Fix Chromium Mac build.

[WebKit.git] / WebCore / ChangeLog

2011-01-05  Darin Adler  <darin@apple.com>

        Fix Chromium Mac build.

        * platform/graphics/Gradient.h: Forward-declare CGContextRef.
        This might even make it possible to remove the include of
        <CoreGraphics/CoreGraphics.h> for PLATFORM(WIN), but I did not
        do that at this time because I don't want to break the Windows
        build when fixing the Chromium build.

2011-01-05  Xianzhu Wang  <wangxianzhu@google.com>

        Reviewed by David Levin.

        Implement RemoteFontStream's skip behavior (in FontCustomPlatformData.cpp)
        https://bugs.webkit.org/show_bug.cgi?id=51736

        No new tests. This change has potential effect on @font-face CSS feature
        on ports using Skia. Tested on Chromium that this change won't break
        fast/css/font-face-remote.html. This change can make the test pass on
        other ports using Skia.

        * platform/graphics/skia/FontCustomPlatformData.cpp:
        (WebCore::RemoteFontStream::read):

2011-01-05  Helder Correia  <helder@sencha.com>

        Reviewed by Simon Fraser.

        Shadows are not drawn when filling a rect with a gradient fillStyle on Mac and Chromium-Mac
        https://bugs.webkit.org/show_bug.cgi?id=51869

        GraphicsContext::fillRect(const FloatRect&) is clipping to the rect.
        The shadow is being drawn, but getting clipped to the rect being filled.
        The solution is to draw the gradient into a CGLayer (when a shadow is
        required), then render the layer to the GraphicsContext.

        This patch also fixes:
        canvas/philip/tests/2d.shadow.gradient.alpha.html
        canvas/philip/tests/2d.shadow.gradient.basic.html
        canvas/philip/tests/2d.shadow.gradient.transparent.2.html

        Test: fast/canvas/canvas-fillRect-gradient-shadow.html

        * platform/graphics/Gradient.h:
        * platform/graphics/cg/GradientCG.cpp:
        (WebCore::Gradient::paint):
        * platform/graphics/cg/GraphicsContextCG.cpp:
        (WebCore::GraphicsContext::fillRect):

2011-01-05  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Darin Adler.

        Debug and Release builds on Windows clobber each other
        https://bugs.webkit.org/show_bug.cgi?id=49185

        Changes the structure of WebKitBuild build products directory so we
        completely separate each build configuration into independent directories.
        
        Although we previously had per-configuration directories for obj, this change adds
        per-configuration directories for bin, lib, obj, and include. Each configuration's
        build products are stored within a directory inside of WebKitBuild.
        
        Most use of $(WebKitOutputDir) in the build files has been replaced by $(ConfigurationBuildDir),
        defined in common.vsprops to be $(WebKitOutputDir)\$(ConfigurationName).

        * WebCore.vcproj/QTMovieWinCommon.vsprops:
        * WebCore.vcproj/QTMovieWinPostBuild.cmd:
        * WebCore.vcproj/QTMovieWinPreBuild.cmd:
        * WebCore.vcproj/WebCore.make:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.vcproj/WebCoreCommon.vsprops:
        * WebCore.vcproj/WebCoreGeneratedCairo.vsprops:
        * WebCore.vcproj/WebCoreGeneratedCommon.vsprops:
        * WebCore.vcproj/WebCoreMediaQT.vsprops:
        * WebCore.vcproj/WebCorePostBuild.cmd:
        * WebCore.vcproj/WebCorePreBuild.cmd:
        * WebCore.vcproj/copyForwardingHeaders.cmd:
        * WebCore.vcproj/copyInspectorFiles.cmd:

2011-01-05  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Ariya Hidayat.

        Directly composited SVG images fail to render
        https://bugs.webkit.org/show_bug.cgi?id=51975
        
        There was already code, added in r67900, to detect whether an image
        may be directly composited, by checking to see if it's a bitmap image.
        
        However, CachedImage::image() returns the nullImage() if m_image is null,
        which is a BitmapImage, confusing this test.

        Test: compositing/images/direct-svg-image.html

        * loader/cache/CachedImage.cpp: Remove nullImage(), since Image has
        exactly the same method.
        (WebCore::CachedImage::image): Use Image::nullImage().
        * loader/cache/CachedImage.h:
        (WebCore::CachedImage::hasImage): New method to ask whether the m_image is
        non-null.
        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::isDirectlyCompositedImage): Use hasImage()
        so that we don't test the nullImage.

2011-01-05  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Darin Adler.

        Fix percentage values in radius of radial gradients
        https://bugs.webkit.org/show_bug.cgi?id=51979
        
        Handle percentage values for the ending size of radial gradients.

        Test: fast/gradients/css3-radial-gradients3.html

        * css/CSSGradientValue.cpp:
        (WebCore::CSSRadialGradientValue::resolveRadius): Add parameter for height or width, which is
        non-null if percentage values are allowed, and compute percentage values when we see them.
        (WebCore::CSSRadialGradientValue::createGradient): Pass height or width down to say that
        percentages are OK when resolving end radii.
        * css/CSSGradientValue.h: New parameter for resolveRadius.

2011-01-05  James Simonsen  <simonjam@chromium.org>

        Reviewed by Adam Barth.

        [Web Timing] Zero out some values on cross-origin redirects
        https://bugs.webkit.org/show_bug.cgi?id=49294

        Tests: http/tests/misc/webtiming-cross-origin-redirect.php
               http/tests/misc/webtiming-origins.html

        * loader/DocumentLoadTiming.h:
        (WebCore::DocumentLoadTiming::DocumentLoadTiming): Add flags for cross-origin redirects and same-origin navigation.
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::~FrameLoader):
        (WebCore::FrameLoader::stopLoading): Don't overwrite previous unload times.
        (WebCore::FrameLoader::completed):
        (WebCore::FrameLoader::loadWithDocumentLoader): Remember previous URL to see if navigation is same-origin.
        (WebCore::FrameLoader::commitProvisionalLoad): Set flag if navigation is same-origin.
        * loader/FrameLoader.h:
        * loader/MainResourceLoader.cpp:
        (WebCore::MainResourceLoader::willSendRequest): Set flag on cross-origin redirects.
        * page/PerformanceNavigation.cpp:
        (WebCore::PerformanceNavigation::redirectCount): Clear if cross-origin redirect flag is set.
        * page/PerformanceTiming.cpp:
        (WebCore::PerformanceTiming::redirectStart): Ditto.
        (WebCore::PerformanceTiming::redirectEnd): Ditto.
        (WebCore::PerformanceTiming::unloadEventStart): Clear if cross-origin navigation is set.
        (WebCore::PerformanceTiming::unloadEventEnd): Ditto.

2011-01-05  Helder Correia  <helder@sencha.com>

        Reviewed by Ariya Hidayat.

        [Qt] Default focus ring is too wide
        https://bugs.webkit.org/show_bug.cgi?id=51854

        This is an intermediate step to solve this bug. Use QPalette::Highlight
        instead of default half-transparent black as the focus ring color.

        * platform/graphics/qt/GraphicsContextQt.cpp:
        (WebCore::drawFocusRingForPath):
        * platform/qt/RenderThemeQt.cpp:
        (WebCore::RenderThemeQt::platformFocusRingColor):
        * platform/qt/RenderThemeQt.h:

2011-01-05  James Simonsen  <simonjam@chromium.org>

        Reviewed by Darin Fisher.

        [Web Timing] requestStart and responseStart should be available even if the document is still loading
        https://bugs.webkit.org/show_bug.cgi?id=51368

        Test: http/tests/misc/webtiming-slow-load.php

        * page/PerformanceTiming.cpp:
        (WebCore::getPossiblySkewedTimeInKnownRange): Handle case where response is not yet complete.
        (WebCore::PerformanceTiming::resourceLoadTimeRelativeToAbsolute): Use full upper bound.

2011-01-05  Chris Rogers  <crogers@google.com>

        Reviewed by Darin Fisher.

        Add WebKitClient::createAudioDevice() for Chromium port of web audio API
        https://bugs.webkit.org/show_bug.cgi?id=51424

        No new tests since audio API is not yet implemented.

        * WebCore.gyp/WebCore.gyp:
        * WebCore.gypi:

2011-01-05  Anders Carlsson  <andersca@apple.com>

        Try again.

        * WebCore.exp.in:

2011-01-05  Anders Carlsson  <andersca@apple.com>

        Attempt to fix Leopard build.

        * WebCore.exp.in:

2011-01-05  Evan Martin  <evan@chromium.org>

        Reviewed by Tony Chang.

        [chromium] refactor complex text on Linux
        https://bugs.webkit.org/show_bug.cgi?id=51956

        Pull out TextRunWalker from FontLinux.cpp.  Put it in its own file.
        Rename it to ComplexTextController, to match platform/graphics/mac
        more closely.

        No tests because the change is just renaming a class.

        * WebCore.gyp/WebCore.gyp:
        * platform/graphics/chromium/ComplexTextControllerLinux.cpp: Added.
        (WebCore::ComplexTextController::ComplexTextController):
        (WebCore::ComplexTextController::~ComplexTextController):
        (WebCore::ComplexTextController::isWordBreak):
        (WebCore::ComplexTextController::determineWordBreakSpacing):
        (WebCore::ComplexTextController::setPadding):
        (WebCore::ComplexTextController::reset):
        (WebCore::ComplexTextController::setBackwardsIteration):
        (WebCore::ComplexTextController::nextScriptRun):
        (WebCore::ComplexTextController::widthOfFullRun):
        (WebCore::ComplexTextController::setupFontForScriptRun):
        (WebCore::ComplexTextController::allocHarfbuzzFont):
        (WebCore::ComplexTextController::deleteGlyphArrays):
        (WebCore::ComplexTextController::createGlyphArrays):
        (WebCore::ComplexTextController::resetGlyphArrays):
        (WebCore::ComplexTextController::shapeGlyphs):
        (WebCore::ComplexTextController::setGlyphXPositions):
        (WebCore::ComplexTextController::normalizeSpacesAndMirrorChars):
        (WebCore::ComplexTextController::getNormalizedTextRun):
        * platform/graphics/chromium/ComplexTextControllerLinux.h: Added.
        (WebCore::ComplexTextController::setWordSpacingAdjustment):
        (WebCore::ComplexTextController::setLetterSpacingAdjustment):
        (WebCore::ComplexTextController::letterSpacing):
        (WebCore::ComplexTextController::setXOffsetToZero):
        (WebCore::ComplexTextController::rtl):
        (WebCore::ComplexTextController::glyphs):
        (WebCore::ComplexTextController::length):
        (WebCore::ComplexTextController::xPositions):
        (WebCore::ComplexTextController::advances):
        (WebCore::ComplexTextController::width):
        (WebCore::ComplexTextController::logClusters):
        (WebCore::ComplexTextController::numCodePoints):
        (WebCore::ComplexTextController::fontPlatformDataForScriptRun):
        (WebCore::ComplexTextController::isCodepointSpace):
        * platform/graphics/chromium/FontLinux.cpp:
        (WebCore::Font::drawComplexText):
        (WebCore::Font::floatWidthForComplexText):
        (WebCore::glyphIndexForXPositionInScriptRun):
        (WebCore::Font::offsetForPositionForComplexText):
        (WebCore::Font::selectionRectForComplexText):

2011-01-05  Pratik Solanki  <psolanki@apple.com>

        Reviewed by Darin Adler.

        Merge cf and mac header files with same name
        https://bugs.webkit.org/show_bug.cgi?id=51916

        Since most of the code in the header files is identical, we can merge them and have #if
        USE(CFNETWORK) to separate out the network specific parts.

        * WebCore.xcodeproj/project.pbxproj:
        * platform/network/cf/AuthenticationChallenge.h:
        (WebCore::AuthenticationChallenge::sender):
        (WebCore::AuthenticationChallenge::nsURLAuthenticationChallenge):
        * platform/network/cf/ResourceError.h:
        (WebCore::ResourceError::ResourceError):
        * platform/network/cf/ResourceRequest.h:
        (WebCore::ResourceRequest::ResourceRequest):
        * platform/network/cf/ResourceResponse.h:
        (WebCore::ResourceResponse::ResourceResponse):
        * platform/network/mac/AuthenticationChallenge.h: Removed.
        * platform/network/mac/ResourceError.h: Removed.
        * platform/network/mac/ResourceRequest.h: Removed.
        * platform/network/mac/ResourceResponse.h: Removed.

2011-01-05  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adele Peterson.

        Implement word transformation
        https://bugs.webkit.org/show_bug.cgi?id=51943

        Export Editor functions needed by WebKit2.

        * WebCore.exp.in:

2011-01-05  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dave Hyatt.

        Implement repeating CSS3 gradients
        https://bugs.webkit.org/show_bug.cgi?id=51843
        
        Add support for repeating gradients, via the new gradient functions
        -webkit-repeating-linear-gradient and -webkit-repeating-radial-gradient.

        Tests: fast/gradients/css3-repeating-linear-gradients.html
               fast/gradients/css3-repeating-radial-gradients.html

        * css/CSSGradientValue.h:
        (WebCore::CSSGradientValue::isRepeating): Accessor for whethe the gradient repeats.
        (WebCore::CSSGradientValue::CSSGradientValue): Pass CSSGradientRepeat in.
        (WebCore::CSSLinearGradientValue::create): Pass CSSGradientRepeat.
        (WebCore::CSSLinearGradientValue::CSSLinearGradientValue): Ditto
        (WebCore::CSSRadialGradientValue::create): Ditto.
        (WebCore::CSSRadialGradientValue::CSSRadialGradientValue): Ditto.

        * css/CSSGradientValue.cpp:
        (WebCore::CSSGradientValue::addStops): Pass maxLengthForRepeat, which
        is used for repeating radial gradients. Add code to repeat the stops,
        adding stops before the start, and after the end until the 0-N range is
        covered, where N is large enough to ensure the box is covered. Fix an issue
        with repeating gradients where if both stops were < 0, it would fail to clamp
        them to 0.
        
        (WebCore::CSSLinearGradientValue::cssText): Output strings for repeating gradients.
        (WebCore::CSSLinearGradientValue::createGradient): maxExtent for linear gradients is 1,
        because they are guaranteed to fill the box already.
        (WebCore::CSSRadialGradientValue::cssText): Output strings for repeating gradients.
        (WebCore::CSSRadialGradientValue::createGradient): Compute the distance to the furthest
        corner in order to inform addStops() how much additional extent is required.

        * css/CSSParser.cpp:
        (WebCore::CSSParser::parseDeprecatedGradient): Old gradients never repeat.
        (WebCore::CSSParser::parseLinearGradient): Pass down CSSGradientRepeat.
        (WebCore::CSSParser::parseRadialGradient): Ditto.
        (WebCore::CSSParser::isGeneratedImageValue): Add repeating gradient functions.
        (WebCore::CSSParser::parseGeneratedImage): Ditto.
        * css/CSSParser.h: Pass CSSGradientRepeat to gradient functions.
        * platform/graphics/cg/GradientCG.cpp: Remove unused #include.

2011-01-05  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Ariya Hidayat.

        ContextShadow::blurLayerImage() not endian-safe
        https://bugs.webkit.org/show_bug.cgi?id=51822

        Correct the channel map in ContextShadow to reflect the different
        ordering of pixels on processors with different endianness.

        No new tests. This is covered by the existing shadow tests. It was
        not discovered, because all of the bots using ContextShadow are
        little-endian.

        * platform/graphics/ContextShadow.cpp:
        (WebCore::ContextShadow::blurLayerImage): Initialize the channel mapping
        based on machine endianness.

2011-01-05  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Nate Chapin.

        [Chromium] Crash when drawing PDF plug-in scrollbars
        https://bugs.webkit.org/show_bug.cgi?id=51942

        Pepper plug-in scrollbars don't have a parent, guard against that before
        seeing if the parent is a FrameView.

        Not testable via layout tests, verified fix by hand.
        
        * platform/chromium/ScrollbarThemeChromiumMac.mm:
        (WebCore::ScrollbarThemeChromiumMac::paint):

2011-01-05  Abhishek Arya  <inferno@chromium.org>

        Unreviewed.

        Update FIXME comment to sound better. Also initialize the block to fix compile.

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

2011-01-05  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dave Hyatt.

        Do not merge anonymous blocks when one of them is the one getting removed.
        https://bugs.webkit.org/show_bug.cgi?id=51919

        Test: fast/block/merge-anonymous-block-remove-child-crash.html

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

2011-01-05  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Eric Seidel.

        ApplyStyleCommand::applyRelativeFontStyleChange should take EditingStyle*
        https://bugs.webkit.org/show_bug.cgi?id=50641

        Modified applyRelativeFontStyleChange to take EditingStyle instead of CSSMutableStyleDeclaration.
        Also extracted the logic to obtain the font size delta from ApplyStyleCommand to EditingStyle.
        The font size delta is now stored as a member variable of EditingStyle as supposed to a property
        in CSSMutableStyleDeclration as soon as EditingStyle is instantiated.

        No new tests are added since this is a refactoring.

        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::doApply): Calls applyRelativeFontStyleChange.
        (WebCore::ApplyStyleCommand::applyRelativeFontStyleChange): Takes EditingStyle.
        * editing/ApplyStyleCommand.h:
        * editing/EditingStyle.cpp:
        (WebCore::EditingStyle::EditingStyle): Initializes m_fontSizeDelta.
        (WebCore::EditingStyle::init): Ditto.
        (WebCore::EditingStyle::extractFontSizeDelta): Extracted from applyRelativeFontStyleChange.
        (WebCore::EditingStyle::isEmpty): Takes care of m_fontSizeDelta.
        (WebCore::EditingStyle::setStyle): Calls extractFontSizeDelta.
        (WebCore::EditingStyle::overrideWithStyle): Ditto.
        (WebCore::EditingStyle::clear): Initializes m_fontSizeDelta.
        (WebCore::EditingStyle::copy): Calls extractFontSizeDelta.
        * editing/EditingStyle.h:
        (WebCore::EditingStyle::fontSizeDelta): Added.
        (WebCore::EditingStyle::hasFontSizeDelta): Added.
        * editing/Editor.h:

2011-01-05  Dan Bernstein  <mitz@apple.com>

        Reviewed by Darin Adler.

        <rdar://problem/8799835> hyphenate-character: auto should fall back to hyphen-minus if the primary font lacks a hyphen glyph
        https://bugs.webkit.org/show_bug.cgi?id=51915

        This changes the results of existing tests in fast/text.

        * platform/graphics/Font.h:
        * platform/graphics/FontFastPath.cpp:
        (WebCore::Font::primaryFontHasGlyphForCharacter): Added.
        * rendering/style/RenderStyle.cpp:
        * platform/graphics/qt/FontQt.cpp:
        (WebCore::Font::primaryFontHasGlyphForCharacter): Added a stub.
        (WebCore::RenderStyle::hyphenString): For hyphenate-character: auto, return a
        hyphen only if the primary font has a hyphen glyph, and otherwise return a
        hyphen-minus.

2011-01-05  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Ariya Hidayat.

        [Qt] [Symbian] Make CONFIG-=def_files the default configuration for trunk builds
        https://bugs.webkit.org/show_bug.cgi?id=51888

        Overrule Qt's default from qconfig.pri for non-production builds so
        that CONFIG(def_files) evaluates to false and QtWebKit trunk builds succeed.

