Reviewed by Brady

        Fixed: <rdar://problem/4126976> private keys imported by WebKit should not be accessible by all applications

        * libWebKitSystemInterface.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@20003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-07  Mark Rowe  <mrowe@apple.com>

        Add dumpAsText to test to match expected results.

        * fast/dom/Window/window-location-replace-functions.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@20002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-06  Mark Rowe  <mrowe@apple.com>

        Reviewed by Sam Weinig.

        Fix http://bugs.webkit.org/show_bug.cgi?id=12942
        Bug 12942: ASSERTION FAILURE: qantas.com.au changing selected item in <select> via JS

        Test: fast/dom/select-selectedIndex-bug-12942.html.

        * html/HTMLSelectElement.cpp:
        (WebCore::HTMLSelectElement::recalcListItems): Reset m_lastOnChangeIndex when recalculating list items.
        * html/HTMLSelectElement.h:

2007-03-06  Mark Rowe  <mrowe@apple.com>

        Reviewed by Sam Weinig.

        Test for http://bugs.webkit.org/show_bug.cgi?id=12942
        Bug 12942: ASSERTION FAILURE: qantas.com.au changing selected item in <select> via JS

        * fast/dom/select-selectedIndex-bug-12942-expected.txt: Added.
        * fast/dom/select-selectedIndex-bug-12942.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@20001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Rubberstamped by Kevin Decker

        20,000!

        * ChangeLog: Point out revision 20,000

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@20000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Not reviewed.

        Gdk build fix.

        * loader/gdk/FrameLoaderClientGdk.cpp: update userAgent() signature.
        (WebCore::FrameLoaderClientGdk::userAgent):
        * loader/gdk/FrameLoaderClientGdk.h: ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-06  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Adele.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12986
          REGRESSION(NativeListBox): Listboxes not updated when resized dynamically

        Test: fast/forms/select-change-listbox-size.html

        * html/HTMLSelectElement.cpp:
        (WebCore::HTMLSelectElement::parseMappedAttribute): Reattach on list box size change.

2007-03-06  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Adele.

        - test for http://bugs.webkit.org/show_bug.cgi?id=12986
          REGRESSION(NativeListBox): Listboxes not updated when resized dynamically

        * fast/forms/select-change-listbox-size-expected.checksum: Added.
        * fast/forms/select-change-listbox-size-expected.png: Added.
        * fast/forms/select-change-listbox-size-expected.txt: Added.
        * fast/forms/select-change-listbox-size.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-06  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12885
          REGRESSION (r19696): Incomplete background repaint

        Tests: fast/repaint/content-into-overflow.html
               fast/repaint/overflow-into-content.html

        Changed repaintAfterLayoutIfNeeded() to take, in addition to the clipped overflow
        rect, the unclipped border box plus outline, and to repaint any areas that
        were added or removed from that box, in addition to any areas added or removed
        from the clipped overflow rect.

        * platform/graphics/svg/SVGResourceMarker.cpp:
        (WebCore::SVGResourceMarker::draw):
        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::layoutBlock):
        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::absoluteClippedOverflowRect): Renamed getAbsoluteRepaintRect() to
        this.
        * rendering/RenderBox.h:
        * rendering/RenderFlexibleBox.cpp:
        (WebCore::RenderFlexibleBox::layoutBlock):
        * rendering/RenderFlow.cpp:
        (WebCore::RenderFlow::absoluteClippedOverflowRect):
        * rendering/RenderFlow.h:
        * rendering/RenderForeignObject.cpp:
        (WebCore::RenderForeignObject::layout):
        * rendering/RenderHTMLCanvas.cpp:
        (WebCore::RenderHTMLCanvas::layout):
        * rendering/RenderImage.cpp:
        (WebCore::RenderImage::layout):
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::RenderLayer):
        (WebCore::RenderLayer::checkForRepaintOnResize):
        (WebCore::RenderLayer::updateLayerPositions):
        * rendering/RenderLayer.h:
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::repaint):
        (WebCore::RenderObject::repaintAfterLayoutIfNeeded):
        (WebCore::RenderObject::getAbsoluteRepaintRectWithOutline):
        (WebCore::RenderObject::absoluteClippedOverflowRect):
        (WebCore::RenderObject::absoluteOutlineBox):
        * rendering/RenderObject.h:
        * rendering/RenderPath.cpp:
        (WebCore::RenderPath::layout):
        (WebCore::RenderPath::absoluteClippedOverflowRect):
        (WebCore::RenderPath::absoluteRects):
        (WebCore::RenderPath::drawMarkersIfNeeded):
        * rendering/RenderPath.h:
        * rendering/RenderSVGContainer.cpp:
        (WebCore::RenderSVGContainer::layout):
        (WebCore::RenderSVGContainer::absoluteClippedOverflowRect):
        (WebCore::RenderSVGContainer::absoluteRects):
        * rendering/RenderSVGContainer.h:
        * rendering/RenderSVGHiddenContainer.cpp:
        (WebCore::RenderSVGHiddenContainer::absoluteClippedOverflowRect):
        * rendering/RenderSVGHiddenContainer.h:
        * rendering/RenderSVGImage.cpp:
        (WebCore::RenderSVGImage::imageChanged):
        (WebCore::RenderSVGImage::absoluteClippedOverflowRect):
        (WebCore::RenderSVGImage::absoluteRects):
        * rendering/RenderSVGImage.h:
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::absoluteClippedOverflowRect):
        (WebCore::RenderSVGText::layout):
        (WebCore::RenderSVGText::absoluteRects):
        * rendering/RenderSVGText.h:
        * rendering/RenderTable.cpp:
        (WebCore::RenderTable::layout):
        * rendering/RenderTableCell.cpp:
        (WebCore::RenderTableCell::absoluteClippedOverflowRect):
        * rendering/RenderTableCell.h:
        * rendering/RenderTableCol.cpp:
        (WebCore::RenderTableCol::absoluteClippedOverflowRect):
        * rendering/RenderTableCol.h:
        * rendering/RenderTableRow.cpp:
        (WebCore::RenderTableRow::absoluteClippedOverflowRect):
        * rendering/RenderTableRow.h:
        * rendering/RenderText.cpp:
        (WebCore::RenderText::absoluteClippedOverflowRect):
        * rendering/RenderText.h:

