2009-06-17  Brent Fulgham  <bfulgham@webkit.org>

        Reviewed by Gustavo Noronha Silva.

        Fixes: https://bugs.webkit.org/show_bug.cgi?id=26470.
        The use of zero-width or zero-height rectangles in generating
        gradients caused Windows Cairo to crash, and webkitgtk to
        produce invalid images.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-17  Ojan Vafai  <ojan@chromium.org>

        Reviewed by NOBODY (OOPS!).

        If loading a font fails because of the sandbox, we ask the browser process to
        try to load it by calling ensureFontLoaded. If it still fails after
        ensureFontLoaded, we hit a ASSERT_NOT_REACHED.

        This case happens once in a while during browser shutdown. The browser will
        queue a message to the renderer to shutdown, and will then stop answering sync
        messages from the renderer. If the renderer is still loading a page during this
        time, it might try to call the browser process to ask to load a font. The
        browser process will ignore the request, and the font will fail to load, even
        after the second try.

        This is unfortunate, but there is no real risk here, since the renderer will be
        going away as soon as it processes another message.

        This can't be layout tested as it depends on the sandbox.

        https://bugs.webkit.org/show_bug.cgi?id=26484

        * platform/graphics/chromium/FontCacheChromiumWin.cpp:
        (WebCore::fontContainsCharacter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-17  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=26482
        <rdar://problem/6978590> When setting playback rate to 0, the audio element stops
        playing (doesn't resume later)

        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivate::setRate): Always set the rate, even when "paused".

        * platform/graphics/win/MediaPlayerPrivateQuickTimeWin.cpp:
        (WebCore::MediaPlayerPrivate::setRate): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Add all the generated JS*.cpp files back to WebCore.vcproj

        They are not compiled by the vcproj directly. Instead they are
        compiled as part of DerivedSources.cpp. Having them listed in the
        vcproj makes them be included in Project Find, etc.

        Rubber-stamped in advance by Steve Falkenburg.

        * WebCore.vcproj/WebCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Fisher.

        Don't fire redundant 'change' events for a file upload form.
        https://bugs.webkit.org/show_bug.cgi?id=26471

        * platform/FileChooser.cpp:
        (WebCore::FileChooser::chooseFiles): Suppress change event if the
        existing selected files and the incoming selected files are equal.
        (WebCore::FileChooser::chooseIcon): Returns 0 if there is no selected
        files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-17 Adam Treat <adam.treat@torchmobile.com>

Reviewed by George Staikos.

https://bugs.webkit.org/show_bug.cgi?id=23155
Move WIN_CE -> WINCE as previously discussed with Qt WINCE folks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-17 George Staikos <george.staikos@torchmobile.com>

Reviewed by Adam Treat.

https://bugs.webkit.org/show_bug.cgi?id=23155
Move WIN_CE -> WINCE as previously discussed with Qt WINCE folks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Fix Skipped file format after r44519

        * platform/qt/Skipped:
        * platform/win/Skipped:
        Removed bogus ": Added."s in these files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Speculative Mac build fix

        * page/Page.h: Forward-declare RenderTheme instead of including
        RenderTheme.h so that we don't need to make RenderTheme.h a private
        header for WebKit's benefit.

        * editing/SelectionController.cpp:
        * rendering/InlineTextBox.cpp:
        * rendering/RenderObject.cpp:
        Added #includes of RenderTheme.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Speculative Mac build fix

        * rendering/RenderThemeMac.mm:
        (WebCore::RenderTheme::create): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Speculative Mac build fix

        * rendering/RenderThemeMac.mm:
        (WebCore::RenderTheme::themeForPage): Remove the name of an unused
        parameter.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Speculative Mac build fix

        * page/Page.cpp:
        (WebCore::Page::Page): Change the initializer order to match the
        declaration order.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Kenneth Rohde Christiansen  <kenneth.christiansen@openbossa.org>

        Reviewed by Simon Hausmann.

        Follow up to bug https://bugs.webkit.org/show_bug.cgi?id=26278
        Patch that make WebCore have a RenderTheme per page

        Make the Qt implementation of RenderTheme create a theme per page,
        and use the QStyle associated with the view of each page, in order
        to make the QWidget setStyle() method work as advertised.

        * platform/qt/RenderThemeQt.cpp:
        (WebCore::RenderThemeQt::RenderThemeQt):
        (WebCore::RenderThemeQt::qStyle):
        (WebCore::findFrameLineWidth):
        (WebCore::inflateButtonRect):
        (WebCore::RenderThemeQt::adjustRepaintRect):
        (WebCore::RenderThemeQt::isControlStyled):
        (WebCore::RenderThemeQt::computeSizeBasedOnStyle):
        (WebCore::RenderThemeQt::setButtonPadding):
        (WebCore::RenderThemeQt::paintButton):
        (WebCore::RenderThemeQt::paintTextField):
        (WebCore::RenderThemeQt::setPopupPadding):
        * platform/qt/RenderThemeQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16 Kenneth Rohde Christiansen <kenneth.christiansen@openbossa.org>

Reviewed by Dave Hyatt and Adam Roben.

        https://bugs.webkit.org/show_bug.cgi?id=26278
        Patch that make WebCore have a RenderTheme per page

        Create a different RenderTheme per page, so that RenderTheme has
        access to page specific theming. This is needed for the Qt port, as Qt
        supports setting the theme (style) per widget.

        This change was suggested and discussed with Dave Hyatt.

        More detailed:

        1) Create a theme per page or one global one, depending on the needs
           of the platform.
        2) Add an accesser to the theme from RenderObject.
        3) Change all uses of the theming to access the theme through
           RenderObject, using the global default theme as fallback, when the
           document of RenderObject has no page.
           When we don't have access to a RenderObject, use the default theme.
        4) Modify all RenderTheme platform implementations to work with the
           above changes, still creating only one global theme.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

JavaScriptCore:

2009-06-17  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver Hunt.

        <rdar://problem/6974175> ASSERT in JITStubs.cpp at appsaccess.apple.com

        Remove PropertySlot::putValue - PropertySlots should only be used for getting,
        not putting.  Rename JSGlobalObject::getOwnPropertySlot to hasOwnPropertyForWrite,
        which is what it really was being used to ask, and remove some other getOwnPropertySlot
        & getOwnPropertySlotForWrite methods, which were unused and likely to lead to confusion.

        * runtime/JSGlobalObject.h:
        (JSC::JSGlobalObject::hasOwnPropertyForWrite):
        * runtime/JSObject.h:
        * runtime/JSStaticScopeObject.cpp:
        * runtime/JSStaticScopeObject.h:
        * runtime/PropertySlot.h:

WebCore:

2009-06-17  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver Hunt.

        <rdar://problem/6974175> ASSERT in JITStubs.cpp at appsaccess.apple.com

        JSDOMWindowCustom was using PropertySlot::putValue, however this interface
        appears to be fundaementally incorrect - PropertySlots are only used to get
        values, all puts use PutPropertySlot.  However PutPropertySlot cannot be
        used in the fashion desired here - it only reports the caching type of a
        write that has been performed.

        (This caused a bug where the put should have triggered a transition, and
        failed to do so.)

        Removing the faulty case from the optimization leads to a ~0.5% progression
        on in-browser SunSpider (presumably the very first case was not being hit
        often, and the simplification here is beneficial).

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::put):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-17  Markus Goetz <Markus.Goetz@nokia.com>

        Reviewed by Simon Hausmann.

        QWebPage: Don't call supportsSsl()

        This stops QWebPage from loading the OpenSSL libs,
        certificates etc. when they are not needed for the non-HTTPS case.