        No new tests as there is no new functionality.

        * WebCore.pro:

2011-01-05  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r74778.
        http://trac.webkit.org/changeset/74778
        https://bugs.webkit.org/show_bug.cgi?id=51930

        Android and QtWebKit-V8 uses V8DOMWindowShell::setContext.
        (Requested by bbandix on #webkit).

        * bindings/v8/V8DOMWindowShell.cpp:
        (WebCore::V8DOMWindowShell::setContext):
        * bindings/v8/V8DOMWindowShell.h:

2011-01-05  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Andreas Kling.

        [Qt] Animated GIF images does not animate 10x as expected by default.
        https://bugs.webkit.org/show_bug.cgi?id=36818

        Add a test animated GIF which should be displayed 2 times and stop.

        * manual-tests/animated-gif-looping.html:
        * manual-tests/resources/animated-2x.gif: Added.

2011-01-05  Carlos Garcia Campos  <cgarcia@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Fork code of ScrollbarThemeGtk that will differ for GTK+ 3
        https://bugs.webkit.org/show_bug.cgi?id=51923

        Split out pieces of ScrollbarThemeGtk into ScrollbarThemeGtk2 and
        ScrollbarThemeGtk3 that will differ between GTK+ 2 and GTK+ 3.

        No new tests. This should not change functionality.

        * GNUmakefile.am: Add ScrollbarThemeGtk2.cpp and ScrollbarThemeGtk3.cpp
        * platform/gtk/ScrollbarThemeGtk.cpp:
        (WebCore::ScrollbarThemeGtk::updateScrollbarsFrameThickness):
        * platform/gtk/ScrollbarThemeGtk.h:
        * platform/gtk/ScrollbarThemeGtk2.cpp: Added.
        (WebCore::ScrollbarThemeGtk::updateThemeProperties):
        (WebCore::ScrollbarThemeGtk::paintTrackBackground):
        (WebCore::ScrollbarThemeGtk::paintScrollbarBackground):
        (WebCore::ScrollbarThemeGtk::paintThumb):
        (WebCore::ScrollbarThemeGtk::paintButton):
        * platform/gtk/ScrollbarThemeGtk3.cpp: Added.
        (WebCore::ScrollbarThemeGtk::updateThemeProperties):
        (WebCore::ScrollbarThemeGtk::paintTrackBackground):
        (WebCore::ScrollbarThemeGtk::paintScrollbarBackground):
        (WebCore::ScrollbarThemeGtk::paintThumb):
        (WebCore::ScrollbarThemeGtk::paintButton):

2011-01-05  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Andreas Kling.

        [CMake] Fix the usage of SOURCE_GROUP
        https://bugs.webkit.org/show_bug.cgi?id=51739

        * CMakeLists.txt:

2011-01-05  Andras Becsi  <abecsi@webkit.org>

        Reviewed by Csaba Osztrogonác.

        [Qt][V8] Fix the build after recent changes.

        No new tests needed.

        * WebCore.pro: Add missing files, and remove obsolete ones.

2011-01-05  Yong Li  <yoli@rim.com>

        Reviewed by Adam Barth.

        Suspend HTMLParserScheduler when page load is deferred to
        avoid potential JS re-entrancy. Otherwise, when JS execution is triggered
        from an external script or by events, and is blocked by a modal dialog,
        WebKit can parse more HTML source and also start another JS execution.
        
        https://bugs.webkit.org/show_bug.cgi?id=48077

        Test case: WebCore/manual-tests/bugzilla-48077.html.

        * dom/DocumentParser.cpp:
        (WebCore::DocumentParser::suspendScheduledTasks):
        (WebCore::DocumentParser::resumeScheduledTasks):
        * dom/DocumentParser.h:
        * html/parser/HTMLDocumentParser.cpp:
        (WebCore::HTMLDocumentParser::suspendScheduledTasks):
        (WebCore::HTMLDocumentParser::resumeScheduledTasks):
        * html/parser/HTMLParserScheduler.cpp:
        (WebCore::HTMLParserScheduler::HTMLParserScheduler):
        (WebCore::HTMLParserScheduler::suspend):
        (WebCore::HTMLParserScheduler::resume):
        * html/parser/HTMLParserScheduler.h:
        (WebCore::HTMLParserScheduler::isScheduledForResume):
        * page/PageGroupLoadDeferrer.cpp:
        (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
        (WebCore::PageGroupLoadDeferrer::~PageGroupLoadDeferrer):

2011-01-05  Simon Hausmann  <simon.hausmann@nokia.com>

        Reviewed by Kenneth Rohde Christiansen, Gustavo Noronha Silva

        [GStreamer] MediaPlayerPrivateGStreamer::totalBytes() does not work reliably
        https://bugs.webkit.org/show_bug.cgi?id=51926

        Work around potential upstream bug: Querying a GstBin for the duration,
        will forward the query to all sink children. Our WebKitWebSrc however
        is a bin with only source children, therefore the query fails. Until
        this is changed upstream, this patch works around it as follows,
        based on suggestion by Philippe Normand and Sebastian Dröge:

        When the initial query fails, attempt the same query on all source
        pads and take the maximum length reported.

        * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
        (WebCore::MediaPlayerPrivateGStreamer::totalBytes):

2011-01-05  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Eric Seidel.

        [GStreamer] build fails with gstreamer core 0.10.31 and -plugins-base 0.10.30
        https://bugs.webkit.org/show_bug.cgi?id=51810

        Based on a patch by Mart Raudsepp.

        No new tests, build fix only.

        * platform/graphics/gstreamer/GStreamerGWorld.cpp:
        (WebCore::GStreamerGWorld::setWindowOverlay): Use the new xoverlay
        API if the GST_CHECK_PLUGINS_BASE_VERSION macro is defined. It was
        introduced in same version as the xoverlay API change.

2011-01-05  Leo Yang  <leo.yang@torchmobile.com.cn>

        Reviewed by Dirk Schulze.

        Malformed SVG causes crash in updateContainerOffset
        https://bugs.webkit.org/show_bug.cgi?id=44610

        We should check recursive <use> at the begining of
        WebCore::SVGUseElement::buildInstanceTree instead
        of at the end of it because the target element's
        children may cause infinite recursive <use>.

        Test: svg/custom/recursive-use.svg

        * svg/SVGUseElement.cpp:
        (WebCore::SVGUseElement::buildInstanceTree):
        (WebCore::SVGUseElement::hasCycleUseReferencing):
        * svg/SVGUseElement.h:

2011-01-04  Antti Koivisto  <antti@apple.com>

        Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=51907
        Style selector should only be recalculated when the pending sheet count goes to zero
        
        Shark says this is a significant performance improvement on some sites with lots of stylesheets.

        * dom/Document.cpp:
        (WebCore::Document::removePendingSheet):

2011-01-05  Carlos Garcia Campos  <cgarcia@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Use GtkStyleContext to get platform colors
        https://bugs.webkit.org/show_bug.cgi?id=51830

        Use GtkStyleContext API to get platform colors when building with
        GTK+ 3.x.

        No new tests. This should not change functionality.

        * platform/graphics/Color.h: Add specialized constructor for GdkRGBA.
        * platform/graphics/gtk/ColorGtk.cpp:
        (WebCore::Color::Color):
        (WebCore::Color::operator GdkRGBA):
        * platform/gtk/RenderThemeGtk3.cpp:
        (WebCore::RenderThemeGtk::platformActiveSelectionBackgroundColor):
        (WebCore::RenderThemeGtk::platformInactiveSelectionBackgroundColor):
        (WebCore::RenderThemeGtk::platformActiveSelectionForegroundColor):
        (WebCore::RenderThemeGtk::platformInactiveSelectionForegroundColor):
        (WebCore::RenderThemeGtk::activeListBoxSelectionBackgroundColor):
        (WebCore::RenderThemeGtk::inactiveListBoxSelectionBackgroundColor):
        (WebCore::RenderThemeGtk::activeListBoxSelectionForegroundColor):
        (WebCore::RenderThemeGtk::inactiveListBoxSelectionForegroundColor):
        (WebCore::RenderThemeGtk::systemColor):

2011-01-05  Carlos Garcia Campos  <cgarcia@igalia.com>

        Unreviewed. Build fix for GTK when using gtk+-3 after r75009.

        * platform/gtk/RenderThemeGtk3.cpp:
        (WebCore::RenderThemeGtk::getStockIcon): Use adoptGRef instead of
        adoptGRefPtr which doesn't exist.

2011-01-05  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Sam Weinig.

        https://bugs.webkit.org/show_bug.cgi?id=51903
        Tell UI process which subframe is largest

        * WebCore.exp.in: Exported FrameView::contentsSize().

2011-01-04  Chris Fleizach  <cfleizach@apple.com>

        Unreviewed. GTK build fix.

        WK2: Support Accessibility
        https://bugs.webkit.org/show_bug.cgi?id=51859

        * page/FrameView.cpp:
        (WebCore::FrameView::axObjectCache):

2011-01-04  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r74992.
        http://trac.webkit.org/changeset/74992
        https://bugs.webkit.org/show_bug.cgi?id=51917

        Crash in HTMLDocumentParser::suspendScheduledTasks()
        (Requested by tkent on #webkit).

        * dom/DocumentParser.cpp:
        * dom/DocumentParser.h:
        * html/parser/HTMLDocumentParser.cpp:
        * html/parser/HTMLDocumentParser.h:
        * html/parser/HTMLParserScheduler.cpp:
        (WebCore::HTMLParserScheduler::HTMLParserScheduler):
        * html/parser/HTMLParserScheduler.h:
        (WebCore::HTMLParserScheduler::isScheduledForResume):
        * page/PageGroupLoadDeferrer.cpp:
        (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
        (WebCore::PageGroupLoadDeferrer::~PageGroupLoadDeferrer):

2011-01-04  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] [WK2] Add support for non-trivial context menu action
        https://bugs.webkit.org/show_bug.cgi?id=51902

        Fix the preprocessor guard for ContextMenuItemTagSelectAll in
        the implementation of ContextMenuController::contextMenuItemSelected(). The guard was
        inconsistent with what is defined in ContextMenuItem.h.

        * page/ContextMenuController.cpp:
        (WebCore::ContextMenuController::contextMenuItemSelected):

2011-01-04  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Ryosuke Niwa.

        Avoid manual ref/deref in AsyncScriptRunner by using PendingScript
        https://bugs.webkit.org/show_bug.cgi?id=51723

        ScriptElement should also be able to keep its Element+CachedScript in
        a PendingScript, and then executeScriptSoon can accept a PendingScript.
        Once HTMLScriptRunner, ScriptElement, and AsyncScriptRunner all use
        PendingScripts, then generic request and execute which operate on
        PendingScripts can be factored out to avoid code duplicate that we have.

        No new tests because no new functionality.

        * dom/AsyncScriptRunner.cpp:
        (WebCore::AsyncScriptRunner::~AsyncScriptRunner):
        (WebCore::AsyncScriptRunner::executeScriptSoon):
        (WebCore::AsyncScriptRunner::timerFired):
        * dom/AsyncScriptRunner.h:
        * dom/PendingScript.h: Add ctor which sets element and cachedScript. Rename adoptElement->setElement.
        * dom/HTMLScriptRunner.cpp:
        (WebCore::HTMLScriptRunner::requestPendingScript):

2011-01-04  Jeff Miller  <jeffm@apple.com>

        Reviewed by Darin Adler.

        Member variable and method naming cleanup to be consistent with WebKit2.
        https://bugs.webkit.org/show_bug.cgi?id=51228

        * page/win/EventHandlerWin.cpp:
        (WebCore::EventHandler::eventActivatedView):
        Rename activatedWebView() to didActivateWebView().
        
        * platform/PlatformMouseEvent.h:
        Rename m_activatedWebView to m_didActivateWebView.
        
        (WebCore::PlatformMouseEvent::PlatformMouseEvent):
        Initialize platform-specific member variables in PlatformMouseEvent constructors.
        
        (WebCore::PlatformMouseEvent::didActivateWebView):
        Rename activatedWebView() to didActivateWebView().

        * platform/win/PlatformMouseEventWin.cpp:
        (WebCore::PlatformMouseEvent::PlatformMouseEvent):
        Rename m_activatedWebView to m_didActivateWebView.

2011-01-04  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Darin Adler.

        CSS3 :nth-child selector and valid whitespace test fails
        https://bugs.webkit.org/show_bug.cgi?id=36279

        Teach the tokenizer to allow white spaces around [+/-] signs in nth constructs.

        Test: fast/css/css3-nth-space.html

        * css/tokenizer.flex:

2011-01-04  Chris Fleizach  <cfleizach@apple.com>

        Unreviewed. Build fix.

        WK2: Support Accessibility
        https://bugs.webkit.org/show_bug.cgi?id=51859

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (fallbackObject):

2011-01-04  Cosmin Truta  <ctruta@chromium.org>

        Reviewed by Eric Seidel.

        [chromium] PNG compression settings optimized for speed
        https://bugs.webkit.org/show_bug.cgi?id=51719

        Although Z_HUFFMAN_ONLY is the fastest on hard-to-compress images,
        it is also the slowest, by a wide margin, on easy-to-compress images.
        Use a more balanced configuration, based on the libpng compression defaults,
        but with a faster compression level (3 instead of 6), and a faster filter
        ("sub" instead of "all").

        No change in behaviour, so no new tests.

        * platform/image-encoders/skia/PNGImageEncoder.cpp:
        (WebCore::PNGImageEncoder::encode):

2011-01-04  Chris Fleizach  <cfleizach@apple.com>

        Reviewed by Sam Weinig.

        WK2: Support Accessibility
        https://bugs.webkit.org/show_bug.cgi?id=51859

        WK2 doesn't use platform scroll views anymore. Accessibility code
        needs to have its own scroll views. 

        Test: platform/mac/accessibility/webkit-scrollarea.html

        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.exp.in:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * accessibility/AXObjectCache.cpp:
        (WebCore::AXObjectCache::AXObjectCache):
        (WebCore::AXObjectCache::get):
        (WebCore::AXObjectCache::getOrCreate):
        (WebCore::AXObjectCache::rootObject):
        (WebCore::AXObjectCache::remove):
        (WebCore::AXObjectCache::handleScrollbarUpdate):
        * accessibility/AXObjectCache.h:
        (WebCore::AXObjectCache::handleScrollbarUpdate):
        * accessibility/AccessibilityARIAGrid.h:
        * accessibility/AccessibilityAllInOne.cpp:
        * accessibility/AccessibilityImageMapLink.cpp:
        * accessibility/AccessibilityImageMapLink.h:
        * accessibility/AccessibilityListBoxOption.cpp:
        * accessibility/AccessibilityListBoxOption.h:
        * accessibility/AccessibilityObject.cpp:
        (WebCore::AccessibilityObject::updateChildrenIfNecessary):
        (WebCore::AccessibilityObject::elementAccessibilityHitTest):
        (WebCore::AccessibilityObject::axObjectCache):
        (WebCore::AccessibilityObject::focusedUIElement):
        * accessibility/AccessibilityObject.h:
        (WebCore::AccessibilityObject::isAccessibilityScrollView):
        (WebCore::AccessibilityObject::isScrollView):
        (WebCore::AccessibilityObject::canSetNumericValue):
        (WebCore::AccessibilityObject::scrollBar):
        (WebCore::AccessibilityObject::size):
        (WebCore::AccessibilityObject::setValue):
        * accessibility/AccessibilityRenderObject.cpp:
        (WebCore::AccessibilityRenderObject::parentObject):
        (WebCore::AccessibilityRenderObject::accessibilityHitTest):
        (WebCore::AccessibilityRenderObject::updateChildrenIfNecessary):
        (WebCore::AccessibilityRenderObject::addChildren):
        * accessibility/AccessibilityRenderObject.h:
        * accessibility/AccessibilityScrollView.cpp: Added.
        (WebCore::AccessibilityScrollView::AccessibilityScrollView):
        (WebCore::AccessibilityScrollView::create):
        (WebCore::AccessibilityScrollView::scrollBar):
        (WebCore::AccessibilityScrollView::children):
        (WebCore::AccessibilityScrollView::updateChildrenIfNecessary):
        (WebCore::AccessibilityScrollView::removeChildScrollbar):
        (WebCore::AccessibilityScrollView::addChildScrollbar):
        (WebCore::AccessibilityScrollView::addChildren):
        (WebCore::AccessibilityScrollView::webAreaObject):
        (WebCore::AccessibilityScrollView::accessibilityHitTest):
        (WebCore::AccessibilityScrollView::document):
        (WebCore::AccessibilityScrollView::elementRect):
        (WebCore::AccessibilityScrollView::parentObject):
        * accessibility/AccessibilityScrollView.h: Added.
        (WebCore::AccessibilityScrollView::roleValue):
        (WebCore::AccessibilityScrollView::scrollView):
        (WebCore::AccessibilityScrollView::accessibilityIsIgnored):
        (WebCore::AccessibilityScrollView::isAccessibilityScrollView):
        (WebCore::toAccessibilityScrollView):
        * accessibility/AccessibilityScrollbar.cpp:
        (WebCore::AccessibilityScrollbar::AccessibilityScrollbar):
        (WebCore::AccessibilityScrollbar::create):
        (WebCore::AccessibilityScrollbar::elementRect):
        (WebCore::AccessibilityScrollbar::document):
        (WebCore::AccessibilityScrollbar::orientation):
        (WebCore::AccessibilityScrollbar::isEnabled):
        (WebCore::AccessibilityScrollbar::valueForRange):
        (WebCore::AccessibilityScrollbar::setValue):
        * accessibility/AccessibilityScrollbar.h:
        (WebCore::AccessibilityScrollbar::setParent):
        (WebCore::AccessibilityScrollbar::canSetValueAttribute):
        (WebCore::AccessibilityScrollbar::canSetNumericValue):
        (WebCore::AccessibilityScrollbar::isAccessibilityScrollbar):
        (WebCore::AccessibilityScrollbar::parentObject):
        (WebCore::AccessibilityScrollbar::roleValue):
        * accessibility/mac/AccessibilityObjectMac.mm:
        (WebCore::AccessibilityObject::accessibilityIgnoreAttachment):
        * accessibility/mac/AccessibilityObjectWrapper.mm:
        (-[AccessibilityObjectWrapper accessibilityAttributeNames]):
        (-[AccessibilityObjectWrapper remoteAccessibilityParentObject]):
        (-[AccessibilityObjectWrapper position]):
        (-[AccessibilityObjectWrapper scrollViewParent]):
        (-[AccessibilityObjectWrapper accessibilityAttributeValue:]):
        (-[AccessibilityObjectWrapper accessibilityHitTest:]):
        (-[AccessibilityObjectWrapper accessibilitySetValue:forAttribute:]):
        * dom/Document.cpp:
        (WebCore::Document::axObjectCache):
        * loader/EmptyClients.h:
        (WebCore::EmptyFrameLoaderClient::accessibilityRemoteObject):
        * loader/FrameLoaderClient.h:
        * page/FrameView.cpp:
        (WebCore::FrameView::~FrameView):
        (WebCore::FrameView::axObjectCache):
        * page/FrameView.h:
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::setHasHorizontalScrollbar):
        (WebCore::ScrollView::setHasVerticalScrollbar):
        * platform/Scrollbar.cpp:
        (WebCore::Scrollbar::~Scrollbar):
        (WebCore::Scrollbar::scroll):
        (WebCore::Scrollbar::axObjectCache):
        * platform/Scrollbar.h:
        * platform/Widget.h:
        (WebCore::Widget::axObjectCache):

2011-01-04  W. James MacLean  <wjmaclean@chromium.org>

        Reviewed by Kenneth Russell.

