WebKit-https.git
8 years agoCache CSSInitialValue instances per-document.
kling@webkit.org [Sat, 3 Dec 2011 21:53:29 +0000 (21:53 +0000)]
Cache CSSInitialValue instances per-document.
<http://webkit.org/b/73745>

Reviewed by Oliver Hunt.

Source/WebCore:

Test: http/tests/security/cross-origin-css-primitive.html

Have CSSValuePool manage the caching of CSSInitialValue objects.

* css/CSSInitialValue.h:
(WebCore::CSSInitialValue::createExplicit):
(WebCore::CSSInitialValue::createImplicit):
(WebCore::CSSInitialValue::CSSInitialValue):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseFillShorthand):
(WebCore::CSSParser::parseAnimationShorthand):
(WebCore::CSSParser::parseTransitionShorthand):
(WebCore::CSSParser::parseShorthand):
* css/CSSValuePool.cpp:
(WebCore::CSSValuePool::CSSValuePool):
(WebCore::CSSValuePool::createImplicitInitialValue):
(WebCore::CSSValuePool::createExplicitInitialValue):
* css/CSSValuePool.h:

LayoutTests:

Test that 'initial' and 'inherit' values are cached per-document.

* http/tests/security/cross-origin-css-primitive-expected.txt:
* http/tests/security/cross-origin-css-primitive.html:
* http/tests/security/resources/cross-origin-css-primitive-iframe.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoKeep CSSInheritedValue in the CSS value pool.
kling@webkit.org [Sat, 3 Dec 2011 21:45:35 +0000 (21:45 +0000)]
Keep CSSInheritedValue in the CSS value pool.
<http://webkit.org/b/73747>

Reviewed by Antti Koivisto.

We only need one CSSInheritedValue instance per document, so cache it
in CSSValuePool and have the parser create it through there.

* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* css/CSSValuePool.cpp:
(WebCore::CSSValuePool::CSSValuePool):
(WebCore::CSSValuePool::createInheritedValue):
* css/CSSValuePool.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Implement History API FrameLoaderClient methods
commit-queue@webkit.org [Sat, 3 Dec 2011 20:06:11 +0000 (20:06 +0000)]
[GTK] Implement History API FrameLoaderClient methods
https://bugs.webkit.org/show_bug.cgi?id=59017

Patch by Xan Lopez <xlopez@igalia.com> on 2011-12-03
Reviewed by Gustavo Noronha Silva.

Fake the load process when the new HTML5 History APIs are
used. Otherwise the UA does not have a chance of updating the URL
entry, back/forward buttons, etc.

* WebCoreSupport/FrameLoaderClientGtk.cpp:
(WebKit::FrameLoaderClient::dispatchDidNavigateWithinPage): fake a
load process.
(WebKit::FrameLoaderClient::dispatchDidPushStateWithinPage): call
dispatchDidNavigateWithinPage.
(WebKit::FrameLoaderClient::dispatchDidReplaceStateWithinPage):
ditto.
(WebKit::FrameLoaderClient::dispatchDidPopStateWithinPage): do
nothing, because push was already called, but note this in a
comment.
* WebCoreSupport/FrameLoaderClientGtk.h: add the new method.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101933 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRename CSSPrimitiveValueCache to CSSValuePool.
kling@webkit.org [Sat, 3 Dec 2011 18:19:44 +0000 (18:19 +0000)]
Rename CSSPrimitiveValueCache to CSSValuePool.
<http://webkit.org/b/73742>

Reviewed by Antti Koivisto.

CSSPrimitiveValueCache -> CSSValuePool
Document::primitiveValueCache() -> Document::cssValuePool()

This is in preparation for sharing more than just primitive values.

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/scripts/CodeGeneratorV8.pm:
(AddIncludesForType):
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::valueForNinePieceImageSlice):
(WebCore::valueForNinePieceImageQuad):
(WebCore::valueForNinePieceImageRepeat):
(WebCore::valueForNinePieceImage):
(WebCore::zoomAdjustedPixelValue):
(WebCore::zoomAdjustedNumberValue):
(WebCore::zoomAdjustedPixelValueForLength):
(WebCore::valueForReflection):
(WebCore::getPositionOffsetValue):
(WebCore::CSSComputedStyleDeclaration::currentColorOrValidColor):
(WebCore::getBorderRadiusCornerValue):
(WebCore::computedTransform):
(WebCore::CSSComputedStyleDeclaration::valueForFilter):
(WebCore::valueForGridTrackList):
(WebCore::getDelayValue):
(WebCore::getDurationValue):
(WebCore::createLineBoxContainValue):
(WebCore::CSSComputedStyleDeclaration::getFontSizeCSSValuePreferringKeyword):
(WebCore::CSSComputedStyleDeclaration::valueForShadow):
(WebCore::valueForFamily):
(WebCore::renderTextDecorationFlagsToCSSValue):
(WebCore::fillRepeatToCSSValue):
(WebCore::fillSizeToCSSValue):
(WebCore::contentToCSSValue):
(WebCore::counterToCSSValue):
(WebCore::fontFamilyFromStyle):
(WebCore::lineHeightFromStyle):
(WebCore::fontSizeFromStyle):
(WebCore::fontStyleFromStyle):
(WebCore::fontVariantFromStyle):
(WebCore::fontWeightFromStyle):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::parseColorValue):
(WebCore::parseSimpleLengthValue):
(WebCore::CSSParser::setStyleSheet):
(WebCore::CSSParser::createPrimitiveNumericValue):
(WebCore::CSSParser::createPrimitiveStringValue):
(WebCore::CSSParser::parseValidPrimitive):
(WebCore::CSSParser::parseValue):
(WebCore::parseBackgroundClip):
(WebCore::CSSParser::parseFillShorthand):
(WebCore::CSSParser::parsePage):
(WebCore::CSSParser::parseSizeParameter):
(WebCore::CSSParser::parseContent):
(WebCore::CSSParser::parseAttr):
(WebCore::CSSParser::parseBackgroundColor):
(WebCore::CSSParser::parseFillPositionX):
(WebCore::CSSParser::parseFillPositionY):
(WebCore::CSSParser::parseFillPositionComponent):
(WebCore::CSSParser::parseFillPosition):
(WebCore::CSSParser::parseFillRepeat):
(WebCore::CSSParser::parseFillSize):
(WebCore::CSSParser::parseFillProperty):
(WebCore::CSSParser::parseAnimationDirection):
(WebCore::CSSParser::parseAnimationFillMode):
(WebCore::CSSParser::parseAnimationIterationCount):
(WebCore::CSSParser::parseAnimationName):
(WebCore::CSSParser::parseAnimationPlayState):
(WebCore::CSSParser::parseAnimationProperty):
(WebCore::CSSParser::parseAnimationTimingFunction):
(WebCore::CSSParser::parseGridTrackList):
(WebCore::CSSParser::parseDashboardRegions):
(WebCore::CSSParser::parseCounterContent):
(WebCore::CSSParser::parseShape):
(WebCore::CSSParser::parseWrapShape):
(WebCore::CSSParser::parseFont):
(WebCore::CSSParser::parseFontFamily):
(WebCore::CSSParser::parseFontStyle):
(WebCore::CSSParser::parseFontVariant):
(WebCore::CSSParser::parseFontWeight):
(WebCore::CSSParser::parseColor):
(WebCore::ShadowParseContext::ShadowParseContext):
(WebCore::ShadowParseContext::commitLength):
(WebCore::ShadowParseContext::commitStyle):
(WebCore::CSSParser::parseShadow):
(WebCore::CSSParser::parseReflect):
(WebCore::CSSParser::parseFlex):
(WebCore::BorderImageParseContext::BorderImageParseContext):
(WebCore::CSSParser::parseBorderImage):
(WebCore::CSSParser::parseBorderImageRepeat):
(WebCore::BorderImageSliceParseContext::BorderImageSliceParseContext):
(WebCore::BorderImageSliceParseContext::commitNumber):
(WebCore::BorderImageSliceParseContext::commitBorderImageSlice):
(WebCore::CSSParser::parseBorderImageSlice):
(WebCore::BorderImageQuadParseContext::BorderImageQuadParseContext):
(WebCore::BorderImageQuadParseContext::commitNumber):
(WebCore::BorderImageQuadParseContext::commitBorderImageQuad):
(WebCore::CSSParser::parseBorderImageQuad):
(WebCore::CSSParser::parseBorderRadius):
(WebCore::CSSParser::parseAspectRatio):
(WebCore::CSSParser::parseCounter):
(WebCore::parseDeprecatedGradientPoint):
(WebCore::parseDeprecatedGradientColorStop):
(WebCore::CSSParser::parseDeprecatedGradient):
(WebCore::valueFromSideKeyword):
(WebCore::parseGradientColorOrKeyword):
(WebCore::CSSParser::parseLinearGradient):
(WebCore::CSSParser::parseRadialGradient):
(WebCore::CSSParser::parseCrossfade):
(WebCore::CSSParser::parseCustomFilter):
(WebCore::CSSParser::parseFilter):
(WebCore::CSSParser::parseFlowThread):
(WebCore::CSSParser::parseRegionThread):
(WebCore::CSSParser::parseTextEmphasisStyle):
(WebCore::CSSParser::parseFontFeatureSettings):
* css/CSSParser.h:
(WebCore::CSSParser::cssValuePool):
* css/CSSValuePool.cpp: Renamed from Source/WebCore/css/CSSPrimitiveValueCache.cpp.
(WebCore::CSSValuePool::CSSValuePool):
(WebCore::CSSValuePool::~CSSValuePool):
(WebCore::CSSValuePool::createIdentifierValue):
(WebCore::CSSValuePool::createColorValue):
(WebCore::CSSValuePool::createValue):
* css/CSSValuePool.h: Renamed from Source/WebCore/css/CSSPrimitiveValueCache.h.
(WebCore::CSSValuePool::create):
(WebCore::CSSValuePool::createValue):
* css/mediaControlsGtk.css:
(audio::-webkit-media-controls-volume-slider, video::-webkit-media-controls-volume-slider):
* dom/Document.cpp:
(WebCore::Document::cssValuePool):
* dom/Document.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAnother GTK build fix after r101922.
philn@webkit.org [Sat, 3 Dec 2011 18:17:52 +0000 (18:17 +0000)]
Another GTK build fix after r101922.

Rubber-stamped by Martin James Robinson.

* DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
(LayoutTestController::pathToLocalResource): return value needs to
be a URI.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSmall fix for my previous patch. Do not try to delete a member
mrobinson@webkit.org [Sat, 3 Dec 2011 16:37:50 +0000 (16:37 +0000)]
Small fix for my previous patch. Do not try to delete a member
of a dictionary that does not exist.

Reviewed by Gustavo Noronha Silva.

* gtk/run-with-jhbuild:
(install_and_run_jhbuild):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix jhbuild installation for people who have unusual MAKE environment
mrobinson@webkit.org [Sat, 3 Dec 2011 15:55:51 +0000 (15:55 +0000)]
Fix jhbuild installation for people who have unusual MAKE environment
variables such as 'make -j4' as the jhbuild build does not handle it.

Reviewed by Gustavo Noronha Silva.

* gtk/run-with-jhbuild:
(install_and_run_jhbuild): Remove any MAKE environment variable before building jhbuild.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix the update-webkitgtk-libs script to properly trigger a jhbuild installation
mrobinson@webkit.org [Sat, 3 Dec 2011 15:39:02 +0000 (15:39 +0000)]
Fix the update-webkitgtk-libs script to properly trigger a jhbuild installation
when run for the first time.

Reviewed by Gustavo Noronha Silva.