2007-03-06  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - repaint tests and updated result for http://bugs.webkit.org/show_bug.cgi?id=12885
          REGRESSION (r19696): Incomplete background repaint

        * fast/repaint/bugzilla-6278-expected.checksum:
        * fast/repaint/bugzilla-6278-expected.png:
        * fast/repaint/content-into-overflow-expected.checksum: Added.
        * fast/repaint/content-into-overflow-expected.png: Added.
        * fast/repaint/content-into-overflow-expected.txt: Added.
        * fast/repaint/content-into-overflow.html: Added.
        * fast/repaint/overflow-into-content-expected.checksum: Added.
        * fast/repaint/overflow-into-content-expected.png: Added.
        * fast/repaint/overflow-into-content-expected.txt: Added.
        * fast/repaint/overflow-into-content.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19997 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-06  Ian Eng <ian.eng.webkit@gmail.com>

        Reviewed by Maciej.

        - Tests for http://bugs.webkit.org/show_bug.cgi?id=12720
        Bug 12720: Re-defining window.location.toString function keeps re-loading forever

        * fast/dom/Window/window-location-replace-functions-expected.txt: Added.
        * fast/dom/Window/window-location-replace-functions.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-06  Ian Eng <ian.eng.webkit@gmail.com>

        Reviewed by Maciej.

        - fixed http://bugs.webkit.org/show_bug.cgi?id=12720
        Bug 12720: Re-defining window.location.toString function keeps re-loading forever

        * bindings/js/kjs_window.cpp:
        Disallow replacing functions in LocationTable, and return early without updating URL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej Stachowiak.

        2% speedup on super accurate JS iBench.

        (KJS::Collector::collect): Removed anti-optimization to call
        pthread_is_threaded_np() before calling pthread_main_np(). Almost all
        apps have more than one thread, so the extra call is actually worse.
        Interestingly, even the single-threaded testkjs shows a speed gain
        from removing the pthread_is_threaded_np() short-circuit. Not sure why.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Darin.

        <http://bugs.webkit.org/show_bug.cgi?id=12686>
        REGRESSION: Bloglines.com Feeds tab cannot expand folders in TOT
        - Now all class constructors implement implementsHasInstance.

        * fast/dom/dom-instanceof-expected.txt: Added.
        * fast/dom/dom-instanceof.html: Added.

WebCore:

        Reviewed by Darin.

        <http://bugs.webkit.org/show_bug.cgi?id=12686>
        REGRESSION: Bloglines.com Feeds tab cannot expand folders in TOT
        - Now all class constructors implement implementsHasInstance.

        * bindings/scripts/CodeGeneratorJS.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Rob & Oliver.

Preparations for the new SVG text engine.

Handle baseline-shift / kerning css values correctly.
Recognize missing svg presentation attribute "font-size-adjust".
Add all missing svg<->css property mappings in mapToEntry().

Doesn't affect any layout test (as these properties are not used w/o my text patch).
While I'm at it, unify the macro names (RS_ -> SVG_RS_) & cleanup style a bit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by kevin

        <http://bugs.webkit.org/show_bug.cgi?id=12245>
        FCKeditor: Remove Format sometimes doesn't work
        <rdar://problem/4786404>
        Underline style is not removed from selection after performing Remove Format

        Added:
        * editing/execCommand/4786404-1-expected.checksum: Added.
        * editing/execCommand/4786404-1-expected.png: Added.
        * editing/execCommand/4786404-1-expected.txt: Added.
        * editing/execCommand/4786404-1.html: Added.
        * editing/execCommand/4786404-2-expected.checksum: Added.
        * editing/execCommand/4786404-2-expected.png: Added.
        * editing/execCommand/4786404-2-expected.txt: Added.
        * editing/execCommand/4786404-2.html: Added.
        Demonstrate fixes:
        * editing/execCommand/4920488-expected.checksum:
        * editing/execCommand/4920488-expected.png:
        * editing/execCommand/4920488-expected.txt:
        * editing/execCommand/4920742-1-expected.checksum:
        * editing/execCommand/4920742-1-expected.png:
        * editing/execCommand/4920742-1-expected.txt:
        * editing/execCommand/remove-formatting-2-expected.checksum:
        * editing/execCommand/remove-formatting-2-expected.png:
        * editing/execCommand/remove-formatting-2-expected.txt:
        Ditto, also added a FIXME:
        * editing/execCommand/remove-formatting-expected.checksum:
        * editing/execCommand/remove-formatting-expected.png:
        * editing/execCommand/remove-formatting-expected.txt:
        * editing/execCommand/remove-formatting.html:

WebCore:

        Reviewed by kevin

        <http://bugs.webkit.org/show_bug.cgi?id=12245>
        FCKeditor: Remove Format sometimes doesn't work
        <rdar://problem/4786404>
        Underline style is not removed from selection after performing Remove Format

        * editing/Editor.cpp:
        (WebCore::Editor::removeFormattingAndStyle): Re-wrote this.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Oliver.

Add new <use> layout tests. (forgot in last commit)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Oliver.

Fixes: http://bugs.webkit.org/show_bug.cgi?id=12979
Fixes: http://bugs.webkit.org/show_bug.cgi?id=12981
Guard against registering pending resources with empty id.
Be careful with calling recalcStyle() in SVGUseElement, when change is "Detach" we
have to use special code, as calling attach() on the shadow tree root element will crash,
because it has no (direct) parent node, only a shadow parent node element.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Adele.

