GitLab

        Reviewed by Simon Hausmann.

        As Qtish implementation of MIMETypeRegistry::getMIMETypeForExtension()
        returns the application/octet-stream mimetype when it can't associate
        extension with any mimetype, it can happen that the application/octet-stream
        mimetype will hit the list of supported image formats. For instance,
        it is possible when QImageReader or QImageWriter support an extension
        that is not in the extensions map.

        Make sure that this mimetype is not treated as displayable image type.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        ASSERT(useTransforms) in SVG mapLocalToContainer implementations
        https://bugs.webkit.org/show_bug.cgi?id=25532

        * rendering/RenderSVGRoot.cpp:
        (WebCore::RenderSVGRoot::mapLocalToContainer):
        * rendering/SVGRenderSupport.cpp:
        (WebCore::SVGRenderBase::mapLocalToContainer):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Remove m_absoluteBounds hack from RenderSVGText
        https://bugs.webkit.org/show_bug.cgi?id=25532

        No functional changes (SVGs inside CSS transformed HTML should theoretically repaint better)

        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::layout):
        * rendering/RenderSVGText.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Remove the vestigial calculateLocalTransform()
        https://bugs.webkit.org/show_bug.cgi?id=25532

        RenderSVGContainer is now the only render which needs this
        function, but it no longer returns the unused bool.

        calculateLocalTransform() previously was called by the DOM before
        transform updates were part of layout().

        * rendering/RenderForeignObject.cpp:
        (WebCore::RenderForeignObject::layout):
        * rendering/RenderForeignObject.h:
        (WebCore::RenderForeignObject::localTransform):
        * rendering/RenderPath.cpp:
        (WebCore::RenderPath::layout):
        * rendering/RenderPath.h:
        * rendering/RenderSVGContainer.cpp:
        * rendering/RenderSVGContainer.h:
        (WebCore::RenderSVGContainer::calculateLocalTransform):
        * rendering/RenderSVGImage.cpp:
        (WebCore::RenderSVGImage::layout):
        * rendering/RenderSVGImage.h:
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::layout):
        * rendering/RenderSVGText.h:
        * rendering/RenderSVGTransformableContainer.cpp:
        (WebCore::RenderSVGTransformableContainer::calculateLocalTransform):
        * rendering/RenderSVGTransformableContainer.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Move more code into SVGRenderBase
        https://bugs.webkit.org/show_bug.cgi?id=25532

        clippedOverflowRectForRepaint, computeRectForRepaint and mapLocalToContainer
        are now all shared via SVGRenderBase.

        RenderForeignObject should also be sharing this code, but I've left it
        alone for now, as changing that would likely cause test changes.

        No test changes.  It's possible that transformed <svg:image> elements will
        now show up with better metrics in the inspector.

        * rendering/RenderSVGImage.cpp:
        (WebCore::RenderSVGImage::clippedOverflowRectForRepaint):
        (WebCore::RenderSVGImage::computeRectForRepaint):
        (WebCore::RenderSVGImage::mapLocalToContainer):
        * rendering/RenderSVGImage.h:
        * rendering/RenderSVGModelObject.cpp:
        (WebCore::RenderSVGModelObject::clippedOverflowRectForRepaint):
        (WebCore::RenderSVGModelObject::computeRectForRepaint):
        (WebCore::RenderSVGModelObject::mapLocalToContainer):
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::clippedOverflowRectForRepaint):
        (WebCore::RenderSVGText::computeRectForRepaint):
        (WebCore::RenderSVGText::mapLocalToContainer):
        * rendering/SVGRenderSupport.cpp:
        (WebCore::SVGRenderBase::clippedOverflowRectForRepaint):
        (WebCore::SVGRenderBase::computeRectForRepaint):
        (WebCore::SVGRenderBase::mapLocalToContainer):
        * rendering/SVGRenderSupport.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Add SVGRenderBase to share logic between SVG renderers
        https://bugs.webkit.org/show_bug.cgi?id=25532

        I've added an SVGRenderBase base class which is shared by
        all of the SVG renders now.  This patch is just moving code
        there are no functional changes in this patch.

        Currently I've left these functions in SVGRenderSupport.h
        In a later patch I will rename SVGRenderSupport.* to SVGRenderBase.*

        clampImageBufferSizeToViewport now takes a FrameView* which makes some code easier to read.

        * rendering/RenderSVGBlock.h:
        * rendering/RenderSVGImage.h:
        * rendering/RenderSVGModelObject.h:
        * rendering/RenderSVGRoot.h:
        * rendering/SVGRenderSupport.cpp:
        (WebCore::SVGRenderBase::prepareToRenderSVGContent):
        (WebCore::SVGRenderBase::finishRenderSVGContent):
        (WebCore::clampImageBufferSizeToViewport):
        (WebCore::SVGRenderBase::computeContainerBoundingBox):
        (WebCore::SVGRenderBase::filterBoundingBoxForRenderer):
        * rendering/SVGRenderSupport.h:
        * rendering/SVGRootInlineBox.cpp:
        (WebCore::SVGRootInlineBoxPaintWalker::chunkStartCallback):
        (WebCore::SVGRootInlineBoxPaintWalker::chunkEndCallback):
        (WebCore::SVGRootInlineBox::paint):
        * svg/SVGMaskElement.cpp:
        (WebCore::SVGMaskElement::drawMaskerContent):
        * svg/SVGPatternElement.cpp:
        (WebCore::SVGPatternElement::buildPattern):
        * svg/graphics/SVGPaintServerGradient.cpp:
        (WebCore::createMaskAndSwapContextForTextGradient):
        (WebCore::clipToTextMask):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Unreviewed SVG build fix.