* Scripts/update-webkitgtk-libs: Fix first run.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101928 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, skip new failing tests on GTK.
philn@webkit.org [Sat, 3 Dec 2011 14:07:44 +0000 (14:07 +0000)]
Unreviewed, skip new failing tests on GTK.

* platform/gtk/Skipped: Skip 3 new
http/tests/security/mixedContent failing partly because of bug 73743.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, GTK build fix after r101922.
philn@webkit.org [Sat, 3 Dec 2011 13:55:07 +0000 (13:55 +0000)]
Unreviewed, GTK build fix after r101922.

* DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
(LayoutTestController::pathToLocalResource): Added sentinel in
g_build_filename call.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoGTK API tests build fix. Run the tests inside jhbuild.
philn@webkit.org [Sat, 3 Dec 2011 13:02:58 +0000 (13:02 +0000)]
GTK API tests build fix. Run the tests inside jhbuild.

Rubber-stamped by Gustavo Noronha Silva.

* Scripts/run-gtk-tests:
(TestRunner.__init__):
(TestRunner.run._error_handler):
(TestRunner):
(TestRunner.run):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoShrink CSSValueList.
kling@webkit.org [Sat, 3 Dec 2011 12:20:26 +0000 (12:20 +0000)]
Shrink CSSValueList.
<http://webkit.org/b/73732>

Reviewed by Antti Koivisto.

Packed CSSValueList::m_isSpaceSeparated into the CSSValue bit field
and renamed it to m_isSpaceSeparatedValue, shrinking CSSValueList
by one CPU word.

Also renamed CSSValue::m_isImplicit to m_isImplicitInitialValue
for good measure.

* css/CSSInitialValue.h:
(WebCore::CSSInitialValue::CSSInitialValue):
* css/CSSValue.h:
(WebCore::CSSValue::isImplicitInitialValue):
(WebCore::CSSValue::CSSValue):
* css/CSSValueList.cpp:
(WebCore::CSSValueList::CSSValueList):
(WebCore::CSSValueList::copy):
(WebCore::CSSValueList::customCssText):
* css/CSSValueList.h:
(WebCore::CSSValueList::isSpaceSeparated):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSmall build fix. Properly specify the path to the locally installed jhbuild.
mrobinson@webkit.org [Sat, 3 Dec 2011 12:12:07 +0000 (12:12 +0000)]
Small build fix. Properly specify the path to the locally installed jhbuild.

* Scripts/update-webkitgtk-libs: Fix jhbuild path.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK][DRT] Normalize file:///tmp/LayoutTests in LayoutTestController::pathToLocalReso...
commit-queue@webkit.org [Sat, 3 Dec 2011 11:33:05 +0000 (11:33 +0000)]
[GTK][DRT] Normalize file:///tmp/LayoutTests in LayoutTestController::pathToLocalResource()
https://bugs.webkit.org/show_bug.cgi?id=67256

Patch by Martin Robinson <mrobinson@igalia.com> on 2011-12-03
Reviewed by Philippe Normand.

Implement LayoutTestController::pathToLocalResource for GTK+. Instead of passing
WEBKIT_TEST_FONTS to the GTK+ test harnesses, pass a more generic WEBKIT_TOP_LEVEL,
which points to the more generic top-level path of the WebKit checkout. This code is
duplicated between WK1 and WK2 harnesses because we do not currently have a way to
share code here.

If WEBKIT_TOP_LEVEL is not provided, we search for the top level based on the binary
location. This will cause the fallback to fail if you build into a non-typical location
or even fake it with a symlink. In this case it's important to use the environment variable.

* DumpRenderTree/gtk/DumpRenderTree.cpp:
(getTopLevelPath): Added.
(initializeFonts): Use the new helper to get the font path.
* DumpRenderTree/gtk/DumpRenderTreeGtk.h: Expose the new helper.
* DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
(LayoutTestController::pathToLocalResource):Use the new helper to normalize
test paths.
* Scripts/webkitpy/layout_tests/port/gtk.py:
(GtkPort.setup_environ_for_server):Pass the top-level path of the
checkout instead of passing the font path.
* WebKitTestRunner/InjectedBundle/gtk/ActivateFontsGtk.cpp:
(WTR::getTopLevelPath): Added this helper.
(WTR::inititializeFontConfigSetting): Use the helper to find the font path.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, GTK rebaseline after r101858.
philn@webkit.org [Sat, 3 Dec 2011 11:06:21 +0000 (11:06 +0000)]
Unreviewed, GTK rebaseline after r101858.

* platform/gtk/fast/writing-mode/broken-ideograph-small-caps-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GStreamer] Fullscreen controller support for the new WebKit Fullscreen API
philn@webkit.org [Sat, 3 Dec 2011 10:26:48 +0000 (10:26 +0000)]
[GStreamer] Fullscreen controller support for the new WebKit Fullscreen API
https://bugs.webkit.org/show_bug.cgi?id=66968

Reviewed by Martin Robinson.

Source/WebCore:

* GNUmakefile.am: Don't use the Quicktime fullscreen CSS anymore,
it wasn't suited for GTK in the first place anyway.
* platform/gtk/RenderThemeGtk.cpp:
(WebCore::RenderThemeGtk::extraFullScreenStyleSheet): Simply reuse
GTK CSS.

Source/WebKit/gtk:

If the element being fullscreened is a video use the native
fullscreen controls.

* WebCoreSupport/ChromeClientGtk.cpp:
(WebKit::ChromeClient::enterFullScreenForElement):
(WebKit::ChromeClient::exitFullScreenForElement):

LayoutTests:

* platform/gtk/Skipped: Skip
fullscreen/video-controls-override.html now failing because of bug 73583.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago2011-12-03 Alejandro G. Castro <alex@igalia.com>
alex@webkit.org [Sat, 3 Dec 2011 09:44:59 +0000 (09:44 +0000)]
2011-12-03  Alejandro G. Castro  <alex@igalia.com>

        [GTK] TextureMapperNode should not use Qt types and functions
        https://bugs.webkit.org/show_bug.cgi?id=73713

        Replace qreal with double and qMin with min.

        Reviewed by Noam Rosenthal.

        * platform/graphics/texmap/TextureMapperNode.cpp:
        (WebCore::solveCubicBezierFunction):
        (WebCore::solveStepsFunction):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Unreviewed weekend gardening.
ossy@webkit.org [Sat, 3 Dec 2011 09:36:54 +0000 (09:36 +0000)]
[Qt] Unreviewed weekend gardening.

* platform/qt/Skipped: Skip new failing tests.
* platform/qt/fast/css/font-face-opentype-expected.png: Updated.
* platform/qt/fast/css/font-face-opentype-expected.txt: Updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago.: [GTK] Update required libsoup version
mrobinson@webkit.org [Sat, 3 Dec 2011 09:22:54 +0000 (09:22 +0000)]
.: [GTK] Update required libsoup version
https://bugs.webkit.org/show_bug.cgi?id=71611

Patch by Dan Winship <danw@gnome.org> on 2011-12-03
Reviewed by Martin Robinson.

* configure.ac: require libsoup 2.37.2.1 for SoupRequestHTTP API
changes

Source/WebCore: [GTK] Remove lots of code that is now unnecessary after
SoupRequestHTTP API changes.
https://bugs.webkit.org/show_bug.cgi?id=71611

Patch by Dan Winship <danw@gnome.org> on 2011-12-03
Reviewed by Martin Robinson.

No new tests; behavior is unchanged

* platform/network/ResourceHandleInternal.h:
(WebCore::ResourceHandleInternal::ResourceHandleInternal): remove
no-longer-needed fields.
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::finishedCallback): remove this, and m_finished, which
we no longer need to explicitly track
(WebCore::statusWillBeHandledBySoup):
(WebCore::soupErrorShouldCauseLoadFailure): No longer needed;
SoupRequestHTTP's logic is now aligned with WebKit's.
(WebCore::gotChunkCallback): remove gotChunkCallback, which
is no longer needed
(WebCore::startHTTPRequest):
(WebCore::sendRequestCallback):
(WebCore::gotHeadersCallback):
(WebCore::contentSniffedCallback): merge the code from
gotHeadersCallback and contentSniffedCallback into
sendRequestCallback
* platform/network/soup/ResourceResponseSoup.cpp:
(WebCore::ResourceResponse::updateFromSoupMessage): move the
sniffing override logic here

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] [EFL] Update some cookie-related tests
mrobinson@webkit.org [Sat, 3 Dec 2011 09:19:29 +0000 (09:19 +0000)]
[GTK] [EFL] Update some cookie-related tests
https://bugs.webkit.org/show_bug.cgi?id=73673

Patch by Dan Winship <danw@gnome.org> on 2011-12-03
Reviewed by Martin Robinson.

* platform/efl/http/tests/cookies/double-quoted-value-with-semi-colon-expected.txt:
* platform/gtk/http/tests/cookies/double-quoted-value-with-semi-colon-expected.txt:
Copied from platform/chromium because libsoup also follows the RFC
6265 behavior.

* platform/efl/Skipped:
* platform/gtk/Skipped: Unskip that, and another cookie test that
has been fixed for a while.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r101904.
ossy@webkit.org [Sat, 3 Dec 2011 09:06:15 +0000 (09:06 +0000)]
Unreviewed, rolling out r101904.
http://trac.webkit.org/changeset/101904
https://bugs.webkit.org/show_bug.cgi?id=73739

It broke zillions of tests on all bot (Requested by
Ossy_weekend on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-12-03

Source/WebCore:

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::handlePostMessage):

LayoutTests:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Refactoring] Use join(", ", @arguments) to build a method argument string in CodeGen...
haraken@chromium.org [Sat, 3 Dec 2011 07:22:10 +0000 (07:22 +0000)]
[Refactoring] Use join(", ", @arguments) to build a method argument string in CodeGenerator*.pm
https://bugs.webkit.org/show_bug.cgi?id=73651

Reviewed by Darin Adler.

The code in CodeGenerator*.pm to build a method argument string is really dirty
and error-prone. It is building an argument string incrementally judging whether
", " is necessary or not, like this:

    my $method = ... ? "func(" : "func(a";
    if (...) {
        $method .= $method =~ /\($/ ? "b" : ", b";
    }
    $method .= ")";

Alternatively, we can refactor the code as follows:

    my $funcName = "func";
    my @arguments;
    push(@arguments, "a") if (...);
    push(@arguments, "b") if (...);
    my $method = $funcName . "(" . join(", ", @arguments) . ")";

This patch just refactors the code, and generated .h and .cpp files should be
exactly the same as the current .h and .cpp files.

Tests: bindings/scripts/test/*

* bindings/scripts/CodeGenerator.pm:
(GetterExpression):
(SetterExpression):
* bindings/scripts/CodeGeneratorCPP.pm:
(GenerateImplementation):
* bindings/scripts/CodeGeneratorGObject.pm:
():
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateImplementation):
* bindings/scripts/CodeGeneratorObjC.pm:
(GenerateImplementation):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GenerateNormalAttrSetter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRename WTF class from TemporarilyChange to TemporaryChange.
levin@chromium.org [Sat, 3 Dec 2011 07:06:54 +0000 (07:06 +0000)]
Rename WTF class from TemporarilyChange to TemporaryChange.
https://bugs.webkit.org/show_bug.cgi?id=73479

Reviewed by Eric Seidel.