        Fixed: <rdar://problem/5041660> REGRESSION: <keygen> element broken, prevents users from signing up for Thawte email certs

        * bindings/objc/DOM.mm:
        (WebCore::createElementClassMap): Added <keygen> to the DOM bindings so it can be accessed from Objective-C.
        * css/html4.css: Apply the look of the <select> element to <keygen>.
        * html/HTMLElementFactory.cpp:
        (WebCore::keygenConstructor): Added.
        (WebCore::createFunctionMap): Added keygen.
        * html/HTMLSelectElement.cpp: Made const typeAheadTimeout variable static const.
        (WebCore::HTMLSelectElement::HTMLSelectElement):  Fix both HTMLSelectElement constructors to initialize the
         same number of member variables. The fact that some fields were not initialized could (and would) crash the
         keygen element when selecting different items. Also removed m_typedString(String()) from the constructor
         initialization because this is not needed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Mitz.
Remove an alert() which got accidently added in my last commit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin

        Made WebAuthenticationHandler.h SPI so Safari can call it directly.

        * WebKit.exp:
        added .objc_class_name_WebPanelAuthenticationHandler
        * WebKit.xcodeproj/project.pbxproj:
        changed status of WebAuthenticationHandler.h from "project" to "private"

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        - Rename a function to clarify its purpose.

        * WebView/WebView.mm:
        (+[WebView registerURLSchemeAsLocal:]):
        * WebView/WebViewPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        - Rename a function to clarify its purpose.

        * WebCore.exp:
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::registerURLSchemeAsLocal):
        * loader/FrameLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Build fix.

        * WebCore.exp: Updated symbols.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Anders.

        Added a parameter to all StringTruncator methods to specify whether
        rounding hacks should be on or off.

        No layout test possible.

        * platform/StringTruncator.cpp:
        (WebCore::stringWidth): Added disableRoundingHacks parameter.
        (WebCore::truncateString): Ditto.
        (WebCore::StringTruncator::centerTruncate): Ditto.
        (WebCore::StringTruncator::rightTruncate): Ditto.
        (WebCore::StringTruncator::width): Ditto.
        * platform/StringTruncator.h: Ditto.
        * platform/mac/FileChooserMac.mm:
        (WebCore::FileChooser::basenameForWidth): Pass in false to
        centerTruncate so that the truncation matches the way the text will be
        rendered.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Maciej.

        <rdar://problem/5035045>
        REGRESSION: WebKit browser doesn't display image at http://www.metoffice.gov.uk/weather/satellite/index.html

        Add some tests where image elements have id _and_ name elements.

        * fast/dom/HTMLDocument/document-special-properties-expected.txt:
        * fast/dom/HTMLDocument/document-special-properties.html:

WebCore:

        Reviewed by Maciej.

        <rdar://problem/5035045>
        REGRESSION: WebKit browser doesn't display image at http://www.metoffice.gov.uk/weather/satellite/index.html

        It turns out WinIE does allow you to access images by their id as special document properties. However, this is only
        allowed when the element also has a name attribute. The value of the name attribute is ignored and can even be empty!

        * bindings/js/kjs_html.cpp:
        (KJS::JSHTMLDocument::namedItemGetter):
        Return jsUndefined() if the collection is empty.

        * html/HTMLImageElement.cpp:
        (WebCore::HTMLImageElement::parseMappedAttribute):
        (WebCore::HTMLImageElement::insertedIntoDocument):
        (WebCore::HTMLImageElement::removedFromDocument):
        * html/HTMLImageElement.h:
        Add the id attribute value to the extra named item map.

        * html/HTMLNameCollection.cpp:
        (WebCore::HTMLNameCollection::traverseNextItem):
        Check for images with name attributes that match, as well as elements with id attributes that match where
        the element also has a name attribute.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

        Reviewed by Adam.

        WebCore part of patch to make it possible to have different user agents for different URLs.

        * bindings/js/kjs_navigator.cpp:
        (KJS::Navigator::getValueProperty):
        * bindings/js/kjs_proxy.cpp:
        (WebCore::KJSProxy::initScriptIfNeeded):
        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::setLoading):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::userAgent):
        (WebCore::FrameLoader::loadResourceSynchronously):
        (WebCore::FrameLoader::applyUserAgent):
        * loader/FrameLoader.h:
        * loader/FrameLoaderClient.h:
        * platform/graphics/svg/SVGImageEmptyClients.h:
        (WebCore::SVGEmptyFrameLoaderClient::userAgent):

WebKit:

        Reviewed by Adam.

        Update for WebCore changes.

        * WebCoreSupport/WebFrameLoaderClient.h:
        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::userAgent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19978 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Mark and Dave H.

- rdar://problem/5038491
        An oversite of the security fix that prevented remote from loading local is that it
        prevents user style sheets when the site is remote.  This fixes that.

        * http/tests/security/local-user-CSS-from-remote-expected.txt: Added.
        * http/tests/security/local-user-CSS-from-remote.html: Added.

WebCore:

        Reviewed by Mark and Dave H.

        - rdar://problem/5038491
        An oversite of the security fix that prevented remote from loading local is that it
        prevents user style sheets when the site is remote.  This fixes that.

        * loader/Cache.cpp: Propogate and check user style sheet flag.
        (WebCore::createResource):
        (WebCore::Cache::requestResource):
        * loader/Cache.h: Propogate user style sheet flag.
        * loader/CachedCSSStyleSheet.cpp: Propogate user style sheet flag.
        (WebCore::CachedCSSStyleSheet::CachedCSSStyleSheet):
        * loader/CachedCSSStyleSheet.h: Propogate user style sheet flag.
        * loader/DocLoader.cpp: Propogate user style sheet flag.
        (WebCore::DocLoader::requestResource):
        * loader/SubresourceLoader.cpp: Propogate and check user style sheet flag.
        (WebCore::SubresourceLoader::create):
        * loader/SubresourceLoader.h: Add check for user style sheet flag.
        * loader/loader.cpp: Propogate user style sheet flag.
        (WebCore::Loader::load):
        (WebCore::Loader::servePendingRequests):
        * loader/loader.h: Propogate user style sheet flag.