        Include MappedAttribute.h after r43187.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Sam 'The Belly' Weinig

        <rdar://problem/6828894> - Facebook photo uploader applet broken on Leopard only

        In r41568 we started adding a default codebase attribute to <applet> tags if they didn't specify one
        themselves.  Leopard's Java plug-in mishandles this case and fails to load code for the applet.

        The spirit of r41568 can be maintained by removing the "default codebase" piece and only performing
        the canLoad() check if a codebase was actually set on the applet tag.

        * html/HTMLAppletElement.cpp:
        (WebCore::HTMLAppletElement::createRenderer): Don't create a default codebase attribute.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::createJavaAppletWidget): Only perform the canLoad() check if the codebase was set.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        * html/HTMLCollection.h:
        * html/HTMLFormElement.h:
        Declare CollectionCache as a struct, not a class, to match its
        definition.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=25557



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Fix an ancient Changelog entry where Sam Weinig flat out lied and confused my exploration of a bug.

        * ChangeLog



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=24883
        24883: Bad success test in parseXMLDocumentFragment in XMLTokenizerLibxml2.cpp

        Fixed test whether all the chunk has been processed to correctly count utf8 bytes.

        Test: fast/innerHTML/innerHTML-nbsp.xhtml

        * dom/XMLTokenizerLibxml2.cpp:
        (WebCore::parseXMLDocumentFragment):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-04  Darin Adler  <darin@apple.com>

        Reviewed by Eric Seidel.

        Bug 24924: remove Document.h include of Attr.h and HTMLCollection.h,
        and NamedMappedAttrMap.h include of MappedAttribute.h
        https://bugs.webkit.org/show_bug.cgi?id=24924

        Make compiles faster, and debugging info smaller.

        * GNUmakefile.am: Added new source files.
        * WebCore.pro: Ditto.
        * WebCore.scons: Ditto.
        * WebCore.vcproj/WebCore.vcproj: Ditto.
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * WebCoreSources.bkl: Ditto.

        * bindings/js/JSHTMLCollectionCustom.cpp:
        (WebCore::toJS): Updated for new collection type names.

        * dom/Document.cpp:
        (WebCore::Document::createAttribute): Added. No longer inline.
        (WebCore::Document::images): Updated for new collection type names.
        (WebCore::Document::applets): Ditto.
        (WebCore::Document::embeds): Ditto.
        (WebCore::Document::plugins): Ditto.
        (WebCore::Document::objects): Ditto.
        (WebCore::Document::scripts): Ditto.
        (WebCore::Document::links): Ditto.
        (WebCore::Document::forms): Ditto.
        (WebCore::Document::anchors): Ditto.
        (WebCore::Document::all): Ditto.
        (WebCore::Document::windowNamedItems): Ditto.
        (WebCore::Document::documentNamedItems): Ditto.
        (WebCore::Document::nameCollectionInfo): Ditto.

        * dom/Document.h: Changed around includes and forward declarations.
        Updated for changes to collection types.

        * dom/NamedMappedAttrMap.h: Removed include of MappedAttribute.h.

        * html/CollectionCache.cpp: Copied from WebCore/html/HTMLCollection.cpp.
        Contains the class that used to be HTMLCollection::CollectionInfo.
        * html/CollectionCache.h: Copied from WebCore/html/HTMLCollection.h.
        Ditto.

        * html/CollectionType.h: Copied from WebCore/html/HTMLCollection.h.
        Has the enum that used to be HTMLCollection::Type.

        * html/HTMLCollection.cpp:
        (WebCore::HTMLCollection::HTMLCollection): Update for collection type change.
        (WebCore::HTMLCollection::create): Ditto.
        (WebCore::HTMLCollection::resetCollectionInfo): Ditto.
        (WebCore::HTMLCollection::itemAfter): Ditto.
        * html/HTMLCollection.h: Ditto.

        * html/HTMLElement.cpp:
        (WebCore::HTMLElement::children): Updated for new collection type names.
        * html/HTMLFormCollection.cpp:
        (WebCore::HTMLFormCollection::formCollectionInfo): Ditto.
        (WebCore::HTMLFormCollection::HTMLFormCollection): Ditto.
        * html/HTMLFormCollection.h: Ditto.
        * html/HTMLFormElement.h: Ditto.
        * html/HTMLMapElement.cpp:
        (WebCore::HTMLMapElement::areas): Ditto.
        * html/HTMLNameCollection.cpp:
        (WebCore::HTMLNameCollection::HTMLNameCollection): Ditto.
        * html/HTMLNameCollection.h:
        (WebCore::HTMLNameCollection::create): Ditto.
        * html/HTMLSelectElement.h:
        (WebCore::HTMLSelectElement::collectionInfo): Ditto.
        * html/HTMLTableElement.cpp:
        (WebCore::HTMLTableElement::tBodies):Ditto.
        * html/HTMLTableRowElement.cpp:
        (WebCore::HTMLTableRowElement::cells):Ditto.
        * html/HTMLTableRowsCollection.cpp:
        (WebCore::HTMLTableRowsCollection::HTMLTableRowsCollection): Ditto.
        * html/HTMLTableSectionElement.cpp:
        (WebCore::HTMLTableSectionElement::rows): Ditto.