        Loading the SSL libraries can be a very slow operation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

2009-06-17  David Levin  <levin@chromium.org>

        Reviewed by NOBODY, layout tests fix.

        https://bugs.webkit.org/show_bug.cgi?id=26326
        This reverts commit r44751.

        Once that change was checked scrollbars/scrollbar-orientation.html started
        crashing on Windows.

        * dom/Document.cpp:
        (WebCore::Document::detach):
        * page/FrameView.cpp:
        * page/FrameView.h:
        * platform/Scrollbar.h:
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::destroy):
        * rendering/RenderScrollbar.h:

LayoutTests:

2009-06-17  David Levin  <levin@chromium.org>

        Reviewed by NOBODY, layout tests fix.

        https://bugs.webkit.org/show_bug.cgi?id=26326
        This reverts commit r44751.

        Once that change was checked scrollbars/scrollbar-orientation.html started
        crashing on Windows.

        * scrollbars/resources/page-with-custom-scrollbars.html: Removed.
        * scrollbars/scrollbar-crash-on-refresh-expected.txt: Removed.
        * scrollbars/scrollbar-crash-on-refresh.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Kevin Ollivier.

In ImageSource::setData, delete the old m_encoder before replacing it with a new one.

https://bugs.webkit.org/show_bug.cgi?id=26458

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein

        <rdar://problem/6976766> Control-click on pages with accelerated compositing
        content does not work.

        Part deux. #ifdef the use of _private->layerHostingView with
        USE(ACCELERATED_COMPOSITING)

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein

        <rdar://problem/6976766> Control-click on pages with accelerated compositing
        content does not work.

        Remove the WebLayerHostingView subclass of NSView, with its attempts to
        forward events to the WebHTMLView, and just fix -[WebHTMLView hitTest:]
        to ignore the layerHostingView.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

2009-06-16  David Levin  <levin@chromium.org>

        Reviewed by David Hyatt.

        REGRESSION: When the main page (ScrollView) has a custom scrollbar, it crashes on destruction.
        https://bugs.webkit.org/show_bug.cgi?id=26326

        Test: scrollbars/scrollbar-crash-on-refresh.html

        * dom/Document.cpp:
        (WebCore::Document::detach):
        Gives the FrameView a change to do any necessary cleanup on
        Document::detach() which is where the renderArena gets detroyed.
        * page/FrameView.cpp:
        (WebCore::FrameView::detachCustomScrollbars):
        Gets rid of any custom scrollbars (if the docment supplied them).
        * page/FrameView.h:
        * platform/Scrollbar.h:
        (WebCore::Scrollbar::isCustomScrollbar):
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::destroy):
        Removed the check for document()->frame().  If frame() is 0 in this code,
        then the call to animation() is also incorrect (since it does document()->frame()->animation()).
        * rendering/RenderScrollbar.h:
        (WebCore::RenderScrollbar::isCustomScrollbar):

LayoutTests:

2009-06-16  David Levin  <levin@chromium.org>

        Reviewed by David Hyatt.

        REGRESSION: When the main page (ScrollView) has a custom scrollbar, it crashes on destruction.
        https://bugs.webkit.org/show_bug.cgi?id=26326

        Test that refreshing an iframe with custom scrollbars works without crashing.

        * scrollbars/resources/page-with-custom-scrollbars.html: Added.
        * scrollbars/scrollbar-crash-on-refresh-expected.txt: Added.
        * scrollbars/scrollbar-crash-on-refresh.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Brian Weinstein  <bweinstein@apple.com>

        Reviewed by Adele Peterson.

        Fix of <rdar://6967547> Ctrl-C copies null value erasing text in clipboard in Safari.
        This patch kept the way events were firing as they were before, and Windows events are
        consistent with Mac.

        * editing/Editor.cpp:
        (WebCore::Editor::tryDHTMLCopy): Added check for canCopy() before clearing PasteBoard
        (WebCore::Editor::tryDHTMLCut): Added check for canCut() before clearing PasteBoard

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Antti Koivisto  <antti@apple.com>

        Reviewed by Brady Eidson.

        <rdar://problem/6660037> CrashTracer: [USER] 46 crashes in Safari at com.apple.WebCore • WebCore::CachedCSSStyleSheet::addClient + 53

        When revalidating a resource, calling addClient() on one client might cause another to get removed.

        - made CachedResource::addClient() non-virtual and added virtual didAddClient()
        - in CachedResource::switchClientsToRevalidatedResource() add all clients to the client set of the revalidated resource first
        - check if the client is still in the set before invoking didAddClient() for it

        No test case, I didn't manage to construct one. You need some combination of 304 revalidation, stylesheets that
        reference each other via @imports and reloading.

        * WebCore.base.exp:
        * loader/CachedCSSStyleSheet.cpp:
        (WebCore::CachedCSSStyleSheet::didAddClient):
        * loader/CachedCSSStyleSheet.h:
        * loader/CachedFont.cpp:
        (WebCore::CachedFont::didAddClient):
        * loader/CachedFont.h:
        * loader/CachedImage.cpp:
        (WebCore::CachedImage::didAddClient):
        * loader/CachedImage.h:
        * loader/CachedResource.cpp:
        (WebCore::CachedResource::addClient):
        (WebCore::CachedResource::addClientToSet):
        (WebCore::CachedResource::switchClientsToRevalidatedResource):
        * loader/CachedResource.h:
        * loader/CachedScript.cpp:
        (WebCore::CachedScript::didAddClient):
        * loader/CachedScript.h:
        * loader/CachedXSLStyleSheet.cpp:
        (WebCore::CachedXSLStyleSheet::didAddClient):
        * loader/CachedXSLStyleSheet.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver hunt.

        Temporarily partially disable r44492, since this is causing some problems on internal builds.

        * jit/JITOpcodes.cpp:
        (JSC::JIT::emit_op_throw):
        * jit/JITStubs.cpp:
        (JSC::JITStubs::DEFINE_STUB_FUNCTION):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Simon Fraser  <simon.fraser@apple.com>

        No Review

        Fix code inside an #ifdef that draws the video framerate.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivate::paint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Sam Weinig  <sam@webkit.org>

        Fix windows build.

        * jit/JIT.cpp:
        (JSC::JIT::JIT):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Jian Li  <jianli@chromium.org>

        Reviewed by Dimitri Glazkov.

        Bug 26456: Hook up V8 bindings for Worker's importScripts functionality.
        https://bugs.webkit.org/show_bug.cgi?id=26456

        * bindings/v8/custom/V8WorkerContextCustom.cpp:
        (WebCore::CALLBACK_FUNC_DECL):
        * bindings/v8/custom/V8WorkerCustom.cpp: Fixed missing exception code
          handling in Worker constructor for V8 bindings.
        (WebCore::CALLBACK_FUNC_DECL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Sam Weinig  <sam@webkit.org>

        Reviewed by Oliver Hunt.

        Initialize m_bytecodeIndex to -1 in JIT, and correctly initialize
        it for each type of stub using the return address to find the correct
        offset.

        * jit/JIT.cpp:
        (JSC::JIT::JIT):
        * jit/JIT.h:
        (JSC::JIT::compileGetByIdProto):
        (JSC::JIT::compileGetByIdSelfList):
        (JSC::JIT::compileGetByIdProtoList):
        (JSC::JIT::compileGetByIdChainList):
        (JSC::JIT::compileGetByIdChain):
        (JSC::JIT::compilePutByIdTransition):
        (JSC::JIT::compileCTIMachineTrampolines):
        (JSC::JIT::compilePatchGetArrayLength):
        * jit/JITStubCall.h:
        (JSC::JITStubCall::call):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Added spec directories from dev.w3.org/cvsweb/csswg/
        which is where they will live. We will edit them here and
        then push them to the W3C cvs server.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Removed all the specs because they are stale

        index.html: Redirected to point at w3c site

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Brent Fulgham  <bfulgham@webkit.org>

        Reviewed by Maciej Stachowiak.