Source/JavaScriptCore:

* JavaScriptCore.gypi:
* JavaScriptCore.vcproj/WTF/WTF.vcproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* wtf/TemporaryChange.h: Renamed from Source/JavaScriptCore/wtf/TemporarilyChange.h.
(WTF::TemporaryChange::TemporaryChange):
(WTF::TemporaryChange::~TemporaryChange):

Source/JavaScriptGlue:

* ForwardingHeaders/wtf/TemporarilyChange.h: Removed.
* ForwardingHeaders/wtf/TemporaryChange.h: Added.

Source/WebCore:

* ForwardingHeaders/wtf/TemporarilyChange.h: Removed.
* ForwardingHeaders/wtf/TemporaryChange.h: Added.
* page/FrameView.cpp:
(WebCore::FrameView::forceLayoutParentViewIfNeeded):
(WebCore::FrameView::layout):
(WebCore::FrameView::setScrollPosition):
(WebCore::FrameView::autoSizeIfEnabled):

Source/WebKit/mac:

* ForwardingHeaders/wtf/TemporarilyChange.h: Removed.
* ForwardingHeaders/wtf/TemporaryChange.h: Added.

Tools:

* DumpRenderTree/ForwardingHeaders/wtf/TemporarilyChange.h: Removed.
* DumpRenderTree/ForwardingHeaders/wtf/TemporaryChange.h: Added.
* TestWebKitAPI/TestWebKitAPI.gypi:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WTF/TemporaryChange.cpp: Renamed from Tools/TestWebKitAPI/Tests/WTF/TemporarilyChange.cpp.
(TestWebKitAPI::TEST):
* TestWebKitAPI/win/TestWebKitAPI.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101913 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r101906.
scheib@chromium.org [Sat, 3 Dec 2011 06:56:09 +0000 (06:56 +0000)]
Unreviewed, rolling out r101906.
http://trac.webkit.org/changeset/101906
https://bugs.webkit.org/show_bug.cgi?id=73736

Failing chromium webkit_unit_tests (Requested by scheib on
#webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-12-02

* WebKit.gypi:
* tests/DispatchEventTest.cpp: Removed.
* tests/data/event_target.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoASSERT fails in updateState ACTION_DRAW case
commit-queue@webkit.org [Sat, 3 Dec 2011 06:47:35 +0000 (06:47 +0000)]
ASSERT fails in updateState ACTION_DRAW case
https://bugs.webkit.org/show_bug.cgi?id=73351

Patch by Armand Navabi <armand.navabi@gmail.com> on 2011-12-02
Reviewed by James Robinson.

Changed ASSERT in CCSchedulerStateMachine to include ( || !m_visible) as discussed in bug
report. m_commitState is set to COMMIT_STATE_WAITING_FOR_FIRST_DRAW if m_needsCommit or
!m_visible, so in ACTION_DRAW the assert should have both conditions.

* platform/graphics/chromium/cc/CCSchedulerStateMachine.cpp:
(WebCore::CCSchedulerStateMachine::updateState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION (r99754): All layout tests crash on Windows
mhahnenberg@apple.com [Sat, 3 Dec 2011 06:31:50 +0000 (06:31 +0000)]
REGRESSION (r99754): All layout tests crash on Windows
https://bugs.webkit.org/show_bug.cgi?id=72305

Reviewed by Geoffrey Garen.

Fixes a crash in release builds on Windows.  Windows was optimizing the out-of-line virtual destructor in
JSFunction away, which left it with no virtual functions.  Its vtable ptr was then identical to that of
a different class, therefore the optimization in the visitChildren helper function in MarkedStack.cpp was calling an
incorrect version of visitChildren on the object, which left its children unmarked, causing them to be
collected when they were still reachable.

* runtime/JSFunction.cpp:
(JSC::JSFunction::vtableAnchor): Add a virtual function to JSFunction that Visual Studio can't optimize away.
* runtime/JSFunction.h:
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::storeVPtrs): Add checks to make sure that all virtual pointers that we rely on for optimization
purposes are distinct from one another.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoremove a stale comment in flexbox code
tony@chromium.org [Sat, 3 Dec 2011 06:21:48 +0000 (06:21 +0000)]
remove a stale comment in flexbox code
https://bugs.webkit.org/show_bug.cgi?id=73720

Reviewed by Ojan Vafai.

This was fixed by Ojan in https://bugs.webkit.org/show_bug.cgi?id=70754.

* css3/flexbox/flex-flow-padding.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Scissor rect optimization for chromium compositor
commit-queue@webkit.org [Sat, 3 Dec 2011 06:14:08 +0000 (06:14 +0000)]
[chromium] Scissor rect optimization for chromium compositor
https://bugs.webkit.org/show_bug.cgi?id=67341

Patch by Shawn Singh <shawnsingh@chromium.org> on 2011-12-02
Reviewed by James Robinson.

Mostly covered by damage tracker tests. Currently this relies on
manually running layout tests, because this patch requires partial
swaps that are not supported by DumpRenderTree. The feature
automatically disables if partial swap is not supported.

* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::initialize):
(WebCore::LayerRendererChromium::trackDamageForAllSurfaces):
(WebCore::LayerRendererChromium::drawLayersOntoRenderSurfaces):
(WebCore::LayerRendererChromium::drawLayersInternal):
(WebCore::LayerRendererChromium::swapBuffers):
(WebCore::LayerRendererChromium::drawLayer):
* platform/graphics/chromium/LayerRendererChromium.h:
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings):
(WebCore::LayerRendererCapabilities::LayerRendererCapabilities):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::draw):
* platform/graphics/chromium/cc/CCRenderSurface.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[NRWT] reftest asserts intermittently on the Qt-WK2 bot
tony@chromium.org [Sat, 3 Dec 2011 06:06:56 +0000 (06:06 +0000)]
[NRWT] reftest asserts intermittently on the Qt-WK2 bot
https://bugs.webkit.org/show_bug.cgi?id=73453

Reviewed by Ryosuke Niwa.

We used to assert if a ref test didn't generate pixel results. Instead, just claim it is a failing test.
The verbose output will include the failure reason.

* Scripts/webkitpy/layout_tests/controllers/single_test_runner.py: Don't assert, return FailureReftestNoImagesGenerated instead.
* Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
(write_test_result):
* Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py: Remove stray letter t.
* Scripts/webkitpy/layout_tests/models/test_failures.py: Add a new failure type: FailureReftestNoImagesGenerated
(determine_result_type):
(FailureReftestNoImagesGenerated.__init__):
(FailureReftestNoImagesGenerated):
(FailureReftestNoImagesGenerated.message):
* Scripts/webkitpy/layout_tests/port/test.py: Add a test case for integration tests.
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: Update test counts.
(MainTest.test_run_singly_actually_runs_tests):
(MainTest.test_unexpected_failures):
(EndToEndTest.test_end_to_end):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd a unit test for creating, listening for, and dispatching events
commit-queue@webkit.org [Sat, 3 Dec 2011 05:56:31 +0000 (05:56 +0000)]
Add a unit test for creating, listening for, and dispatching events
via the chromium API.
https://bugs.webkit.org/show_bug.cgi?id=72988

Patch by Dave Michael <dmichael@chromium.org> on 2011-12-02
Reviewed by Darin Fisher.

* WebKit.gypi:
* tests/DispatchEventTest.cpp: Added.
(WebKit::MockListener::MockListener):
(WebKit::MockListener::~MockListener):
(WebKit::MockListener::events):
(WebKit::DispatchEventTest::DispatchEventTest):
(WebKit::DispatchEventTest::TearDown):
(WebKit::TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoDon't call FocusController:setActive() in
japhet@chromium.org [Sat, 3 Dec 2011 05:51:02 +0000 (05:51 +0000)]
Don't call FocusController:setActive() in
WebViewImpl::setFocus(), since being active and
being focused are different concepts and chromium
now properly sends setActive IPCs on all platforms.
https://bugs.webkit.org/show_bug.cgi?id=73687

Reviewed by Eric Seidel.

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::setFocus):
* tests/WebViewTest.cpp: Test how different combinations of
    setIsActive() and setFocus() affect whether the document
    has focus.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agohttps://bugs.webkit.org/show_bug.cgi?id=73691
dslomov@google.com [Sat, 3 Dec 2011 05:40:46 +0000 (05:40 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=73691
[JSC] Implement correct order of window.postMessage arguments.

Reviewed by Geoffrey Garen.

Source/WebCore:

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::handlePostMessage):

LayoutTests:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101904 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION (r91125): Polyline tool in google docs is broken
commit-queue@webkit.org [Sat, 3 Dec 2011 05:34:42 +0000 (05:34 +0000)]
REGRESSION (r91125): Polyline tool in google docs is broken
https://bugs.webkit.org/show_bug.cgi?id=65796

Source/WebCore:

Patch by Stephen Chenney <schenney@chromium.org> on 2011-12-02
Reviewed by Darin Adler.

Work around a bug in CoreGraphics, that caused incorrect bounds for paths
consisting only of move-to elements. This causes problems in SVG, when the enormous
bounds prevented the drawing of things behind.

Tests: svg/custom/path-moveto-only-rendering.svg
       svg/custom/subpaths-moveto-only-rendering.svg

* platform/graphics/cg/PathCG.cpp:
(WebCore::PathIsEmptyOrSingleMoveTester::PathIsEmptyOrSingleMoveTester): Class to
test for isEmpty accoridng ot the same rules as other platforms.
(WebCore::PathIsEmptyOrSingleMoveTester::isEmpty): Query the result
(WebCore::PathIsEmptyOrSingleMoveTester::testPathElement): Path iterator method
(WebCore::PathHasOnlyMoveToTester::PathHasOnlyMoveToTester): Class to test whether a
path contains only move-to elements, and hence should have null bounds.
(WebCore::PathHasOnlyMoveToTester::hasOnlyMoveTo): Query the result
(WebCore::PathHasOnlyMoveToTester::testPathElement): Path iterator method.
(WebCore::Path::boundingRect): Modified to check for move-to only paths
(WebCore::Path::fastBoundingRect): Modified to check for move-to only paths
(WebCore::Path::isEmpty): Now uses the method that matches other platforms.
(WebCore::Path::hasCurrentPoint): Now uses CGPathIsEmpty directly
(WebCore::Path::transform) : Now uses CGPathIsEmpty directly

LayoutTests:

Work around a bug in CoreGraphics, that caused incorrect bounds for paths
consisting only of move-to elements. This causes problems in SVG, when the enormous
bounds prevented the drawing of things behind.

Will revert expectation file when expectations are stable.

Patch by Stephen Chenney <schenney@chromium.org> on 2011-12-02
Reviewed by Darin Adler.

* platform/chromium-mac/svg/custom/zero-path-square-cap-rendering2-expected.txt: Removed.
* platform/chromium-win/svg/custom/zero-path-square-cap-rendering2-expected.txt: Removed.
* platform/mac/svg/custom/path-moveto-only-rendering-expected.png: Added.
* platform/mac/svg/custom/path-moveto-only-rendering-expected.txt: Added.
* platform/mac/svg/custom/subpaths-moveto-only-rendering-expected.png: Added.
* platform/mac/svg/custom/subpaths-moveto-only-rendering-expected.txt: Added.
* svg/custom/path-moveto-only-rendering.svg: Added.
* svg/custom/subpaths-moveto-only-rendering.svg: Added.
* svg/custom/zero-path-square-cap-rendering2-expected.txt: Modified text output
* platform/chromium/test_expectations.txt: Added flakiness for new tests from this patch

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101903 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] [WK2] MiniBrowser --touch asserts in debug build
commit-queue@webkit.org [Sat, 3 Dec 2011 05:26:52 +0000 (05:26 +0000)]
[Qt] [WK2] MiniBrowser --touch asserts in debug build
https://bugs.webkit.org/show_bug.cgi?id=73697

Patch by Viatcheslav Ostapenko <ostapenko.viatcheslav@nokia.com> on 2011-12-02
Reviewed by Eric Seidel.