        * bindings/js/JSDOMWindowBase.cpp: Added newly-needed includes.
        * bindings/js/JSHTMLFrameSetElementCustom.cpp: Ditto.
        * css/CSSStyleSelector.cpp: Ditto.
        * dom/Element.cpp: Ditto.
        * dom/InputElement.cpp: Ditto.
        * dom/NamedAttrMap.cpp: Ditto.
        * dom/NamedMappedAttrMap.cpp: Ditto.
        * dom/Node.cpp: Ditto.
        * dom/StyledElement.cpp: Ditto.
        * dom/StyledElement.h: Ditto.
        * editing/ApplyStyleCommand.cpp: Ditto.
        * editing/DeleteSelectionCommand.cpp: Ditto.
        * editing/Editor.cpp: Ditto.
        * editing/EditorCommand.cpp: Ditto.
        * editing/InsertParagraphSeparatorCommand.cpp: Ditto.
        * editing/ReplaceSelectionCommand.cpp: Ditto.
        * editing/markup.cpp: Ditto.
        * html/CanvasRenderingContext2D.cpp: Ditto.
        * html/HTMLAnchorElement.cpp: Ditto.
        * html/HTMLAppletElement.cpp: Ditto.
        * html/HTMLAreaElement.cpp: Ditto.
        * html/HTMLBRElement.cpp: Ditto.
        * html/HTMLBaseElement.cpp: Ditto.
        * html/HTMLBodyElement.cpp: Ditto.
        * html/HTMLButtonElement.cpp: Ditto.
        * html/HTMLCanvasElement.cpp: Ditto.
        * html/HTMLDivElement.cpp: Ditto.
        * html/HTMLEmbedElement.cpp: Ditto.
        * html/HTMLFontElement.cpp: Ditto.
        * html/HTMLFormControlElement.cpp: Ditto.
        * html/HTMLFormElement.cpp: Ditto.
        * html/HTMLFrameElement.cpp: Ditto.
        * html/HTMLFrameElementBase.cpp: Ditto.
        * html/HTMLFrameSetElement.cpp: Ditto.
        * html/HTMLHRElement.cpp: Ditto.
        * html/HTMLIFrameElement.cpp: Ditto.
        * html/HTMLImageElement.cpp: Ditto.
        * html/HTMLInputElement.cpp: Ditto.
        * html/HTMLIsIndexElement.cpp: Ditto.
        * html/HTMLKeygenElement.cpp: Ditto.
        * html/HTMLLIElement.cpp: Ditto.
        * html/HTMLLinkElement.cpp: Ditto.
        * html/HTMLMarqueeElement.cpp: Ditto.
        * html/HTMLMetaElement.cpp: Ditto.
        * html/HTMLOListElement.cpp: Ditto.
        * html/HTMLObjectElement.cpp: Ditto.
        * html/HTMLOptionElement.cpp: Ditto.
        * html/HTMLParagraphElement.cpp: Ditto.
        * html/HTMLParamElement.cpp: Ditto.
        * html/HTMLPlugInElement.cpp: Ditto.
        * html/HTMLPreElement.cpp: Ditto.
        * html/HTMLScriptElement.cpp: Ditto.
        * html/HTMLSelectElement.cpp: Ditto.
        * html/HTMLStyleElement.cpp: Ditto.
        * html/HTMLTableCaptionElement.cpp: Ditto.
        * html/HTMLTableCellElement.cpp: Ditto.
        * html/HTMLTableColElement.cpp: Ditto.
        * html/HTMLTablePartElement.cpp: Ditto.
        * html/HTMLTextAreaElement.cpp: Ditto.
        * html/HTMLTokenizer.cpp: Ditto.
        * html/HTMLUListElement.cpp: Ditto.
        * html/HTMLVideoElement.cpp: Ditto.
        * html/HTMLViewSourceDocument.cpp: Ditto.
        * loader/ImageDocument.cpp: Ditto.
        * page/Frame.cpp: Ditto.
        * rendering/RenderTreeAsText.cpp: Ditto.
        * svg/SVGAElement.cpp: Ditto.
        * svg/SVGAnimateMotionElement.cpp: Ditto.
        * svg/SVGAnimateTransformElement.cpp: Ditto.
        * svg/SVGAnimationElement.cpp: Ditto.
        * svg/SVGCircleElement.cpp: Ditto.
        * svg/SVGClipPathElement.cpp: Ditto.
        * svg/SVGCursorElement.cpp: Ditto.
        * svg/SVGElement.cpp: Ditto.
        * svg/SVGEllipseElement.cpp: Ditto.
        * svg/SVGExternalResourcesRequired.cpp: Ditto.
        * svg/SVGFitToViewBox.cpp: Ditto.
        * svg/SVGFontFaceElement.cpp: Ditto.
        * svg/SVGFontFaceUriElement.cpp: Ditto.
        * svg/SVGForeignObjectElement.cpp: Ditto.
        * svg/SVGGlyphElement.cpp: Ditto.
        * svg/SVGGradientElement.cpp: Ditto.
        * svg/SVGImageElement.cpp: Ditto.
        * svg/SVGLangSpace.cpp: Ditto.
        * svg/SVGLineElement.cpp: Ditto.
        * svg/SVGLinearGradientElement.cpp: Ditto.
        * svg/SVGMarkerElement.cpp: Ditto.
        * svg/SVGMaskElement.cpp: Ditto.
        * svg/SVGPathElement.cpp: Ditto.
        * svg/SVGPatternElement.cpp: Ditto.
        * svg/SVGPolyElement.cpp: Ditto.
        * svg/SVGRadialGradientElement.cpp: Ditto.
        * svg/SVGRectElement.cpp: Ditto.
        * svg/SVGSVGElement.cpp: Ditto.
        * svg/SVGScriptElement.cpp: Ditto.
        * svg/SVGStopElement.cpp: Ditto.
        * svg/SVGStyleElement.cpp: Ditto.
        * svg/SVGStyledElement.cpp: Ditto.
        * svg/SVGStyledTransformableElement.cpp: Ditto.
        * svg/SVGTests.cpp: Ditto.
        * svg/SVGTextContentElement.cpp: Ditto.
        * svg/SVGTextElement.cpp: Ditto.
        * svg/SVGTextPathElement.cpp: Ditto.
        * svg/SVGTextPositioningElement.cpp: Ditto.
        * svg/SVGURIReference.cpp: Ditto.
        * svg/SVGUseElement.cpp: Ditto.
        * svg/SVGViewElement.cpp: Ditto.
        * svg/animation/SVGSMILElement.cpp: Ditto.
        * xml/XPathStep.cpp: Ditto.