        Update of https://bugs.webkit.org/show_bug.cgi?id=26353.
        Provide an assignment operator to avoid improper reference
        counts on the Cairo font objects.  This brings the Windows
        Cairo port in line with the GTK+ port.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Brian Weinstein  <bweinstein@apple.com>

        Reviewed by Adam Roben.

        Switch Control+Mousewheel Zooming definitions to match other browsers on Windows.
        Control+Mousewheel-Down now zooms out, Control+Mousewheel-Up now zooms in.
        https://bugs.webkit.org/show_bug.cgi?id=25875.
        <rdar://problem/6903976>

        * WebView.cpp:
        (WebView::mouseWheel):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        Fix for <rdar://problem/6890126> Theme code should fetch the MediaControlElementType from
        the MediaControlInputElement, rather than computing it again

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::percentLoaded): New, utility function for controller implementation.
        * html/HTMLMediaElement.h:

        * rendering/MediaControlElements.h:
        (WebCore::MediaControlInputElement::displayType): New, return m_displayType.

        * rendering/RenderMediaControls.cpp:
        (WebCore::RenderMediaControls::paintMediaControlsPart): Stop using MediaPlayer object, get button
        state from the button itself and get movie state from HTMLMediaElement.

        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::paintMediaMuteButton): Get state from button instead of MediaPlayer.
        (WebCore::RenderThemeMac::paintMediaPlayButton): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Jian Li  <jianli@chromium.org>

        Reviewed by David Levin.

        Bug 26450: Rename values of enum RedirectOriginCheck to make them
        clearer.
        https://bugs.webkit.org/show_bug.cgi?id=26450

        * loader/DocumentThreadableLoader.cpp:
        (WebCore::DocumentThreadableLoader::create):
        (WebCore::DocumentThreadableLoader::DocumentThreadableLoader):
        (WebCore::DocumentThreadableLoader::willSendRequest):
        * loader/DocumentThreadableLoader.h:
        * loader/ThreadableLoader.cpp:
        (WebCore::ThreadableLoader::create):
        (WebCore::ThreadableLoader::loadResourceSynchronously):
        * loader/ThreadableLoader.h:
        * workers/Worker.cpp:
        (WebCore::Worker::Worker):
        * workers/WorkerContext.cpp:
        (WebCore::WorkerContext::importScripts):
        * loader/WorkerThreadableLoader.cpp:
        (WebCore::WorkerThreadableLoader::WorkerThreadableLoader):
        (WebCore::WorkerThreadableLoader::loadResourceSynchronously):
        (WebCore::WorkerThreadableLoader::MainThreadBridge::MainThreadBridge):
        (WebCore::WorkerThreadableLoader::MainThreadBridge::mainThreadCreateLoader):
        * loader/WorkerThreadableLoader.h:
        (WebCore::WorkerThreadableLoader::create):
        * workers/WorkerScriptLoader.cpp:
        (WebCore::WorkerScriptLoader::loadSynchronously):
        (WebCore::WorkerScriptLoader::loadAsynchronously):
        * workers/WorkerScriptLoader.h:
        (WebCore::WorkerContext::importScripts):
        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::loadRequestAsynchronously):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Tony Chang  <tony@chromium.org>

        Reviewed by Darin Fisher.

        Fix a UMR in WebCore::BitStack by initializing new memory to 0.
        https://bugs.webkit.org/show_bug.cgi?id=26449
        No new tests, covered by purify.

        * editing/TextIterator.cpp:
        (WebCore::BitStack::push):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

009-06-16  Brent Fulgham  <bfulgham@webkit.org>

        Reviewed by Xan Lopez.

        Fixes https://bugs.webkit.org/show_bug.cgi?id=26453.
        Null Cairo contextwill crash Windows Cairo build.

        Test: fast/multicol/columns-shorthand-parsing.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Peter Kasting  <pkasting@google.com>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=26447
        Fix animated GIF breakage in Cairo/wx ports.

        * platform/image-decoders/cairo/ImageDecoderCairo.cpp:
        (WebCore::RGBA32Buffer::copyBitmapData):
        (WebCore::RGBA32Buffer::operator=):
        * platform/image-decoders/wx/ImageDecoderWx.cpp:
        (WebCore::RGBA32Buffer::copyBitmapData):
        (WebCore::RGBA32Buffer::operator=):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Simon Fraser  <simon.fraser@apple.com>

        Rubber-stamped by Anders Carlsson.

        Update WebKitSystemInterface for <rdar://problem/6937882>.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Disable another new test.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by John Sullivan

        <rdar://problem/6937882>

        Tweak "time remaining" and "time elapsed" fields in the overlay video controller.

        * css/mediaControlsQT.css:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Do not use g_assert_cmpint to test for NULL pointer, just
        g_assert.

        * tests/testnetworkrequest.c:
        (test_network_request_create_destroy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        editing/deleting/4875189.html was changed from pixel to
        dumpAsText, so remove it from Skipped list.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=26437

        Make the commit-log-editor match the ambient line endings in commit
        messages.

        * Scripts/commit-log-editor:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

2009-06-16  Jian Li  <jianli@chromium.org>

        Reviewed by Adam Barth and David Levin.

        Bug 26146: Change to use ThreadableLoader to load the worker script
        in order to check URL origin for redirection.
        https://bugs.webkit.org/show_bug.cgi?id=26146

        Test: http/tests/workers/worker-redirect.html

        * GNUmakefile.am:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * workers/Worker.cpp:
        (WebCore::Worker::Worker):
        (WebCore::Worker::notifyFinished):
        * workers/Worker.h:
        * workers/WorkerContext.cpp:
        (WebCore::WorkerContext::importScripts):
        * workers/WorkerImportScriptsClient.cpp: Removed.
        * workers/WorkerImportScriptsClient.h: Removed.
        * workers/WorkerScriptLoader.cpp: Renamed from workers/WorkerImportScriptsClient.cpp.
          This to make it more generic so worker script loading could use it.
        (WebCore::WorkerScriptLoader::loadSynchronously):
        (WebCore::WorkerScriptLoader::loadAsynchronously):
        (WebCore::WorkerScriptLoader::didFinishLoading):
        (WebCore::WorkerScriptLoader::didFail):
        (WebCore::WorkerScriptLoader::didFailRedirectCheck):
        (WebCore::WorkerScriptLoader::didReceiveAuthenticationCancellation):
        (WebCore::WorkerScriptLoader::notifyFinished):
        * workers/WorkerScriptLoader.h: Renamed from workers/WorkerImportScriptsClient.h.
          This to make it more generic so worker script loading could use it.
        * workers/WorkerScriptLoaderClient.h: Added.

LayoutTests:

2009-06-16  Jian Li  <jianli@chromium.org>

        Reviewed by Adam Barth and David Levin.

        Bug 26146: Change to use ThreadableLoader to load the worker script
        in order to check URL origin for redirection.
        https://bugs.webkit.org/show_bug.cgi?id=26146

        Add a layout test to cover the scenario that the worker script is
        loaded from different redirect origin.

        * http/tests/workers/resources/worker-redirect-target.js: Added.
        * http/tests/workers/worker-redirect-expected.txt: Added.
        * http/tests/workers/worker-redirect.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Fix typo in test name.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Disable new tests.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

009-06-16  Brent Fulgham  <bfulgham@gmail.com>

        Reviewed by Darin Adler.