Add missing guard to QtViewportInteractionEngine::itemSizeChanged().

* UIProcess/qt/QtViewportInteractionEngine.cpp:
(WebKit::QtViewportInteractionEngine::itemSizeChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoChangeLog should be able to parse annotated changelogs
rniwa@webkit.org [Sat, 3 Dec 2011 04:59:22 +0000 (04:59 +0000)]
ChangeLog should be able to parse annotated changelogs
https://bugs.webkit.org/show_bug.cgi?id=72703

Reviewed by Eric Seidel.

Make parse_entries_from_file aware of svn blame annotation such as "100000 ossy@webkit.org ",
and extract the revision number for each entry.

Also add a support to extract "svn blame" result from svn/git checkout to scm classes.

* Scripts/webkitpy/common/checkout/changelog.py:
* Scripts/webkitpy/common/checkout/changelog_unittest.py:
* Scripts/webkitpy/common/checkout/scm/git.py:
* Scripts/webkitpy/common/checkout/scm/scm.py:
* Scripts/webkitpy/common/checkout/scm/svn.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] test_expectations typo fix for chunked-progress-event-expectedLength.html
scheib@chromium.org [Sat, 3 Dec 2011 04:54:22 +0000 (04:54 +0000)]
[Chromium] test_expectations typo fix for chunked-progress-event-expectedLength.html

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CSSRegions]More renaming for region style rules.
mihnea@adobe.com [Sat, 3 Dec 2011 04:46:07 +0000 (04:46 +0000)]
[CSSRegions]More renaming for region style rules.
https://bugs.webkit.org/show_bug.cgi?id=73526

Reviewed by Eric Seidel.

Source/WebCore:

Rename isRegionStyleRule() -> isRegionRule()
WEBKIT_REGION_STYLE_RULE -> WEBKIT_REGION_RULE.

* bindings/objc/DOMCSS.mm:
(kitClass):
* bindings/v8/custom/V8CSSRuleCustom.cpp:
(WebCore::toV8):
* css/CSSGrammar.y:
* css/CSSParser.cpp:
(WebCore::CSSParser::createRegionRule):
* css/CSSParser.h:
* css/CSSRule.cpp:
(WebCore::CSSRule::cssText):
(WebCore::CSSRule::destroy):
* css/CSSRule.h:
(WebCore::CSSRule::isRegionRule):
* css/CSSRule.idl:
* css/CSSStyleSelector.cpp:
(WebCore::CSSStyleSelector::addRegionRule):
(WebCore::RuleSet::addRulesFromSheet):
* css/CSSStyleSelector.h:
* css/WebKitCSSRegionRule.cpp:
(WebCore::WebKitCSSRegionRule::WebKitCSSRegionRule):
* css/tokenizer.flex:

LayoutTests:

Rename WEBKIT_REGION_STYLE_RULE -> WEBKIT_REGION_RULE.

* fast/dom/Window/window-properties-expected.txt:
* fast/regions/parsing-region-style-rule.html:
* platform/gtk/fast/dom/Window/window-properties-expected.txt:
* platform/mac/fast/dom/Window/window-properties-expected.txt:
* platform/qt-wk2/fast/dom/Window/window-properties-expected.txt:
* platform/qt/fast/dom/Window/window-properties-expected.txt:
* platform/win/fast/dom/Window/window-properties-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoXSSAuditor includes more terminating characters when truncating
tsepez@chromium.org [Sat, 3 Dec 2011 04:06:42 +0000 (04:06 +0000)]
XSSAuditor includes more terminating characters when truncating
attribute values.
https://bugs.webkit.org/show_bug.cgi?id=73684

Reviewed by Daniel Bates.

Source/WebCore:

* html/parser/XSSAuditor.cpp:
(WebCore::isTerminatingCharacter):
(WebCore::XSSAuditor::eraseDangerousAttributesIfInjected):

LayoutTests:

* http/tests/security/xssAuditor/property-escape-comment-expected.txt:
* http/tests/security/xssAuditor/property-escape-comment.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoNeed to implement flex-flow: column-reverse
tony@chromium.org [Sat, 3 Dec 2011 03:50:24 +0000 (03:50 +0000)]
Need to implement flex-flow: column-reverse
https://bugs.webkit.org/show_bug.cgi?id=73504

Reviewed by David Hyatt.

Source/WebCore:

We can't just change the direction of the FlexOrderIterator because we want the overflow to be
on the top. We can't just position them in the reverse location since we don't know where the
bottom edge is until we've layed out the flex items. So we do an extra pass, but it should be
fast since we're not laying out, just moving.

* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::flowAwareBorderEnd):
(WebCore::RenderFlexibleBox::flowAwarePaddingEnd):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren):
(WebCore::RenderFlexibleBox::layoutColumnReverse): Do an extra pass to reposition flexitems in the reverse order.
* rendering/RenderFlexibleBox.h:

LayoutTests:

Enable testing of column-reverse.

* css3/flexbox/flex-flow-expected.txt:
* css3/flexbox/flex-flow.html: Added test cases to make sure start and end still apply to the text flow direction.
* css3/flexbox/flex-flow-border-expected.txt:
* css3/flexbox/flex-flow-border.html: Single flexitem, so reverse does nothing.
* css3/flexbox/flex-flow-margins-auto-size-expected.txt:
* css3/flexbox/flex-flow-margins-auto-size.html: Ditto.  The vertical tests cases (orthogonal directions) have always failed.
* css3/flexbox/flex-flow-margins-expected.txt:
* css3/flexbox/flex-flow-margins.html: Ditto.
* css3/flexbox/flex-flow-orientations-expected.txt:
* css3/flexbox/flex-flow-orientations.html: This test has 2 flexitems, the positions of which are now flipped.
* css3/flexbox/flex-flow-overflow-expected.txt:
* css3/flexbox/flex-flow-overflow.html: The overflow should go in the opposite direction vertically, but the same direction horizontally.
* css3/flexbox/flex-flow-padding-expected.txt:
* css3/flexbox/flex-flow-padding.html: Single flexitem, no change.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd WebArrayBuffer to chromium API
commit-queue@webkit.org [Sat, 3 Dec 2011 03:13:43 +0000 (03:13 +0000)]
Add WebArrayBuffer to chromium API
https://bugs.webkit.org/show_bug.cgi?id=73593

Patch by Dave Michael <dmichael@chromium.org> on 2011-12-02
Reviewed by Darin Fisher.

* WebKit.gyp:
* public/WebArrayBuffer.h: Added.
(WebKit::WebArrayBuffer::~WebArrayBuffer):
(WebKit::WebArrayBuffer::WebArrayBuffer):
(WebKit::WebArrayBuffer::operator=):
(WebKit::WebArrayBuffer::isNull):
* public/WebBindings.h:
* src/WebArrayBuffer.cpp: Added.
(WebKit::WebArrayBuffer::create):
(WebKit::WebArrayBuffer::reset):
(WebKit::WebArrayBuffer::assign):
(WebKit::WebArrayBuffer::data):
(WebKit::WebArrayBuffer::byteLength):
(WebKit::WebArrayBuffer::toV8Value):
(WebKit::WebArrayBuffer::WebArrayBuffer):
(WebKit::WebArrayBuffer::operator=):
(WebKit::WebArrayBuffer::operator WTF::PassRefPtr<WTF::ArrayBuffer>):
* src/WebBindings.cpp:
(WebKit::getArrayBufferImpl):
(WebKit::WebBindings::getArrayBuffer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoDivide by zero for zero-length arcs
commit-queue@webkit.org [Sat, 3 Dec 2011 03:12:34 +0000 (03:12 +0000)]
Divide by zero for zero-length arcs
https://bugs.webkit.org/show_bug.cgi?id=73021

Source/WebCore:

Patch by Stephen Chenney <schenney@chromium.org> on 2011-12-02
Reviewed by Nikolas Zimmermann.

A zero-length path produces divide by zero, resulting in nothing being
drawn. This change modifies the behavior to produce a zero-length line
which will generate correct linecaps.

Test: svg/stroke/zero-length-arc-linecaps-rendering.svg

* svg/SVGPathParser.cpp:
(WebCore::SVGPathParser::parseArcToSegment): Catch the case of
zero-length arcs and convert them to lines (the same as would happen
for zero arc radii).

LayoutTests:

A zero-length path produces divide by zero, resulting in nothing being
drawn. This change modifies the behavior to produce a zero-length line
which will generate correct linecaps.

Patch by Stephen Chenney <schenney@chromium.org> on 2011-12-02
Reviewed by Nikolas Zimmermann.

* platform/chromium/test_expectations.txt: Added flakiness for the new test until wehave expectations.
* platform/mac/svg/stroke: Added.
* platform/mac/svg/stroke/zero-length-arc-linecaps-rendering-expected.png: Added.
* platform/mac/svg/stroke/zero-length-arc-linecaps-rendering-expected.txt: Added.
* svg/stroke/zero-length-arc-linecaps-rendering.svg: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago-Implement the SSE optimization for vsmul and vadd.
commit-queue@webkit.org [Sat, 3 Dec 2011 02:43:43 +0000 (02:43 +0000)]
-Implement the SSE optimization for vsmul and vadd.
https://bugs.webkit.org/show_bug.cgi?id=73182

Patch by James Wei <james.wei@intel.com> & Xingnan Wang <xingnan.wang@intel.com> on 2011-12-02
Reviewed by Kenneth Russell.

* platform/audio/VectorMath.cpp:
(WebCore:VectorMath):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoGrant workers experimental access to IndexedDB.
dgrogan@chromium.org [Sat, 3 Dec 2011 02:20:10 +0000 (02:20 +0000)]
Grant workers experimental access to IndexedDB.
https://bugs.webkit.org/show_bug.cgi?id=73609

Reviewed by Nate Chapin.

Source/WebCore:

No new tests - there will be chromium ui tests that depend on
webkit.org/b/73297.

* storage/IDBFactory.cpp:
(WebCore::IDBFactory::open): Call new function,
IDBFactoryBackendInterface::openFromWorker.
* storage/IDBFactoryBackendImpl.cpp:
(WebCore::IDBFactoryBackendImpl::open):
(WebCore::IDBFactoryBackendImpl::openFromWorker):
(WebCore::IDBFactoryBackendImpl::openInternal):
* storage/IDBFactoryBackendImpl.h:
* storage/IDBFactoryBackendInterface.h:
* workers/WorkerContext.cpp:
(WebCore::WorkerContext::webkitIndexedDB): Stores
IDBFactoryBackendInterface, implemented by IDBFactoryBackendProxy in
chromium, in the WorkerContext.  For the Document case it is stored in
the PageGroup.  Storing it in the WorkerContext causes more memory
churn, but that should be trivial.  I don't know of any better
alternatives.
* workers/WorkerContext.h:
* workers/WorkerContext.idl:

Source/WebKit/chromium:

* src/IDBFactoryBackendProxy.cpp:
(WebKit::IDBFactoryBackendProxy::allowIDBFromWorkerThread): Return
true while behind a runtime flag, and while we figure out how to check
permission in a thread-safe manner.
(WebKit::IDBFactoryBackendProxy::openFromWorker): Using the webFrame
from the worker seems to satisfy the weak condition in chromium that
neither it nor it's associated RenderView is NULL.
* src/IDBFactoryBackendProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoVersioning.
lforschler@apple.com [Sat, 3 Dec 2011 02:12:22 +0000 (02:12 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Remove entry pointer before freeing it from matrix.
commit-queue@webkit.org [Sat, 3 Dec 2011 01:59:48 +0000 (01:59 +0000)]
[EFL] Remove entry pointer before freeing it from matrix.
https://bugs.webkit.org/show_bug.cgi?id=73625

Patch by JungJik Lee <jungjik.lee@samsung.com> on 2011-12-02
Reviewed by Ryosuke Niwa.