WebKit/mac:

2009-05-04  Darin Adler  <darin@apple.com>

        Reviewed by Eric Seidel.

        Bug 24924: remove Document.h include of Attr.h and HTMLCollection.h,
        and NamedMappedAttrMap.h include of MappedAttribute.h
        https://bugs.webkit.org/show_bug.cgi?id=24924

        * WebView/WebFrame.mm: Added include of CSSMutableStyleDeclaration.h
        and ScriptValue.h.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Timothy Hatcher.

        Do not instantiate hidden WebInspector panels. Otherwise they are hitting unimplemented InspectorController methods.
        https://bugs.webkit.org/show_bug.cgi?id=25520

        * inspector/front-end/inspector.js:
        (WebInspector.loaded):



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Darin Adler.

        <rdar://problem/6839881> With "Mail Contents of This Page" in Safari images don't appear in the Tiger Mail compose window

        DocumentLoader::subresource() is the implementation for the WebKit API [WebDataSource subresourceForURL:] and has 
        particularly sticky behavior.  

        If the DocumentLoader represents a WebArchive, this method should return ArchiveResources from that archive.  However, we 
        prefer CachedResources over ArchiveResources because they might represent fresher data than what was originally loaded 
        from the WebArchive.

        In some instances, CachedResources are created and associated with the DocumentLoader but not immediately run through the 
        loading delegate machinery.  This was always a possibility, but the Preload Scanner made this significantly more likely.

        When this happens and someone asks the WebDataSource for a subresource, it would prefer the CachedResource over an 
        ArchiveResource it has even if the CachedResource hasn't been loaded yet.  The CachedResource has nil data, so no
        WebResource is returned, and the client thinks the subresource doesn't exist even though it does.

        This broke Tiger Mail and probably various other WebKit applications but has a fairly straightforward fix.

        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::subresource): If there is a CachedResource but it hasn't been loaded yet, fallback to the
          ArchiveResource.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Adam Roben.
        
        https://bugs.webkit.org/show_bug.cgi?id=25545
        Bug 25545: HTMLMediaElement: display anamorphic video correctly
        
        Tell QuickTime to use clean aperture mode so movies with non-square pixels are sized
        and displayed correctly.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivate::createQTMovie): Ask QuickTime to use clean aperture mode.
        * platform/graphics/win/QTMovieWin.cpp:
        (QTMovieWin::load): Ditto. Add ASSERT to ensure that static movie property array isn't 
        filled beyond capacity.

2009-05-04  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Adam Roben.

        https://bugs.webkit.org/show_bug.cgi?id=25545
        Bug 25545: HTMLMediaElement: display anamorphic video correctly
        
        Added 720x576i PAL anamorphic movie, extended video-size-intrinsic-scale
        test to check that it is sized correctly.