WebKitTools:

        Reviewed by Mark and Dave H.

        - rdar://problem/4922454
        - This fixes a security issue by making remote referrers not able to access local
        resources, unless they register their schemes to be treated as local. The result is
        that those schemes can access local resources and cannot be accessed by remote
        referrers.
        Because this behavior is new a link-on-or-after check is made to determine if the
        app should use the older, less safe, behavior.

        * DumpRenderTree/DumpRenderTree.m: Add ability to set user style sheet to DRT.
        (+[LayoutTestController isSelectorExcludedFromWebScript:]):
        (+[LayoutTestController webScriptNameForSelector:]):
        (-[LayoutTestController setUserStyleSheetLocation:]):
        (-[LayoutTestController setUserStyleSheetEnabled:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19977 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Darin.

Fix all (except one) known <use> bug. The remaining bug (12630) is not crucical at all.
Enable <use> again as default SVG feature, as discussed on webkit-dev.

Fixes: http://bugs.webkit.org/show_bug.cgi?id=12936 (Master bug used to track all current use problems.)
Fixes: http://bugs.webkit.org/show_bug.cgi?id=12926 (transform attribute not respected in nested <use> elements)
Fixes: http://bugs.webkit.org/show_bug.cgi?id=12267 (getElementById broken for <use>)
Fixes: http://bugs.webkit.org/show_bug.cgi?id=12916 (use instance in symbol definition does not work)
Fixes: http://bugs.webkit.org/show_bug.cgi?id=12917 (mouseout event does not occur after scaling use instance)
Fixes: http://bugs.webkit.org/show_bug.cgi?id=12838 (SVG <use> DOM broken for script)
Added test: svg/custom/use-nested-transform.svg
Added test: svg/custom/struct-use-09-b.svg (not yet official W3C-SVG-1.1 testcase)
Fixes test: svg/custom/use-elementInstance-event-target.svg (rectangle now turns green on first click)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by John

        Fixes <rdar://problem/4974258>
        Adds some key null checking

        * bindings/js/kjs_html.cpp:
        (KJS::JSHTMLElement::implementsCall): Null check doc/frame
        * bindings/objc/DOMInternal.mm:
        (-[WebScriptObject _initializeScriptDOMNodeImp]): Null check doc/frame

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Darin.

Path::normalAngleAtLength() / Path::pointAtLength() don't work correctly.
pointAtLength() was not implemented, basically and normalAngleAtLength()
had a bug in the tangent slope calculation.

The normalAngleAtLength() stuff can only be tested with my local textPath support.
New LayoutTest: svg/custom/path-textPath-simulation.svg

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-05  Darin Adler  <darin@apple.com>

        Reviewed by Oliver.

        Text encoding test case

        * fast/encoding/char-encoding-expected.txt: Added.
        * fast/encoding/char-encoding.html: Added.
        * fast/encoding/charset-names-expected.txt: Moved.
        * fast/encoding/charset-names.html: Moved.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        http://bugs.webkit.org/show_bug.cgi?id=12970
        Fix and import 4XPath test_core_functions.html test

WebCore:

        * xml/XPathExpression.cpp:
        (WebCore::XPathExpression::evaluate): Fully initialize the evaluation context.

        * xml/XPathFunctions.cpp:
        (WebCore::XPath::FunSubstring::doEvaluate): Fixed handling of edge cases.
        (WebCore::XPath::FunRound::round): Reimplemented to match the spec; exposed FunRound::round() to be used in
        other functions.

LayoutTests:
        * fast/xpath/4XPath/Core/test_core_functions-expected.txt: Added.
        * fast/xpath/4XPath/Core/test_core_functions.html: Added.
        The remaining failures with id() are caused by an XML DOM problem - ID attributes declared
        in an internal subset are not treated as such.

        * fast/xpath/4XPath/Core/test.js: Common data for XPath tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    * plugins/get-url-with-blank-target-expected.txt:
        Update test result.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Adam, Darin.

        <rdar://problem/5025212>
        In Mail, a crash occurs at WebCore::Frame::tree() when clicking on embedded flash object

        * plugins/get-url-with-blank-target-expected.txt: Added.
        * plugins/get-url-with-blank-target.html: Added.

WebKit:

        Reviewed by Adam, Darin.

        <rdar://problem/5025212>
        In Mail, a crash occurs at WebCore::Frame::tree() when clicking on embedded flash object

        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView loadPluginRequest:]):
        Handle the case where the web view returned from the delegate method is null. Also, send out an error notification
        in that case so we can catch it.

WebKitTools:

        Reviewed by Adam, Darin.

        <rdar://problem/5025212>
        In Mail, a crash occurs at WebCore::Frame::tree() when clicking on embedded flash object

        Add a "getURLNotify" method to the plugin object. This lets you pass a URL, a target and a callback function
        to be run when the URL has finished (or failed) loading.

        * DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.c:
        (pluginInvoke):
        (handleCallback):
        * DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.h:
        * DumpRenderTree/TestNetscapePlugIn.subproj/main.c:
        (NPP_URLNotify):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin and Kevin D

        - fixed <rdar://problem/5038087> Header and footer on printed page are too large after certain steps

        * WebView/WebView.mm:
        (-[WebView _adjustPrintingMarginsForHeaderAndFooter]):
        This method was modifying the margins in the NSPrintInfo object without any sort of check whether
        this had already been done. In some cases this can be called multiple times with the same
        NSPrintInfo, so now we stash information in the NSPrintInfo's dictionary such that we always
        start with a fresh copy of the original margins.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove accidentally committed tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        http://bugs.webkit.org/show_bug.cgi?id=12954
        XPath relative operations are implemented incorrectly

WebCore:
        * xml/XPathPredicate.cpp:
        (WebCore::XPath::NumericOp::doEvaluate):
        (WebCore::XPath::EqTestOp::compare):
        (WebCore::XPath::EqTestOp::doEvaluate):
        Reimplemented relative equality operations to match the spec.