        Use consistent GUID comparison functions.
        https://bugs.webkit.org/show_bug.cgi?id=26427

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Fix compiler warning.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Remove dummy AtkStreamableContent implementation.

        It's completely empty, we'll add it back (and conditionally
        instead of unconditionally) when it does something.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (GetAtkInterfaceTypeFromWAIType):
        (getInterfaceMaskFromObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Update GtkLauncher to recent API changes in the progress property,
        which now goes from 0.0 to 1.0.

        * GtkLauncher/main.c:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  David Boddie  <dboddie@trolltech.com>

        Reviewed by Simon Hausmann.

        Doc: Fixed QWebPage::forwardUnsupportedContent documentation and added
        more information about Web plugins.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Morten Engvoldsen  <morten.engvoldsen@nokia.com>

        Reviewed by Ariya Hidayat.

        Clearifying QWebFrame docs

        Adding docs to toHtml() and toPlainText()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix ChangeLog typo.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-16  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Alder.

        https://bugs.webkit.org/show_bug.cgi?id=26000

        Teach prepare-ChangeLog to match the line ends that are already present
        in ChangeLog files.  This helps folks whose use cygwin perl with CR LF
        line endings on Windows.

        Also, teach prepare-ChangeLog to normalize backslashes in paths.  This
        helps folks who use Windows SVN prepare correct ChangeLogs.

        * Scripts/prepare-ChangeLog:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Roll over some ChangeLog files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add a script to automate the rolling over of ChangeLog files when they grow too large.

Reviewed by Oliver Hunt.

* Scripts/roll-over-ChangeLogs: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gavin Barraclough  <barraclough@apple.com>

        Rubber Stamped by Sam Weinig.

        Rename PatchBuffer to LinkBuffer.  Previously our terminology has been a little
        mixed up, but we have decided to fix on refering to the process that takes place
        at the end of code generation as 'linking', and on any modifications that take
        place later (and once the code has potentially already been executed) as 'patching'.

        However, the term 'PatchBuffer' is already in use, and needs to be repurposed.

        To try to minimize confusion, we're going to switch the terminology over in stages,
        so for now we'll refer to later modifications as 'repatching'.  This means that the
        new 'PatchBuffer' has been introduced with the name 'RepatchBuffer' instead.

        This patch renames the old 'PatchBuffer' to 'LinkBuffer'.  We'll leave ToT in this
        state for a week or so to try to avoid to much overlap of the meaning of the term
        'PatchBuffer', then will come back and rename 'RepatchBuffer'.

        * assembler/ARMv7Assembler.h:
        * assembler/AbstractMacroAssembler.h:
        (JSC::AbstractMacroAssembler::LinkBuffer::LinkBuffer):
        (JSC::AbstractMacroAssembler::LinkBuffer::~LinkBuffer):
        * jit/JIT.cpp:
        (JSC::JIT::privateCompile):
        * jit/JITPropertyAccess.cpp:
        (JSC::JIT::privateCompilePutByIdTransition):
        (JSC::JIT::privateCompilePatchGetArrayLength):
        (JSC::JIT::privateCompileGetByIdProto):
        (JSC::JIT::privateCompileGetByIdSelfList):
        (JSC::JIT::privateCompileGetByIdProtoList):
        (JSC::JIT::privateCompileGetByIdChainList):
        (JSC::JIT::privateCompileGetByIdChain):
        * yarr/RegexJIT.cpp:
        (JSC::Yarr::RegexGenerator::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

2009-06-15  Adele Peterson  <adele@apple.com>

        Reviewed by Darin Adler.

        Fix for <rdar://problem/6907228> Pasting a paragraph as quotation leaves you typing in blue instead of black; repros

        * editing/InsertParagraphSeparatorCommand.cpp: (WebCore::InsertParagraphSeparatorCommand::doApply):
        When inserting a newline after the blockquote, we don't want to apply the original style after the insertion.

LayoutTests:

2009-06-15  Adele Peterson  <adele@apple.com>

        Reviewed by Darin Adler.

        Updating test for:
        <rdar://problem/6907228> Pasting a paragraph as quotation leaves you typing in blue instead of black; repros

        This test case covered inserting a newline after a blockquote, but outside of the blockquote.  I added some
        text insertion to verify that the newline doesn't get any old styles copied from the blockquote.

        * editing/pasteboard/5368833.html:
        * platform/mac/editing/pasteboard/5368833-expected.checksum:
        * platform/mac/editing/pasteboard/5368833-expected.png:
        * platform/mac/editing/pasteboard/5368833-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Sam Weinig.

        Having moved most of their functionality into the RepatchBuffer class,
        we can simplify the CodeLocation* classes.

        The CodeLocation* classes are currently a tangle of templatey and friendly
        badness, burried in the middle of AbstractMacroAssembler.  Having moved
        the ability to repatch out into RepatchBufer they are now do-nothing wrappers
        on CodePtr (MacroAssemblerCodePtr), that only exist to provide type-safety.

        Simplify the code, and move them off into their own header.

        * JavaScriptCore.xcodeproj/project.pbxproj:
        * assembler/AbstractMacroAssembler.h:
        (JSC::AbstractMacroAssembler::PatchBuffer::patch):
        * assembler/CodeLocation.h: Copied from assembler/AbstractMacroAssembler.h.
        (JSC::CodeLocationCommon::CodeLocationCommon):
        (JSC::CodeLocationInstruction::CodeLocationInstruction):
        (JSC::CodeLocationLabel::CodeLocationLabel):
        (JSC::CodeLocationJump::CodeLocationJump):
        (JSC::CodeLocationCall::CodeLocationCall):
        (JSC::CodeLocationNearCall::CodeLocationNearCall):
        (JSC::CodeLocationDataLabel32::CodeLocationDataLabel32):
        (JSC::CodeLocationDataLabelPtr::CodeLocationDataLabelPtr):
        (JSC::CodeLocationCommon::instructionAtOffset):
        (JSC::CodeLocationCommon::labelAtOffset):
        (JSC::CodeLocationCommon::jumpAtOffset):
        (JSC::CodeLocationCommon::callAtOffset):
        (JSC::CodeLocationCommon::nearCallAtOffset):
        (JSC::CodeLocationCommon::dataLabelPtrAtOffset):
        (JSC::CodeLocationCommon::dataLabel32AtOffset):
        * assembler/MacroAssemblerCodeRef.h:
        (JSC::MacroAssemblerCodePtr::operator!):
        * bytecode/CodeBlock.h:
        (JSC::getStructureStubInfoReturnLocation):
        (JSC::getCallLinkInfoReturnLocation):
        (JSC::getMethodCallLinkInfoReturnLocation):
        * bytecode/Instruction.h:
        * bytecode/JumpTable.h:
        (JSC::StringJumpTable::ctiForValue):
        (JSC::SimpleJumpTable::ctiForValue):
        * bytecode/StructureStubInfo.h:
        * bytecompiler/BytecodeGenerator.cpp:
        (JSC::BytecodeGenerator::emitCatch):
        * jit/JIT.cpp:
        (JSC::JIT::privateCompile):
        * jit/JITStubs.cpp:
        (JSC::JITStubs::DEFINE_STUB_FUNCTION):
        (JSC::JITStubs::getPolymorphicAccessStructureListSlot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Remove outdated comment.

        * bindings/js/JSDOMWindowBase.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Bring CustomGetPropertyNames more inline with the other
        extended attributes prefixed with Custom, by making it
        require a complete implementation.