        * media/content/720x576i-anamorphic.mov: Added.
        * media/video-size-intrinsic-scale-expected.txt:
        * media/video-size-intrinsic-scale.html:



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Bug 25332: Plug-in inserted in to DOM of PluginDocument fails to load

Reviewed by Darin Adler.

Only the first plug-in in a PluginDocument can use the main resource data.  For all
subsequent plug-ins, such as the plug-in that ClickToFlash inserts when swapping itself
out for the real Flash plug-in, we need to load the resource data in the same manner
as for other embedded plug-ins.

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Rubber-stamped by Ariya Hidayat.

        Update the qrc file for the Web Inspector in the Qt build
        with missing files.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Move absoluteRects and absoluteQuads into RenderSVGInline and remove absoluteTransform() usage
        https://bugs.webkit.org/show_bug.cgi?id=25532

        * rendering/RenderSVGInline.cpp:
        (WebCore::RenderSVGInline::absoluteRects):
        (WebCore::RenderSVGInline::absoluteQuads):
        * rendering/RenderSVGInline.h:
        * rendering/RenderSVGTSpan.cpp:
        * rendering/RenderSVGTSpan.h:
        * rendering/RenderSVGTextPath.cpp:
        * rendering/RenderSVGTextPath.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Move RenderSVGText off of localToAbsolute()
        https://bugs.webkit.org/show_bug.cgi?id=25532

        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::absoluteRects):
        (WebCore::RenderSVGText::absoluteQuads):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Remove broken absoluteTransform() code from RenderSVGInlineText
        https://bugs.webkit.org/show_bug.cgi?id=25532

        This improved SVG text focus rings.
        This also improved SVG.getScreenCTM() results.

        SVGSVGElement::getScreenCTM was passing "true" for the second argument in
        localToAbsolute, which is "fixed" it meant to pass "true" for the
        third argument (useTransforms).  That change alone caused no layout test changes, however
        before I fixed this, it was hitting ASSERT(!fixed) in RenderSVGRoot::mapLocalToContainer

        mapLocalToContainer implementations really should be shared,
        but I'll do that in a later patch.

        * rendering/RenderSVGInlineText.cpp:
        (WebCore::RenderSVGInlineText::styleDidChange):
        (WebCore::RenderSVGInlineText::absoluteQuads):
        (WebCore::RenderSVGInlineText::computeRepaintRectForRange):
        (WebCore::RenderSVGInlineText::computeRepaintQuadForRange):
        * rendering/RenderSVGInlineText.h:
        * rendering/RenderSVGRoot.cpp:
        (WebCore::RenderSVGRoot::mapLocalToContainer):
        * rendering/RenderSVGRoot.h:
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::mapLocalToContainer):
        * rendering/RenderSVGText.h:
        * svg/SVGSVGElement.cpp:
        (WebCore::SVGSVGElement::getScreenCTM):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Remove dead code from RenderPath
        https://bugs.webkit.org/show_bug.cgi?id=25532

        * rendering/RenderPath.cpp:
        * rendering/RenderPath.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Share layout code between RenderSVGViewportContainer and RenderSVGContainer
        https://bugs.webkit.org/show_bug.cgi?id=25532

        * rendering/RenderSVGContainer.cpp:
        (WebCore::RenderSVGContainer::layout):
        * rendering/RenderSVGContainer.h:
        (WebCore::RenderSVGContainer::calcViewport):
        * rendering/RenderSVGViewportContainer.cpp:
        (WebCore::RenderSVGViewportContainer::applyViewportClip):
        * rendering/RenderSVGViewportContainer.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Remove redundant disableLayoutState() calls
        https://bugs.webkit.org/show_bug.cgi?id=25532

        * rendering/RenderForeignObject.cpp:
        (WebCore::RenderForeignObject::layout):
        * rendering/RenderSVGContainer.cpp:
        (WebCore::RenderSVGContainer::layout):
        * rendering/RenderSVGViewportContainer.cpp:
        (WebCore::RenderSVGViewportContainer::layout):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Simplify RenderForeignObject::paint and fix a repaint bug.
        Paint now calls applyTransformToPaintInfo which correctly transforms the damage rect.
        https://bugs.webkit.org/show_bug.cgi?id=16939

        * rendering/RenderForeignObject.cpp:
        (WebCore::RenderForeignObject::paint):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Eric Seidel.

        Test: editing/spelling/spellcheck-attribute.html

        Bug 14552: Add a way to disable spell checking for specific element
        https://bugs.webkit.org/show_bug.cgi?id=14552

        For compatibility with Firefox, this code ascends the DOM tree when an element does
        not have its "spellcheck" attribute.

        * editing/Editor.cpp:
        (WebCore::markMisspellingsOrBadGrammar): Retrieve the value of the "spellcheck"
        attribute of an element before calling a spell checker or a grammar checker.
        * html/HTMLAttributeNames.in: Add a "spellcheck" attribute.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-02  Sam Weinig  <sam@webkit.org>

        Roll JSC API number marshaling back in one last time (I hope).

WebCore:

2009-05-03  Sam Weinig  <sam@webkit.org>

        Roll JSC API number marshaling back in one last time (I hope).



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-03  Sam Weinig  <sam@webkit.org>

        Roll JSC API number marshaling back out. It still breaks windows.