        [chromium] Add support to compositor to composite to offscreen texture.
        https://bugs.webkit.org/show_bug.cgi?id=50833

        A patch to extend compositor to be able to composite into an offscreen texture instead
        of just directly to the display buffer. Builds on RenderSurfaceChromium support.
        Explicitly resets alpha channel before rendering to display.

        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::LayerRendererChromium):
        (WebCore::LayerRendererChromium::updateAndDrawRootLayer):
        (WebCore::LayerRendererChromium::drawLayers):
        (WebCore::LayerRendererChromium::setCompositeOffscreen):
        (WebCore::LayerRendererChromium::useRenderSurface):
        (WebCore::LayerRendererChromium::setScissorToRect):
        * platform/graphics/chromium/LayerRendererChromium.h:
        (WebCore::LayerRendererChromium::isCompositingOffscreen):
        (WebCore::LayerRendererChromium::getOffscreenLayerTexture):

2011-01-04  Evan Martin  <evan@chromium.org>

        Reviewed by Tony Chang.

        [chromium] depend on harfbuzz explicitly
        https://bugs.webkit.org/show_bug.cgi?id=51895

        platform/graphics/chromium uses Harfbuzz directly.  Rather than
        expecting Skia to adjust the include paths, be explicit about
        the dependency.

        * WebCore.gyp/WebCore.gyp:

2011-01-04  Darin Adler  <darin@apple.com>

        Reviewed by Brady Eidson.

        Add back/forward encoding and decoding to WebKit2
        https://bugs.webkit.org/show_bug.cgi?id=51901

        * WebCore.exp.in: Export functions used in WebKit2.

        * history/HistoryItem.cpp:
        (WebCore::HistoryItem::encodeBackForwardTree): Use references instead of pointers.
        (WebCore::HistoryItem::encodeBackForwardTreeNode): Ditto.
        (WebCore::HistoryItem::decodeBackForwardTree): Ditto.
        * history/HistoryItem.h: Ditto.
        * platform/network/FormData.cpp:
        (WebCore::encode): Ditto.
        (WebCore::decode): Ditto.
        (WebCore::FormData::encodeForBackForward): Ditto.
        (WebCore::FormData::decodeForBackForward): Ditto.
        * platform/network/FormData.h: Ditto.

2011-01-04  Xiaomei Ji  <xji@chromium.org>

        Reviewed by Dan Bernstein.

        Implement sel.modify('move', 'left'/'right', 'lineBoundary').
        https://bugs.webkit.org/show_bug.cgi?id=33435

        * editing/SelectionController.cpp:
        (WebCore::SelectionController::modifyMovingRight):
        (WebCore::SelectionController::modifyMovingLeft):
        * editing/visible_units.cpp:
        (WebCore::logicalStartOfLine):
        (WebCore::logicalEndOfLine):
        (WebCore::leftBoundaryOfLine):
        (WebCore::rightBoundaryOfLine):
        * editing/visible_units.h:

2011-01-04  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Use the default CSS text input control border width of relying on the GTK+ theme
        https://bugs.webkit.org/show_bug.cgi?id=51887

        Do not rely on the GTK+ theme to determine text control borders. This simplifies
        the RenderTheme code, prevents metrics differences between GTK+ 2.x and 3.x and
        doesn't break the rendering of any theme that I've encountered.

        * platform/gtk/RenderThemeGtk.cpp:
        (WebCore::RenderThemeGtk::adjustSearchFieldStyle): No longer call setTextInputBorders, but
        still disable any border radius.
        * platform/gtk/RenderThemeGtk.h: Remove setTextInputBorders and no longer override the
        default style of text areas and text input controls.
        * platform/gtk/RenderThemeGtk2.cpp: Ditto.
        * platform/gtk/RenderThemeGtk3.cpp: Ditto.

2011-01-03  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Darin Adler.

        Remove the last non-GObject usage of PlatformRefPtr and move the code to GRefPtr
        https://bugs.webkit.org/show_bug.cgi?id=51846

        No new tests. This should not introduce any functional changes.

        Changed all references to PlatformRefPtr in GTK+ code to GRefPtr. Changed
        all references to PlatformRefPtr in Brew code to RefPtr.

2011-01-04  David Kilzer  <ddkilzer@apple.com>

        <http://webkit.org/b/51890> Move ResourceLoadPriority.h into WebCore/platform/network

        Reviewed by Antti Koivisto.

        This is needed to prevent a WebCore/platform policy violation in
        my next patch.

        * WebCore.gypi: Updated for moved header.
        * WebCore.pro: Ditto.
        * WebCore.vcproj/WebCore.vcproj: Ditto.
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * platform/network/ResourceLoadPriority.h: Renamed from WebCore/loader/ResourceLoadPriority.h.

2011-01-04  Brady Eidson  <beidson@apple.com>

        Reviewed by a spell checker.

        * storage/StorageMap.h: mesured?  Really?

2011-01-04  Carlos Garcia Campos  <cgarcia@igalia.com> and Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Port stock icon painting to GtkStyleContext
        https://bugs.webkit.org/show_bug.cgi?id=51764

        Port stock icon painting for media and search input elements to
        GtkStyleContext. Also create the initial machinery for accessing
        style contexts for all GTK+ 3.x based widgets.

        No new tests. This should not change functionality.

        * platform/gtk/RenderThemeGtk.cpp:
        (WebCore::paintGdkPixbuf):
        (WebCore::RenderThemeGtk::paintSearchFieldResultsDecoration):
        (WebCore::RenderThemeGtk::paintSearchFieldCancelButton):
        (WebCore::RenderThemeGtk::paintMediaButton):
        * platform/gtk/RenderThemeGtk.h:
        * platform/gtk/RenderThemeGtk2.cpp:
        (WebCore::RenderThemeGtk::getStockIcon):
        * platform/gtk/RenderThemeGtk3.cpp:
        (WebCore::gtkStyleChangedCallback):
        (WebCore::styleContextMap):
        (WebCore::getStyleContext):
        (WebCore::RenderThemeGtk::paintMenuList):

2010-12-31  Antti Koivisto  <antti@apple.com>

        Reviewed by Dave Hyatt.

        Scripts should not be executed before preceding stylesheets are loaded
        https://bugs.webkit.org/show_bug.cgi?id=8852

        Block inline script execution on pending stylesheet loads. This matches
        other browsers and HTML5.

        Tests: fast/tokenizer/inline-script-stylesheet-write.html
               fast/tokenizer/inline-script-stylesheet.html

        * dom/PendingScript.cpp:
        (WebCore::PendingScript::releaseElementAndClear):
        * dom/PendingScript.h:
        (WebCore::PendingScript::PendingScript):
        (WebCore::PendingScript::operator=):
        (WebCore::PendingScript::startingPosition):
        (WebCore::PendingScript::setStartingPosition):
        * html/HTMLLinkElement.cpp:
        (WebCore::HTMLLinkElement::process):
        
        This fixes an unrelated bug with beforeload events that was exposed by the other
        changes (fast/dom/beforeload/remove-link-in-beforeload-listener.html).
        
        * html/parser/HTMLScriptRunner.cpp:
        (WebCore::HTMLScriptRunner::sourceFromPendingScript):
        (WebCore::HTMLScriptRunner::runScript):

2011-01-04  Xianzhu Wang  <wangxianzhu@google.com>

        Reviewed by Darin Adler.

        Let DataView.getInt8() and DataView.setInt8() use int8_t instead of
        ambiguous 'char' to fix fast/canvas/webgl/data-view-test.html failure
        on ARM.
        https://bugs.webkit.org/show_bug.cgi?id=51861

        By the way, changed other getIntXX() and setIntXX() methods to use
        more precise integer types.

        * bindings/js/JSDataViewCustom.cpp:
        (WebCore::setDataViewMember):
        * bindings/v8/custom/V8DataViewCustom.cpp:
        (WebCore::V8DataView::getInt8Callback):
        (WebCore::V8DataView::setInt8Callback):
        * html/canvas/DataView.cpp:
        (WebCore::DataView::getInt8):
        (WebCore::DataView::setInt8):
        * html/canvas/DataView.h:
        * html/canvas/DataView.idl:

2011-01-04  Yong Li  <yoli@rim.com>

        Reviewed by Adam Barth.

        Suspend HTMLParserScheduler when page load is deferred to
        avoid potential JS re-entrancy. Otherwise, when JS execution is triggered
        from an external script or by events, and is blocked by a modal dialog,
        WebKit can parse more HTML source and also start another JS execution.
        
        https://bugs.webkit.org/show_bug.cgi?id=48077

        Test case: WebCore/manual-tests/bugzilla-48077.html.

        * dom/DocumentParser.cpp:
        (WebCore::DocumentParser::suspendScheduledTasks):
        (WebCore::DocumentParser::resumeScheduledTasks):
        * dom/DocumentParser.h:
        * html/parser/HTMLDocumentParser.cpp:
        (WebCore::HTMLDocumentParser::suspendScheduledTasks):
        (WebCore::HTMLDocumentParser::resumeScheduledTasks):
        * html/parser/HTMLParserScheduler.cpp:
        (WebCore::HTMLParserScheduler::HTMLParserScheduler):
        (WebCore::HTMLParserScheduler::suspend):
        (WebCore::HTMLParserScheduler::resume):
        * html/parser/HTMLParserScheduler.h:
        (WebCore::HTMLParserScheduler::isScheduledForResume):
        * page/PageGroupLoadDeferrer.cpp:
        (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
        (WebCore::PageGroupLoadDeferrer::~PageGroupLoadDeferrer):

2011-01-04  Tony Gentilcore  <tonyg@chromium.org>

        Unreviewed build fix.

        Include glib-object.h for GObject
        https://bugs.webkit.org/show_bug.cgi?id=51876

        * bindings/gobject/DOMObjectCache.cpp:

2011-01-04  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Dan Bernstein.

        Clicking on the first or the last letter of LTR/RTL text in a RTL/LTR block puts caret on the opposite side.
        https://bugs.webkit.org/show_bug.cgi?id=50992

        Fixed the bug by interchanging the offset when the direction of inline text box and the containing block
        does not match. Reused the code added by http://trac.webkit.org/changeset/73553.

        Test: editing/selection/caret-bidi-first-and-last-letters.html

        * rendering/InlineTextBox.cpp:
        (WebCore::InlineTextBox::offsetForPosition):

2010-12-29  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Darin Adler.

        Remove 2 unnecessary includes from Document.h
        https://bugs.webkit.org/show_bug.cgi?id=51727

        No new tests because no changed functionality.

        * dom/Document.cpp:
        * dom/Document.h:
        * editing/Editor.cpp:
        * editing/SpellChecker.h:
        * editing/SplitTextNodeCommand.cpp:
        * editing/TextCheckingHelper.cpp:
        * page/FrameView.cpp:
        * page/Page.cpp:
        * rendering/HitTestResult.cpp:
        * rendering/InlineTextBox.cpp:
        * rendering/InlineTextBox.h:
        * rendering/svg/SVGInlineFlowBox.cpp:

2011-01-04  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed WinCE build fix.

        * CMakeListsWinCE.txt: Added missing include directories, files and libraries.

2011-01-04  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Simon Hausmann.

        Provide an interface to require using fullscreen mediaplayer
        https://bugs.webkit.org/show_bug.cgi?id=51133

        Only video should be displayed in fullscreen when 
        requiresFullscreenForVideoPlayback is specified.

        No new tests because no client implements requiresFullscreenForVideoPlayback yet.

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::updatePlayState):

2011-01-04  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] [WK2] create an initial implementation of the context menu handling for WebKit 2
        https://bugs.webkit.org/show_bug.cgi?id=51825

        Implement contextMenuItemVector() for Qt, this method is used in WebKit 2 to get the
        platform independent menu informations.

        * platform/qt/ContextMenuQt.cpp:
        (WebCore::contextMenuItemVector):

2011-01-04  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r74960.
        http://trac.webkit.org/changeset/74960
        https://bugs.webkit.org/show_bug.cgi?id=51868

        Rollout since we hit an assert after this patch. (Requested by
        krit on #webkit).

        * svg/SVGUseElement.cpp:
        (WebCore::SVGUseElement::buildInstanceTree):

2011-01-04  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r74961.
        http://trac.webkit.org/changeset/74961
        https://bugs.webkit.org/show_bug.cgi?id=51867

        breaks 32/64-bits debug bots build (Requested by philn-tp on
        #webkit).

        * platform/graphics/gstreamer/GStreamerGWorld.cpp:
        (WebCore::GStreamerGWorld::setWindowOverlay):

2011-01-03  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Eric Seidel.

        [GStreamer] build fails with gstreamer core 0.10.31 and -plugins-base 0.10.30
        https://bugs.webkit.org/show_bug.cgi?id=51810

        Based on a patch by Mart Raudsepp.

        No new tests, build fix only.

        * platform/graphics/gstreamer/GStreamerGWorld.cpp:
        (WebCore::GStreamerGWorld::setWindowOverlay): Use the new xoverlay
        API if the GST_CHECK_PLUGINS_BASE_VERSION macro is defined. It was
        introduced in same version as the xoverlay API change.

2011-01-04  Leo Yang  <leo.yang@torchmobile.com.cn>

        Reviewed by Dirk Schulze.

        Malformed SVG causes crash in updateContainerOffset
        https://bugs.webkit.org/show_bug.cgi?id=44610

        We should check recursive <use> at the begining of
        WebCore::SVGUseElement::buildInstanceTree instead
        of at the end of it because the target element's
        children may cause infinite recursive <use>.

        Test: svg/custom/recursive-use.svg

        * svg/SVGUseElement.cpp:
        (WebCore::SVGUseElement::buildInstanceTree):

2011-01-04  Kent Tamura  <tkent@chromium.org>

        Unreviewed, build fix.

        Try to fix Clang build.

        * css/CSSParser.cpp:
        (WebCore::valueFromSideKeyword):
        (WebCore::CSSParser::parseLinearGradient):

2011-01-03  Julie-Jeongeun-Kim  <Jeongeun.Kim@obigo.com>

        Reviewed by Darin Adler.

        HTML5 Conformance Test failure: approved/xhtml5/object_border_pixel.xhtml
        https://bugs.webkit.org/show_bug.cgi?id=48596
        
        border attribute was not parsed in HTMLObjectElement.
        I added code for it as HTMLImageElement handled.

        new test:fast/html/object-border.html

        * html/HTMLObjectElement.cpp:
        (WebCore::HTMLObjectElement::parseMappedAttribute):

2011-01-03  Darin Adler  <darin@apple.com>

        Try to fix Chromium build.

        * history/HistoryItem.cpp:
        (WebCore::HistoryItem::encodeBackForwardTreeNode): Added a V8 case here.
        Not sure this ever will be used with V8.
        (WebCore::HistoryItem::decodeBackForwardTree): Ditto.

2011-01-03  Darin Adler  <darin@apple.com>

        Try to fix Leopard build.

        * history/HistoryItem.cpp: Use a consistent type, uint64_t, for the
        size variable.

2011-01-03  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Darin Adler.

        Fixes before child calculation when adding anonymous childs to table parts.
        https://bugs.webkit.org/show_bug.cgi?id=50932

        Fix in r74364 was incomplete. When before child is equal to the table part(to
        which the new child is getting added), it confuses the table part to add it
        incorrectly as an after child. The patch fixes by passing the before child as
        the table part's first child.

        Tests: fast/css-generated-content/table-before-after-child-add.html
               fast/css-generated-content/table-cell-before-after-child-add.html
               fast/css-generated-content/table-row-before-after-child-add.html
               fast/css-generated-content/table-row-before-after-child-add.html

        * rendering/RenderTable.cpp:
        (WebCore::RenderTable::addChild):
        * rendering/RenderTableRow.cpp:
        (WebCore::RenderTableRow::addChild):
        * rendering/RenderTableSection.cpp:
        (WebCore::RenderTableSection::addChild):

2011-01-03  Dan Bernstein  <mitz@apple.com>

        Reviewed by Simon Fraser.

        Shrink HTMLTableCellElement
        https://bugs.webkit.org/show_bug.cgi?id=51858

        * html/HTMLTableCellElement.cpp:
        (WebCore::HTMLTableCellElement::HTMLTableCellElement): Removed intialization of m_row and
        m_col.
        * html/HTMLTableCellElement.h: Removed m_row and m_col and their accessors.

2011-01-03  Darin Adler  <darin@apple.com>

        Fix build.

        * platform/network/FormData.cpp:
        (WebCore::FormData::create): Roll back uint8_t change.
        (WebCore::FormData::flatten): Ditto.
        (WebCore::FormData::flattenToString): Ditto.
        (WebCore::encode): Ditto.
        (WebCore::decode): Ditto.
        (WebCore::FormData::encodeForBackForward):
        * platform/network/FormData.h: Ditto.
        * platform/network/mac/FormDataStreamMac.mm:
        (WebCore::advanceCurrentStream): Ditto.

2011-01-03  Darin Adler  <darin@apple.com>

        Reviewed by Brady Eidson.

        Add code to encode/decode the back/forward tree
        https://bugs.webkit.org/show_bug.cgi?id=51853

        * history/HistoryItem.cpp:
        (WebCore::HistoryItem::encodeBackForwardTree): Added.
        (WebCore::HistoryItem::encodeBackForwardTreeNode): Added.
        (WebCore::HistoryItem::decodeBackForwardTree): Added.
        * history/HistoryItem.h: Added functions above.

        * platform/network/FormData.cpp:
        (WebCore::FormData::create): Changed type to Vector<uint8_t> instead of
        Vector<char>. This type is used in our newer code for vectors of bytes.
        (WebCore::FormData::flatten): Ditto.
        (WebCore::FormData::flattenToString): Ditto. Had to add a cast here.
        (WebCore::FormData::encodeForBackForward): Added.
        (WebCore::FormData::decodeForBackForward): Added.
        * platform/network/FormData.h: Changed type to Vector<uint8_t> instead of
        Vector<char>. Also made some single-argument constructors explicit.
        Also added the new functions.

        * platform/network/mac/FormDataStreamMac.mm:
        (WebCore::advanceCurrentStream): Changed to use uint8_t to match above.

2011-01-03  Simon Fraser  <simon.fraser@apple.com>

        Attempt to fix GTK build.

        * platform/graphics/gtk/FontGtk.cpp:
        (WebCore::drawGlyphsShadow):

2011-01-03  Helder Correia  <helder@sencha.com>

        Reviewed by Simon Fraser.

        ContextShadow can avoid code duplication for getting the CTM
        https://bugs.webkit.org/show_bug.cgi?id=51799

        ContextShadow defines its own getTransformationMatrixFromContext()
        which does exactly the same as GraphicsContext::getCTM().