To prevent leaving dangling pointer on the matrix, remove the entry before freeing it.

* ewk/ewk_tiled_matrix.cpp:
(ewk_tile_matrix_free):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoImprove float array support in the DFG JIT
oliver@apple.com [Sat, 3 Dec 2011 01:47:27 +0000 (01:47 +0000)]
Improve float array support in the DFG JIT
https://bugs.webkit.org/show_bug.cgi?id=73722

Reviewed by Gavin Barraclough.

Add basic support for float typed arrays in JSC.  This is currently
less optimal than it could be in the following ways:
 * float32Array1[0] = float32Array2[0] (eg. an element by element copy)
   promotes float to double and then back to float.
 * float64Array[0] will always perform NaN tests in order to prevent
   signalling NaNs from entering the engine.

We also don't support Float32Array on ARMv7

* assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::loadDouble):
(JSC::MacroAssemblerARMv7::loadFloat):
(JSC::MacroAssemblerARMv7::storeDouble):
(JSC::MacroAssemblerARMv7::storeFloat):
(JSC::MacroAssemblerARMv7::convertFloatToDouble):
(JSC::MacroAssemblerARMv7::convertDoubleToFloat):
* assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::loadDouble):
(JSC::MacroAssemblerX86Common::loadFloat):
(JSC::MacroAssemblerX86Common::storeDouble):
(JSC::MacroAssemblerX86Common::storeFloat):
(JSC::MacroAssemblerX86Common::convertDoubleToFloat):
(JSC::MacroAssemblerX86Common::convertFloatToDouble):
* assembler/X86Assembler.h:
(JSC::X86Assembler::cvtsd2ss_rr):
(JSC::X86Assembler::cvtss2sd_rr):
(JSC::X86Assembler::movsd_rm):
(JSC::X86Assembler::movss_rm):
(JSC::X86Assembler::movsd_mr):
(JSC::X86Assembler::movss_mr):
* dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::execute):
* dfg/DFGNode.h:
(JSC::DFG::Node::shouldSpeculateFloat32Array):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray):
(JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray):
(JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray):
* dfg/DFGSpeculativeJIT.h:
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal):
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Move WebLayer APIs to platform directory
jamesr@google.com [Sat, 3 Dec 2011 01:42:23 +0000 (01:42 +0000)]
[chromium] Move WebLayer APIs to platform directory
https://bugs.webkit.org/show_bug.cgi?id=73610

Reviewed by Darin Fisher.

* public/WebContentLayer.h:
* public/WebContentLayerClient.h:
* public/WebExternalTextureLayer.h:
* public/WebLayer.h:
* public/WebLayerClient.h:
* public/WebLayerTreeView.h:
* public/WebLayerTreeViewClient.h:
* public/platform/WebContentLayer.h: Copied from Source/WebKit/chromium/public/WebContentLayer.h.
(WebKit::WebContentLayer::WebContentLayer):
(WebKit::WebContentLayer::~WebContentLayer):
(WebKit::WebContentLayer::operator=):
* public/platform/WebContentLayerClient.h: Copied from Source/WebKit/chromium/public/WebContentLayerClient.h.
(WebKit::WebContentLayerClient::~WebContentLayerClient):
* public/platform/WebExternalTextureLayer.h: Copied from Source/WebKit/chromium/public/WebExternalTextureLayer.h.
(WebKit::WebExternalTextureLayer::WebExternalTextureLayer):
(WebKit::WebExternalTextureLayer::~WebExternalTextureLayer):
(WebKit::WebExternalTextureLayer::operator=):
* public/platform/WebLayer.h: Copied from Source/WebKit/chromium/public/WebLayer.h.
(WebKit::WebLayer::WebLayer):
(WebKit::WebLayer::~WebLayer):
(WebKit::WebLayer::operator=):
(WebKit::WebLayer::isNull):
(WebKit::WebLayer::to):
(WebKit::WebLayer::toConst):
(WebKit::WebLayer::unwrap):
(WebKit::WebLayer::constUnwrap):
(WebKit::operator==):
(WebKit::operator!=):
* public/platform/WebLayerClient.h: Copied from Source/WebKit/chromium/public/WebLayerClient.h.
(WebKit::WebLayerClient::~WebLayerClient):
* public/platform/WebLayerTreeView.h: Copied from Source/WebKit/chromium/public/WebLayerTreeView.h.
(WebKit::WebLayerTreeView::Settings::Settings):
(WebKit::WebLayerTreeView::WebLayerTreeView):
(WebKit::WebLayerTreeView::~WebLayerTreeView):
(WebKit::WebLayerTreeView::operator=):
(WebKit::operator==):
(WebKit::operator!=):
* public/platform/WebLayerTreeViewClient.h: Copied from Source/WebKit/chromium/public/WebLayerTreeViewClient.h.
(WebKit::WebLayerTreeViewClient::~WebLayerTreeViewClient):
* public/platform/WebPrivatePtr.h:
* src/WebContentLayer.cpp:
* src/WebContentLayerImpl.cpp:
* src/WebExternalTextureLayer.cpp:
* src/WebExternalTextureLayerImpl.cpp:
* src/WebLayer.cpp:
* src/WebLayerImpl.cpp:
* src/WebLayerTreeView.cpp:
* src/WebLayerTreeViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSource/WebCore: Fix mixed content handling for video in Chromium by having
commit-queue@webkit.org [Sat, 3 Dec 2011 01:38:26 +0000 (01:38 +0000)]
Source/WebCore: Fix mixed content handling for video in Chromium by having
CachedResourceLoader & SubresourceLoader use the
ResourceRequest::TargetType when determining if a RawResource can
be requested.
https://bugs.webkit.org/show_bug.cgi?id=72178

Patch by Aaron Colwell <acolwell@chromium.org> on 2011-12-02
Reviewed by Adam Barth.

Tests: http/tests/security/mixedContent/insecure-video-in-iframe.html
       http/tests/security/mixedContent/insecure-video-in-main-frame.html
       http/tests/security/mixedContent/redirect-http-to-https-video-in-main-frame.html
       http/tests/security/mixedContent/redirect-https-to-http-video-in-main-frame.html

* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::willSendRequest):
* loader/cache/CachedRawResource.cpp:
(WebCore::CachedRawResource::CachedRawResource):
* loader/cache/CachedRawResource.h:
* loader/cache/CachedResource.cpp:
(WebCore::defaultPriorityForResourceType):
(WebCore::cachedResourceTypeToTargetType):
(WebCore::CachedResource::targetTypeToCachedResourceType):
* loader/cache/CachedResource.h:
* loader/cache/CachedResourceLoader.cpp:
(WebCore::createResource):
(WebCore::CachedResourceLoader::checkInsecureContent):
(WebCore::CachedResourceLoader::canRequest):
(WebCore::CachedResourceLoader::requestResource):

LayoutTests: Adding mixed content tests for video tag.
https://bugs.webkit.org/show_bug.cgi?id=72178

Patch by Aaron Colwell <acolwell@chromium.org> on 2011-12-02
Reviewed by Adam Barth.

* http/tests/security/mixedContent/insecure-video-in-iframe-expected.txt: Added.
* http/tests/security/mixedContent/insecure-video-in-iframe.html: Added.
* http/tests/security/mixedContent/insecure-video-in-main-frame-expected.txt: Added.
* http/tests/security/mixedContent/insecure-video-in-main-frame.html: Added.
* http/tests/security/mixedContent/redirect-http-to-https-video-in-main-frame-expected.txt: Added.
* http/tests/security/mixedContent/redirect-http-to-https-video-in-main-frame.html: Added.
* http/tests/security/mixedContent/redirect-https-to-http-video-in-main-frame-expected.txt: Added.
* http/tests/security/mixedContent/redirect-https-to-http-video-in-main-frame.html: Added.
* http/tests/security/mixedContent/resources/frame-with-insecure-video.html: Added.
* http/tests/security/mixedContent/resources/frame-with-redirect-http-to-https-video.html: Added.
* http/tests/security/mixedContent/resources/frame-with-redirect-https-to-http-video.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Rebaselining svg/custom/svg-fonts-word-spacing.html
scheib@chromium.org [Sat, 3 Dec 2011 01:35:56 +0000 (01:35 +0000)]
[Chromium] Rebaselining svg/custom/svg-fonts-word-spacing.html

* platform/chromium-cg-mac-snowleopard/accessibility/aria-checkbox-sends-notification-expected.txt: Copied from LayoutTests/platform/chromium/accessibility/aria-checkbox-sends-notification-expected.txt.
* platform/chromium-mac-snowleopard/accessibility/aria-checkbox-sends-notification-expected.txt: Copied from LayoutTests/platform/chromium/accessibility/aria-checkbox-sends-notification-expected.txt.
* platform/chromium-win/accessibility/aria-checkbox-sends-notification-expected.txt: Renamed from LayoutTests/platform/chromium/accessibility/aria-checkbox-sends-notification-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Marking svg/W3C-SVG-1.1/fonts-elem-04-b.svg flaky for image+text
scheib@chromium.org [Sat, 3 Dec 2011 01:28:19 +0000 (01:28 +0000)]
[Chromium] Marking svg/W3C-SVG-1.1/fonts-elem-04-b.svg flaky for image+text

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoEnable parallel testing for Win and Gtk bots
eric@webkit.org [Sat, 3 Dec 2011 01:04:36 +0000 (01:04 +0000)]
Enable parallel testing for Win and Gtk bots
after confirming with Lucas and Martin the bots should be OK.

Unreviewed. (I ran the idea by Lucas and Martin.)

* Scripts/run-webkit-tests:
(platformIsReadyForParallelTesting):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] add setOpener method to WebFrame
commit-queue@webkit.org [Sat, 3 Dec 2011 00:53:47 +0000 (00:53 +0000)]
[chromium] add setOpener method to WebFrame
https://bugs.webkit.org/show_bug.cgi?id=73379

Patch by Karl Koscher <supersat@chromium.org> on 2011-12-02
Reviewed by Darin Fisher.

* public/WebFrame.h:
(WebKit::WebFrame::clearOpener): now an inline call to setOpener(0)
* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::setOpener): sets window.opener
* src/WebFrameImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r101731.
dcheng@chromium.org [Sat, 3 Dec 2011 00:13:51 +0000 (00:13 +0000)]
Unreviewed, rolling out r101731.
http://trac.webkit.org/changeset/101731
https://bugs.webkit.org/show_bug.cgi?id=73706

Broke copy and paste in chromium. (Requested by dcheng on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-12-02

Source/WebCore:

* editing/SpellChecker.cpp:
(WebCore::SpellChecker::SpellChecker):
(WebCore::SpellChecker::initRequest):
(WebCore::SpellChecker::clearRequest):
(WebCore::SpellChecker::canCheckAsynchronously):
(WebCore::SpellChecker::isBusy):
(WebCore::SpellChecker::isValid):
(WebCore::SpellChecker::requestCheckingFor):
(WebCore::SpellChecker::doRequestCheckingFor):
(WebCore::SpellChecker::didCheck):
* editing/SpellChecker.h:

LayoutTests:

* editing/spelling/spellcheck-queue-expected.txt: Removed.
* editing/spelling/spellcheck-queue.html: Removed.
* platform/gtk/Skipped:
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWhen page scaling is in use position:fixed has incorrect results
fsamuel@chromium.org [Sat, 3 Dec 2011 00:04:46 +0000 (00:04 +0000)]
When page scaling is in use position:fixed has incorrect results
https://bugs.webkit.org/show_bug.cgi?id=68617

Reviewed by Simon Fraser.

