnown):
        (WebCore::MediaPlayerPrivate::totalBytes): Ditto.
        (WebCore::MediaPlayerPrivate::updateStates): Return if error occured, update
        after seek ended with a successful change
        (WebCore::MediaPlayerPrivate::didEnd): Do not pause the playbin when end is reached.
        (WebCore::MediaPlayerPrivate::loadingFailed): Update network state with the given error.
        (WebCore::mimeTypeCache): Gather supported mime types from GStreamer.
        (WebCore::MediaPlayerPrivate::getSupportedTypes):
        (WebCore::MediaPlayerPrivate::supportsType):
        (WebCore::MediaPlayerPrivate::hasSingleSecurityOrigin):
        (WebCore::MediaPlayerPrivate::supportsFullscreen): First step towards fullscreen
        support.
        (WebCore::MediaPlayerPrivate::createGSTPlayBin): Use playbin2 instead of playbin,
        do not set playbin's audio sink.
        * platform/graphics/gtk/MediaPlayerPrivateGStreamer.h:
        * platform/graphics/gtk/VideoSinkGStreamer.cpp:
        (webkit_video_sink_idle_func): Use C++ casts, decrease reference count of async
        queue at the end of the idle function.
        (webkit_video_sink_render): Increase reference count of async queue.
        (webkit_video_sink_stop): Remove any idle functions with the sink as data.

LayoutTests:

2009-08-24  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        [GTK] Integrate GStreamer video with the graphics backend
        https://bugs.webkit.org/show_bug.cgi?id=16356

        Enable media tests that now pass.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Add results with updated font metrics for passing tests added in
kenneth@webkit.org [Mon, 24 Aug 2009 14:26:05 +0000 (14:26 +0000)]
[Qt] Add results with updated font metrics for passing tests added in
r28637 and remove them from Skipped.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-08-24
Reviewed by Ariya Hidayat.

* platform/qt/Skipped:
* platform/qt/fast/backgrounds/svg-as-background-1-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-2-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-3-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-4-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-5-expected.txt: Added.
* platform/qt/fast/backgrounds/svg-as-background-6-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@47709 268f45cc-cd09-0410-ab3c-d52691b4dbfc