        * xml/XPathPredicate.h:
        (WebCore::XPath::NumericOp::):
        (WebCore::XPath::EqTestOp::):
        Moved relative operations to EqTestOp.

        * xml/XPathGrammar.y:
        * xml/XPathParser.cpp:
        (WebCore::XPath::Parser::nextTokenInternal):
        (WebCore::XPath::Parser::lex):
        Adapted for the above changes.

LayoutTests:
        * fast/xpath/4XPath/Core/test_boolean_expr-expected.txt: Added.
        * fast/xpath/4XPath/Core/test_boolean_expr.html: Added.

        * fast/xpath/4XPath/Borrowed/cz_20030217-expected.txt: Added.
        * fast/xpath/4XPath/Borrowed/cz_20030217.html: Added.
        * fast/xpath/4XPath/Borrowed/kd_20010423-expected.txt: Added.
        * fast/xpath/4XPath/Borrowed/kd_20010423.html: Added.
        * fast/xpath/4XPath/Borrowed/od_20000608-expected.txt: Added.
        * fast/xpath/4XPath/Borrowed/od_20000608.html: Added.
        * fast/xpath/4XPath/Borrowed/resources/od_20000608.xml: Added.
        * fast/xpath/4XPath/Borrowed/rs_20010831-expected.txt: Added.
        * fast/xpath/4XPath/Borrowed/rs_20010831.html: Added.
        * fast/xpath/4XPath/Borrowed/sr_20021217-expected.txt: Added.
        * fast/xpath/4XPath/Borrowed/sr_20021217.html: Added.
        Added a bunch of XPath tests that already passed.

        * fast/xpath/4XPath/Borrowed/namespace-nodes-expected.txt: Added.
        * fast/xpath/4XPath/Borrowed/namespace-nodes.html: Added.
        * fast/xpath/namespace-nodes-expected.txt: Removed.
        * fast/xpath/namespace-nodes.html: Removed.
        Moved a previously imported 4XPath test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Pesky apostrophe.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-05  Mark Rowe  <mrowe@apple.com>

        Reviewed by Lars.

        Fix http://bugs.webkit.org/show_bug.cgi?id=12947
        Bug 12947: REGRESSION: ASSERTION FAILED: maxWidth >= 0 in StringTruncator.cpp:109 in WebCore::truncateString()

        Handle nil window correctly in toUserSpace and toDeviceSpace.  On Intel Mac's a message to nil that returns a
        float will return 0.0.  We use this as the divisor in calculating a scale factor, which results in NaN being
        introduced into our rect.

        * platform/mac/ScreenMac.mm:
        (WebCore::toUserSpace):
        (WebCore::toDeviceSpace):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        http://bugs.webkit.org/show_bug.cgi?id=12868
        parts of the CSS classes in this simple SVG example are not applied

        Make sure the xml stylesheets are parsed in strict mode.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        http://bugs.webkit.org/show_bug.cgi?id=12961
        shouldBe() should support NaN

        * fast/js/resources/js-test-pre.js: Make shouldBe() succeed if both sides are NaN numbers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Darin.

        * fast/events/keydown-keypress-preventDefault-expected.txt: Added.
        * fast/events/keydown-keypress-preventDefault.html: Added.

WebCore:

        Reviewed by Darin.

        Change to dispatch the keypress event during the defaultEventHandler for keydown events.  This matches IE behavior.
        This is preparation for fixing event dispatch with input methods (http://bugs.webkit.org/show_bug.cgi?id=10871)

        Test: fast/events/keydown-keypress-preventDefault.html

        * dom/EventTargetNode.cpp: (WebCore::EventTargetNode::defaultEventHandler): Call the defaultKeyboardEventHandler for keydown events.
        * page/EventHandler.cpp:
        (WebCore::eventTargetNodeForDocument): Return 0 instead of false since the return type in EventTargetNode.
        (WebCore::EventHandler::keyEvent): Removed dispatch of keypress event, since this is now done in the default event handler.
        (WebCore::EventHandler::defaultKeyboardEventHandler): For keydown events, create and dispatch a keypress event.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Nikolas Zimmermann (yay!).

        http://bugs.webkit.org/show_bug.cgi?id=12962
        4XPath tests crash on lang() function

        Covered by 4XPath tests, to be landed later.

        * platform/StringImpl.cpp:
        (WebCore::StringImpl::reverseFind): Do not crash with empty strings.
        * xml/XPathFunctions.cpp:
        (WebCore::XPath::FunLang::doEvaluate): Do not crash when an element has no
        attributes. Use a proper namespace for xml:lang (not sure where "xms" came from).
        Rewrote the algorithm for suffix removing to match the spec.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19957 268f45cc-cd09-0410-ab3c-d52691b4dbfc

JavaScriptCore:

        Reviewed by Nikolas Zimmermann.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12950
          Assertions.cpp should not #define macros that are already defined

        * wtf/Assertions.cpp: Don't #define WINVER and _WIN32_WINNT if they
        are already defined.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Darin.

        <rdar://problem/5028165>
        http://bugs.webkit.org/show_bug.cgi?id=12915
        REGRESSION: XMLHttpRequest.abort() does not stop loading (12915)

        * http/tests/xmlhttprequest/abort-should-cancel-load-expected.txt: Added.
        * http/tests/xmlhttprequest/abort-should-cancel-load.html: Added.

WebCore:

        Reviewed by Darin.

        <rdar://problem/5028165>
        http://bugs.webkit.org/show_bug.cgi?id=12915
        REGRESSION: XMLHttpRequest.abort() does not stop loading (12915)

        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::stopLoading):
        Save the value of m_loading since calling FrameLoader::stopLoading could set it to false.

        * loader/SubresourceLoader.cpp:
        (WebCore::SubresourceLoader::didCancel):
        * loader/SubresourceLoader.h:
        Get rid of didCancel now, it's not needed anymore.