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::getPropertyNames):
        * bindings/js/JSHistoryCustom.cpp:
        (WebCore::JSHistory::getPropertyNames):
        * bindings/js/JSLocationCustom.cpp:
        (WebCore::JSLocation::getPropertyNames):
        * bindings/js/JSStorageCustom.cpp:
        (WebCore::JSStorage::getPropertyNames):
        * bindings/scripts/CodeGeneratorJS.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Remove commented out ondataunavailable attribute event listener which
        has now been removed from HTML 5.

        * page/DOMWindow.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Move entire custom implementation of getOwnPropertySlot and put
        for JSDOMWindow into JSDOMWindowCustom, eliminating the bit in
        JSDOMWindowBase.

        * bindings/js/JSDOMWindowBase.cpp:
        * bindings/js/JSDOMWindowBase.h:
        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::nonCachingStaticFunctionGetter):
        (WebCore::childFrameGetter):
        (WebCore::indexGetter):
        (WebCore::namedItemGetter):
        (WebCore::JSDOMWindow::getOwnPropertySlot):
        (WebCore::JSDOMWindow::put):
        (WebCore::JSDOMWindow::getPropertyAttributes):
        * bindings/js/JSDOMWindowCustom.h:
        * page/DOMWindow.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Jeremy Orlow (jorlow@chromium.org)

        Reviewed by Dimitri Glazkov.

        https://bugs.webkit.org/show_bug.cgi?id=26356
        Define the functions in V8CustomBindings.h.  "Forgot" this in my
        earlier patch since it wasn't fully upstreamed when I wrote the
        patch.

        * bindings/v8/custom/V8CustomBinding.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Sam Weinig.

        Having introduced the RepatchBuffer, ProcessorReturnAddress is now a do-nothing
        wrapper around ReturnAddressPtr.  Remove it.  In tugging on this piece of string
        it made sense to roll out the use of ReturnAddressPtr a little further into
        JITStubs (which had always been the intention).

        No performance impact.

        * assembler/AbstractMacroAssembler.h:
        (JSC::AbstractMacroAssembler::RepatchBuffer::relinkCallerToTrampoline):
        (JSC::AbstractMacroAssembler::RepatchBuffer::relinkCallerToFunction):
        (JSC::AbstractMacroAssembler::RepatchBuffer::relinkNearCallerToTrampoline):
        * assembler/MacroAssemblerCodeRef.h:
        (JSC::ReturnAddressPtr::ReturnAddressPtr):
        * bytecode/CodeBlock.h:
        (JSC::CodeBlock::getStubInfo):
        (JSC::CodeBlock::getCallLinkInfo):
        (JSC::CodeBlock::getMethodCallLinkInfo):
        (JSC::CodeBlock::getBytecodeIndex):
        * interpreter/Interpreter.cpp:
        (JSC::bytecodeOffsetForPC):
        * jit/JIT.cpp:
        (JSC::ctiPatchNearCallByReturnAddress):
        (JSC::ctiPatchCallByReturnAddress):
        * jit/JIT.h:
        (JSC::JIT::compileGetByIdProto):
        (JSC::JIT::compileGetByIdChain):
        (JSC::JIT::compilePutByIdTransition):
        (JSC::JIT::compilePatchGetArrayLength):
        * jit/JITPropertyAccess.cpp:
        (JSC::JIT::privateCompilePutByIdTransition):
        (JSC::JIT::patchGetByIdSelf):
        (JSC::JIT::patchPutByIdReplace):
        (JSC::JIT::privateCompilePatchGetArrayLength):
        (JSC::JIT::privateCompileGetByIdProto):
        (JSC::JIT::privateCompileGetByIdChain):
        * jit/JITStubs.cpp:
        (JSC::JITThunks::tryCachePutByID):
        (JSC::JITThunks::tryCacheGetByID):
        (JSC::StackHack::StackHack):
        (JSC::returnToThrowTrampoline):
        (JSC::throwStackOverflowError):
        (JSC::JITStubs::DEFINE_STUB_FUNCTION):
        * jit/JITStubs.h:
        (JSC::):
        (JSC::JITStackFrame::returnAddressSlot):
        * runtime/JSGlobalData.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Benjamin C Meyer  <benjamin.meyer@torchmobile.com>

        Reviewed by Adam Treat.

        Support the back/forward/stop/refresh multimedia keys and accept the
        event when handling backspace and shift backspace as we should.

        * Api/qwebpage.cpp:
        (QWebPagePrivate::keyPressEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Andre Boule  <aboule@apple.com>

        Reviewed by Cameron Zwarich and Dan Bernstein.

        - fix <rdar://problem/6967975> crash with NSString drawAtPoint

        No test possible because this crash can only occur if the client uses
        WebKit text drawing without ever using WebCore text, and DumpRenderTree
        cannot do either.

        * platform/graphics/FontCache.cpp:
        (WebCore::FontCache::purgeInactiveFontData): Null-check
            gFontPlatformDataCache.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Mark Rowe.

        <rdar://problem/6974857>

        Define ENABLE_3D_RENDERING and WTF_USE_ACCELERATED_COMPOSITING when building on 10.6.
        Move the ENABLE_3D_RENDERING switch from config.h and WebKitPrefix.h to wtf/Platform.h

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Make WebInspector's console evaluation/completion asynchronous.
        This way of evaluation allows serializing calls between
        InspectorController and its frontend.

        https://bugs.webkit.org/show_bug.cgi?id=26350

        * inspector/front-end/Console.js:
        (WebInspector.Console.prototype.completions):
        (WebInspector.Console.prototype._reportCompletions):
        (WebInspector.Console.prototype._evalInInspectedWindow):
        (WebInspector.Console.prototype._enterKeyPressed.printResult):
        (WebInspector.Console.prototype._enterKeyPressed):
        * inspector/front-end/DatabaseQueryView.js:
        * inspector/front-end/ObjectPropertiesSection.js:
        (WebInspector.ObjectPropertyTreeElement.prototype.evaluateExpression):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype.evaluateInSelectedCallFrame):
        * inspector/front-end/TextPrompt.js:
        (WebInspector.TextPrompt.prototype.complete):
        (WebInspector.TextPrompt.prototype._completionsReady):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver Hunt.

        Move repatching methods into a set of methods on a class.  This will allow us to
        coallesce memory reprotection calls.  Really, we want this class to be called
        PatchBuffer, we want the class PatchBuffer to be called LinkBuffer, we want both
        to be memblers of MacroAssembler rather then AbstractMacroAssembler, we don't
        want the CodeLocationFoo types anymore (they are now only really there to provide
        type safety, and that is completely undermined by the way we use offsets).  Then
        the link & patch buffers should delegate the actual patching calls to the
        architecture-specific layer of the MacroAssembler.  Landing all these changes as a
        sequence of patches.

        No performance impact.