.:

* Source/autotools/symbols.filter:

Source/WebCore:

Add the option for position:fixed elements to be fixed to the frame
instead of the layout rectangle of the document.

Tests: fast/repaint/fixed-in-page-scale.html
       fast/repaint/fixed-right-bottom-in-page-scale.html
       fast/repaint/fixed-right-in-page-scale.html

* WebCore.exp.in:
* page/FrameView.cpp:
(WebCore::FrameView::reset):
(WebCore::FrameView::scrollXForFixedPosition):
(WebCore::FrameView::scrollYForFixedPosition):
  If position:fixed elements are relative to the frame, disregard the
  drag factor.
(WebCore::FrameView::setShouldLayoutFixedElementsRelativeToFrame):
* page/FrameView.h:
(WebCore::FrameView::shouldLayoutFixedElementsRelativeToFrame):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::shouldLayoutFixedElementRelativeToFrame):
(WebCore::RenderBox::containingBlockLogicalWidthForPositioned):
(WebCore::RenderBox::containingBlockLogicalHeightForPositioned):
  If position:fixed elements are relative to the frame, their container
  is the frame instead of the layout rect of the document.
  This allows proper positioning of these elements to the right and
  bottom.
* rendering/RenderBox.h:
* testing/Internals.cpp:
(WebCore::Internals::setShouldLayoutFixedElementsRelativeToFrame):
* testing/Internals.h:
* testing/Internals.idl:
  Allow enabling and disabling the new behavior in layout tests.

Source/WebKit2:

* win/WebKit2.def:
* win/WebKit2CFLite.def:

LayoutTests:

* fast/repaint/fixed-in-page-scale-expected.png: Added.
* fast/repaint/fixed-in-page-scale-expected.txt: Added.
* fast/repaint/fixed-in-page-scale.html: Added.
* fast/repaint/fixed-right-bottom-in-page-scale-expected.png: Added.
* fast/repaint/fixed-right-bottom-in-page-scale-expected.txt: Added.
* fast/repaint/fixed-right-bottom-in-page-scale.html: Added.
* fast/repaint/fixed-right-in-page-scale-expected.png: Added.
* fast/repaint/fixed-right-in-page-scale-expected.txt: Added.
* fast/repaint/fixed-right-in-page-scale.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoBuild fix for SubresourceLoader when building with Core Foundation
benjamin@webkit.org [Fri, 2 Dec 2011 23:55:21 +0000 (23:55 +0000)]
Build fix for SubresourceLoader when building with Core Foundation
https://bugs.webkit.org/show_bug.cgi?id=73709

Patch by Benjamin Poulain <bpoulain@apple.com> on 2011-12-02
Reviewed by David Kilzer.

The patch r100311 removed SubresourceClient and merged it in SubresourceLoader.
Consequently, m_client does not exist anymore and there is no need to do the check
before invoking didReceiveData().

* loader/cf/SubresourceLoaderCF.cpp:
(WebCore::SubresourceLoader::didReceiveDataArray):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago<rdar://problem/10423627> Layers are flipped in WebKit1 views
simon.fraser@apple.com [Fri, 2 Dec 2011 23:54:43 +0000 (23:54 +0000)]
<rdar://problem/10423627> Layers are flipped in WebKit1 views

Reviewed by Sam Weinig.

Host the compositing layers in an NSView which is flipped, which helps AppKit
get the geometry right in some situations.

* WebView/WebHTMLView.mm:
(-[WebLayerHostingFlippedView isFlipped]):
(-[WebHTMLView attachRootLayer:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFixed a typo
mitz@apple.com [Fri, 2 Dec 2011 23:48:50 +0000 (23:48 +0000)]
Fixed a typo

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101871 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoStyledElement: Simplify addCSSColor().
kling@webkit.org [Fri, 2 Dec 2011 23:48:34 +0000 (23:48 +0000)]
StyledElement: Simplify addCSSColor().
<http://webkit.org/b/73703>

Reviewed by Darin Adler.

The Color(const String&) constructor handles both named and 3/6-digit
hex colors, so there's no need to handle those separately here.
Also tweaked some comments and minor things.

* dom/StyledElement.cpp:
(WebCore::StyledElement::addCSSColor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101870 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTry to fix the GTK+ debug bot. As discussed with other GTK+
mrobinson@webkit.org [Fri, 2 Dec 2011 23:45:39 +0000 (23:45 +0000)]
Try to fix the GTK+ debug bot. As discussed with other GTK+
hackers recently. It makes sense to just remove G_DISABLE_DEPRECATIONS
now in WebKit. The flag itself is now deprecated in GLib in favor of
the gcc attribute.

* GNUmakefile.am: Remove G_DISABLE_DEPRECATED.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101869 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Marking svg/W3C-SVG-1.1/fonts-elem-04-b.svg flaky for text
scheib@chromium.org [Fri, 2 Dec 2011 23:38:23 +0000 (23:38 +0000)]
[Chromium] Marking svg/W3C-SVG-1.1/fonts-elem-04-b.svg flaky for text

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101868 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUpdate platform/iphone to platform/ios
benjamin@webkit.org [Fri, 2 Dec 2011 23:36:06 +0000 (23:36 +0000)]
Update platform/iphone to platform/ios
https://bugs.webkit.org/show_bug.cgi?id=73708

Patch by Benjamin Poulain <bpoulain@apple.com> on 2011-12-02
Reviewed by Darin Adler.

The platform is now best known as iOS, update the platform layer accordingly.

* Configurations/WebCore.xcconfig:
* WebCore.gypi:
* WebCore.xcodeproj/project.pbxproj:
* platform/cocoa/KeyEventCocoa.mm:
* platform/ios/KeyEventCodesIOS.h: Renamed from Source/WebCore/platform/iphone/KeyEventCodesIPhone.h.
* platform/ios/KeyEventIOS.mm: Renamed from Source/WebCore/platform/iphone/KeyEventIPhone.mm.
(WebCore::keyIdentifierForKeyEvent):
(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
(WebCore::PlatformKeyboardEvent::disambiguateKeyDownEvent):
(WebCore::PlatformKeyboardEvent::currentCapsLockState):
(WebCore::PlatformKeyboardEvent::getCurrentModifierState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101867 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Adding test expectation that svg/animations/svginteger-animation-1.html...
scheib@chromium.org [Fri, 2 Dec 2011 23:31:35 +0000 (23:31 +0000)]
[Chromium] Adding test expectation that svg/animations/svginteger-animation-1.html crashes leopard cg

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101866 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Marking http/tests/appcache/video.html TIMEOUT
scheib@chromium.org [Fri, 2 Dec 2011 23:18:09 +0000 (23:18 +0000)]
[Chromium] Marking http/tests/appcache/video.html TIMEOUT

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101865 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Eliminate unnecessary state on previous CL
commit-queue@webkit.org [Fri, 2 Dec 2011 23:03:06 +0000 (23:03 +0000)]
[chromium] Eliminate unnecessary state on previous CL
https://bugs.webkit.org/show_bug.cgi?id=73661

Patch by Jonathan Backer <backer@chromium.org> on 2011-12-02
Reviewed by Kenneth Russell.

* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::initialize):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::initialize):
(WebCore::CCLayerTreeHost::didBecomeInvisibleOnImplThread):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings):
(WebCore::LayerRendererCapabilities::LayerRendererCapabilities):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101864 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Suppressing failures in test_expectations for 101844
scheib@chromium.org [Fri, 2 Dec 2011 23:02:58 +0000 (23:02 +0000)]
[Chromium] Suppressing failures in test_expectations for 101844

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101863 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTry to fix the GTK+ build.
mrobinson@webkit.org [Fri, 2 Dec 2011 22:59:58 +0000 (22:59 +0000)]
Try to fix the GTK+ build.

* gtk/jhbuild.modules: Properly specify the glib-networking
dependency for soup.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101862 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix a syntax error in the committers.py file.
mrobinson@webkit.org [Fri, 2 Dec 2011 22:58:21 +0000 (22:58 +0000)]
Fix a syntax error in the committers.py file.

* Scripts/webkitpy/common/config/committers.py: Add a missing comma.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101861 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago Update sandbox rules
ap@apple.com [Fri, 2 Dec 2011 22:56:23 +0000 (22:56 +0000)]
    Update sandbox rules
        https://bugs.webkit.org/show_bug.cgi?id=73675
        <rdar://problem/9276430>

        Reviewed by Sam Weinig.

        * WebProcess/com.apple.WebProcess.sb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101860 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd Alan Stearns and Peter Linss to the contributors list.
rniwa@webkit.org [Fri, 2 Dec 2011 22:49:58 +0000 (22:49 +0000)]
Add Alan Stearns and Peter Linss to the contributors list.

* Scripts/webkitpy/common/config/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101859 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago<rdar://problem/10520670> REGRESSION (r91738): didFinishLoad is called before custom...
mitz@apple.com [Fri, 2 Dec 2011 22:49:53 +0000 (22:49 +0000)]
<rdar://problem/10520670> REGRESSION (r91738): didFinishLoad is called before custom fonts have finished loading
https://bugs.webkit.org/show_bug.cgi?id=73688

Reviewed by Darin Adler.

Source/WebCore:

The problem was that after CSSFontFaceSource::getFontData() had scheduled a 0-delay timer to
begin loading the font, but before that timer fired, the subresource loader appeared to have
had no resources waiting to be loaded, and therefore didFinishLoad could be called. This change
reworks the fix for <http://webkit.org/b/65123> so that while the load is still started on a
0-dealy timer, the subresource loader’s request count is incremented immediately, preventing
it from hitting 0 while the font load is scheduled to begin. The delayed load mechanism is
moved from CSSFontFaceSource into CSSFontSelector in order to safely handle the possibility of
the latter being decommissioned while waiting for font loading to begin.

* css/CSSFontFaceSource.cpp:
(WebCore::CSSFontFaceSource::CSSFontFaceSource): Removed initializer for m_loadStartTimer.
(WebCore::CSSFontFaceSource::~CSSFontFaceSource): Removed stopping of m_loadStartTimer.
(WebCore::CSSFontFaceSource::getFontData): Replaced code to schedule loading on a timer with
a call to CSSFontSelector::beginLoadingFontSoon.
* css/CSSFontFaceSource.h: Removed m_loadStartTimer and m_fontSelector member variables.
* css/CSSFontSelector.cpp:
(WebCore::CSSFontSelector::CSSFontSelector): Added initialized for m_beginLoadingTimer.
(WebCore::CSSFontSelector::~CSSFontSelector): Added call to clearDocument(), to deal with
anything remaining in m_fontsToBeginLoading at this time.
(WebCore::CSSFontSelector::clearDocument): Now stops m_beginLoadingTimer and balances
incrementRequestCount() calls for anything remaining in m_fontsToBeginLoading.
(WebCore::CSSFontSelector::beginLoadingFontSoon): Added. Schedules the actual call to
CachedFont::beginLoadingIfNeeded on a 0-delay timer, and meanwhile increments the request count
on the CachedResourceLoader, which ensures that didFinishLoad will not be called while waiting
for the timer to fire.
(WebCore::CSSFontSelector::beginLoadTimerFired): Added. Actually calls
CachedFont::beginLoadIfNeeded and balances the incrementRequestCount() made when the timer was
scheduled.
* css/CSSFontSelector.h:

LayoutTests:

Reverted test chnages from r91738. These tests were made to wait for the fonts to finish loading
using a DOM. Now again they do not need to.