        Existing tests: fast/canvas/canvas*shadow*html

        * platform/graphics/ContextShadow.cpp:
        (WebCore::ContextShadow::mustUseContextShadow):
        (WebCore::ContextShadow::adjustBlurDistance):
        (WebCore::ContextShadow::calculateLayerBoundingRect):
        * platform/graphics/ContextShadow.h:
        * platform/graphics/cairo/ContextShadowCairo.cpp:
        (WebCore::ContextShadow::beginShadowLayer):
        (WebCore::ContextShadow::endShadowLayer):
        (WebCore::ContextShadow::drawRectShadowWithoutTiling):
        * platform/graphics/cairo/FontCairo.cpp:
        (WebCore::drawGlyphsShadow):
        * platform/graphics/cairo/GraphicsContextCairo.cpp:
        (WebCore::drawPathShadow):
        * platform/graphics/cairo/ImageCairo.cpp:
        (WebCore::BitmapImage::draw):
        * platform/graphics/gtk/FontGtk.cpp:
        (WebCore::drawGlyphsShadow):
        * platform/graphics/qt/ContextShadowQt.cpp:
        (WebCore::ContextShadow::beginShadowLayer):
        (WebCore::ContextShadow::endShadowLayer):
        * platform/graphics/qt/FontQt.cpp:
        (WebCore::drawTextCommon):
        * platform/graphics/qt/GraphicsContextQt.cpp:
        (WebCore::GraphicsContext::fillPath):
        (WebCore::GraphicsContext::strokePath):
        (WebCore::GraphicsContext::fillRect):
        (WebCore::GraphicsContext::fillRoundedRect):
        * platform/graphics/qt/ImageQt.cpp:
        (WebCore::BitmapImage::draw):
        * platform/graphics/qt/StillImageQt.cpp:
        (WebCore::StillImage::draw):

2011-01-03  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        InsertTextCommand::m_charactersAdded is unused
        https://bugs.webkit.org/show_bug.cgi?id=51849

        * editing/InsertTextCommand.cpp:
        (WebCore::InsertTextCommand::InsertTextCommand): Don't initialize m_charactersAdded.
        (WebCore::InsertTextCommand::input): Don't update m_charactersAdded.
        * editing/InsertTextCommand.h: Removed charactersAdded function and
        m_charactersAdded data member.

2011-01-03  Carlos Garcia Campos  <cgarcia@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Move paintMenuListButton() to RenderThemeGtk.cpp
        https://bugs.webkit.org/show_bug.cgi?id=51832

        Method paintMenuListButton() contains code common to both gtk 2
        and 3, so it can be moved to RenderThemeGtk.cpp instead of
        duplicating it.

        * platform/gtk/RenderThemeGtk.cpp:
        (WebCore::RenderThemeGtk::paintMenuListButton):
        * platform/gtk/RenderThemeGtk2.cpp:
        * platform/gtk/RenderThemeGtk3.cpp:

2011-01-03  Pratik Solanki  <psolanki@apple.com>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=51850
        Put CFNetwork loader code inside USE(CFNETWORK)

        Add USE(CFNETWORK) macros around all the cf/mac network files. Also added all CFNetwork
        files to the WebCore Xcode project.

        * WebCore.xcodeproj/project.pbxproj:
        * loader/cf/ResourceLoaderCFNet.cpp:
        * loader/mac/ResourceLoaderMac.mm:
        * platform/network/cf/AuthenticationCF.cpp:
        * platform/network/cf/AuthenticationCF.h:
        * platform/network/cf/CookieJarCFNet.cpp:
        * platform/network/cf/CookieStorageCFNet.cpp:
        * platform/network/cf/CookieStorageCFNet.h:
        * platform/network/cf/CredentialStorageCFNet.cpp:
        * platform/network/cf/FormDataStreamCFNet.cpp:
        * platform/network/cf/FormDataStreamCFNet.h:
        * platform/network/cf/LoaderRunLoopCF.cpp:
        * platform/network/cf/LoaderRunLoopCF.h:
        * platform/network/cf/ResourceError.h:
        * platform/network/cf/ResourceHandleCFNet.cpp:
        * platform/network/cf/ResourceRequest.h:
        * platform/network/cf/ResourceRequestCFNet.cpp:
        * platform/network/cf/ResourceRequestCFNet.h:
        * platform/network/cf/ResourceResponse.h:
        * platform/network/cf/ResourceResponseCFNet.cpp:
        * platform/network/mac/AuthenticationMac.h:
        * platform/network/mac/AuthenticationMac.mm:
        * platform/network/mac/FormDataStreamMac.h:
        * platform/network/mac/FormDataStreamMac.mm:
        * platform/network/mac/ResourceError.h:
        * platform/network/mac/ResourceHandleMac.mm:
        * platform/network/mac/ResourceRequest.h:
        * platform/network/mac/ResourceRequestMac.mm:
        * platform/network/mac/ResourceResponse.h:
        * platform/network/mac/ResourceResponseMac.mm:

2011-01-03  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Adam Barth.

        [Qt] Add SelectAll option to the context menu for the editor
        https://bugs.webkit.org/show_bug.cgi?id=50049

        Enable SelectAll for the Qt context menu.

        * page/ContextMenuController.cpp:
        (WebCore::ContextMenuController::populate):
        * platform/ContextMenuItem.h:
        * platform/LocalizationStrategy.h:
        * platform/LocalizedStrings.cpp:
        (WebCore::contextMenuItemTagSelectAll):
        * platform/LocalizedStrings.h:

2011-01-03  Dan Bernstein  <mitz@apple.com>

        Reviewed by Darin Adler.

        Shrink RootInlineBox
        https://bugs.webkit.org/show_bug.cgi?id=51847

        * rendering/RootInlineBox.cpp:
        (WebCore::RootInlineBox::alignBoxesInBlockDirection): Changed to use the baselineType()
        accessor.
        * rendering/RootInlineBox.h: Rolled m_baselineType into a bitfield. Changed its type to
        unsigned because otherwise MSVC treats it as signed.
        (WebCore::RootInlineBox::baselineType): Added a cast.

2011-01-03  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Martin Robinson.

        [GTK] Split webkitprivate.{cpp,h} in more manageable chunks
        https://bugs.webkit.org/show_bug.cgi?id=50698

        Fix DOM bindings generation to no longer include the now-gone
        webkitprivate.h header, including the appropriate replacements
        instead.

        * bindings/scripts/CodeGeneratorGObject.pm:

2011-01-03  Brady Eidson  <beidson@apple.com>

        Reviewed by Darin Adler.

        Add Encode/Decode machinery Darin and I plan to work with for back/forward stuff in WebKit2.

        * ForwardingHeaders/wtf/Decoder.h: Added.
        * ForwardingHeaders/wtf/Encoder.h: Added.

2011-01-03  Chris Fleizach  <cfleizach@apple.com>

        Reviewed by Darin Adler.

        AX: Regression: WebKit no longer auto-scrolls with VoiceOver navigation
        https://bugs.webkit.org/show_bug.cgi?id=51833

        Make sure the web area returns the contentsSize, not the same size as what is displayed.

        Test: platform/mac/accessibility/webarea-size-equals-content-size.html

        * accessibility/AccessibilityRenderObject.cpp:
        (WebCore::AccessibilityRenderObject::boundingBoxRect):

2011-01-03  Simon Fraser  <simon.fraser@apple.com>

        Fix warning in 32-bit builds.

        * css/CSSGradientValue.cpp:
        (WebCore::CSSLinearGradientValue::createGradient):

2011-01-03  Benjamin C Meyer  <bmeyer@rim.com>

        Reviewed by Daniel Bates.

        In the cmake build system move logic related to ENABLE_NOTIFICATIONS into its own block.
        https://bugs.webkit.org/show_bug.cgi?id=51827

        * CMakeLists.txt:

2011-01-03  Dan Bernstein  <mitz@apple.com>

        Reviewed by Simon Fraser.

        Remove the unused RenderTableCell::m_percentageHeight
        https://bugs.webkit.org/show_bug.cgi?id=51837

        * rendering/RenderTableCell.cpp:
        (WebCore::RenderTableCell::RenderTableCell): Removed initialization of m_percentageHeight.
        * rendering/RenderTableCell.h: Removed m_percentageHeight.

2011-01-03  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by David Hyatt.

        Implement -webkit-linear-gradient and -webkit-radial-gradient
        https://bugs.webkit.org/show_bug.cgi?id=28152

        Add support for -webkit-radial-gradients.

        Tests: fast/gradients/css3-radial-gradients.html
               fast/gradients/css3-radial-gradients2.html

        * css/CSSGradientValue.cpp:
        (WebCore::blend): Used to blend colors, which is necessary when truncating
        the start of radial gradients.
        (WebCore::GradientStop): Small struct to aid gradient stop processing.
        (WebCore::CSSGradientValue::addStops): Rename 'positions' to 'stops', and store
        the color so that we can blend it when truncating radial gradients. Changed
        to handle both linear and radial gradients.
        (WebCore::CSSRadialGradientValue::cssText): Update to follow the spec.
        (WebCore::CSSRadialGradientValue::resolveRadius): Stylistic change.
        (WebCore::distanceToClosestCorner): New utility method.
        (WebCore::distanceToFarthestCorner): Ditto.
        (WebCore::CSSRadialGradientValue::createGradient): New logic to deal with shape and
        fill rules.

        * css/CSSGradientValue.h:
        (WebCore::CSSGradientValue::isLinearGradient): Color-stop processing needs to know what
        kind of gradient it's dealing with, so add new isFooGradient methods.
        (WebCore::CSSGradientValue::isRadialGradient): Ditto.
        (WebCore::CSSLinearGradientValue::isLinearGradient):
        (WebCore::CSSRadialGradientValue::setShape): New setters for shape, size etc.
        (WebCore::CSSRadialGradientValue::setSizingBehavior):
        (WebCore::CSSRadialGradientValue::setEndHorizontalSize):
        (WebCore::CSSRadialGradientValue::setEndVerticalSize):
        (WebCore::CSSRadialGradientValue::isRadialGradient):

        * css/CSSParser.h:
        * css/CSSParser.h: Pass a CSSParserValueList* into parseFillPositionXY() and parseFillPosition()
        so we can use parseFillPosition() when parsing a gradient function.
        * css/CSSParser.cpp:
        (WebCore::CSSParser::parseFillPositionXY): Pass in a CSSParserValueList*
        (WebCore::CSSParser::parseFillPosition): Ditto
        (WebCore::CSSParser::parseFillProperty): Pass m_valueList to parseFillPosition.
        (WebCore::CSSParser::parseTransformOriginShorthand): Ditto
        (WebCore::CSSParser::parseRadialGradient): Flesh out radial gradient parsing.
        (WebCore::CSSParser::parseTransformOrigin): Pass m_valueList to parseFillPosition.
        (WebCore::CSSParser::parsePerspectiveOrigin): Ditto

        * css/CSSValueKeywords.in: New keywords for radial gradient shape and fill behaviors.

        * platform/graphics/Gradient.cpp:
        (WebCore::Gradient::Gradient): Pass aspect ratio for elliptical gradients.
        * platform/graphics/Gradient.h: Add aspect ratio for elliptical gradients.
        (WebCore::Gradient::create):
        (WebCore::Gradient::startRadius): Expose radii.
        (WebCore::Gradient::endRadius):
        (WebCore::Gradient::setStartRadius): Setters are required for when we need to scale color stops.
        (WebCore::Gradient::setEndRadius):
        (WebCore::Gradient::aspectRatio):
        * platform/graphics/cg/GradientCG.cpp:
        (WebCore::Gradient::paint): For elliptical gradients, scale the CTM.
        * platform/graphics/wince/GraphicsContextWinCE.cpp:
        (WebCore::GraphicsContext::fillRect): Use start/endRadius() rather than r0() and r1().

2011-01-03  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by David Hyatt.

        Implement -webkit-linear-gradient and -webkit-radial-gradient
        https://bugs.webkit.org/show_bug.cgi?id=28152
        
        Add support for the parsing and rendering of non-repeating CSS3 linear gradients,
        according to <http://dev.w3.org/csswg/css3-images/#linear-gradients>.

        Tests: fast/gradients/css3-color-stop-units.html
               fast/gradients/css3-color-stops.html
               fast/gradients/css3-gradient-parsing.html
               fast/gradients/css3-linear-angle-gradients.html

        * css/CSSGradientValue.h:
        Add subclasses of CSSGradientValue for linear and radial gradients.
        
        * css/CSSGradientValue.cpp:
        (WebCore::CSSGradientValue::image): Add FIXME comment about why we can't cache
        the gradient sometimes.
        (WebCore::compareStops): positions are CSSValues now.
        (WebCore::CSSGradientValue::sortStopsIfNeeded): Add assertion that this is only called
        for old-style gradients.
        (WebCore::CSSGradientValue::addStops): New method that handles the spec rules about
        missing positions, ensuring that stop position is ordered, and spacing out stops
        without positions. Also normalize gradients which use stops outside the 0-1 range
        so that platforms can render them correctly.
        (WebCore::positionFromValue): Determin the position on the gradient line for percentage,
        number and length values.
        (WebCore::CSSGradientValue::computeEndPoint):
        (WebCore::CSSLinearGradientValue::cssText):
        (WebCore::endPointsFromAngle): Compute endpoints for a gradient which covers the rectangle
        given an angle.
        (WebCore::CSSLinearGradientValue::createGradient):
        (WebCore::CSSRadialGradientValue::cssText):
        (WebCore::CSSRadialGradientValue::resolveRadius):
        (WebCore::CSSRadialGradientValue::createGradient):

        * css/CSSParser.cpp:
        (WebCore::CSSParser::parseValue): Style fix.
        (WebCore::CSSParser::parseContent): Use isGeneratedImageValue() and parseGeneratedImage()
        (WebCore::CSSParser::parseFillImage): Use isGeneratedImageValue() and parseGeneratedImage()
        (WebCore::CSSParser::parseBorderImage): Use isGeneratedImageValue() and parseGeneratedImage()
        (WebCore::parseDeprecatedGradientPoint): Renamed to indicate that it's only used for -webkit-gradient.
        (WebCore::parseDeprecatedGradientColorStop): Renamed to indicate that it's only used for -webkit-gradient,
        and fixed to store CSSValues for the stop position.
        (WebCore::CSSParser::parseDeprecatedGradient): Renamed to indicate that it's only used for -webkit-gradient,
        and create CSSLinearGradientValue and CSSRadialGradientValue as appropriate.
        (WebCore::valueFromSideKeyword):
        (WebCore::parseGradientColorOrKeyword): Color parsing utility.
        (WebCore::CSSParser::parseLinearGradient): New method to parse linear gradients.
        (WebCore::CSSParser::parseRadialGradient): Stub for new method to parse radial gradients.
        (WebCore::CSSParser::isGeneratedImageValue): Detect gradients and CSS canvas.
        (WebCore::CSSParser::parseGeneratedImage): Parser for generated images like gradients
        and CSS canvas.
        * css/CSSParser.h:
        * platform/graphics/Gradient.h:
        (WebCore::Gradient::setP0): We need to be able to change the points when normalizing gradients.
        (WebCore::Gradient::setP1): Ditto.

2011-01-03  Noam Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Andreas Kling.

        [Qt] document.getElementById(...) doesn't return the right object in combination with QGraphicsWidget
        https://bugs.webkit.org/show_bug.cgi?id=51464

        Added a way to bind any QObject created as a plugin to JavaScript, 
        by adding a custom member (qtObject) to WebCore::Widget. 
        Added a test to tst_qwebpage to make sure plugins created 
        as QGraphicsWidget are accessible through JavaScript.

        * bindings/js/ScriptControllerQt.cpp:
        (WebCore::ScriptController::createScriptInstanceForWidget):
        * platform/Widget.h:
        * platform/qt/WidgetQt.cpp:
        (WebCore::Widget::Widget):
        (WebCore::Widget::setBindingObject):
        (WebCore::Widget::bindingObject):

2011-01-03  David Hyatt  <hyatt@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=51328

        Printing on a page with body { height: 100% } clips out the overflow contents.  This happened
        because the computePageRects function was never correct.  It was using the layer's width and height
        for the document's width and height, and the layer no longer includes overflow in its width and
        height.  It was also incorrectly assuming the first page would begin at (0,0), and this is an invalid
        assumption in RTL or vertical text environments.

        Added printing/page-count-percentage-height.html

        * page/PrintContext.cpp:
        (WebCore::PrintContext::computePageRects):
        (WebCore::PrintContext::computePageRectsWithPageSizeInternal):

2011-01-03  Csaba Osztrogonác  <ossy@webkit.org>

        [Qt] Unreviewed buildfix for --minimal build after r74895.

        * html/InputType.cpp: Missing include added.
        * html/NumberInputType.cpp: Missing include added.

2011-01-02  Darin Adler  <darin@apple.com>

        Reviewed by Kent Tamura.

        Eliminate HTMLInputElement::m_deprecatedTypeNumber, other input refactoring and renaming
        https://bugs.webkit.org/show_bug.cgi?id=51791

        * dom/InputElement.h: Make sanitizeValueForTextField public so it can
        be called by InputType.

        * html/BaseButtonInputType.cpp:
        (WebCore::BaseButtonInputType::accessKeyAction): Added.
        (WebCore::BaseButtonInputType::storesValueSeparateFromAttribute): Added.
        * html/BaseButtonInputType.h: Ditto.

        * html/BaseCheckableInputType.cpp:
        (WebCore::BaseCheckableInputType::canSetStringValue): Added.
        (WebCore::BaseCheckableInputType::accessKeyAction): Added.
        (WebCore::BaseCheckableInputType::fallbackValue): Added.
        (WebCore::BaseCheckableInputType::storesValueSeparateFromAttribute): Added.
        (WebCore::BaseCheckableInputType::isCheckable): Added.
        * html/BaseCheckableInputType.h: Ditto.

        * html/BaseDateAndTimeInputType.cpp:
        (WebCore::BaseDateAndTimeInputType::hasSpinButton): Added.
        * html/BaseDateAndTimeInputType.h: Ditto.

        * html/BaseTextInputType.h: Made more functions private.

        * html/ButtonInputType.cpp:
        (WebCore::ButtonInputType::isTextButton): Added.
        * html/ButtonInputType.h: Ditto.

        * html/CheckboxInputType.cpp:
        (WebCore::CheckboxInputType::willDispatchClick): Added.
        (WebCore::CheckboxInputType::didDispatchClick): Added.
        (WebCore::CheckboxInputType::isCheckbox): Added.
        * html/CheckboxInputType.h: Ditto.

        * html/DateTimeInputType.cpp:
        (WebCore::DateTimeInputType::scaledStepValueShouldBeInteger): Fixed
        typo in name of this function.
        * html/DateTimeInputType.h: Ditto.
        * html/DateTimeLocalInputType.cpp:
        (WebCore::DateTimeLocalInputType::scaledStepValueShouldBeInteger): Ditto.
        * html/DateTimeLocalInputType.h: Ditto.

        * html/EmailInputType.cpp:
        (WebCore::EmailInputType::isEmailField): Added.
        * html/EmailInputType.h: Ditto.

        * html/FileInputType.cpp:
        (WebCore::FileInputType::FileInputType): Added.
        (WebCore::FileInputType::canSetStringValue): Added.
        (WebCore::FileInputType::canChangeFromAnotherType): Added.
        (WebCore::FileInputType::files): Added.
        (WebCore::FileInputType::canSetValue): Added.
        (WebCore::FileInputType::getTypeSpecificValue): Added.
        (WebCore::FileInputType::storesValueSeparateFromAttribute): Added.
        (WebCore::FileInputType::setFileList): Added.
        (WebCore::FileInputType::isFileUpload): Added.
        * html/FileInputType.h: Ditto.