WebCore:

2009-05-03  Sam Weinig  <sam@webkit.org>

        Roll JSC API number marshaling back out. It still breaks windows.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-02  Sam Weinig  <sam@webkit.org>

        Roll JSC API number marshaling back in.

WebCore:

2009-05-02  Sam Weinig  <sam@webkit.org>

        Roll JSC API number marshaling back in.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Darin Adler.

        - another fix for <rdar://problem/6617298> Typing delete on an unquoted
          blank line unquotes the preceding, quoted blank line

        Test: editing/deleting/type-delete-after-quote-2.html

        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::DeleteSelectionCommand): Initialize
        m_startsAtEmptyLine.
        (WebCore::DeleteSelectionCommand::handleSpecialCaseBRDelete): When the
        selection starts at an empty line, do not prevent the merging of blocks.
        This is what allows the text after the line break to be merged into the
        block containing the line break. Also set m_startsAtEmptyLine to true.
        (WebCore::DeleteSelectionCommand::mergeParagraphs): If
        m_startsAtEmptyLine is true, create a placeholder BR to serve as the
        merge destination.
        * editing/DeleteSelectionCommand.h:

LayoutTests:

        Reviewed by Darin Adler.

        - another test for <rdar://problem/6617298> Typing delete on an unquoted
          blank line unquotes the preceding, quoted blank line

        * editing/deleting/type-delete-after-quote-2-expected.txt: Added.
        * editing/deleting/type-delete-after-quote-2.html: Added.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-02  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Simplified null-ish JSValues.
        
        Replaced calls to noValue() with calls to JSValue() (which is what
        noValue() returned). Removed noValue().
        
        Replaced almost all uses of jsImpossibleValue() with uses of JSValue().
        Its one remaining use is for construction of hash table deleted values.
        For that specific task, I made a new, private constructor with a special
        tag. Removed jsImpossibleValue().
        
        Removed "JSValue()" initialiazers, since default construction happens...
        by default.

        * API/JSCallbackObjectFunctions.h:
        (JSC::::call):
        * bytecompiler/BytecodeGenerator.cpp:
        (JSC::BytecodeGenerator::emitLoad):
        * bytecompiler/BytecodeGenerator.h:
        * debugger/DebuggerCallFrame.cpp:
        (JSC::DebuggerCallFrame::evaluate):
        * debugger/DebuggerCallFrame.h:
        (JSC::DebuggerCallFrame::DebuggerCallFrame):
        * interpreter/CallFrame.h:
        (JSC::ExecState::clearException):
        * interpreter/Interpreter.cpp:
        (JSC::Interpreter::privateExecute):
        (JSC::Interpreter::retrieveLastCaller):
        * interpreter/Register.h:
        (JSC::Register::Register):
        * jit/JITCall.cpp:
        (JSC::JIT::unlinkCall):
        (JSC::JIT::compileOpCallInitializeCallFrame):
        (JSC::JIT::compileOpCall):
        * jit/JITStubs.cpp:
        (JSC::JITStubs::cti_op_call_eval):
        (JSC::JITStubs::cti_vm_throw):
        * profiler/Profiler.cpp:
        (JSC::Profiler::willExecute):
        (JSC::Profiler::didExecute):
        * runtime/ArrayPrototype.cpp:
        (JSC::getProperty):
        * runtime/Completion.cpp:
        (JSC::evaluate):
        * runtime/Completion.h:
        (JSC::Completion::Completion):
        * runtime/GetterSetter.cpp:
        (JSC::GetterSetter::getPrimitiveNumber):
        * runtime/JSArray.cpp:
        (JSC::JSArray::putSlowCase):
        (JSC::JSArray::deleteProperty):
        (JSC::JSArray::increaseVectorLength):
        (JSC::JSArray::setLength):
        (JSC::JSArray::pop):
        (JSC::JSArray::sort):
        (JSC::JSArray::compactForSorting):
        * runtime/JSCell.cpp:
        (JSC::JSCell::getJSNumber):
        * runtime/JSCell.h:
        (JSC::JSValue::getJSNumber):
        * runtime/JSGlobalData.cpp:
        (JSC::JSGlobalData::JSGlobalData):
        * runtime/JSImmediate.h:
        (JSC::JSImmediate::fromNumberOutsideIntegerRange):
        (JSC::JSImmediate::from):
        * runtime/JSNumberCell.cpp:
        (JSC::jsNumberCell):
        * runtime/JSObject.cpp:
        (JSC::callDefaultValueFunction):
        * runtime/JSObject.h:
        (JSC::JSObject::getDirect):
        * runtime/JSPropertyNameIterator.cpp:
        (JSC::JSPropertyNameIterator::toPrimitive):
        * runtime/JSPropertyNameIterator.h:
        (JSC::JSPropertyNameIterator::next):
        * runtime/JSValue.h:
        (JSC::JSValue::):
        (JSC::JSValueHashTraits::constructDeletedValue):
        (JSC::JSValueHashTraits::isDeletedValue):
        (JSC::JSValue::JSValue):
        * runtime/JSWrapperObject.h:
        (JSC::JSWrapperObject::JSWrapperObject):
        * runtime/Operations.h:
        (JSC::resolveBase):
        * runtime/PropertySlot.h:
        (JSC::PropertySlot::clearBase):
        (JSC::PropertySlot::clearValue):

WebCore:

2009-05-02  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Simplified null-ish JSValues.
        