* fast/css/color-leakage.html:
* fast/css/custom-font-xheight.html:
* fast/css/font-face-multiple-faces.html:
* fast/css/font-face-multiple-remote-sources.html:
* fast/css/font-face-remote.html:
* fast/css/font-face-woff.html:
* fast/writing-mode/broken-ideograph-small-caps.html:
* svg/W3C-SVG-1.1-SE/text-intro-09-b.svg:
* svg/W3C-SVG-1.1/fonts-elem-07-b.svg:
* svg/custom/svg-fonts-fallback.xhtml:
* svg/custom/svg-fonts-in-html.html:
* svg/custom/svg-fonts-segmented.xhtml:
* svg/custom/svg-fonts-with-no-element-reference.html:
* svg/custom/svg-fonts-without-missing-glyph.xhtml:
* svg/text/text-overflow-ellipsis-svgfont.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101858 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Fix compilation warnings for ASSERT()
tkent@chromium.org [Fri, 2 Dec 2011 22:42:23 +0000 (22:42 +0000)]
[Chromium] Fix compilation warnings for ASSERT()
https://bugs.webkit.org/show_bug.cgi?id=73623

Reviewed by Tony Chang.

* DumpRenderTree/chromium/ImageDiff.cpp:
 - Fix fprintf format: %s -> %d for __LINE__
 - Do nothing for ASSERT() if NDEBUG.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101857 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoEnable parallel testing for all Mac bots.
eric@webkit.org [Fri, 2 Dec 2011 22:37:13 +0000 (22:37 +0000)]
Enable parallel testing for all Mac bots.

Unreviewed.

* Scripts/run-webkit-tests:
(platformIsReadyForParallelTesting):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101856 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSend an AXCheckedStateChanged notification when the aria-checked attribute changes.
commit-queue@webkit.org [Fri, 2 Dec 2011 22:34:53 +0000 (22:34 +0000)]
Send an AXCheckedStateChanged notification when the aria-checked attribute changes.
https://bugs.webkit.org/show_bug.cgi?id=72754

Patch by David Tseng <dtseng@google.com> on 2011-12-02
Reviewed by Chris Fleizach.

Source/WebCore:

Test: accessibility/aria-checkbox-sends-notification.html

* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::checkedStateChanged):
* accessibility/AXObjectCache.h:
* dom/Element.cpp:
(WebCore::Element::updateAfterAttributeChanged):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::setChecked):

LayoutTests:

* accessibility/aria-checkbox-sends-notification.html: Added.
* platform/chromium/accessibility/aria-checkbox-sends-notification-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101855 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Recycle tile-sized textures during commit to prevent reallocations
commit-queue@webkit.org [Fri, 2 Dec 2011 22:30:08 +0000 (22:30 +0000)]
[chromium] Recycle tile-sized textures during commit to prevent reallocations
https://bugs.webkit.org/show_bug.cgi?id=70645

Patch by Grace Kloba <klobag@chromium.org> on 2011-12-02
Reviewed by James Robinson.

Currently texture request is capped by the high limit while we reclaim the
textures in each commit. This triggers new tiles always allocated when scrolling.
The proposal is to recycle the texture during request if the total used memory
is about to exceed the reclaim limit.

* platform/graphics/chromium/ManagedTexture.cpp:
(WebCore::ManagedTexture::reserve):
* platform/graphics/chromium/TextureManager.cpp:
(WebCore::TextureManager::setMemoryLimitBytes):
(WebCore::TextureManager::replaceTexture):
(WebCore::TextureManager::requestTexture):
* platform/graphics/chromium/TextureManager.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101854 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoEnable parallel testing for run-webkit-tests on Mac Lion.
eric@webkit.org [Fri, 2 Dec 2011 22:26:56 +0000 (22:26 +0000)]
Enable parallel testing for run-webkit-tests on Mac Lion.
I plan to move all the machines to parallel testing shortly
but we're starting with Lion.

Unreviewed.

* Scripts/run-webkit-tests:
(platformIsReadyForParallelTesting):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101853 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r101337.
tkent@chromium.org [Fri, 2 Dec 2011 22:17:27 +0000 (22:17 +0000)]
Unreviewed, rolling out r101337.
http://trac.webkit.org/changeset/101337
https://bugs.webkit.org/show_bug.cgi?id=73189

It's a wrong way to fix the problem

Source/WebKit/chromium:

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::popupOpened):

LayoutTests:

* fast/forms/select-popup-crash-expected.txt: Removed.
* fast/forms/select-popup-crash.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101852 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdded helper method to identify whether the page has custom page size style.
tkent@chromium.org [Fri, 2 Dec 2011 22:07:11 +0000 (22:07 +0000)]
Added helper method to identify whether the page has custom page size style.
https://bugs.webkit.org/show_bug.cgi?id=73585

Patch by Kausalya Madhusudhanan <kmadhusu@chromium.org> on 2011-12-02
Reviewed by Darin Fisher.

Source/WebKit/chromium:

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

Tools:

* DumpRenderTree/chromium/LayoutTestController.cpp:
(LayoutTestController::LayoutTestController):
(LayoutTestController::hasCustomPageSizeStyle):
* DumpRenderTree/chromium/LayoutTestController.h:

LayoutTests:

* platform/chromium/printing/custom-page-size-style-expected.txt: Added.
* platform/chromium/printing/custom-page-size-style.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101851 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Adding flaky tests to test_expectations
scheib@chromium.org [Fri, 2 Dec 2011 22:02:58 +0000 (22:02 +0000)]
[Chromium] Adding flaky tests to test_expectations

inspector/extensions tests even flakier
newly flaky media/event-attributes.html

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101850 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed: reset expectation for tests until https://bugs.webkit.org/show_bug.cgi...
dslomov@google.com [Fri, 2 Dec 2011 21:59:05 +0000 (21:59 +0000)]
Unreviewed: reset expectation for tests until https://bugs.webkit.org/show_bug.cgi?id=73691
is fixed.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101849 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Show placeholder even if the element is focused
tkent@chromium.org [Fri, 2 Dec 2011 21:55:41 +0000 (21:55 +0000)]
[Chromium] Show placeholder even if the element is focused
https://bugs.webkit.org/show_bug.cgi?id=73629

Reviewed by Hajime Morita.

No new tests. Need to update some existing placeholder tests.

* rendering/RenderThemeChromiumMac.h: Add shouldShowPlaceholderWhenFocused().
* rendering/RenderThemeChromiumMac.mm:
(WebCore::RenderThemeChromiumMac::shouldShowPlaceholderWhenFocused):
Returns true.
* rendering/RenderThemeChromiumSkia.cpp:
(WebCore::RenderThemeChromiumSkia::shouldShowPlaceholderWhenFocused): ditto.
* rendering/RenderThemeChromiumSkia.h: Add shouldShowPlaceholderWhenFocused().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101848 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Lion][Windows] Both of placeholder and input text are shown in <input type=number>
tkent@chromium.org [Fri, 2 Dec 2011 21:49:30 +0000 (21:49 +0000)]
[Lion][Windows] Both of placeholder and input text are shown in <input type=number>
https://bugs.webkit.org/show_bug.cgi?id=73615

Reviewed by Joseph Pecoraro.

Source/WebCore:

Placeholder visibility was checked by HTMLInputElement::value
emptiness. It should be innerTextValue emptiness because it is
possible that a number field has empty HTMLInputElement::value and
non-empty innerTextValue.

Tests: fast/forms/number/number-placeholder-with-unacceptable-value.html

* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::updateInnerTextValue):
We should update placeholder visiblity when the innerTextValue is updated.
(WebCore::HTMLInputElement::subtreeHasChanged): ditto.
(WebCore::HTMLInputElement::setSuggestedValue):
We don't need to call updatePlaceholderVisibility() because updateInnerTextValue() calls it.
(WebCore::HTMLInputElement::setValueFromRenderer):
We don't need to call updatePlaceholderVisibility() because subtreeHasChanged() calls it.
* html/HTMLInputElement.h: Checks innerTextValue emptiness.
* html/TextFieldInputType.cpp:
(WebCore::TextFieldInputType::setValue):
We don't need to call updatePlaceholderVisibility() because updateInnerTextValue() calls it.

LayoutTests:

* fast/forms/number/number-placeholder-with-unacceptable-value-expected.html: Added.
* fast/forms/number/number-placeholder-with-unacceptable-value.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101847 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] GYP fix for Windows 2010 build
tony@chromium.org [Fri, 2 Dec 2011 21:49:17 +0000 (21:49 +0000)]
[Chromium] GYP fix for Windows 2010 build
https://bugs.webkit.org/show_bug.cgi?id=73483

Reviewed by Adam Barth.

Patch by Carlos Pizano.

* WebKit.gyp: Exclude the android directory and remove a file match that doesn't match anything.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[NRWT] reftest should support having multiple references per test
rniwa@webkit.org [Fri, 2 Dec 2011 21:42:32 +0000 (21:42 +0000)]
[NRWT] reftest should support having multiple references per test
https://bugs.webkit.org/show_bug.cgi?id=73613

Reviewed by Dirk Pranke.

Add a support for having multiple reference files for a single test.

Because a reftest succeeds when it matches at least one of expected matches and fails when it matches
at least one of expected mismatches, we compare expected mismatches first in order to minimize
the number of reference files to open on DRT.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(interpret_test_failures): Remove checks no longer applicable.
* Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
(ResultSummaryTest.test_interpret_test_failures): Ditto.
* Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
(SingleTestRunner.__init__): Remove a bunch of code and just call port.reference_files.
(SingleTestRunner._driver_input):
(SingleTestRunner.run):
(SingleTestRunner._run_reftest): Compare the output of the test to each reference file.
* Scripts/webkitpy/layout_tests/models/test_input.py:
(TestInput.__init__): Remove ref_file and is_mismatch_reftest because they are no longer used.
* Scripts/webkitpy/layout_tests/port/base.py:
(Port.reference_files): Renamed from _reference_file_for. Returns a list of expectation, filename pairs.
(_parse_reftest_list): Now supports parsing multiple entries for a single test.
* Scripts/webkitpy/layout_tests/port/base_unittest.py:
(PortTest.test_parse_reftest_list):
* Scripts/webkitpy/layout_tests/port/test.py:
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
(MainTest.test_unexpected_failures):
(MainTest.test_reftest_skipped_if_unlisted): Renamed from test_missing_and_unexpected_results.
(EndToEndTest.test_end_to_end):
(EndToEndTest.test_reftest_with_two_notrefs): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101845 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSource/WebCore: [chromium] CCLayerQuad does not return FloatQuad in correct order
alokp@chromium.org [Fri, 2 Dec 2011 21:08:40 +0000 (21:08 +0000)]
Source/WebCore: [chromium] CCLayerQuad does not return FloatQuad in correct order
https://bugs.webkit.org/show_bug.cgi?id=73247

Reviewed by James Robinson.

Returned the FloatQuad coordinates in correct order.

Covered by new unit tests in CCLayerQuadTest.cpp.

* platform/graphics/chromium/cc/CCLayerQuad.cpp:
(WebCore::CCLayerQuad::floatQuad):

Source/WebKit/chromium: [chromium] CCLayerQuad does not return FloatQuad in correct order
https://bugs.webkit.org/show_bug.cgi?id=73247

Reviewed by James Robinson.