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::HTMLInputElement): Updated for data member
        renaming and removal.
        (WebCore::HTMLInputElement::isValidValue): Call canSetStringValue
        instead of checking specific type numbers.
        (WebCore::HTMLInputElement::getAllowedValueStepWithDecimalPlaces):
        Use fastGetAttribute and fixed typo in the name of the
        scaledStepValueShouldBeInteger function.
        (WebCore::HTMLInputElement::isKeyboardFocusable): Call
        InputType::isKeyboardFocusable; moved most of the type-specific
        logic there.
        (WebCore::HTMLInputElement::shouldUseInputMethod): Call
        InputType::shouldUseInputMethod; moved the type-specific logic there.
        (WebCore::HTMLInputElement::handleBlurEvent): Call
        InputType::handleBlurEvent; moved the type-specific logic there.
        (WebCore::HTMLInputElement::setType): Tweak the names and types
        in this function.
        (WebCore::HTMLInputElement::updateType): Removed the code that
        set m_deprecatedTypeNumber. Compare types by comparing the value
        from formControlType. Removed file-specific logic and instead call
        InputType::canChangeFromAnotherType. Streamlined the logic for
        activation callbacks.
        (WebCore::HTMLInputElement::accessKeyAction): Call
        InputType::accessKeyAction; moved the type-specific logic there.
        (WebCore::HTMLInputElement::mapToEntry): Call
        InputType::shouldRespectHeightAndWidthAttributes and
        InputType::shouldRespectAlignAttribute.
        (WebCore::HTMLInputElement::parseMappedAttribute): Updated
        for name changes and the removal of m_defaultChecked. Changed the
        alt attribute logic to be based on the class of the renderer.
        Call InputType::srcAttributeChanged and moved the type-specific
        logic there. Call shouldRespectAlignAttribute and
        shouldRespectHeightAndWidthAttributes rather than having
        type-specific logic here. Call minOrMaxAttributeChanged.
        (WebCore::HTMLInputElement::rendererIsNeeded): Call
        InputType::rendererIsNeeded; moved the type-specific logic there.
        (WebCore::HTMLInputElement::attach): Removed the unneeded m_inited
        flag, which was redundant with the m_haveType flag. Call
        InputType::attach; moved the type-specific logic there.
        (WebCore::HTMLInputElement::altText): Use fastGetAttribute.
        (WebCore::HTMLInputElement::isSuccessfulSubmitButton): Call
        InputType::canBeSuccessfulSubmitButton; moved the type-specific
        logic there.
        (WebCore::HTMLInputElement::isActivatedSubmit): Updated for
        name changes.
        (WebCore::HTMLInputElement::setActivatedSubmit): Ditto.
        (WebCore::HTMLInputElement::reset): Updated to use
        InputType::storesValueSeparateFromAttribute. Also use
        hasAttribute(checkedAttr) rather than m_defaultChecked,
        since both are the same. Also updated for name change.
        (WebCore::HTMLInputElement::setChecked): Updated for name
        changes. Also added call to
        InputType::shouldSendChangeEventAfterCheckedChanged and
        moved the type-specific logic there.
        (WebCore::HTMLInputElement::setIndeterminate): Call
        isCheckable to see if indeterminate is allowed.
        Also updated for name changes.
        (WebCore::HTMLInputElement::copyNonAttributeProperties):
        Updated for name changes and removed data members.
        (WebCore::HTMLInputElement::value): Call
        InputType::getTypeSpecificValue and InputType::fallbackValue.
        Moved the type-specific logic there.
        (WebCore::HTMLInputElement::valueWithDefault): Call
        InputType::defaultValue. Moved the type-specific logic there.
        (WebCore::HTMLInputElement::setSuggestedValue): Call
        InputType::canSetSuggestedValue. Moved the type-specific logic
        there.
        (WebCore::HTMLInputElement::setValue): Call
        InputType::canSetValue. Moved the type-specific logic there.
        Clear the file list if it exists rather than checking the
        type explicitly. We could probably do that a more elegant way.
        (WebCore::HTMLInputElement::placeholder): Use fastGetAttribute.
        (WebCore::HTMLInputElement::setValueFromRenderer): Use the
        isFileUpload function in the assertion.
        (WebCore::HTMLInputElement::setFileListFromRenderer): Put most
        of the code from this function into FileInputType::setFileList.
        (WebCore::HTMLInputElement::preDispatchEventHandler): Changed
        to call InputType::willDispatchClick, which does the type-specific
        work. Still need to use a single struct for all types, because
        the type of the element could change during the event handling.
        (WebCore::HTMLInputElement::postDispatchEventHandler): Changed
        to call InputType::didDispatchClick.
        (WebCore::HTMLInputElement::defaultValue): Use fastGetAttribute.
        (WebCore::HTMLInputElement::accept): Use fastGetAttribute.
        (WebCore::HTMLInputElement::alt): Use fastGetAttribute.
        (WebCore::HTMLInputElement::src): Use fastGetAttribute.
        (WebCore::HTMLInputElement::setAutofilled): Updated for name
        changes.
        (WebCore::HTMLInputElement::files): Use InputType::files.
        (WebCore::HTMLInputElement::isAcceptableValue): Use
        InputType::isAcceptableValue.
        (WebCore::HTMLInputElement::sanitizeValue): Use
        InputType::sanitizeValue.
        (WebCore::HTMLInputElement::hasUnacceptableValue): Use
        InputType::hasUnacceptableValue.
        (WebCore::HTMLInputElement::needsActivationCallback): Use
        InputType::shouldResetOnDocumentActivation.
        (WebCore::HTMLInputElement::willMoveToNewOwnerDocument): Use
        InputType::willMoveToNewOwnerDocument.
        (WebCore::HTMLInputElement::dataList): Use
        InputType::shouldRespectListAttribute.
        (WebCore::HTMLInputElement::selectedOption): Tweaked logic
        and naming a bit.
        (WebCore::HTMLInputElement::stepUpFromRenderer): Use
        fastGetAttribute.
        (WebCore::HTMLInputElement::isSpeechEnabled): Use
        InputType::shouldRespectSpeechAttribute.
        (WebCore::HTMLInputElement::isTextButton): No longer inline.
        Calls through to InputType.
        (WebCore::HTMLInputElement::isRadioButton): Ditto.
        (WebCore::HTMLInputElement::isSearchField): Ditto.
        (WebCore::HTMLInputElement::isInputTypeHidden): Ditto.
        (WebCore::HTMLInputElement::isPasswordField): Ditto.
        (WebCore::HTMLInputElement::isCheckbox): Ditto.
        (WebCore::HTMLInputElement::isText): Ditto.
        (WebCore::HTMLInputElement::isEmailField): Ditto.
        (WebCore::HTMLInputElement::isFileUpload): Ditto.
        (WebCore::HTMLInputElement::isImageButton): Ditto.
        (WebCore::HTMLInputElement::isNumberField): Ditto.
        (WebCore::HTMLInputElement::isSubmitButton): Ditto.
        (WebCore::HTMLInputElement::isTelephoneField): Ditto.
        (WebCore::HTMLInputElement::isURLField): Ditto.
        (WebCore::HTMLInputElement::isEnumeratable): Ditto.
        (WebCore::HTMLInputElement::isChecked): No longer inline.
        Calls InputType::isCheckable.
        (WebCore::HTMLInputElement::hasSpinButton): No longer inline.
        Calls through to InputType.
        (WebCore::HTMLInputElement::supportsPlaceholder): No longer
        inline. Calls isTextType as before, but with no special case
        for ISINDEX. That special case was unneeded since it returns
        true from isTextType.
        (WebCore::HTMLInputElement::checkedRadioButtons): No longer
        inline.
        (WebCore::HTMLInputElement::handleBeforeTextInsertedEvent):
        No longer inline.

        * html/HTMLInputElement.h: Removed multiple unneeded includes.
        The big one is probably InputType.h, but also HTMLFormElement.h.
        Removed unneeded forward declarations. Made many functions
        no longer inline since they now need to make virtual function
        dispatches. Removed the unneeded defaultChecked and
        webkitdirectory functions. Both are handled by HTML attribute
        reflection for DOM bindings, and the callers instead can get
        at the attribute. Made updateCheckedRadioButtons public so it
        can be called by an InputType class. Removed
        DeprecatedInputType, InputTypeMap, createTypeMap,
        deprecatedInputType, allowsIndeterminate,
        respectHeightAndWidthAttrs, storesValueSeparateFromAttribute,
        m_imageLoader, m_fileList, m_deprecatedTypeNumber,
        m_defaultChecked, and m_inited. Renamed m_checked to
        m_isChecked, m_useDefaultChecked to m_reflectsCheckedAttribute,
        m_indeterminate tom_isIndeterminate, m_haveType to m_hasType,
        m_activeSubmit to m_isActivatedSubmit, and m_autofilled to
        m_isAutofilled.
        
        * html/HiddenInputType.cpp:
        (WebCore::HiddenInputType::accessKeyAction): Added.
        (WebCore::HiddenInputType::rendererIsNeeded) Added.:
        (WebCore::HiddenInputType::storesValueSeparateFromAttribute): Added.
        (WebCore::HiddenInputType::isHiddenType): Added.
        (WebCore::HiddenInputType::shouldRespectHeightAndWidthAttributes): Added.
        * html/HiddenInputType.h: Ditto.

        * html/ImageInputType.cpp:
        (WebCore::ImageInputType::ImageInputType): Moved out of the header file.
        (WebCore::ImageInputType::srcAttributeChanged): Added.
        (WebCore::ImageInputType::attach): Added.
        (WebCore::ImageInputType::willMoveToNewOwnerDocument): Added.
        (WebCore::ImageInputType::shouldRespectAlignAttribute): Added.
        (WebCore::ImageInputType::canBeSuccessfulSubmitButton): Added.
        (WebCore::ImageInputType::isImageButton): Added.
        (WebCore::ImageInputType::isEnumeratable): Added.
        (WebCore::ImageInputType::shouldRespectHeightAndWidthAttributes): Added.
        * html/ImageInputType.h: Ditto.

        * html/InputType.cpp:
        (WebCore::InputType::create): Changed argument type to String. There
        is no value in taking an AtomicString here.
        (WebCore::InputType::scaledStepValueShouldBeInteger): Fixed typo in
        this function name.
        (WebCore::InputType::canSetStringValue): Added.
        (WebCore::InputType::isKeyboardFocusable): Added.
        (WebCore::InputType::shouldUseInputMethod): Added.
        (WebCore::InputType::handleBlurEvent): Added.
        (WebCore::InputType::accessKeyAction): Added.
        (WebCore::InputType::attach): Added.
        (WebCore::InputType::srcAttributeChanged): Added.
        (WebCore::InputType::willMoveToNewOwnerDocument): Added.
        (WebCore::InputType::shouldRespectAlignAttribute): Added.
        (WebCore::InputType::canChangeFromAnotherType): Added.
        (WebCore::InputType::minOrMaxAttributeChanged): Added.
        (WebCore::InputType::canBeSuccessfulSubmitButton): Added.
        (WebCore::InputType::rendererIsNeeded): Added.
        (WebCore::InputType::files): Added.
        (WebCore::InputType::getTypeSpecificValue): Added.
        (WebCore::InputType::fallbackValue): Added.
        (WebCore::InputType::defaultValue): Added.
        (WebCore::InputType::canSetSuggestedValue): Added.
        (WebCore::InputType::shouldSendChangeEventAfterCheckedChanged): Added.
        (WebCore::InputType::storesValueSeparateFromAttribute): Added.
        (WebCore::InputType::canSetValue): Added.
        (WebCore::InputType::willDispatchClick): Added.
        (WebCore::InputType::didDispatchClick): Added.
        (WebCore::InputType::isAcceptableValue): Added.
        (WebCore::InputType::sanitizeValue): Added.
        (WebCore::InputType::hasUnacceptableValue): Added.
        (WebCore::InputType::setFileList): Added.
        (WebCore::InputType::shouldResetOnDocumentActivation): Added.
        (WebCore::InputType::shouldRespectListAttribute): Added.
        (WebCore::InputType::shouldRespectSpeechAttribute): Added.
        (WebCore::InputType::isTextButton): Added.
        (WebCore::InputType::isRadioButton): Added.
        (WebCore::InputType::isSearchField): Added.
        (WebCore::InputType::isHiddenType): Added.
        (WebCore::InputType::isPasswordField): Added.
        (WebCore::InputType::isCheckbox): Added.
        (WebCore::InputType::isEmailField): Added.
        (WebCore::InputType::isFileUpload): Added.
        (WebCore::InputType::isImageButton): Added.
        (WebCore::InputType::isNumberField): Added.
        (WebCore::InputType::isSubmitButton): Added.
        (WebCore::InputType::isTelephoneField): Added.
        (WebCore::InputType::isURLField): Added.
        (WebCore::InputType::isEnumeratable): Added.
        (WebCore::InputType::isCheckable): Added.
        (WebCore::InputType::hasSpinButton): Added.
        (WebCore::InputType::shouldRespectHeightAndWidthAttributes): Added.

        * html/InputType.h: Removed unneeded ExceptionCode.h include and
        used a typedef instead as we do when we want for forward declare it.
        Added includes and forward declarations needed for new functions.
        Added ClickHandlingState structure. Changed the argument to create
        to a String rather than AtomicString. Added many virtual functions.
        
        * html/IsIndexInputType.cpp:
        (WebCore::IsIndexInputType::shouldRespectListAttribute): Added.
        * html/IsIndexInputType.h: Ditto.

        * html/NumberInputType.cpp:
        (WebCore::NumberInputType::handleBlurEvent): Added.
        (WebCore::NumberInputType::isAcceptableValue): Added.
        (WebCore::NumberInputType::sanitizeValue): Added.
        (WebCore::NumberInputType::hasUnacceptableValue): Added.
        (WebCore::NumberInputType::shouldRespectSpeechAttribute): Added.
        (WebCore::NumberInputType::isNumberField): Added.
        (WebCore::NumberInputType::hasSpinButton): Added.
        * html/NumberInputType.h: Ditto.

        * html/PasswordInputType.cpp:
        (WebCore::PasswordInputType::shouldUseInputMethod): Added.
        (WebCore::PasswordInputType::shouldResetOnDocumentActivation): Added.
        (WebCore::PasswordInputType::shouldRespectListAttribute): Added.
        (WebCore::PasswordInputType::shouldRespectSpeechAttribute): Added.
        (WebCore::PasswordInputType::isPasswordField): Added.
        * html/PasswordInputType.h: Ditto.

        * html/RadioInputType.cpp:
        (WebCore::RadioInputType::isKeyboardFocusable): Added.
        (WebCore::RadioInputType::attach): Added.
        (WebCore::RadioInputType::shouldSendChangeEventAfterCheckedChanged): Added.
        (WebCore::RadioInputType::willDispatchClick): Added.
        (WebCore::RadioInputType::didDispatchClick): Added.
        (WebCore::RadioInputType::isRadioButton): Added.
        * html/RadioInputType.h: Ditto.

        * html/RangeInputType.cpp:
        (WebCore::RangeInputType::accessKeyAction): Added.
        (WebCore::RangeInputType::minOrMaxAttributeChanged): Added.
        (WebCore::RangeInputType::fallbackValue): Added.
        (WebCore::RangeInputType::sanitizeValue): Added.
        (WebCore::RangeInputType::shouldRespectListAttribute): Added.
        * html/RangeInputType.h: Ditto.

        * html/ResetInputType.cpp:
        (WebCore::ResetInputType::defaultValue): Added.
        (WebCore::ResetInputType::isTextButton): Added.
        * html/ResetInputType.h: Ditto.

        * html/SearchInputType.cpp:
        (WebCore::SearchInputType::shouldRespectSpeechAttribute): Added.
        (WebCore::SearchInputType::isSearchField): Added.
        * html/SearchInputType.h: Ditto.

        * html/SubmitInputType.cpp:
        (WebCore::SubmitInputType::canBeSuccessfulSubmitButton): Added.
        (WebCore::SubmitInputType::defaultValue): Added.
        (WebCore::SubmitInputType::isSubmitButton): Added.
        (WebCore::SubmitInputType::isTextButton): Added.
        * html/SubmitInputType.h: Ditto.

        * html/TelephoneInputType.cpp:
        (WebCore::TelephoneInputType::shouldRespectSpeechAttribute): Added.
        (WebCore::TelephoneInputType::isTelephoneField): Added.
        * html/TelephoneInputType.h: Ditto.

        * html/TextFieldInputType.cpp:
        (WebCore::TextFieldInputType::shouldUseInputMethod): Added.
        (WebCore::TextFieldInputType::sanitizeValue): Added.
        (WebCore::TextFieldInputType::shouldRespectListAttribute): Added.
        * html/TextFieldInputType.h: Ditto.

        * html/TextInputType.cpp:
        (WebCore::TextInputType::canSetSuggestedValue): Added.
        (WebCore::TextInputType::shouldRespectSpeechAttribute): Added.
        * html/TextInputType.h: Ditto.

        * html/TimeInputType.cpp:
        (WebCore::TimeInputType::scaledStepValueShouldBeInteger): Fixed
        typo in the name of this function.
        * html/TimeInputType.h: Ditto.

        * html/URLInputType.cpp:
        (WebCore::URLInputType::isURLField): Added.
        * html/URLInputType.h: Ditto.

2011-01-03  Mihai Parparita  <mihaip@chromium.org>

        Unreviewed.

        NSScrollView-based scrollbars and ScrollBarTheme-based scrollbars don't look the same
        https://bugs.webkit.org/show_bug.cgi?id=51151

        Flip switch to use WebThemeEngine for Chromium/Mac scrollbar rendering,
        so that it can match Safari's (and thus use the same baselines).

        No new tests, should be covered by layout tests (which will need
        expectation changes)

        * platform/chromium/ScrollbarThemeChromiumMac.mm:

2011-01-03  Antonio Gomes  <agomes@rim.com>

        Unreviewed build fix for WebKit/Gtk+.

        RenderThemeGtk::initMediaColors() in RenderThemeGtk2.cpp needs to be
        guarded by ENABLE(VIDEO), as it is in its header.

        No new tests.

        * platform/gtk/RenderThemeGtk2.cpp:

2011-01-02  Dan Bernstein  <mitz@apple.com>

        Rubber-stamped by Simon Fraser.

        <rdar://problem/8812159> Update copyright strings

        * Info.plist:

2011-01-02  Dan Bernstein  <mitz@apple.com>

        Reviewed by Sam Weinig.

        <rdar://problem/8806570> REGRESSION (r73337): Incorrect text match count when a match occurs in a text field
        https://bugs.webkit.org/show_bug.cgi?id=51623

        * editing/Editor.cpp:
        (WebCore::Editor::countMatchesForText): Actually remember the original search range end
        container and offset and restore them when needed, instead of using two references to the
        same Range.
        * manual-tests/find-count-matches-after-text-control.html: Added.

2011-01-02  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Csaba Osztrogonác.

        WebKit 2 does not build on 64bits with a recent GCC
        https://bugs.webkit.org/show_bug.cgi?id=51754

        Use explicitely 64 bits types for the attributes of DatabaseDetails.

        * storage/DatabaseDetails.h:
        (WebCore::DatabaseDetails::expectedUsage):
        (WebCore::DatabaseDetails::currentUsage):

2011-01-02  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed WinCE build fix for r74868.