        * assembler/AbstractMacroAssembler.h:
        (JSC::AbstractMacroAssembler::CodeLocationCall::CodeLocationCall):
        (JSC::AbstractMacroAssembler::CodeLocationNearCall::CodeLocationNearCall):
        (JSC::AbstractMacroAssembler::CodeLocationNearCall::calleeReturnAddressValue):
        (JSC::AbstractMacroAssembler::RepatchBuffer::RepatchBuffer):
        (JSC::AbstractMacroAssembler::RepatchBuffer::relink):
        (JSC::AbstractMacroAssembler::RepatchBuffer::repatch):
        (JSC::AbstractMacroAssembler::RepatchBuffer::relinkCallerToTrampoline):
        (JSC::AbstractMacroAssembler::RepatchBuffer::relinkCallerToFunction):
        (JSC::AbstractMacroAssembler::RepatchBuffer::relinkNearCallerToTrampoline):
        (JSC::AbstractMacroAssembler::RepatchBuffer::repatchLoadPtrToLEA):
        * jit/JIT.cpp:
        (JSC::ctiPatchNearCallByReturnAddress):
        (JSC::ctiPatchCallByReturnAddress):
        (JSC::JIT::unlinkCall):
        (JSC::JIT::linkCall):
        * jit/JITPropertyAccess.cpp:
        (JSC::JIT::privateCompilePutByIdTransition):
        (JSC::JIT::patchGetByIdSelf):
        (JSC::JIT::patchMethodCallProto):
        (JSC::JIT::patchPutByIdReplace):
        (JSC::JIT::privateCompilePatchGetArrayLength):
        (JSC::JIT::privateCompileGetByIdProto):
        (JSC::JIT::privateCompileGetByIdSelfList):
        (JSC::JIT::privateCompileGetByIdProtoList):
        (JSC::JIT::privateCompileGetByIdChainList):
        (JSC::JIT::privateCompileGetByIdChain):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6967569>
        CrashTracer: 15 crashes in Safari at com.apple.WebKit • WebKit::NetscapePluginHostManager::didCreateWindow + 85

        Make sure to remove the entry from the plug-in host map so we won't end up with an entry that has a null value.

        * Plugins/Hosted/NetscapePluginHostManager.mm:
        (WebKit::NetscapePluginHostManager::hostForPackage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcher.

        Reverted previous change as the old behavior was correct for heavy view.

        * inspector/front-end/BottomUpProfileDataGridTree.js:
        (WebInspector.BottomUpProfileDataGridNode.prototype._populate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcher.

        <rdar://problem/6968125> Profiler 'Call' count is incorrect in Heavy
        view (26418)

        - Removed code that copied call and time data from it's ancestor.

        * inspector/front-end/BottomUpProfileDataGridTree.js:
        (WebInspector.BottomUpProfileDataGridNode.prototype._populate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Simon Fraser  <simon.fraser@apple.com>.

        https://bugs.webkit.org/show_bug.cgi?id=26163

        Added logic to updateTransitions similar to that in updateKeyframeAnimations
        which detects and removes stale transitions.

        Test: transitions/cancel-transition.html

        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::updateTransitions):
        * page/animation/ImplicitAnimation.cpp:
        (WebCore::ImplicitAnimation::ImplicitAnimation):
        * page/animation/ImplicitAnimation.h:
        (WebCore::ImplicitAnimation::active):
        (WebCore::ImplicitAnimation::setActive):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

        Reviewed by Tim Hatcher.

        - allow the source code font in the Web Inspector to be customized

        * css/view-source.css: Use the default monospace font family and size
            for source code.
        * inspector/front-end/inspector.css: Use the default monospace font
            family and size for console I/O.

WebKit/mac:

        Reviewed by Tim Hatcher.

        - make the source code font in the Web Inspector match Mac defaults

        * WebCoreSupport/WebInspectorClient.mm:
        (-[WebInspectorWindowController init]): Set the default monospace font
        to 11px Menlo, except on Leopard and Tiger where it is 10px Monaco.

WebKit/win:

            Reviewed by Tim Hatcher.

        - use a nice source code font in the Web Inspector

        * WebCoreSupport/WebInspectorClient.cpp:
        (WebInspectorClient::createPage): Set the default monospace font to
            13px Courier New, which is the default source code font in Visual
            Studio.

LayoutTests:

        Reviewed by Tim Hatcher.

        - updated results after changing source view to use the default
          monospace font family and size

        * platform/mac/fast/frames/viewsource-attribute-expected.checksum:
        * platform/mac/fast/frames/viewsource-attribute-expected.png:
        * platform/mac/fast/frames/viewsource-attribute-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        https://bugs.webkit.org/show_bug.cgi?id=26377
        [GTK] Confusion about range of 'progress' property

        Range for the progress property is 0.0 .. 1.0, don't multiply it
        by 100 to make it a percentage.

        * webkit/webkitwebview.cpp:
        (webkit_web_view_get_progress):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Geoff Hunt & Oliver Garen.

        We are currently generating two copies of the slow path for op_call for no reason.  Stop that.

        Originally op_call used two slow paths since the first set up the pointer to the CallLinkInfo
        for use when linking.  However this is now looked up using the return address (as we do for
        property accesses) so the two paths are now identical.

        No performance impact, reduces memory footprint.

        * bytecode/CodeBlock.h:
        * jit/JIT.cpp:
        (JSC::JIT::privateCompile):
        (JSC::JIT::linkCall):
        * jit/JIT.h:
        * jit/JITCall.cpp:
        (JSC::JIT::compileOpCallSlowCase):
        * jit/JITStubs.cpp:
        (JSC::JITStubs::DEFINE_STUB_FUNCTION):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15 Andre Pedralho <andre.pedralho@openbossa.org>

Reviewed by Adam Treat

https://bugs.webkit.org/show_bug.cgi?id=26351
Remove bool QWebHitTestResult::isScrollBar() const and make sure a null QWebHitTestResult is returned instead.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Anders Carlsson  <andersca@apple.com>

        Reviewed by Kevin Decker.

        Add Core Animation movie plug-in example.

        * NetscapeCoreAnimationMoviePlugin: Added.
        * NetscapeCoreAnimationMoviePlugin/English.lproj: Added.
        * NetscapeCoreAnimationMoviePlugin/English.lproj/InfoPlist.strings: Added.
        * NetscapeCoreAnimationMoviePlugin/Info.plist: Added.
        * NetscapeCoreAnimationMoviePlugin/MovieControllerLayer.h: Added.
        * NetscapeCoreAnimationMoviePlugin/MovieControllerLayer.m: Added.
        (createImageNamed):
        (-[MovieControllerLayer init]):
        (-[MovieControllerLayer dealloc]):
        (-[MovieControllerLayer _playPauseButtonRect]):
        (-[MovieControllerLayer _sliderRect]):
        (-[MovieControllerLayer _sliderThumbRect]):
        (-[MovieControllerLayer _innerSliderRect]):
        (-[MovieControllerLayer _drawPlayPauseButtonInContext:]):
        (-[MovieControllerLayer _drawSliderInContext:]):
        (-[MovieControllerLayer drawInContext:]):
        (-[MovieControllerLayer _currentTime]):
        (-[MovieControllerLayer _duration]):
        (-[MovieControllerLayer _isPlaying]):
        (-[MovieControllerLayer _updateTime:]):
        (-[MovieControllerLayer _rateDidChange:]):
        (-[MovieControllerLayer _timeDidChange:]):
        (-[MovieControllerLayer actionForKey:]):
        (-[MovieControllerLayer setMovie:]):
        (-[MovieControllerLayer _setNewTimeForThumbCenterX:]):
        (-[MovieControllerLayer handleMouseDown:]):
        (-[MovieControllerLayer handleMouseUp:]):
        (-[MovieControllerLayer handleMouseDragged:]):
        * NetscapeCoreAnimationMoviePlugin/NetscapeCoreAnimationMoviePlugin.xcodeproj: Added.
        * NetscapeCoreAnimationMoviePlugin/NetscapeCoreAnimationMoviePlugin.xcodeproj/project.pbxproj: Added.
        * NetscapeCoreAnimationMoviePlugin/Pause.tiff: Added.
        * NetscapeCoreAnimationMoviePlugin/Play.tiff: Added.
        * NetscapeCoreAnimationMoviePlugin/SliderTrackCenter.tiff: Added.
        * NetscapeCoreAnimationMoviePlugin/SliderTrackLeft.tiff: Added.
        * NetscapeCoreAnimationMoviePlugin/SliderTrackRight.tiff: Added.
        * NetscapeCoreAnimationMoviePlugin/Thumb.tiff: Added.
        * NetscapeCoreAnimationMoviePlugin/main.m: Added.
        (NP_Initialize):
        (NP_GetEntryPoints):
        (NP_Shutdown):
        (NPP_New):
        (NPP_Destroy):
        (NPP_SetWindow):
        (NPP_NewStream):
        (NPP_DestroyStream):
        (NPP_WriteReady):
        (NPP_Write):
        (NPP_StreamAsFile):
        (NPP_Print):
        (handleMouseDown):
        (togglePlayPause):
        (handleMouseUp):
        (handleMouseDragged):
        (handleMouseEntered):
        (handleMouseExited):
        (handleKeyDown):
        (handleScrollEvent):
        (NPP_HandleEvent):
        (NPP_URLNotify):
        (NPP_GetValue):
        (NPP_SetValue):
        * NetscapeCoreAnimationMoviePlugin/test.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed. Add new symbols list for 1.1.10 release.