* WebKit.gypi:
* tests/CCLayerQuadTest.cpp: Added.
(WebCore::TEST):

LayoutTests: [chromium] CCLayerQuad does not return FloatQuad in correct order
https://bugs.webkit.org/show_bug.cgi?id=73247

Reviewed by James Robinson.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Marking slow test: http/tests/appcache/video.html
scheib@chromium.org [Fri, 2 Dec 2011 19:57:28 +0000 (19:57 +0000)]
[Chromium] Marking slow test: http/tests/appcache/video.html

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101843 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] WebKitMutationObserver::deliverAllMutations should be exposed through...
adamk@chromium.org [Fri, 2 Dec 2011 19:39:04 +0000 (19:39 +0000)]
[chromium] WebKitMutationObserver::deliverAllMutations should be exposed through the Chromium API
https://bugs.webkit.org/show_bug.cgi?id=71242

Reviewed by Darin Fisher.

Relanding, depends on http://codereview.chromium.org/8748015/
to fix a sandboxing issue on Chromium/Linux.

Add addTaskObserver and removeTaskObserver to WebThread,
along with a new WebThread::TaskObserver interface.

For mutation observers, add a TaskObserver to the main thread
to deliver mutations after each task runs.

The Chromium implementation of the new WebThread methods is in
http://codereview.chromium.org/8586038/.

* public/platform/WebThread.h:
(WebKit::WebThread::TaskObserver::~TaskObserver):
* src/WebKit.cpp:
(WebKit::initialize):
(WebKit::shutdown):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101842 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r101833.
scheib@chromium.org [Fri, 2 Dec 2011 19:31:10 +0000 (19:31 +0000)]
Unreviewed, rolling out r101833.
http://trac.webkit.org/changeset/101833
https://bugs.webkit.org/show_bug.cgi?id=73678

test_expectations file invalid: run Tools/Scripts/new-run-
webkit-tests --lint-test-files (Requested by scheib on
#webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-12-02

Source/WebCore:

* platform/graphics/chromium/cc/CCLayerQuad.cpp:
(WebCore::CCLayerQuad::floatQuad):

Source/WebKit/chromium:

* WebKit.gypi:
* tests/CCLayerQuadTest.cpp: Removed.

LayoutTests:

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101841 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoBuild libsoup without gnome dependencies (like keyring).
kov@webkit.org [Fri, 2 Dec 2011 19:24:08 +0000 (19:24 +0000)]
Build libsoup without gnome dependencies (like keyring).

* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101840 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Fixe test_expectations.txt parse error
scheib@chromium.org [Fri, 2 Dec 2011 19:21:58 +0000 (19:21 +0000)]
[Chromium] Fixe test_expectations.txt parse error

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101839 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWebKit2: Freeze the state of the layer tree until frame load completion if incrementa...
aestes@apple.com [Fri, 2 Dec 2011 19:21:19 +0000 (19:21 +0000)]
WebKit2: Freeze the state of the layer tree until frame load completion if incremental rendering is suppressed
https://bugs.webkit.org/show_bug.cgi?id=73641

Reviewed by Darin Adler.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDidFirstLayout): Do not
un-freeze the layer tree state if incremental rendering is suppressed.
* WebProcess/WebPage/DrawingArea.h:
(WebKit::DrawingArea::layerTreeStateIsFrozen): Add a getter for
m_layerTreeStateIsFrozen.
* WebProcess/WebPage/DrawingAreaImpl.h:
(WebKit::DrawingAreaImpl::layerTreeStateIsFrozen): Ditto.
* WebProcess/WebPage/ca/LayerTreeHostCA.cpp:
(WebKit::LayerTreeHostCA::initialize): WebFrameLoaderClient might have
tried to freeze the state of the layer tree before the layer tree host
was created. When creating the layer tree host, only schedule a layer
flush if the layer tree state isn't frozen.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101838 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAlso build gcrypt and p11-kit, and make them dependencies of
kov@webkit.org [Fri, 2 Dec 2011 19:19:49 +0000 (19:19 +0000)]
Also build gcrypt and p11-kit, and make them dependencies of
gnutls.

* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101837 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Expected flakey tests after fix for 73021 is committed
scheib@chromium.org [Fri, 2 Dec 2011 19:19:41 +0000 (19:19 +0000)]
[Chromium] Expected flakey tests after fix for 73021 is committed
https://bugs.webkit.org/show_bug.cgi?id=73677

Unreviewed test_expectations update.

Patch by Stephen Chenney <schenney@chromium.org> on 2011-12-02

* platform/chromium/test_expectations.txt: Added flakey for svg/stroke/...

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101836 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[wx] Unreviewed build fixes for Windows build.
kevino@webkit.org [Fri, 2 Dec 2011 19:16:46 +0000 (19:16 +0000)]
[wx] Unreviewed build fixes for Windows build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101835 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoReviewed by Adam Barth.
eric@webkit.org [Fri, 2 Dec 2011 19:10:30 +0000 (19:10 +0000)]
Reviewed by Adam Barth.

webkit-patch post, post-commits, upload should warn when posting to a closed bug, and offer to reopen it
https://bugs.webkit.org/show_bug.cgi?id=32006

I decided not to make it warn, and just have it re-open the bug.
That's not that different from today's behavior which will
just silently attach the patch.

This patch makes behavior between upload and land-safely consistent
(previously one would assign patches and the other would not)
as well as adds the ability for both to ensure that the bug is open.

To test this I had to add a few more methods to MockBugzilla which
(positively) affected a few other test results.

I also made AbstractStep keep a cached copy of the Bug object
and used the cached copy where appropriate (including for 'bug_title').
This should reduce the number of bug fetches we perform.

* Scripts/webkitpy/tool/commands/download_unittest.py:
* Scripts/webkitpy/tool/commands/upload.py:
* Scripts/webkitpy/tool/commands/upload_unittest.py:
* Scripts/webkitpy/tool/mocktool.py:
* Scripts/webkitpy/tool/steps/__init__.py:
* Scripts/webkitpy/tool/steps/abstractstep.py:
* Scripts/webkitpy/tool/steps/closebug.py:
* Scripts/webkitpy/tool/steps/ensurebugisopenandassigned.py: Added.
* Scripts/webkitpy/tool/steps/postdiff.py:
* Scripts/webkitpy/tool/steps/postdiffforcommit.py:
* Scripts/webkitpy/tool/steps/preparechangelog.py:
* Scripts/webkitpy/tool/steps/steps_unittest.py:
* Scripts/webkitpy/tool/steps/updatechangelogswithreviewer.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101834 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] CCLayerQuad does not return FloatQuad in correct orientation
alokp@chromium.org [Fri, 2 Dec 2011 19:08:22 +0000 (19:08 +0000)]
[chromium] CCLayerQuad does not return FloatQuad in correct orientation
https://bugs.webkit.org/show_bug.cgi?id=73247

Reviewed by James Robinson.

Source/WebCore:

Returned the FloatQuad coordinates in correct order.

Covered by new unit tests in CCLayerQuadTest.cpp.

* platform/graphics/chromium/cc/CCLayerQuad.cpp:
(WebCore::CCLayerQuad::floatQuad):

Source/WebKit/chromium:

* WebKit.gypi:
* tests/CCLayerQuadTest.cpp: Added.
(WebCore::TEST):

LayoutTests:

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101833 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAlso pass --no-interact to jhbuild when updating dependencies.
kov@webkit.org [Fri, 2 Dec 2011 19:05:20 +0000 (19:05 +0000)]
Also pass --no-interact to jhbuild when updating dependencies.

* Scripts/update-webkitgtk-libs:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101832 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agohttps://bugs.webkit.org/show_bug.cgi?id=73589
dslomov@google.com [Fri, 2 Dec 2011 19:00:27 +0000 (19:00 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=73589
[V8][Chromium] Adjust postMessage to the latest "implementation-ready" spec.
- postMessage should support transfer of MessagePorts
- the order of arguments to Window::postMessage and Window::webkitPostMessage should be (msg, targetOrigin [, transfer])

Reviewed by David Levin.

Source/WebCore:

* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::handlePostMessageCallback):
* bindings/v8/custom/V8DedicatedWorkerContextCustom.cpp:
(WebCore::handlePostMessageCallback):
* bindings/v8/custom/V8MessagePortCustom.cpp:
(WebCore::handlePostMessageCallback):
* bindings/v8/custom/V8WorkerCustom.cpp:
(WebCore::handlePostMessageCallback):

LayoutTests:

* fast/canvas/webgl/script-tests/arraybuffer-transfer-of-control.js:
(wrapSend):
(wrapFailSend):
* fast/dom/Window/window-postmessage-args.html:
* platform/chromium/fast/dom/Window/window-postmessage-args-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101831 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agohttps://bugs.webkit.org/show_bug.cgi?id=73497
enrica@apple.com [Fri, 2 Dec 2011 18:57:09 +0000 (18:57 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=73497
This is a followup to the patch submitted for the bug above.
Tha patch was landed in r101575 and I missed to make one of the changes
suggested by the reviewer that had pointed out that the code was still
making use of the deprecatedNode method.

Unreviewed.

* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::removeRedundantBlocks): Changed deprecatedNode to containerNode.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101830 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoContent-security-policy script-src not enforced on workers.
tsepez@chromium.org [Fri, 2 Dec 2011 18:55:05 +0000 (18:55 +0000)]
Content-security-policy script-src not enforced on workers.
https://bugs.webkit.org/show_bug.cgi?id=73240

Reviewed by Adam Barth.

Source/WebCore:

Add a CSP check in AbstractWorker.cpp as part of resolving URL.

Test: http/tests/security/contentSecurityPolicy/worker-script-src.html

* workers/AbstractWorker.cpp:
(WebCore::AbstractWorker::resolveURL):

LayoutTests:

Added test to verify bug.

* http/tests/security/contentSecurityPolicy/worker-script-src-expected.txt: Added.
* http/tests/security/contentSecurityPolicy/worker-script-src.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101829 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Add plumbing for supporting custom MIME types in DataTransfer.
dcheng@chromium.org [Fri, 2 Dec 2011 18:52:52 +0000 (18:52 +0000)]
[chromium] Add plumbing for supporting custom MIME types in DataTransfer.
https://bugs.webkit.org/show_bug.cgi?id=73594

Reviewed by David Levin.

Source/WebCore:

Tests: editing/pasteboard/clipboard-customData.html
       fast/events/drag-customData.html

* platform/chromium/ChromiumDataObject.cpp:
(WebCore::ChromiumDataObject::types):
(WebCore::ChromiumDataObject::getData):
(WebCore::ChromiumDataObject::setData):
* platform/chromium/ChromiumDataObject.h:
(WebCore::ChromiumDataObject::customData):
* platform/chromium/PlatformSupport.h:

Source/WebKit/chromium:

* public/platform/WebClipboard.h:
(WebKit::WebClipboard::readCustomData):
* public/platform/WebDragData.h:
* src/PlatformSupport.cpp:
(WebCore::PlatformSupport::clipboardReadCustomData):
* src/WebDragData.cpp:
(WebKit::WebDragData::customData):
(WebKit::WebDragData::setCustomData):

LayoutTests:

* editing/pasteboard/clipboard-customData-expected.txt: Added.
* editing/pasteboard/clipboard-customData.html: Added.
* fast/events/drag-customData-expected.txt: Added.
* fast/events/drag-customData.html: Added.
* platform/chromium/test_expectations.txt:
* platform/efl/Skipped:
* platform/gtk/Skipped:
* platform/mac/Skipped:
* platform/qt/Skipped:
* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@101828 268f45cc-cd09-0410-ab3c-d52691b4dbfc