        * platform/graphics/wince/ImageBufferWinCE.cpp:
        (WebCore::getImageData):

2011-01-02  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r74847.
        http://trac.webkit.org/changeset/74847
        https://bugs.webkit.org/show_bug.cgi?id=51803

        "Preprocessor guards should be in header files" (Requested by
        ap on #webkit).

        * bindings/js/JSDOMWindowBase.cpp:
        * dom/Document.cpp:
        (WebCore::Document::recalcStyle):
        * html/HTMLDocument.cpp:
        * loader/FrameLoader.cpp:
        * loader/ResourceLoadNotifier.cpp:
        * page/Chrome.cpp:
        * page/Console.cpp:
        * page/ContextMenuController.cpp:
        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::dispatchEvent):
        * page/EventHandler.cpp:
        * page/Page.cpp:
        * storage/Database.cpp:
        * workers/AbstractWorker.cpp:
        * workers/DefaultSharedWorkerRepository.cpp:
        * workers/SharedWorker.cpp:
        * workers/Worker.cpp:
        * workers/WorkerContext.cpp:
        * workers/WorkerMessagingProxy.cpp:
        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::callReadyStateChangeListener):

2011-01-02  Patrick Gansterer  <paroga@webkit.org>

        [CMake] Unreviewed build fix.

        * CMakeLists.txt:

2011-01-02  Dirk Schulze  <krit@webkit.org>

        Reviewed by Simon Fraser.

        Clarify ImageBuffer and ImageData relationship
        https://bugs.webkit.org/show_bug.cgi?id=51297

        Remove references to ImageData and CanvasPixelArray from ImageBuffer and SVG filter/masker code,
        since this was a layering violation, and replaced it by ByteArray.

        No change of functionality, so no new test added.

        * html/ImageData.cpp:
        (WebCore::ImageData::create):
        (WebCore::ImageData::ImageData):
        * html/ImageData.h: Added new constructor with ByteArray input.
        (WebCore::ImageData::width):
        (WebCore::ImageData::height):
        (WebCore::ImageData::data):
        * html/canvas/CanvasPixelArray.cpp:
        (WebCore::CanvasPixelArray::create):
        (WebCore::CanvasPixelArray::CanvasPixelArray):
        * html/canvas/CanvasPixelArray.h: New conctructor with ByteArray input, used by HTML Canvas.
        (WebCore::CanvasPixelArray::data):
        (WebCore::CanvasPixelArray::length):
        (WebCore::CanvasPixelArray::set):
        (WebCore::CanvasPixelArray::get):
        * html/canvas/CanvasRenderingContext2D.cpp:
        (WebCore::CanvasRenderingContext2D::getImageData):
        (WebCore::CanvasRenderingContext2D::putImageData):
        * platform/graphics/ImageBuffer.h:
        * platform/graphics/cairo/ImageBufferCairo.cpp:
        (WebCore::getImageData):
        (WebCore::ImageBuffer::getUnmultipliedImageData):
        (WebCore::ImageBuffer::getPremultipliedImageData):
        (WebCore::putImageData):
        (WebCore::ImageBuffer::putUnmultipliedImageData):
        (WebCore::ImageBuffer::putPremultipliedImageData):
        * platform/graphics/cg/ImageBufferCG.cpp:
        (WebCore::getImageData):
        (WebCore::ImageBuffer::getUnmultipliedImageData):
        (WebCore::ImageBuffer::getPremultipliedImageData):
        (WebCore::putImageData):
        (WebCore::ImageBuffer::putUnmultipliedImageData):
        (WebCore::ImageBuffer::putPremultipliedImageData):
        * platform/graphics/filters/FEBlend.cpp:
        (WebCore::FEBlend::apply):
        * platform/graphics/filters/FEColorMatrix.cpp:
        (WebCore::FEColorMatrix::apply):
        * platform/graphics/filters/FEComponentTransfer.cpp:
        (WebCore::FEComponentTransfer::apply):
        * platform/graphics/filters/FEComposite.cpp:
        (WebCore::FEComposite::apply):
        * platform/graphics/filters/FEConvolveMatrix.cpp:
        (WebCore::setDestinationPixels):
        (WebCore::FEConvolveMatrix::apply):
        * platform/graphics/filters/FEConvolveMatrix.h:
        * platform/graphics/filters/FEDisplacementMap.cpp:
        (WebCore::FEDisplacementMap::apply):
        * platform/graphics/filters/FEGaussianBlur.cpp:
        (WebCore::FEGaussianBlur::apply):
        * platform/graphics/filters/FELighting.cpp:
        (WebCore::FELighting::apply):
        * platform/graphics/filters/FEMorphology.cpp:
        (WebCore::FEMorphology::apply):
        * platform/graphics/filters/FETurbulence.cpp:
        (WebCore::FETurbulence::apply):
        * platform/graphics/filters/FilterEffect.cpp:
        (WebCore::FilterEffect::asImageBuffer):
        (WebCore::FilterEffect::asUnmultipliedImage):
        (WebCore::FilterEffect::asPremultipliedImage):
        (WebCore::FilterEffect::copyImageBytes):
        (WebCore::FilterEffect::copyUnmultipliedImage):
        (WebCore::FilterEffect::copyPremultipliedImage):
        (WebCore::FilterEffect::createUnmultipliedImageResult):
        (WebCore::FilterEffect::createPremultipliedImageResult):
        * platform/graphics/filters/FilterEffect.h:
        * platform/graphics/qt/ImageBufferQt.cpp:
        (WebCore::getImageData):
        (WebCore::ImageBuffer::getUnmultipliedImageData):
        (WebCore::ImageBuffer::getPremultipliedImageData):
        (WebCore::putImageData):
        (WebCore::ImageBuffer::putUnmultipliedImageData):
        (WebCore::ImageBuffer::putPremultipliedImageData):
        * platform/graphics/skia/ImageBufferSkia.cpp:
        (WebCore::getImageData):
        (WebCore::ImageBuffer::getUnmultipliedImageData):
        (WebCore::ImageBuffer::getPremultipliedImageData):
        (WebCore::putImageData):
        (WebCore::ImageBuffer::putUnmultipliedImageData):
        (WebCore::ImageBuffer::putPremultipliedImageData):
        * platform/graphics/wince/ImageBufferWinCE.cpp:
        (WebCore::getImageData):
        (WebCore::ImageBuffer::getUnmultipliedImageData):
        (WebCore::ImageBuffer::getPremultipliedImageData):
        (WebCore::putImageData):
        (WebCore::ImageBuffer::putUnmultipliedImageData):
        (WebCore::ImageBuffer::putPremultipliedImageData):
        * platform/graphics/wx/ImageBufferWx.cpp:
        (WebCore::ImageBuffer::getUnmultipliedImageData):
        (WebCore::ImageBuffer::getPremultipliedImageData):
        (WebCore::ImageBuffer::putUnmultipliedImageData):
        (WebCore::ImageBuffer::putPremultipliedImageData):
        * rendering/RenderSVGResourceMasker.cpp:
        (WebCore::RenderSVGResourceMasker::drawContentIntoMaskImage):

2011-01-01  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Move JavaScriptCore to Source
        https://bugs.webkit.org/show_bug.cgi?id=51604

        Update references to JavaScriptCore.

        * Android.derived.jscbindings.mk:
        * Android.v8bindings.mk:
        * CMakeLists.txt:
        * WebCore.gyp/WebCore.gyp:
        * WebCore.pro:
          - These changes are subtle and might not be 100% correct.
        * move-js-headers.sh:

2011-01-01  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        forbid sandboxed frames to call top.close() when allow-same-origin is not setted
        https://bugs.webkit.org/show_bug.cgi?id=38340

        We now pass the ScriptExecutionContext to window.close so it can find
        the Frame and check whether navigation is allowed.  This check will
        almost always pass because you can only close top-level frames, but the
        check will fail when the calling script is sandboxed.

        Tests: fast/frames/sandboxed-iframe-close-top-noclose.html
               fast/frames/sandboxed-iframe-close-top.html

        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::close):
        * page/DOMWindow.h:
        * page/DOMWindow.idl:

2011-01-01  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        sandbox iframes have access to top.history methods
        https://bugs.webkit.org/show_bug.cgi?id=38152

        To enforce the sandbox restrictions on History, we need to pass the
        ScriptExecutionContext to WebCore.  This patch leaves the original
        History methods in place because they are used directly by folks who
        don't care about security checks.

        Test: fast/frames/sandboxed-iframe-history-denied.html

        * page/History.cpp:
        (WebCore::History::back):
        (WebCore::History::forward):
        (WebCore::History::go):
        * page/History.h:
        * page/History.idl:

2011-01-01  Adam Barth  <abarth@webkit.org>

        Remove empty file.

        * Android.mk:
        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * dom/DOMAllInOne.cpp:
        * dom/Entity.cpp: Removed.

2011-01-01  Adam Barth  <abarth@webkit.org>

        Remove empty file.

        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/JSBindingsAllInOne.cpp:
        * bindings/js/JSPopStateEventCustom.cpp: Removed.

2011-01-01  Adam Barth  <abarth@webkit.org>

        Remove empty files.

        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/generic/BindingFrame.h: Removed.
        * bindings/generic/BindingLocation.h: Removed.

2011-01-01  Konstantin Tokarev  <annulen@yandex.ru>

        Reviewed by Darin Adler.

        Don't include Inspector headers when Inspector is disabled
        https://bugs.webkit.org/show_bug.cgi?id=51789

        * bindings/js/JSDOMWindowBase.cpp: Include Inspector headers when
        Inspector is enabled
        * dom/Document.cpp: Include Inspector headers when
        Inspector is enabled
        (WebCore::Document::recalcStyle): Disabled InspectorInstrumentationCookie
        handling
        * html/HTMLDocument.cpp: Include Inspector headers when
        Inspector is enabled
        * loader/FrameLoader.cpp: Include Inspector headers when
        Inspector is enabled
        * loader/ResourceLoadNotifier.cpp: Include Inspector headers when
        Inspector is enabled
        * page/Chrome.cpp: Include Inspector headers when
        Inspector is enabled
        * page/Console.cpp: Include Inspector headers when
        Inspector is enabled
        * page/ContextMenuController.cpp: Include Inspector headers when
        Inspector is enabled
        * page/DOMWindow.cpp: Include Inspector headers when
        Inspector is enabled
        (WebCore::DOMWindow::dispatchEvent): Disabled InspectorInstrumentationCookie
        handling
        * page/EventHandler.cpp: Include Inspector headers when
        Inspector is enabled
        * page/Page.cpp: Include Inspector headers when
        Inspector is enabled
        * storage/Database.cpp: Include Inspector headers when
        Inspector is enabled
        * workers/AbstractWorker.cpp: Include Inspector headers when
        Inspector is enabled
        * workers/DefaultSharedWorkerRepository.cpp: Include Inspector headers when
        Inspector is enabled
        * workers/SharedWorker.cpp: Include Inspector headers when
        Inspector is enabled
        * workers/Worker.cpp: Include Inspector headers when
        Inspector is enabled
        * workers/WorkerContext.cpp: Include Inspector headers when
        Inspector is enabled
        * workers/WorkerMessagingProxy.cpp: Include Inspector headers when
        Inspector is enabled
        * xml/XMLHttpRequest.cpp: Include Inspector headers when
        Inspector is enabled
        (WebCore::XMLHttpRequest::callReadyStateChangeListener): Disabled
        InspectorInstrumentationCookie handling

2010-12-31  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        fast/loader/about-blank-hash-change.html fails on Chromium
        https://bugs.webkit.org/show_bug.cgi?id=51788

        I'm unable to reproduce the failure locally, so this patch is slightly
        speculative.  We need to be slighly more careful how we compare the
        fragment identifiers for Chromium because Chromium uses a URL parser
        that is more agressive about canonicalization.

        * page/Location.cpp:
        (WebCore::Location::setHash):

2010-12-31  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Use GLib/GIO API for FileSystemGtk implementation
        https://bugs.webkit.org/show_bug.cgi?id=51617

        Complete implementation of FileSystem platform code for GLib. This
        converts remaining API points using POSIX calls to platform-independent
        GLib calls and fixes other compilation errors.

        No new tests. This code will be tested when an implementation
        of beginDragWithFiles is complete.

        (WebCore::JSDirectoryEntry::getFile):
        (WebCore::JSDirectoryEntry::getDirectory):
        * platform/FileSystem.h: Added forward declarations for GLib types to
        avoid GLib includes and changed the PlatformFileHandle to be a GIOStream.
        * platform/gtk/FileSystemGtk.cpp:
        (WebCore::openTemporaryFile): Reimplement using only GLib calls.
        (WebCore::openFile): Ditto.
        (WebCore::closeFile): Ditto.
        (WebCore::seekFile): Added implementation.
        (WebCore::writeToFile): Reimplement using only GLib calls.
        (WebCore::readFromFile): Ditto.

2010-12-31  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Some renaming and refactoring of form element code
        https://bugs.webkit.org/show_bug.cgi?id=51784

        * bindings/js/JSHTMLSelectElementCustom.cpp:
        (WebCore::JSHTMLSelectElement::remove): Call new overload of remove
        for option elements instead of putting the logic in the binding.

        * html/HTMLButtonElement.cpp:
        (WebCore::HTMLButtonElement::HTMLButtonElement): Updated for name changes.
        (WebCore::HTMLButtonElement::defaultEventHandler): Renamed argument to
        "event" instead of "evt". Also updated for name changes.
        (WebCore::HTMLButtonElement::isActivatedSubmit): Ditto.
        (WebCore::HTMLButtonElement::setActivatedSubmit): Ditto.
        (WebCore::HTMLButtonElement::appendFormData): Ditto.
        * html/HTMLButtonElement.h: Renamed m_activeSubmit to m_isActivatedSubmit
        to match the getter function name. The name still doesn't seem great.

        * html/HTMLFormElement.cpp:
        (WebCore::HTMLFormElement::HTMLFormElement): Updated for name changes.
        (WebCore::HTMLFormElement::~HTMLFormElement): Use autoComplete function
        instead of m_autocomplete data member.
        (WebCore::HTMLFormElement::rendererIsNeeded): Use m_wasDemoted data member
        instead of otherwise-unused isDemoted function.
        (WebCore::HTMLFormElement::submitImplicitly): Updated for name changes.
        (WebCore::HTMLFormElement::validateInteractively): Removed code to clear
        m_insubmit. This is now handled by the caller, prepareForSubmission.
        (WebCore::HTMLFormElement::prepareForSubmission): Renamed. Updated for
        name changes. Moved code to clear m_isSubmittingOrPreparingForSubmission
        here from validateInteractively. Removed unneeded check of m_doingsubmit
        boolean before setting it.
        (WebCore::HTMLFormElement::submit): Factored this function into two.
        One for JavaScript and one for non-JavaScript. Neither function needs a frame
        argument, because the question being asked, anyPageIsProcessingUserGesture,
        is a question asked of an entire page group, not a specific frame or page,
        so it's not important which is the active frame.
        (WebCore::HTMLFormElement::submitFromJavaScript): Ditto.
        (WebCore::HTMLFormElement::reset): Updated for name changes.
        (WebCore::HTMLFormElement::parseMappedAttribute): Removed code to parse
        acceptAttr and got rid of code to set m_autocomplete.
        (WebCore::HTMLFormElement::elementForAlias): Changed return type to a raw
        pointer.
        (WebCore::HTMLFormElement::getNamedElements): Updated to use raw pointer
        and the vector find function. Added a FIXME about the comment.
        (WebCore::HTMLFormElement::documentDidBecomeActive): Use autoComplete
        function instead of m_autocomplete data member.
        (WebCore::HTMLFormElement::willMoveToNewOwnerDocument): Ditto.
        (WebCore::HTMLFormElement::didMoveToNewOwnerDocument): Ditto.
        (WebCore::HTMLFormElement::autoComplete): Ditto.
        * html/HTMLFormElement.h: Added a FIXME about renaming the autoComplete
        function. Renamed prepareSubmit to prepareForSubmission. Got rid of the
        frame argument from the submit function and added a new submitFromJavaScript
        variant. Removed the unneeded isDemoted function. Changed the return type
        of elementForAlias to a raw pointer. Removed m_autocomplete. Renamed
        m_insubmit to m_isSubmittingOrPreparingForSubmission, m_doingsubmit to
        m_shouldSubmit, m_inreset to m_inResetFunction, m_malformed to m_wasMalformed,
        m_demoted to m_wasDemoted. Use plain bool instead of bitfields.

        * html/HTMLFormElement.idl: Use the function named submitFromJavaScript
        instead of using CallWith=DynamicFrame because we don't need to know
        what frame is calling. If we did, it's not clear that the "dynamic"
        frame would be the right one anyway.

        * html/HTMLSelectElement.cpp:
        (WebCore::HTMLSelectElement::remove): Added an overload that takes an option
        element, since the JavaScript binding supports this.
        * html/HTMLSelectElement.h: Ditto.

        * html/ImageInputType.cpp:
        (WebCore::ImageInputType::handleDOMActivateEvent): Updated for name changes.
        * html/ImageInputType.h: Ditto.
        * html/SubmitInputType.cpp:
        (WebCore::SubmitInputType::handleDOMActivateEvent): Ditto.

2010-12-31  Daniel Bates  <dbates@rim.com>

        Reviewed by Darin Adler.

        Pass style for <area> instead of associated <img> when querying whether
        the theme draws the focus ring for an <area>
        https://bugs.webkit.org/show_bug.cgi?id=51632

        Fixes an issue where the style for the associated <img> of an <area>
        was passed when considering whether the theme should draw a focus ring
        for the <area>. Instead, we should pass the theme the style for the <area>.

        It's not possible to test this at this time since there are no themes
        that override focus ring drawing for <area>s.

        * rendering/RenderImage.cpp:
        (WebCore::RenderImage::paintFocusRing): Moved call to RenderTheme::supportsFocusRing()
        such that it's called with respect to the focused <area>.

2010-12-31  Darin Adler  <darin@apple.com>

        * WebCore.xcodeproj/project.pbxproj: Added back the language.
        Developers should be using Xcode 3.2.5 or newer.

2010-12-31  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Move V8 to WebCore Location implementation
        https://bugs.webkit.org/show_bug.cgi?id=51768

        * bindings/generic/BindingFrame.h:
        * bindings/generic/BindingLocation.h:
        * bindings/v8/V8Binding.h:
        * bindings/v8/V8DOMWindowShell.cpp:
        (WebCore::V8DOMWindowShell::setLocation):
        * bindings/v8/V8Utilities.cpp:
        * bindings/v8/V8Utilities.h:
        * bindings/v8/custom/V8LocationCustom.cpp:
        (WebCore::V8Location::hashAccessorSetter):
        (WebCore::V8Location::hostAccessorSetter):
        (WebCore::V8Location::hostnameAccessorSetter):
        (WebCore::V8Location::hrefAccessorSetter):
        (WebCore::V8Location::pathnameAccessorSetter):
        (WebCore::V8Location::portAccessorSetter):
        (WebCore::V8Location::protocolAccessorSetter):
        (WebCore::V8Location::searchAccessorSetter):
        (WebCore::V8Location::reloadCallback):
        (WebCore::V8Location::replaceCallback):
        (WebCore::V8Location::assignCallback):

2010-12-31  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Eric Seidel.

        Make retrieving extraMediaControlsStyleSheet be page dependent.
        https://bugs.webkit.org/show_bug.cgi?id=51752

        Use themeForPage to retrieve extraMediaControlsStyleSheet for media
        controls.