        * xml/xmlhttprequest.cpp:
        (WebCore::XMLHttpRequest::abort):
        Call cancel() instead of stopLoading(). Also, set m_aborted to true so the XMLHttpRequest object
        won't be dereferenced in didFail when aborting.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

  WebCore:

        Reviewed by Nikolas Zimmermann.

        Move ScrollView stubs to ScrollViewGdk.cpp

        * platform/gdk/FrameGdk.h: remove comment that no longer makes sense
        * platform/gdk/ScrollViewGdk.cpp:
        (WebCore::ScrollView::addChild):
        (WebCore::ScrollView::removeChild):
        (WebCore::ScrollView::scrollPointRecursively):
        (WebCore::ScrollView::inWindow):
        (WebCore::ScrollView::wheelEvent):
        (WebCore::ScrollView::updateScrollbars):
        (WebCore::ScrollView::updateScrollInfo):
        (WebCore::ScrollView::windowToContents):
        (WebCore::ScrollView::contentsToWindow):
        (WebCore::ScrollView::scrollbarUnderMouse):
        * platform/gdk/TemporaryLinkStubs.cpp:

  WebKitTools:

        Reviewed by Nikolas Zimmermann.

        Load a url given on a command line in a way that also
        works for local (file://) urls.

        * GdkLauncher/main.cpp:
        (main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19954 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Nikolas Zimmermann.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12951
          REGRESSION (r19949): LayoutTests/fast/xsl/xslt-processer.html fails

        Fixed spelling of "processer" to "processor" in file names.

        * fast/xsl/xslt-processer-expected.txt: Removed.
        * fast/xsl/xslt-processer.html: Removed.
        * fast/xsl/xslt-processor-expected.txt: Added. Results progressed after fixing
        Bugzilla Bug 12735 (<rdar://problem/4545040>).
        * fast/xsl/xslt-processor.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Geoff.

        - rdar://problem/4922454
        - This fixes a security issue by making remote referrers not able to access local
        resources, unless they register their schemes to be treated as local. The result is
        that those schemes can access local resources and cannot be accessed by remote
        referrers.
        Because this behavior is new a link-on-or-after check is made to determine if the
        app should use the older, less safe, behavior.

        * fast/loader/local-CSS-from-local-expected.txt: Added.
        * fast/loader/local-CSS-from-local.html: Added.
        * fast/loader/local-JavaScript-from-local-expected.txt: Added.
        * fast/loader/local-JavaScript-from-local.html: Added.
        * fast/loader/local-iFrame-source-from-local-expected.txt: Added.
        * fast/loader/local-iFrame-source-from-local.html: Added.
        * fast/loader/local-image-from-local-expected.txt: Added.
        * fast/loader/local-image-from-local.html: Added.
        * http/tests/security/local-CSS-from-remote-expected.txt: Added.
        * http/tests/security/local-CSS-from-remote.html: Added.
        * http/tests/security/local-JavaScript-from-remote-expected.txt: Added.
        * http/tests/security/local-JavaScript-from-remote.html: Added.
        * http/tests/security/local-iFrame-from-remote-expected.txt: Added.
        * http/tests/security/local-iFrame-from-remote.html: Added.
        * http/tests/security/local-image-from-remote-expected.txt: Added.
        * http/tests/security/local-image-from-remote.html: Added.
        * http/tests/security/resources/compass.jpg: Added.
        * http/tests/security/resources/cssStyle.css: Added.
        * http/tests/security/resources/localPage.html: Added.
        * http/tests/security/resources/localScript.js: Added.

WebCore:

        Reviewed by Geoff.

        - rdar://problem/4922454
        - This fixes a security issue by making remote referrers not able to access local
        resources, unless they register their schemes to be treated as local. The result is
        that those schemes can access local resources and cannot be accessed by remote
        referrers.
        Because this behavior is new a link-on-or-after check is made to determine if the
        app should use the older, less safe, behavior.