        Replaced calls to noValue() with calls to JSValue() (which is what
        noValue() returned). Removed noValue().
        
        Removed "JSValue()" initialiazers, since default construction happens...
        by default.

        * bindings/js/JSDOMBinding.cpp:
        (WebCore::setDOMException):
        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::open):
        (WebCore::JSDOMWindow::showModalDialog):
        * bindings/js/JSEventListener.cpp:
        (WebCore::JSEventListener::handleEvent):
        * bindings/js/JSJavaScriptCallFrameCustom.cpp:
        (WebCore::JSJavaScriptCallFrame::evaluate):
        * bindings/js/JSSQLResultSetRowListCustom.cpp:
        (WebCore::JSSQLResultSetRowList::item):
        * bindings/js/ScriptController.cpp:
        (WebCore::ScriptController::evaluate):
        * bindings/js/ScriptValue.h:
        (WebCore::ScriptValue::ScriptValue):
        (WebCore::ScriptValue::hasNoValue):
        * bindings/js/WorkerScriptController.cpp:
        (WebCore::WorkerScriptController::evaluate):
        * bridge/jni/jni_instance.cpp:
        (JavaInstance::invokeMethod):
        * bridge/jni/jni_runtime.cpp:
        (JavaField::dispatchValueFromInstance):
        (JavaField::dispatchSetValueToInstance):
        * bridge/runtime.h:
        (JSC::Bindings::Instance::invokeConstruct):

WebKit/mac:

2009-05-02  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Simplified null-ish JSValues.
        
        Replaced calls to noValue() with calls to JSValue() (which is what
        noValue() returned). Removed noValue().
        
        Removed "JSValue()" initialiazers, since default construction happens...
        by default.

        * WebView/WebScriptDebugDelegate.mm:
        (-[WebScriptCallFrame evaluateWebScript:]):



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Dan Bernstein.

        <rdar://problem/6631735> Crash in WebCore::CSSStyleSelector::applyProperty(int, WebCore::CSSValue*) (RenderStyle.h:454)
        
        Avoid re-entering style selector from load delegates by not issuing resource loads synchronously during attach().

        * dom/ContainerNode.cpp:
        (WebCore::ContainerNode::suspendPostAttachCallbacks):
        (WebCore::ContainerNode::resumePostAttachCallbacks):
        (WebCore::ContainerNode::attach):
        * dom/Element.cpp:
        (WebCore::Element::attach):
        * loader/loader.cpp:
        (WebCore::Loader::Loader):
        (WebCore::Loader::servePendingRequests):
        (WebCore::Loader::suspendPendingRequests):
        (WebCore::Loader::resumePendingRequests):
        (WebCore::Loader::Host::servePendingRequests):
        * loader/loader.h:
        (WebCore::Loader::isSuspendingPendingRequests):



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-02  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Bug 25352: REGRESSION(r42322): style isn't applied at bmwusa.com
        https://bugs.webkit.org/show_bug.cgi?id=25352
        rdar://problem/6823239

        Test: http/tests/mime/standard-mode-loads-stylesheet-with-text-css-and-invalid-type.html

        * platform/network/HTTPParsers.cpp:
        (WebCore::extractMIMETypeFromMediaType): Allow comma as a separator.

LayoutTests:

2009-05-02  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Bug 25352: REGRESSION(r42322): style isn't applied at bmwusa.com
        https://bugs.webkit.org/show_bug.cgi?id=25352
        rdar://problem/6823239

        * http/tests/mime/resources/style-with-text-css-and-invalid-type.php:
        Copied from LayoutTests/http/tests/mime/resources/style-with-charset.php and adapted.
        * http/tests/mime/standard-mode-loads-stylesheet-with-text-css-and-invalid-type-expected.txt: Added.
        * http/tests/mime/standard-mode-loads-stylesheet-with-text-css-and-invalid-type.html:
        Copied from LayoutTests/http/tests/mime/standard-mode-loads-stylesheet-with-empty-content-type.html and adapted.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-02  Darin Adler  <darin@apple.com>

        Reviewed by Brady Eidson.

        Bug 25491: WebFrame leak when a subframe removes itself
        https://bugs.webkit.org/show_bug.cgi?id=25491
        rdar://problem/6833859

        Test: fast/loading/subframe-removes-itself.html

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::checkLoadCompleteForThisFrame): When isStopping is true,
        treat load as complete, even if isLoadingInAPISense is still returning true.

LayoutTests:

2009-05-02  Darin Adler  <darin@apple.com>

        Reviewed by Brady Eidson (except for the non-fast/loading test fixes).

        Bug 25491: WebFrame leak when a subframe removes itself
        https://bugs.webkit.org/show_bug.cgi?id=25491
        rdar://problem/6833859

        * fast/loading: Added.
        * fast/loading/resources: Added.
        * fast/loading/resources/subframe-that-removes-itself.html: Added.
        * fast/loading/subframe-removes-itself-expected.txt: Added.
        * fast/loading/subframe-removes-itself.html: Added.