        No new tests because no platform-specific theme implementation uses
        the passed page pointer yet.

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::styleForElement):

2010-12-31  Ilya Tikhonovsky  <loislo@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: [REGRESSION] WebInspector.ResourcesPanel.prototype.canShowSourceLine() broken.

        https://bugs.webkit.org/show_bug.cgi?id=51709

        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel.prototype.canShowSourceLine):

2010-12-31  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed build fix after r74800.

        * bindings/js/JSDOMWindowCustom.cpp: Add missing header.
        * bindings/js/JSLocationCustom.cpp: Ditto.

2010-12-31  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move HTML and XML parser benchmarks into PerformanceTests/Parser
        https://bugs.webkit.org/show_bug.cgi?id=51772

        Remove the benchmarks folder now that it's empty.  Future benchmarks
        should go in the PerformanceTests directory.

        * benchmarks: Removed.
        * benchmarks/parser: Removed.
        * benchmarks/parser/html-parser.html: Removed.
        * benchmarks/parser/resources: Removed.
        * benchmarks/parser/resources/html5.html: Removed.
        * benchmarks/parser/resources/runner.js: Removed.
        * benchmarks/parser/xml-parser.html: Removed.

2010-12-31  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Adam Barth.

        Add XML parser benchmark
        https://bugs.webkit.org/show_bug.cgi?id=51612

        Move the JavaScript code of html parser into a separate file to share it with xml parser.

        * benchmarks/parser/html-parser.html:
        * benchmarks/parser/resources/runner.js: Copied from WebCore/benchmarks/parser/html-parser.html.
        (runFunction):
        (run):
        (start):
        * benchmarks/parser/xml-parser.html: Copied from WebCore/benchmarks/parser/html-parser.html.

2010-12-29  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        Update validation of stencil mask and ref values
        https://bugs.webkit.org/show_bug.cgi?id=50716

        * html/canvas/WebGLRenderingContext.cpp:
        (WebCore::WebGLRenderingContext::initializeNewContext): Initialize cached stencil settings.
        (WebCore::WebGLRenderingContext::drawArrays): Validate stencil settings.
        (WebCore::WebGLRenderingContext::drawElements): Ditto.
        (WebCore::WebGLRenderingContext::stencilFunc): Cache stencil settings, but generate no error.
        (WebCore::WebGLRenderingContext::stencilFuncSeparate): Ditto.
        (WebCore::WebGLRenderingContext::stencilMask): Ditto.
        (WebCore::WebGLRenderingContext::stencilMaskSeparate): Ditto.
        (WebCore::WebGLRenderingContext::validateStencilSettings): Helper function to validate stencil settings.
        * html/canvas/WebGLRenderingContext.h: Declare stencil setting members.

2010-12-30  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Fork pieces of RenderThemeGtk that will differ for GTK+ 3
        https://bugs.webkit.org/show_bug.cgi?id=51755

        Split out pieces of RenderThemeGtk into RenderThemeGtk2 and RenderThemeGtk3
        if they are going to differ between GTK+ 2.x and GTK+ 3.x. This is the preliminary
        step to landing support for the new GTK+ 3.x theming API.

        No new tests. This should not change functionality.

        * GNUmakefile.am: Add RenderThemeGtk2 and RenderThemeGtk3 to the sources list.
        * platform/gtk/RenderThemeGtk.cpp:
        (WebCore::RenderThemeGtk::adjustTextFieldStyle): Call the setTextInputBorders static method.
        (WebCore::RenderThemeGtk::adjustTextAreaStyle): Ditto.
        (WebCore::RenderThemeGtk::paintMediaButton): Ditto.
        (WebCore::RenderThemeGtk::adjustProgressBarStyle): ditto.
        * platform/gtk/RenderThemeGtk.h: Make setTextInputBorders a method, so that
        it can be exist in RenderThemeGtk{2,3} and be called from RenderThemeGtk.
        * platform/gtk/RenderThemeGtk2.cpp: Added. A fork of the bits of RenderThemeGtk
        that will eventually be platform-dependent.
        * platform/gtk/RenderThemeGtk3.cpp: Added. Ditto.

2010-12-30  Justin Schuh  <jschuh@chromium.org>

        Reviewed by James Robinson.

        Reverting r74292 because it introduced a crash with ruby text.
        https://bugs.webkit.org/show_bug.cgi?id=51637

        Test: fast/css/counters/counter-ruby-text-cleared.html

        * rendering/RenderCounter.cpp:
        (WebCore::findPlaceForCounter):
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::addChild):
        * rendering/RenderObjectChildList.cpp:
        (WebCore::RenderObjectChildList::appendChildNode):
        (WebCore::RenderObjectChildList::insertChildNode):

2010-12-30  Darin Adler  <darin@apple.com>

        Reviewed by David Kilzer.

        Remove unused JSBinding source files
        https://bugs.webkit.org/show_bug.cgi?id=51762

        * bindings/js/JSBinding.h: Removed.
        * bindings/js/specialization: Removed.
        * bindings/js/specialization/JSBindingState.cpp: Removed.
        * bindings/js/specialization/JSBindingState.h: Removed.

        * CMakeLists.txt: Removed references to files.
        * GNUmakefile.am: Ditto.
        * WebCore.gypi: Ditto.
        * WebCore.pro: Ditto.
        * WebCore.vcproj/WebCore.vcproj: Ditto.
        * bindings/js/JSBindingsAllInOne.cpp: Ditto.

2010-12-30  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Darin Adler.

        Remove Document.h include from some headers where it isn't necessary
        https://bugs.webkit.org/show_bug.cgi?id=51730

        No new tests because no new functionality.

        * editing/EditingStyle.cpp:
        * editing/EditingStyle.h:
        * editing/SelectionController.h:
        * editing/visible_units.h:
        * html/HTMLAudioElement.h:
        * html/HTMLBodyElement.h:
        * inspector/InspectorCSSAgent.h:
        * inspector/InspectorDOMAgent.h:
        * inspector/InspectorInstrumentation.h:
        * xml/DOMParser.cpp:
        * xml/DOMParser.h:
        (WebCore::DOMParser::create):
        (WebCore::DOMParser::DOMParser):

2010-12-30  Benjamin C Meyer  <bmeyer@rim.com>

        Reviewed by Darin Adler.

        bindings/js/JSDesktopNotificationsCustom.cpp is not included in the cmake build
        which is required when NOTIFICATIONS is enabled.

        https://bugs.webkit.org/show_bug.cgi?id=51761

        * CMakeLists.txt:

2010-12-30  Antti Koivisto  <antti@apple.com>

        Not reviewed.

        Revert accidental project file change.

        * WebCore.xcodeproj/project.pbxproj:

2010-12-30  Antti Koivisto  <antti@apple.com>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=51134
        Move loading related code from MemoryCache to CachedResourceLoader

        - Merge MemoryCache::requestResource to CachedResourceLoader::requestResource
        - Merge MemoryCache::requestUserCSSStyleSheet to CachedResourceLoader::requestUserCSSStyleSheet
        - Move MemoryCache::revalidateResource to CachedResourceLoader::revalidateResource
        - Add MemoryCache::add
        - Refactor the decision on whether to reload, revalidate or use the existing resource to 
          a single function, CachedResourceLoader::determineRevalidationPolicy

        * css/CSSImageValue.cpp:
        (WebCore::CSSImageValue::cachedImage):
            
            Remove a code path that called MemoryCache::requestResource directly. This code path would have crashed
            if ever taken (since it passes null CachedResourceLoader pointer).
            
        * loader/ImageLoader.cpp:
        (WebCore::ImageLoader::updateFromElement):
        * loader/cache/CachedImage.cpp:
        * loader/cache/CachedResource.cpp:
        (WebCore::CachedResource::CachedResource):
        (WebCore::CachedResource::~CachedResource):
        (WebCore::CachedResource::mustRevalidateDueToCacheHeaders):
        
            Moved tests that were not about cache headers to CachedResourceLoader::determineRevalidationPolicy and renamed.
        
        (WebCore::CachedResource::setLoadPriority):
        
            Check for Unresolved value before setting.
        
        * loader/cache/CachedResource.h:
        (WebCore::CachedResource::setOwningCachedResourceLoader):
        
            Rename to be bit less mysterious.
        
        * loader/cache/CachedResourceLoader.cpp:
        (WebCore::createResource):
        
            This was moved from MemoryCache.
        
        (WebCore::CachedResourceLoader::~CachedResourceLoader):
        (WebCore::CachedResourceLoader::determineRevalidationPolicy):
        (WebCore::CachedResourceLoader::requestUserCSSStyleSheet):
        
            This was moved/merged from MemoryCache.
        
        (WebCore::CachedResourceLoader::canRequest):
        (WebCore::CachedResourceLoader::requestResource):
        
            This combines MemoryCache::requestResource and the existing method.
        
        (WebCore::CachedResourceLoader::revalidateResource):
        
            This was moved from MemoryCache.
        
        (WebCore::CachedResourceLoader::loadResource):
        
            New method for initiating loading.
        
        (WebCore::CachedResourceLoader::notifyLoadedFromMemoryCache):
        
            Renamed the mysterious CachedResourceLoader::checkCacheObjectStatus
        
        * loader/cache/CachedResourceLoader.h:
        * loader/cache/MemoryCache.cpp:
        (WebCore::MemoryCache::add):
        * loader/cache/MemoryCache.h:
        (WebCore::MemoryCache::remove):

2010-12-30  Steve Block  <steveblock@google.com>

        Reviewed by Gavin Barraclough.

        Enable JSC Rich Source Info for Android even when Inspector and Debugger are disabled
        https://bugs.webkit.org/show_bug.cgi?id=51740

        This is a policy decision for Android.

        No new tests, simply enabling existing, tested code on Android.

        * bindings/js/JSDOMWindowBase.cpp:
        (WebCore::JSDOMWindowBase::supportsRichSourceInfo):

2010-12-30  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Adam Barth.

        REGRESSION: fast/dom/cssTarget-crash.html fails
        https://bugs.webkit.org/show_bug.cgi?id=20342

        Do not reload the page when submitting a form, using "GET" method, and the
        form action url matches the location url, except for the fragment.
      
        Test: fast/forms/submit-change-fragment.html

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::loadURL):
        (WebCore::FrameLoader::loadWithDocumentLoader):
        (WebCore::FrameLoader::shouldScrollToAnchor):
        * loader/FrameLoader.h:

2010-12-30  Darin Adler  <darin@apple.com>

        Reviewed by Adam Barth.

        Move security logic out of the JavaScript binding for location into the DOM class
        https://bugs.webkit.org/show_bug.cgi?id=51714

        * WebCore.xcodeproj/project.pbxproj: Removed JSBindingState.cpp, JSBindingState.h,
        and JSBinding.h. Later, we'll remove them from other project files and delete
        the source files.

        * bindings/js/JSBinding.h: Emptied out, marked for later deletion.

        * bindings/js/JSDOMBinding.cpp: Removed unused overload of the jsOwnedStringOrNull
        function, unused shouldAllowNavigation function, unused allowSettingSrcToJavascriptURL
        function, unused toLexicalFrame function, unused completeURL function, unused
        getCachedDOMStructure and cacheDOMStructure overloads, and unused
        getCachedDOMConstructor and cacheDOMConstruction functions.
        (WebCore::activeDOMWindow): Added.
        (WebCore::firstDOMWindow): Added.
        (WebCore::toDynamicFrame): Changed to use firstDOMWindow instead of calling through
        JSBindingState, since the latter is an unneeded abstraction.
        (WebCore::processingUserGesture): Changed to call ScriptController's function
        rather than calling through JSBindingState, which is an unneeded abstraction

        * bindings/js/JSDOMBinding.h: Removed the functions mentioned above. Added
        activeDOMWindow and firstDOMWindow. Also added a FIXME for the extremely poorly
        named allowsAccessFromFrame functions, which answer the question of whether
        script is allowed access *to* a frame.

        * bindings/js/JSDOMWindowCustom.cpp: Removed many unneeded includes.
        (WebCore::JSDOMWindow::setLocation): Streamlined by using the new
        activeDOMWindow and firstDOMWindow functions.
        (WebCore::JSDOMWindow::open): Ditto.
        (WebCore::JSDOMWindow::showModalDialog): Ditto.
        (WebCore::JSDOMWindow::postMessage): Ditto.

        * bindings/js/JSLocationCustom.cpp:
        (WebCore::JSLocation::setHref): Changed to pass the active and first windows through
        to the Location::setHref function, with it doing the rest of the work. This moves
        work out of the DOM binding that belongs in the DOM itself. Eventually such functions
        could even be automatically generated by the bindings script.
        (WebCore::JSLocation::setProtocol): Ditto.
        (WebCore::JSLocation::setHost): Ditto.
        (WebCore::JSLocation::setHostname): Ditto.
        (WebCore::JSLocation::setPort): Ditto.
        (WebCore::JSLocation::setPathname): Ditto.
        (WebCore::JSLocation::setSearch): Ditto.
        (WebCore::JSLocation::setHash): Ditto.
        (WebCore::JSLocation::replace): Ditto.
        (WebCore::JSLocation::reload): Ditto.
        (WebCore::JSLocation::assign): Ditto.

        * bindings/js/specialization/JSBindingState.cpp: Emptied out, marked for later deletion.
        * bindings/js/specialization/JSBindingState.h: Emptied out, marked for later deletion.

        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::setLocation): Added a SetLocationLocking argument that allows us
        to use this function for Location::replace, which is the same as setLocation except that
        it locks both history and the back/forward list.
        * page/DOMWindow.h: Ditto.

        * page/Location.cpp:
        (WebCore::Location::setHref): Added. Uses DOMWindow::setLocation so that the security
        logic there does not have to be repeated or even refactored.
        (WebCore::Location::setProtocol): Ditto.
        (WebCore::Location::setHost): Ditto.
        (WebCore::Location::setHostname): Ditto.
        (WebCore::Location::setPort): Ditto.
        (WebCore::Location::setPathname): Ditto.
        (WebCore::Location::setSearch): Ditto.
        (WebCore::Location::setHash): Ditto.
        (WebCore::Location::assign): Ditto.
        (WebCore::Location::replace): Ditto.
        (WebCore::Location::reload): Added. Security logic was refactored from the code in
        JSLocationCustom.cpp. Added a FIXME about the fact that this security logic seems
        possibly unneeded.

        * page/Location.h: Fixed indentation on the whole file. Added the new functions above.
        (WebCore::Location::create):
        (WebCore::Location::frame):

2010-12-30  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Pasting a style property with value should automatically split it into prop/value
        https://bugs.webkit.org/show_bug.cgi?id=51581

        The "paste" DOM event is handled for the CSS property name field to parse out the name and value parts
        of a CSS property being pasted (by the first ':' found). The property is committed (if not a new one),
        and the edit focus is transferred to the value field.

        * inspector/front-end/StylesSidebarPane.js:
        (WebInspector.StylePropertyTreeElement.prototype.selectElement):
        (WebInspector.StylePropertyTreeElement.prototype):
        * inspector/front-end/inspector.js:
        (WebInspector.completeURL): Drive-by: return full URLs as-is.
        (WebInspector.startEditing.cleanUpAfterEditing):
        (WebInspector.startEditing):
        (WebInspector.startEditing.pasteEventListener):
        (WebInspector.startEditing.keyDownEventListener):

2010-12-30  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: breakpoint is not disabled when clicking on breakpoints sidebar pane checkbox.
        https://bugs.webkit.org/show_bug.cgi?id=51745

        * inspector/front-end/Breakpoint.js:
        (WebInspector.Breakpoint.prototype.set enabled):

2010-12-30  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: call stack shows "Paused on a JavaScript breakpoint" when stepping.
        https://bugs.webkit.org/show_bug.cgi?id=51748

        * inspector/front-end/CallStackSidebarPane.js:
        (WebInspector.CallStackSidebarPane.prototype.update):
        (WebInspector.CallStackSidebarPane.prototype.registerShortcuts):
        (WebInspector.CallStackSidebarPane.prototype._scriptBreakpointHit):
        (WebInspector.CallStackSidebarPane.prototype._nativeBreakpointHit):

2010-12-30  Mikhail Naganov  <mnaganov@chromium.org>

        Reviewed by Pavel Feldman.

        Fix leak of MemoryInfo wrappers introduced in r57004 (see issue 51712).

        https://bugs.webkit.org/show_bug.cgi?id=51713

        * WebCore.gypi:
        * WebCore.pro:
        * bindings/v8/custom/V8ConsoleCustom.cpp:
        (WebCore::V8Console::memoryAccessorGetter):
        * bindings/v8/custom/V8PerformanceCustom.cpp: Copied from V8ConsoleCustom.cpp.
        (WebCore::V8Performance::memoryAccessorGetter):
        * page/Console.idl:
        * page/Performance.idl:

2010-12-30  Steve Block  <steveblock@google.com>

        Reviewed by Sam Weinig.

        Visiting macnn.com often causes SQL spew via geolocation database
        https://bugs.webkit.org/show_bug.cgi?id=51557

        If the Geolocation position cache database path has not been set, early-out
        rather than using an empty path and thus failing to open the database.
        This avoids SQL log spew.

        Also, avoid starting the database thread until the path has been set, and
        shorten the thread name to avoid warnings due to exceeding 30 characters.

        No new tests, implementation clean-up only.

        * page/GeolocationPositionCache.cpp:
        (WebCore::GeolocationPositionCache::addUser):
        (WebCore::GeolocationPositionCache::removeUser):
        (WebCore::GeolocationPositionCache::setDatabasePath):
        (WebCore::GeolocationPositionCache::startBackgroundThread):

2010-12-29  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] minimal build fails at link time due to missing sqlite3 symbols
        https://bugs.webkit.org/show_bug.cgi?id=51327

        Guard the code calling sqlite3 with the ENABLE(DATABASE) check.

        No new tests, build fix only.

        * platform/sql/SQLiteAuthorizer.cpp:
        * platform/sql/SQLiteDatabase.cpp:
        * platform/sql/SQLiteFileSystem.cpp:
        * platform/sql/SQLiteStatement.cpp:
        * platform/sql/SQLiteTransaction.cpp:

2010-12-30  Justin Schuh  <jschuh@chromium.org>

        Reviewed by Dirk Schulze.

        Prevent stringToLengthType() from skipping past end-of-string on invalid length unit. 
        https://bugs.webkit.org/show_bug.cgi?id=51692

        * svg/SVGLength.cpp:
        (WebCore::stringToLengthType):
        (WebCore::SVGLength::setValueAsString):

2010-12-30  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Eric Seidel.

        Create a helper function for finding descendent video elements for a node.
        https://bugs.webkit.org/show_bug.cgi?id=51696

        Test: media/video-element-other-namespace-crash.html

        * html/MediaDocument.cpp:
        (WebCore::descendentVideoElement): helper function.
        (WebCore::MediaDocument::defaultEventHandler): use the new helper function. fix code repetitions.
        (WebCore::MediaDocument::replaceMediaElementTimerFired): use the new helper function.

2010-12-29  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: extract Database & DOM Storage agents; remove InspectorBackend.
        https://bugs.webkit.org/show_bug.cgi?id=51707

        This change brushes up storage agents + removes redundant InspectorBackend class.