        * WebCore.exp: added exported functions
        * bindings/objc/DOM.mm: consolodated function to base class
        (-[DOMElement image]):
        (-[DOMElement _imageTIFFRepresentation]):
        * dom/Document.cpp: Cache the document's ability to load local resources.
        (WebCore::Document::Document):
        (WebCore::Document::setURL):
        (WebCore::Document::shouldBeAllowedToLoadLocalResources):
        (WebCore::Document::stylesheetLoaded):
        * dom/Document.h: Cache the docuent's ability to load local resources.
        (WebCore::Document::getPendingSheet):
        (WebCore::Document::isAllowedToLoadLocalResources):
        * html/HTMLImageLoader.cpp: Moved functionality into base class.
        (WebCore::HTMLImageLoader::updateFromElement):
        (WebCore::HTMLImageLoader::dispatchLoadEvent):
        * html/HTMLLinkElement.cpp: Handles null returns correctly now.
        * html/HTMLTokenizer.cpp: Moved functionality into base class.
        (WebCore::HTMLTokenizer::notifyFinished):
        * ksvg2/misc/SVGImageLoader.cpp: Moved functionality into base class.
        (WebCore::SVGImageLoader::dispatchLoadEvent):
        * loader/Cache.cpp: Checks if the cached resource can be loaded.
        (WebCore::Cache::requestResource):
        * loader/CachedCSSStyleSheet.cpp: Moved functionality into base class.
        (WebCore::CachedCSSStyleSheet::ref):
        (WebCore::CachedCSSStyleSheet::error):
        * loader/CachedImage.cpp: Moved functionality into base class.
        (WebCore::CachedImage::CachedImage):
        * loader/CachedImage.h: Moved functionality into base class.
        (WebCore::CachedImage::canRender):
        * loader/CachedResource.cpp: Cache if the CachedResource should be treated as local
        (WebCore::CachedResource::CachedResource):
        * loader/CachedResource.h: Moved functionality into base class.
        (WebCore::CachedResource::errorOccurred):
        (WebCore::CachedResource::shouldTreatAsLocal):
        * loader/CachedScript.cpp: Moved functionality into base class.
        (WebCore::CachedScript::CachedScript):
        * loader/CachedScript.h: Moved functionality into base class.
        (WebCore::CachedScript::schedule):
        * loader/CachedXBLDocument.cpp: Moved functionality into base class.
        (WebCore::CachedXBLDocument::error):
        * loader/CachedXSLStyleSheet.cpp: Moved functionality into base class.
        (WebCore::CachedXSLStyleSheet::error):
        * loader/FrameLoader.cpp: See comments for each function below.
        (WebCore::FrameLoader::loadSubframe): Use new canLoad.
        (WebCore::FrameLoader::restrictAccessToLocal): return value of linked-on-or-after check.
        (WebCore::FrameLoader::setRestrictAccessToLocal): set value for linked-on-or-after check.
        (WebCore::localSchemes): Return set of schemes that are to be treated as local.
        (WebCore::FrameLoader::loadPlugin): Use new canLoad.
        (WebCore::FrameLoader::canLoad): Now multiple functions that each do the same work but some can take advantage of the cached values, if they were computed previously.
        (WebCore::FrameLoader::shouldHideReferrer): Extracted out the logic to determine if the referrer should be hidden so it is only calculated when needed.
        (WebCore::FrameLoader::loadResourceSynchronously): No longer calls canLoad to get hideReferrer info.
        (WebCore::FrameLoader::registerSchemeAsLocal): Functionality to register a scheme to be treated as local.
        (WebCore::FrameLoader::treatURLAsLocal): Given a URL this function determines if it should be treated as local.
        * loader/FrameLoader.h: Declared functions for this security fix.  See above.
        * loader/MainResourceLoader.cpp: Optized order of bools to regain performance.
        (WebCore::MainResourceLoader::continueAfterContentPolicy):
        * loader/SubresourceLoader.cpp: Now restricts remote from loading local resources.
        (WebCore::SubresourceLoader::create):
        * page/EventHandler.cpp: Moved functionality into base class.
        (WebCore::selectCursor):
        * platform/KURL.cpp: KURLs need to check all the registered schemes now.
        (WebCore::KURL::isLocalFile):
        * rendering/HitTestResult.cpp: Moved functionality into base class.
        (WebCore::HitTestResult::image):
        * rendering/RenderImage.cpp: Moved functionality into base class.
        (WebCore::RenderImage::setCachedImage):
        (WebCore::RenderImage::imageChanged):
        (WebCore::RenderImage::paint):
        (WebCore::RenderImage::layout):
        (WebCore::RenderImage::calcAspectRatioWidth):
        (WebCore::RenderImage::calcAspectRatioHeight):
        * rendering/RenderImage.h: Moved functionality into base class.
        (WebCore::RenderImage::errorOccurred):
        * rendering/RenderListItem.cpp: Moved functionality into base class.
        (WebCore::RenderListItem::setStyle):
        * rendering/RenderListMarker.cpp: Moved functionality into base class.
        (WebCore::RenderListMarker::isImage):
        * xml/xmlhttprequest.cpp: Check doc's cached value instead of determining independently.
        (WebCore::XMLHttpRequest::urlMatchesDocumentDomain):

WebKit:

        Reviewed by Geoff.

        - rdar://problem/4922454
        - This fixes a security issue by making remote referrers not able to access local
        resources, unless they register their schemes to be treated as local. The result is
        that those schemes can access local resources and cannot be accessed by remote
        referrers.
        Because this behavior is new a link-on-or-after check is made to determine if the
        app should use the older, less safe, behavior.

        * Misc/WebKitVersionChecks.h: added linked-on-or-after check
        * Misc/WebNSAttributedStringExtras.mm: Moved functionalit into the base class.
        (fileWrapperForElement):
        * Plugins/WebNetscapePluginStream.mm: uses new canLoad functions
        * Plugins/WebPluginContainerCheck.mm: uses new canLoad functions
        (-[WebPluginContainerCheck _isForbiddenFileLoad]):
        * WebView/WebView.mm: make linked-on-or-after check and cache value, exposes SPI
        for registering a scheme as local.
        (-[WebView _commonInitializationWithFrameName:groupName:]):
        (+[WebView registerSchemeAsLocal:]):
        * WebView/WebViewPrivate.h: exposes SPI for registering a scheme as local.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by kevin

        <rdar://problem/5028447>
        REGRESSION: Gmail Editor: Copied message text pastes at the wrong font size

        * editing/pasteboard/5028447-expected.checksum: Added.
        * editing/pasteboard/5028447-expected.png: Added.
        * editing/pasteboard/5028447-expected.txt: Added.
        * editing/pasteboard/5028447.html: Added.

WebCore:

        Reviewed by kevin

        <rdar://problem/5028447>
        REGRESSION: Gmail Editor: Copied message text pastes at the wrong font size

        * editing/markup.cpp:
        (WebCore::createMarkup): The style of the div that holds
        a fully selected body's styles didn't include styles inherited
        from the body's ancestors.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Tweaked parse-malloc-history to work with new malloc_history output format.

        * Scripts/parse-malloc-history:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by harrison

        <rdar://problem/4545040>
        innerHTML does not HTML-escape text nodes inside PRE elements
        <rdar://problem/5027857>
        Pasting into Mail from Safari's view-source window renders the HTML

        * editing/pasteboard/5027857-expected.checksum: Added.
        * editing/pasteboard/5027857-expected.png: Added.
        * editing/pasteboard/5027857-expected.txt: Added.
        * editing/pasteboard/5027857.html: Added.