        * fast/frames/location-put-after-removal-expected.txt: Updated.
        For some reason, fixing the loader sequence results in output that has one more blank line.

        * http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body.html:
        * http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny.html:
        * http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag.html:
        Updated tests so they don't terminate when the partial load of the subframe completes, but
        rather wait until the subsequent load of about:blank occurs.

        * platform/gtk/Skipped: Skip loading tests.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2009-05-01  Sam Weinig  <sam@webkit.org>

        Roll out JavaScriptCore API number marshaling.

        * API/APICast.h:
        (toJS):
        (toRef):
        * API/JSBase.cpp:
        (JSEvaluateScript):
        (JSCheckScriptSyntax):
        * API/JSCallbackConstructor.cpp:
        (JSC::constructJSCallback):
        * API/JSCallbackFunction.cpp:
        (JSC::JSCallbackFunction::call):
        * API/JSCallbackObjectFunctions.h:
        (JSC::::getOwnPropertySlot):
        (JSC::::put):
        (JSC::::deleteProperty):
        (JSC::::construct):
        (JSC::::hasInstance):
        (JSC::::call):
        (JSC::::toNumber):
        (JSC::::toString):
        (JSC::::staticValueGetter):
        (JSC::::callbackGetter):
        * API/JSObjectRef.cpp:
        (JSObjectMakeFunction):
        (JSObjectMakeArray):
        (JSObjectMakeDate):
        (JSObjectMakeError):
        (JSObjectMakeRegExp):
        (JSObjectGetPrototype):
        (JSObjectSetPrototype):
        (JSObjectGetProperty):
        (JSObjectSetProperty):
        (JSObjectGetPropertyAtIndex):
        (JSObjectSetPropertyAtIndex):
        (JSObjectDeleteProperty):
        (JSObjectCallAsFunction):
        (JSObjectCallAsConstructor):
        * API/JSValueRef.cpp:
        (JSValueGetType):
        (JSValueIsUndefined):
        (JSValueIsNull):
        (JSValueIsBoolean):
        (JSValueIsNumber):
        (JSValueIsString):
        (JSValueIsObject):
        (JSValueIsObjectOfClass):
        (JSValueIsEqual):
        (JSValueIsStrictEqual):
        (JSValueIsInstanceOfConstructor):
        (JSValueMakeUndefined):
        (JSValueMakeNull):
        (JSValueMakeBoolean):
        (JSValueMakeNumber):
        (JSValueMakeString):
        (JSValueToBoolean):
        (JSValueToNumber):
        (JSValueToStringCopy):
        (JSValueToObject):
        (JSValueProtect):
        (JSValueUnprotect):
        * JavaScriptCore.exp:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:
        * runtime/JSNumberCell.cpp:
        * runtime/JSNumberCell.h:
        * runtime/JSValue.h:

WebCore:

2009-05-01  Sam Weinig  <sam@webkit.org>

        Roll out JavaScriptCore API number marshaling.

        * bindings/js/ScriptValue.cpp:
        (WebCore::ScriptValue::isEqual):
        * inspector/JavaScriptProfile.cpp:
        (WebCore::getHeadCallback):
        * inspector/JavaScriptProfileNode.cpp:
        (WebCore::getChildren):
        (WebCore::getParent):
        (WebCore::getHead):



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43138 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Fix build.

        * workers/WorkerContext.idl:



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43137 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Oliver Hunt.

        Cleanup workers code a bit.

        * bindings/js/JSWorkerContextBase.cpp:
        (WebCore::toJS):
        * bindings/js/JSWorkerContextBase.h:
        * bindings/js/JSWorkerContextCustom.cpp:
        (WebCore::JSWorkerContext::customGetOwnPropertySlot):
        * workers/WorkerContext.h:
        (WebCore::WorkerContext::toWorkerContext):
        (WebCore::WorkerContext::self):
        (WebCore::WorkerContext::setOnmessage):
        (WebCore::WorkerContext::onmessage):
        * workers/WorkerContext.idl:



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Darin Adler.

        A little clean up.

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



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Fix the build.

        * bindings/js/ScriptValue.cpp:
        (WebCore::ScriptValue::isEqual):
        * inspector/JavaScriptProfile.cpp:
        (WebCore::getHeadCallback):
        * inspector/JavaScriptProfileNode.cpp:
        (WebCore::getChildren):
        (WebCore::getParent):
        (WebCore::getHead):



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

        Reviewed by Oliver Hunt and Sam "Jon 'The Belly' Honeycutt" Weinig

        Fix <rdar://problem/6848867> Addition of DOCTYPE node breaks Tiger Mail

        * bindings/objc/DOMHTML.mm:
        (-[DOMHTMLDocument firstChild]): Add a Tiger Mail version of this method that skips DOCTYPE nodes.



git-svn-id: http://svn.webkit.org/repository/webkit/trunk@43123 268f45cc-cd09-0410-ab3c-d52691b4dbfc