        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * inspector/CodeGeneratorInspector.pm:
        * inspector/InjectedScriptHost.cpp:
        (WebCore::InjectedScriptHost::databaseForId):
        (WebCore::InjectedScriptHost::selectDatabase):
        (WebCore::InjectedScriptHost::selectDOMStorage):
        (WebCore::InjectedScriptHost::inspectorDOMAgent):
        * inspector/Inspector.idl:
        * inspector/InspectorBackend.cpp: Removed.
        * inspector/InspectorBackend.h: Removed.
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::InspectorController):
        (WebCore::InspectorController::~InspectorController):
        (WebCore::InspectorController::connectFrontend):
        (WebCore::InspectorController::releaseFrontendLifetimeAgents):
        (WebCore::InspectorController::didUseDOMStorage):
        (WebCore::InspectorController::setInjectedScriptSource):
        (WebCore::InspectorController::dispatchOnInjectedScript):
        (WebCore::InspectorController::releaseWrapperObjectGroup):
        * inspector/InspectorController.h:
        * inspector/InspectorDOMStorageAgent.cpp: Added.
        (WebCore::InspectorDOMStorageAgent::~InspectorDOMStorageAgent):
        (WebCore::InspectorDOMStorageAgent::getDOMStorageEntries):
        (WebCore::InspectorDOMStorageAgent::setDOMStorageItem):
        (WebCore::InspectorDOMStorageAgent::removeDOMStorageItem):
        (WebCore::InspectorDOMStorageAgent::selectDOMStorage):
        (WebCore::InspectorDOMStorageAgent::InspectorDOMStorageAgent):
        (WebCore::InspectorDOMStorageAgent::getDOMStorageResourceForId):
        * inspector/InspectorDOMStorageAgent.h: Copied from WebCore/inspector/InspectorStorageAgent.h.
        (WebCore::InspectorDOMStorageAgent::create):
        (WebCore::InspectorDOMStorageAgent::frontend):
        * inspector/InspectorDatabaseAgent.cpp: Renamed from WebCore/inspector/InspectorStorageAgent.cpp.
        (WebCore::InspectorDatabaseAgent::~InspectorDatabaseAgent):
        (WebCore::InspectorDatabaseAgent::getDatabaseTableNames):
        (WebCore::InspectorDatabaseAgent::executeSQL):
        (WebCore::InspectorDatabaseAgent::databaseForId):
        (WebCore::InspectorDatabaseAgent::selectDatabase):
        (WebCore::InspectorDatabaseAgent::clearFrontend):
        (WebCore::InspectorDatabaseAgent::InspectorDatabaseAgent):
        * inspector/InspectorDatabaseAgent.h: Renamed from WebCore/inspector/InspectorStorageAgent.h.
        (WebCore::InspectorDatabaseAgent::create):
        (WebCore::InspectorDatabaseAgent::frontend):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype.show):
        (WebInspector.ScriptsPanel.prototype.attachDebuggerWhenShown):
        (WebInspector.ScriptsPanel.prototype._toggleDebugging):

2010-12-29  Dirk Schulze  <krit@webkit.org>

        Reviewed by Darin Adler.

        Cleanup SVG code according to the webkit style rules 3
        https://bugs.webkit.org/show_bug.cgi?id=51490

        Last patch to fix indention and other style issues according to the WebKit style rules in the SVG code.
        Just one file can't be fixed for check-webkit-style.
        * SVGAllInOne.cpp: check-webkit-style wants a config.h at the beginning

        No changes of functionality, so no new tests. 

        * svg/SVGAllInOne.cpp:
        * svg/SVGImage.cpp:
        * svg/SVGLength.cpp:
        (WebCore::SVGLength::setValue):
        * svg/SVGPolygonElement.h:
        * svg/SVGPolylineElement.h:
        * svg/SVGPreserveAspectRatio.cpp:
        (WebCore::SVGPreserveAspectRatio::parsePreserveAspectRatio):
        (WebCore::SVGPreserveAspectRatio::transformRect):
        * svg/SVGSVGElement.cpp:
        (WebCore::SVGSVGElement::viewport):
        * svg/SVGSetElement.h:
        * svg/SVGStylable.h:
        (WebCore::SVGStylable::~SVGStylable):
        * svg/SVGStyledLocatableElement.h:
        (WebCore::SVGStyledLocatableElement::localCoordinateSpaceTransform):
        (WebCore::SVGStyledLocatableElement::isStyledLocatable):
        * svg/SVGStyledTransformableElement.cpp:
        (WebCore::SVGStyledTransformableElement::isKnownAttribute):
        * svg/SVGSwitchElement.cpp:
        (WebCore::SVGSwitchElement::childShouldCreateRenderer):
        * svg/SVGTests.cpp:
        (WebCore::SVGTests::parseMappedAttribute):
        * svg/SVGTextContentElement.cpp:
        (WebCore::SVGTextContentElement::isKnownAttribute):
        * svg/SVGTextPathElement.cpp:
        * svg/SVGTextPathElement.h:
        * svg/SVGTitleElement.h:
        (WebCore::SVGTitleElement::rendererIsNeeded):
        * svg/SVGTransformDistance.cpp:
        (WebCore::SVGTransformDistance::SVGTransformDistance):
        (WebCore::SVGTransformDistance::scaledDistance):
        (WebCore::SVGTransformDistance::addSVGTransforms):
        (WebCore::SVGTransformDistance::addSVGTransform):
        (WebCore::SVGTransformDistance::addToSVGTransform):
        (WebCore::SVGTransformDistance::isZero):
        * svg/SVGTransformList.cpp:
        * svg/SVGURIReference.cpp:
        (WebCore::SVGURIReference::getTarget):
        * svg/SVGVKernElement.h:
        * svg/SVGViewSpec.cpp:
        (WebCore::SVGViewSpec::parseViewSpec):
        * svg/SVGZoomAndPan.h:
        (WebCore::SVGZoomAndPan::SVGZoomAndPan):
        (WebCore::SVGZoomAndPan::~SVGZoomAndPan):
        (WebCore::SVGZoomAndPan::zoomAndPan):
        * svg/SVGZoomEvent.h:
        (WebCore::SVGZoomEvent::create):
        * svg/animation/SMILTime.cpp:
        (WebCore::operator*):
        * svg/animation/SMILTime.h:
        (WebCore::SMILTime::SMILTime):
        (WebCore::SMILTime::unresolved):
        (WebCore::SMILTime::indefinite):
        (WebCore::SMILTime::operator=):
        (WebCore::SMILTime::value):
        (WebCore::SMILTime::isFinite):
        (WebCore::SMILTime::isIndefinite):
        (WebCore::SMILTime::isUnresolved):
        (WebCore::operator==):
        (WebCore::operator!): new operator checks for 0 or infinite values.
        (WebCore::operator!=):
        (WebCore::operator>):
        (WebCore::operator<):
        (WebCore::operator>=):
        (WebCore::operator<=):
        * svg/animation/SMILTimeContainer.h:
        (WebCore::SMILTimeContainer::create):
        (WebCore::SMILTimeContainer::setDocumentOrderIndexesDirty):
        * svg/animation/SVGSMILElement.cpp:
        (WebCore::ConditionEventListener::operator==):
        (WebCore::SVGSMILElement::repeatingDuration):
        (WebCore::SVGSMILElement::resolveInterval):
        (WebCore::SVGSMILElement::calculateAnimationPercentAndRepeat):
        * svg/animation/SVGSMILElement.h:
        (WebCore::SVGSMILElement::timeContainer):
        (WebCore::SVGSMILElement::intervalBegin):
        (WebCore::SVGSMILElement::intervalEnd):
        (WebCore::SVGSMILElement::previousIntervalBegin):
        (WebCore::SVGSMILElement::documentOrderIndex):
        (WebCore::SVGSMILElement::setDocumentOrderIndex):
        * svg/graphics/SVGImage.h:
        (WebCore::SVGImage::create):
        (WebCore::SVGImage::destroyDecodedData):
        (WebCore::SVGImage::decodedSize):
        (WebCore::SVGImage::frameAtIndex):
        * svg/graphics/filters/SVGFilterBuilder.h:
        (WebCore::SVGFilterBuilder::create):
        (WebCore::SVGFilterBuilder::lastEffect):
        (WebCore::SVGFilterBuilder::getEffectReferences):
        (WebCore::SVGFilterBuilder::addBuiltinEffects):

2010-12-29  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Darin Adler.

        ASSERT(oldchild->parent() == owner) fails.
        https://bugs.webkit.org/show_bug.cgi?id=50480

        In RenderBlock removeChild function, when the inlineChildrenBlock(equal to prev or next)
        is reparented to blockChildrenBlock, it is no longer a child of "this". This causes the
        assertion failure when removeChildNode executes on the child(equal to prev or next).
        Fix a typo in canMergeContiguousAnonymousBlocks.

        Test: fast/multicol/span/double-merge-anonymous-block-crash.html

        * rendering/RenderBlock.cpp:
        (WebCore::canMergeContiguousAnonymousBlocks): fix typo, change prev to next.
        (WebCore::RenderBlock::removeChild): if prev or not is reparented, then set it to zero.

2010-12-29  Justin Schuh  <jschuh@chromium.org>

        Reviewed by Darin Adler.

        Check SVG element type in FrameView::scrollToAnchor
        https://bugs.webkit.org/show_bug.cgi?id=51718

        Test: svg/custom/scroll-to-anchor-in-symbol.svg

        * page/FrameView.cpp:
        (WebCore::FrameView::scrollToAnchor):

2010-12-29  Anton Muhin  <antonm@chromium.org>

        Reviewed by Eric Seidel.

        [v8] Minor cleanup: remove unused method (the only method which can mutate context of V8DOMWindowShell)
        https://bugs.webkit.org/show_bug.cgi?id=51704

        * bindings/v8/V8DOMWindowShell.cpp:
        * bindings/v8/V8DOMWindowShell.h:

2010-12-29  Kenichi Ishibashi  <bashi@google.com>

        Reviewed by Darin Adler.

        Use a HashMap for m_continuation to save memory
        https://bugs.webkit.org/show_bug.cgi?id=43716

        Saving memory consumption by applying a HashMap convention to
        continuation pointers of RenderInline and RenderBlock classes.

        * rendering/RenderBlock.cpp: Removed m_continuation.
        (WebCore::RenderBlock::RenderBlock):
        (WebCore::RenderBlock::destroy):
        (WebCore::RenderBlock::inlineElementContinuation):
        (WebCore::RenderBlock::blockElementContinuation):
        * rendering/RenderBlock.h: Removed m_continuation.
        * rendering/RenderBoxModelObject.cpp: Added a hash map for continuations.
        (WebCore::RenderBoxModelObject::destroy): Added an assertion.
        (WebCore::RenderBoxModelObject::continuation): Added.
        (WebCore::RenderBoxModelObject::setContinuation): Added.
        * rendering/RenderBoxModelObject.h:
        * rendering/RenderInline.cpp: Removed m_continuation.
        (WebCore::RenderInline::RenderInline):
        (WebCore::RenderInline::destroy):
        (WebCore::RenderInline::inlineElementContinuation):
        * rendering/RenderInline.h: Removed m_continuation.

2010-12-27  Amruth Raj  <amruthraj@motorola.com> and Ravi Kasibhatla  <ravi.kasibhatla@motorola.com>

        Reviewed by Martin Robinson.

        [GTK] Enable building whatever already exists of WebKit2
        https://bugs.webkit.org/show_bug.cgi?id=37369

        No new functionality added or deleted. Only makefile change. Hence, no tests added.

        * GNUmakefile.am: Removed bindings/gobject from webcore_sources & webcore_cppflags and
        added them to WebKit/gtk/GNUmakefile.am
        * platform/network/soup/cache/webkit/soup-cache.h: Remove include <webkit/webkitdefines.h>
        and declare WEBKIT_API directly

2010-12-29  Eric Seidel  <eric@webkit.org>

        Unreviewed.

        Simplify make-hash-tools.pl
        https://bugs.webkit.org/show_bug.cgi?id=49922

        Added HashTools.h to the project file now that its not autogenerated.
        I also sorted the project file using sort-xcode-project-file

        * WebCore.xcodeproj/project.pbxproj:

2010-12-29  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: revert r74755 and 74757.

2010-12-29  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Sam Weinig.

        JSDataViewCustom.cpp gives the fastcall calling convention to functions called via C++
        https://bugs.webkit.org/show_bug.cgi?id=51722

        Remove the JSC_HOST_CALL from methods that are called from C++. JSC_HOST_CALL gives
        methods the fastcall calling convention, which leads to runtime errors when they are
        called from C++. Also remove a bit of unnecessary code duplication.

        No new tests. This is covered by fast/canvas/webgl/data-view-test.html.

        * bindings/js/JSDataViewCustom.cpp:
        (WebCore::getDataViewMember): Remove duplicated code.
        (WebCore::JSDataView::getInt8): Remove JSC_HOST_CALL.
        (WebCore::JSDataView::getUint8): Ditto.
        (WebCore::JSDataView::getFloat32): Ditto.
        (WebCore::JSDataView::getFloat64): Ditto.
        (WebCore::setDataViewMember): Remove duplicated code.
        (WebCore::JSDataView::setInt8): Remove JSC_HOST_CALL.
        (WebCore::JSDataView::setUint8): Ditto.

2010-12-29  Dan Bernstein  <mitz@apple.com>

        Reviewed by Kenneth Russel.

        A more robust fix for https://bugs.webkit.org/show_bug.cgi?id=51681

        * html/HTMLCanvasElement.cpp:
        (WebCore::HTMLCanvasElement::attach): Added. Like recalcStyle(), calls
        CanvasRenderingContext2D::updateFont() if necessary. This covers the case of a detach/
        attach-type style recalc.
        * html/HTMLCanvasElement.h:
        * html/canvas/CanvasRenderingContext2D.cpp:
        (WebCore::CanvasRenderingContext2D::drawTextInternal): Removed the early return added in
        r74716. A font that is loading custom fonts is okay to use, as long as it is valid.
        (WebCore::CanvasRenderingContext2D::accessFont): Added a call to
        Document::updateStyleIfNeeded(). This ensures that any pending style recalc will take place
        and update the font if it is invalid.
        * platform/graphics/Font.h:
        (WebCore::Font::loadingCustomFonts): Made this private.

2010-12-29  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: Qt build fix.

        * inspector/InspectorController.cpp:

2010-12-29  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: extract Database & DOM Storage
        agents; remove InspectorBackend.
        https://bugs.webkit.org/show_bug.cgi?id=51707

        This change brushes up storage agents + removes redundant
        InspectorBackend class.

        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * inspector/CodeGeneratorInspector.pm:
        * inspector/InjectedScriptHost.cpp:
        (WebCore::InjectedScriptHost::databaseForId):
        (WebCore::InjectedScriptHost::selectDatabase):
        (WebCore::InjectedScriptHost::selectDOMStorage):
        (WebCore::InjectedScriptHost::inspectorDOMAgent):
        * inspector/Inspector.idl:
        * inspector/InspectorBackend.cpp: Removed.
        * inspector/InspectorBackend.h: Removed.
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::InspectorController):
        (WebCore::InspectorController::~InspectorController):
        (WebCore::InspectorController::connectFrontend):
        (WebCore::InspectorController::releaseFrontendLifetimeAgents):
        (WebCore::InspectorController::didUseDOMStorage):
        (WebCore::InspectorController::setInjectedScriptSource):
        (WebCore::InspectorController::dispatchOnInjectedScript):
        (WebCore::InspectorController::releaseWrapperObjectGroup):
        * inspector/InspectorController.h:
        * inspector/InspectorDOMStorageAgent.cpp: Added.
        (WebCore::InspectorDOMStorageAgent::~InspectorDOMStorageAgent):
        (WebCore::InspectorDOMStorageAgent::getDOMStorageEntries):
        (WebCore::InspectorDOMStorageAgent::setDOMStorageItem):
        (WebCore::InspectorDOMStorageAgent::removeDOMStorageItem):
        (WebCore::InspectorDOMStorageAgent::selectDOMStorage):
        (WebCore::InspectorDOMStorageAgent::InspectorDOMStorageAgent):
        (WebCore::InspectorDOMStorageAgent::getDOMStorageResourceForId):
        * inspector/InspectorDOMStorageAgent.h: Copied from WebCore/inspector/InspectorStorageAgent.h.
        (WebCore::InspectorDOMStorageAgent::create):
        (WebCore::InspectorDOMStorageAgent::frontend):
        * inspector/InspectorDatabaseAgent.cpp: Renamed from WebCore/inspector/InspectorStorageAgent.cpp.
        (WebCore::InspectorDatabaseAgent::~InspectorDatabaseAgent):
        (WebCore::InspectorDatabaseAgent::getDatabaseTableNames):
        (WebCore::InspectorDatabaseAgent::executeSQL):
        (WebCore::InspectorDatabaseAgent::databaseForId):
        (WebCore::InspectorDatabaseAgent::selectDatabase):
        (WebCore::InspectorDatabaseAgent::clearFrontend):
        (WebCore::InspectorDatabaseAgent::InspectorDatabaseAgent):
        * inspector/InspectorDatabaseAgent.h: Renamed from WebCore/inspector/InspectorStorageAgent.h.
        (WebCore::InspectorDatabaseAgent::create):
        (WebCore::InspectorDatabaseAgent::frontend):
        * inspector/InspectorFrontendClientLocal.cpp:
        * inspector/InspectorFrontendClientLocal.h:
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype.show):

2010-12-29  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Eric Seidel.

        Assertion failure: element->inDocument() in AsyncScriptRunner::executeScriptSoon()
        https://bugs.webkit.org/show_bug.cgi?id=51067

        Typically when a script element is removed from the document, the cached script
        client is removed. However, during the before load event, the cached script client
        hasn't been created yet so it can't be removed.

        This patch handles that case by explicitly checking if the script element was
        removed during the beforeload event. Also, it avoids caching the Document references
        over the arbitrary script execution in the before load event.

        Test: fast/dom/HTMLScriptElement/move-in-beforeload.html
              fast/dom/HTMLScriptElement/remove-in-beforeload.html

        * dom/ScriptElement.cpp:
        (WebCore::ScriptElement::requestScript):

2010-12-29  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: Element highlight tooltip displays wrong element size when zoomed in
        https://bugs.webkit.org/show_bug.cgi?id=51703

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::drawElementTitle):

2010-12-29  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Kent Tamura.

        [HTML5][Forms] Support for :in-range and :out-of-range CSS selectors
        https://bugs.webkit.org/show_bug.cgi?id=29071

        Use InputType::rangeUnderflow() and InputType::rangeOverflow()
        to determine if a control is in-range or out-of-range.

        Tests: fast/css/pseudo-in-range-invalid-value.html
               fast/css/pseudo-in-range.html
               fast/css/pseudo-out-of-range.html

        * css/CSSSelector.cpp:
        (WebCore::CSSSelector::pseudoId):
        (WebCore::nameToPseudoTypeMap):
        (WebCore::CSSSelector::extractPseudoType):
        * css/CSSSelector.h:
        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::canShareStyleWithElement):
        (WebCore::CSSStyleSelector::SelectorChecker::checkOneSelector):
        * dom/Element.h:
        (WebCore::Element::isInRange):
        (WebCore::Element::isOutOfRange):
        * html/BaseDateAndTimeInputType.cpp:
        (WebCore::BaseDateAndTimeInputType::supportsRangeLimitation):
&