        * docs/webkitgtk-docs.sgml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Rubber-stamped by Xan Lopez.

        Fix Since tags for the new uri and message properties.

        * webkit/webkitnetworkrequest.cpp:
        (webkit_network_request_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Simon Hausmann  <simon.hausmann@nokia.com>

        Reviewed by and done with Tor Arne Vestbø.

        Fix initialization of uninitialized members in FontFallbackListQt, leading
        to random crashes when rendering text.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed. Adding new documented API to the generated
        documentation.

        * docs/webkitgtk-sections.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Simon Hausmann  <simon.hausmann@nokia.com>

        Reviewed by Adam Treat.

        Fix the logic for disabling the fixed layout feature, when an invalid
        QSize is set.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Version bump in preparation for 1.1.10 release.

        * configure.ac:

WebKit/gtk:

2009-06-15  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Update NEWS for 1.1.10 release.

        * NEWS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Holger Freyther.

        https://bugs.webkit.org/show_bug.cgi?id=25895
        [Gtk] Segfault when deleting the last/only character in an entry if a11y is enabled

        Use computeOffsetInContainerNode instead of the inline
        offsetInContainerNode to return the caret offset, since the former
        can handle all types of anchoring. Otherwise we'll hit an ASSERT
        in some situations in debug builds, and crash.

        * editing/gtk/SelectionControllerGtk.cpp:
        (WebCore::SelectionController::notifyAccessibilityForSelectionChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-15  Joe Ligman  <joseph.ligman@nokia.com>

        Reviewed by Simon Hausmann.

        [Qt] Fix build break
        Rename JSONObject to InspectorJSONObject

        https://bugs.webkit.org/show_bug.cgi?id=26340

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-12  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>

        Reviewed by Mark Rowe.
        Reviewed by Simon Hausmann.
        Reviewed by Xan Lopez.

        Refactor handling of options in the build-webkit script

        Options are now defined in one place, and then reused when creating
        the usage help text, the arguments to GetOptions(), and when passing
        the options on to the underlying port-dependent build systems.

        This allows the Qt port to read the defaults for the options from the
        pro file (dynamically), and to pass the options on to qmake at build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix <https://bugs.webkit.org/show_bug.cgi?id=26312>.
Bug 26312: Multiple copies of Safari opened with WebKit nightly and Safari 3.2.3

Reviewed by Cameron Zwarich.

Skip the trampoline detection for versions of Safari < 4.0 as they don't play these
tricks, and having WebKitNightlyEnabler assume that we were in the trampoline was
preventing our LaunchServices trickery from being run.  This led to fresh copies of
WebKit.app being launched when a URL or file was opened from an external application
rather than the existing instance being reused.

* WebKitLauncher/WebKitNightlyEnabler.m:
(insideSafari4OnTigerTrampoline):
(enableWebKitNightlyBehaviour):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix Qt build break after r44672

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Bug 26399: REGRESSION: java applet accessibility broken
https://bugs.webkit.org/show_bug.cgi?id=26399

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-14  Sam Weinig  <sam@webkit.org>

        Reviewed by Cameron Zwarich.

        Rename the CustomGetOwnPropertySlot and CustomPutFunction extended attributes
        to DelegatingGetOwnPropertySlot and DelegatingPutFunction to more clearly describe
        their functionality (customGetOwnPropertySlot and customPut member functions have
        been renamed getOwnPropertySlotDelegate and putDelegate).  CustomGetOwnPropertySlot
        and CustomPutFunction have been repurposed to mean a complete custom implementation
        of the function which is inline with other extended attributes prefixed with Custom.

        * bindings/js/JSCSSStyleDeclarationCustom.cpp:
        (WebCore::JSCSSStyleDeclaration::putDelegate):
        * bindings/js/JSDOMWindowCustom.h:
        (WebCore::JSDOMWindow::getOwnPropertySlotDelegate):
        (WebCore::JSDOMWindow::putDelegate):
        * bindings/js/JSHTMLAppletElementCustom.cpp:
        (WebCore::JSHTMLAppletElement::getOwnPropertySlotDelegate):
        (WebCore::JSHTMLAppletElement::putDelegate):
        * bindings/js/JSHTMLEmbedElementCustom.cpp:
        (WebCore::JSHTMLEmbedElement::getOwnPropertySlotDelegate):
        (WebCore::JSHTMLEmbedElement::putDelegate):
        * bindings/js/JSHTMLObjectElementCustom.cpp:
        (WebCore::JSHTMLObjectElement::getOwnPropertySlotDelegate):
        (WebCore::JSHTMLObjectElement::putDelegate):
        * bindings/js/JSHistoryCustom.cpp:
        (WebCore::JSHistory::getOwnPropertySlotDelegate):
        (WebCore::JSHistory::putDelegate):
        * bindings/js/JSLocationCustom.cpp:
        (WebCore::JSLocation::getOwnPropertySlotDelegate):
        (WebCore::JSLocation::putDelegate):
        (WebCore::JSLocationPrototype::putDelegate):
        * bindings/js/JSStorageCustom.cpp:
        (WebCore::JSStorage::putDelegate):
        * bindings/js/JSWorkerContextCustom.cpp:
        (WebCore::JSWorkerContext::getOwnPropertySlotDelegate):
        * bindings/scripts/CodeGeneratorJS.pm:
        * css/CSSStyleDeclaration.idl:
        * html/HTMLAppletElement.idl:
        * html/HTMLEmbedElement.idl:
        * html/HTMLObjectElement.idl:
        * page/DOMWindow.idl:
        * page/History.idl:
        * page/Location.idl:
        * storage/Storage.idl:
        * workers/WorkerContext.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin Adler.

        - fix https://bugs.webkit.org/show_bug.cgi?id=26395
          <rdar://problem/6971333> REGRESSION (r40813-r40884): No caret after
          going back/forward to page with focused text field

        No test, since the back/forward cache is disabled in DumpRenderTree

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::open): Added a call to CachedPage::restore(),
        which was removed in r40870.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-14  Darin Adler  <darin@apple.com>

        * editing/TextIterator.cpp:
        (WebCore::BitStack::push): Use named constant instead of hardcoded numbers.
        (WebCore::BitStack::top): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

2009-06-14  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Bug 26364: Search can find text that's hidden by overflow:hidden
        https://bugs.webkit.org/show_bug.cgi?id=26364
        <rdar://problem/6952081>

        Test: fast/text/find-hidden-text.html

        * editing/TextIterator.cpp:
        (WebCore::BitStack::BitStack): Added.
        (WebCore::BitStack::push): Added.
        (WebCore::BitStack::pop): Added.
        (WebCore::BitStack::top): Added.
        (WebCore::BitStack::size): Added.
        (WebCore::parentOrShadowParent): Added. Helper function for walking up
        the parent node chain, crossing shadow tree boundaries.
        (WebCore::depthCrossingShadowBoundaries): Added for use in assertions.
        Counts the depth of a node using the parentOrShadowParent function.
        (WebCore::fullyClipsContents): Added. Returns true for an element that
        fully clips its contents, currently defined as a box that has zero width
        or height and hides overflow. We can add other cases here later.
        (WebCore::ignoresContainerClip): Added. Returns true for an element that
        ignores its container clip, currently defined as an element with absolute
        or fixed positioning.
        (WebCore::pushFullyClippedState): Added. Pushes a bit on the stack indicating
        if the node in question fully clips its contents.
        (WebCore::setUpFullyClippedStack): Added. Pushes a bit for each ancestor of
        a node. Used when creating an iterator.
        (WebCore::TextIterator::TextIterator): Removed code to initialize
        m_inShadowContent. Call setUpFullyClippedStack.
        (WebCore::TextIterator::advance): Use parentOrShadowParent. Call pop when moving
        up to a parent node, and also one extra time when moving to a sibling node.
        Call pushFullyClippedState when moving to a child node, or when moving to a
        sibling node after the aforementioned call to pop.
        (WebCore::TextIterator::handleTextNode): Return early if the node is fully clipped.
        (WebCore::TextIterator::handleReplacedElement): Ditto.
        (WebCore::SimplifiedBackwardsTextIterator::SimplifiedBackwardsTextIterator):
        Call setUpFullyClippedStack.
        (WebCore::SimplifiedBackwardsTextIterator::advance): Same changes as
        TextIterator::advance above.

        * editing/TextIterator.h: Added BitStack class and added m_fullyClippedStack to both
        TextIterator and SimplifiedBackwardsTextIterator. Removed unneeded m_inShadowContent.

LayoutTests:

2009-06-14  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Bug 26364: Search can find text that's hidden by overflow:hidden
        https://bugs.webkit.org/show_bug.cgi?id=26364
        <rdar://problem/6952081>

        * fast/text/find-hidden-text-expected.txt: Added.
        * fast/text/find-hidden-text.html: Added.

        * fast/block/float/crash-replaced-display-block-expected.txt: Removed a space.
        * fast/dom/Range/acid3-surround-contents-expected.txt: Removed a blank line.
        * fast/dom/object-embed-plugin-scripting-expected.txt: Removed four spaces.
        * fast/parser/badentity-expected.txt: Removed a blank line.
        * http/tests/security/local-video-poster-from-remote-expected.txt: Removed a blank line.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-14  David Hyatt  <hyatt@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26362

        Back out Simon's fix and replace it with a better one that doesn't needlessly destroy RenderObjects.
        Use NoInherit rather than Detach for child index changes, and don't worry about other pseudo-class state
        changes, since they were handled already with setNeedsStyleRecalcs on the affected objects themselves.

        * dom/Element.cpp:
        (WebCore::Element::recalcStyle):
        * dom/Node.cpp:
        (WebCore::Node::diff):
        * rendering/style/RenderStyle.h:
        (WebCore::InheritedFlags::setChildIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-14  David Hyatt  <hyatt@apple.com>

        Reviewed by Dan Bernstein.

        Move all of the mutable member variables from Font into the refcounted FontFallbackList.  This makes
        copying of Fonts more efficient and also effectively makes all of the mutations of a Font object happen
        inside the FontFallbackList.

        * WebCore.base.exp:
        * platform/graphics/Font.cpp:
        (WebCore::Font::Font):
        (WebCore::Font::operator=):
        (WebCore::Font::primaryFont):
        (WebCore::Font::update):
        * platform/graphics/Font.h:
        (WebCore::Font::setWordSpacing):
        (WebCore::Font::setLetterSpacing):
        * platform/graphics/FontFallbackList.cpp:
        (WebCore::FontFallbackList::FontFallbackList):
        (WebCore::FontFallbackList::invalidate):
        (WebCore::FontFallbackList::determinePitch):
        * platform/graphics/FontFallbackList.h:
        (WebCore::FontFallbackList::primarySimpleFontData):
        (WebCore::FontFallbackList::primaryFontData):
        * platform/graphics/FontFastPath.cpp:
        (WebCore::Font::glyphDataForCharacter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44672 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-14  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Darin Adler

        https://bugs.webkit.org/show_bug.cgi?id=26362

        Fix an issue noted on quirksmode.com where dynamically adding siblings does not
        update :nth-child style correctly.

        Test: fast/css/nth-child-dynamic.html

        * dom/Node.h:
        Change diff() to take const RenderStyle*
        * dom/Node.cpp:
        (WebCore::Node::diff):
        When comparing styles to decide whether to detach and reattach, we
        need to look at whether the pseudoclass-related data in the style
        changed (since RenderStyle::diff doesn't check this).

        * rendering/style/RenderStyle.cpp:
        (WebCore::RenderStyle::getCachedPseudoStyle):
        (WebCore::RenderStyle::inheritedNotEqual):
        Make these methods const.
        * rendering/style/RenderStyle.h:
        (WebCore::InheritedFlags::styleType):
        (WebCore::InheritedFlags::pseudoClassStateEquivalent):
        Make methods const.
        Add new method that compares the pseudoclass-related members.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Update V8 benchmark to version 4.

        Reviewed by Cameron Zwarich.

        * tests/LIST-V8:
        * tests/v8-crypto.js:
        * tests/v8-deltablue.js:
        * tests/v8-earley-boyer.js:
        * tests/v8-raytrace.js:
        * tests/v8-regexp.js: Added.
        * tests/v8-richards.js:
        * tests/v8-splay.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-06-13  Peter Kasting  <pkasting@google.com>

        Reviewed by Kevin Ollivier.

        https://bugs.webkit.org/show_bug.cgi?id=25709 part eleven
        Write a wx implementation so the wx port can compile.  For now this is
        mostly a copy of the Cairo port, but in the future it can be changed to
        use a wxBitmap as the storage type everywhere (just as Cairo can
        probably be changed).

        Also ASSERT for any other ports who call getAddr() (I don't think there
        are any, and besides, they probably won't link, but this doesn't hurt).

        Also fixes a dumb Cairo build bustage typo introduced in part ten.

        * platform/graphics/wx/ImageSourceWx.cpp:
        (WebCore::ImageSource::createFrameAtIndex):
        * platform/image-decoders/ImageDecoder.h:
        (WebCore::RGBA32Buffer::getAddr):
        * platform/image-decoders/cairo/ImageDecoderCairo.cpp:
        (WebCore::RGBA32Buffer::setStatus):
        * platform/image-decoders/wx: Added.
        * platform/image-decoders/wx/ImageDecoderWx.cpp: Added.
        (WebCore::RGBA32Buffer::RGBA32Buffer):
        (WebCore::RGBA32Buffer::clear):
        (WebCore::RGBA32Buffer::zeroFill):
        (WebCore::RGBA32Buffer::copyBitmapData):
        (WebCore::RGBA32Buffer::setSize):
        (WebCore::RGBA32Buffer::asNewNativeImage):
        (WebCore::RGBA32Buffer::hasAlpha):
        (WebCore::RGBA32Buffer::setHasAlpha):
        (WebCore::RGBA32Buffer::setStatus):
        (WebCore::RGBA32Buffer::operator=):
        (WebCore::RGBA32Buffer::width):
        (WebCore::RGBA32Buffer::height):
        * webcore-wx.bkl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44669 268f45cc-cd09-0410-ab3c-d52691b4dbfc