WebCore:

        Reviewed by harrison

        <rdar://problem/4545040>
        innerHTML does not HTML-escape text nodes inside PRE elements
        <rdar://problem/5027857>
        Pasting into Mail from Safari's view-source window renders the HTML

        * editing/HTMLInterchange.cpp:
        (WebCore::convertHTMLTextToInterchangeFormat): Send this function
        the node that the text comes from as a parameter.  It shouldn't convert
        '\n's to spaces/nbsps if the text is coming from text where newlines are
        preserved.
        * editing/HTMLInterchange.h:
        * editing/markup.cpp:
        (WebCore::startMarkup): Escape text inside the children of PREs.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

        Reviewed by Anders.

        Try to fix the Qt build.

        * platform/qt/TemporaryLinkStubs.cpp: Add stubs.
        (WebCore::searchMenuNoRecentSearchesText):
        (WebCore::searchMenuRecentSearchesText):
        (WebCore::searchMenuClearRecentSearchesText):
        (WebCore::AXWebAreaText):
        (WebCore::AXLinkText):
        (WebCore::AXListMarkerText):
        (WebCore::AXImageMapText):
        (WebCore::AXHeadingText):

WebKitQt:

        Reviewed by Anders.

        Try to fix the Qt build

        * WebCoreSupport/DragClientQt.cpp:
        (WebCore::DragClientQt::willPerformDragDestinationAction):
        (WebCore::DragClientQt::dragControllerDestroyed):
        (WebCore::DragClientQt::createDragImageForLink):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19948 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-03-02  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Anders.

        Add unsigned int hash traits (matches existing unsigned long version)

        * wtf/HashTraits.h:
        (WTF::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19947 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Suggested by Darin.

        A more efficient solution to rdar://4961431.

        * bridge/mac/WebCoreAXObject.mm:
        (-[WebCoreAXObject accessibilityIsIgnored]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Kevin M.

        Try to fix the Qt build.

        * kjs/DateMath.cpp:
        (KJS::msToGregorianDateTime): Removed unnecessary "struct" keyword.
        * kjs/DateMath.h: Moved forward declarations to the top of the file
        before they are used.
        * kjs/date_object.cpp:
        (KJS::formatLocaleDate): Changed to take a const GregorianDateTime&
        since GregorianDateTime is Noncopyable.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by john

        <rdar://problem/5032095>
        Gmail Editor: Copied text pastes on a new line instead of current line

        * editing/pasteboard/5032095-expected.checksum: Added.
        * editing/pasteboard/5032095-expected.png: Added.
        * editing/pasteboard/5032095-expected.txt: Added.
        * editing/pasteboard/5032095.html: Added.

WebCore:

        Reviewed by john

        <rdar://problem/5032095>
        Gmail Editor: Copied text pastes on a new line instead of current line

        Start merge failed to occur because positionAtStartOfInsertedContent
        had a bug.

        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::positionAtStartOfInsertedContent):
        Was failing when inserting <span><div>foo/div></span>. Return the
        inserted content's first VisiblePosition.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Kevin McCullough.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12867
          REGRESSION: BenchJS test 7 (dates) is 220% slower than in Safari 2.0.4

        * kjs/DateMath.h: Marked GregorianDateTime as noncopyable, since it has a non-trivial
        destructor and not the correspoding copy constructor or assignment operator.
        Changed the GregorianDateTime constructor to use member initialization syntax.
        Fixed the destructor to use the array delete operator, since timeZone is an array.

        * kjs/DateMath.cpp:
        (KJS::daysInYear): Changed to call isLeapYear so the rule is not repeated twice.
        (KJS::getUTCOffset): Added caching on PLATFORM(DARWIN), since we can rely on the
        notify_check function and "com.apple.system.timezone" to let us know when the
        offset has changed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix glyph map crasher.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej.

        Fix for http://bugs.webkit.org/show_bug.cgi?id=12895
        REGRESSION: imagemap: pointer cursor is shown everywhere
        <rdar://problem/5028163>

        Image with imagemap should never itself be URLElement in hit test results.

        No layout test, cursor state can't be captured.

        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::hitTest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Maciej.

        HTTP layout test for http://bugs.webkit.org/show_bug.cgi?id=12690
        REGRESSION: can not log in to bank of america with TOT webkit
        <rdar://problem/4990044>

        also for
        http://bugs.webkit.org/show_bug.cgi?id=12604
        http://bugs.webkit.org/show_bug.cgi?id=12020

        * http/tests/misc/multiple-submit-expected.txt: Added.
        * http/tests/misc/multiple-submit.html: Added.
        * http/tests/misc/resources/post-echo-and-notify-done.cgi: Added.

WebCore:

        Reviewed by Maciej.

        Fix http://bugs.webkit.org/show_bug.cgi?id=12690
        REGRESSION: can not log in to bank of america with TOT webkit
        <rdar://problem/4990044>

        and http://bugs.webkit.org/show_bug.cgi?id=12604
        REGRESSION: After closing the "Would you like to save password" sheet, the form fails
        to submit automatically at http://www.mac.com/WebObjects/HomePage.woa
        <rdar://problem/4871752>

        and http://bugs.webkit.org/show_bug.cgi?id=12020
        REGRESSION: Flickr uploading broken
        <rdar://problem/4928662>

        Turn protection against multiple forms submission back on. This approach is buggy
        but it is way better than not having it at all. Not protecting against this
        breaks number of major sites.

        * bridge/mac/WebCoreAXObject.mm:
        (-[WebCoreAXObject accessibilityPerformAction:]):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::resetMultipleFormSubmissionProtection):
        (WebCore::FrameLoader::submitForm):
        (WebCore::FrameLoader::receivedMainResourceError):
        * loader/FrameLoader.h:
        * page/EventHandler.cpp:
        (WebCore::EventHandler::keyEvent):
        * page/Frame.cpp:
        (WebCore::Frame::setView):
        * page/mac/EventHandlerMac.mm:
        (WebCore::EventHandler::mouseDown):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Adam.

        - Tests that the url of a document after an open is what it was before the open.

        * fast/loader/external-script-URL-location-expected.txt: Added.
        * fast/loader/external-script-URL-location.html: Added.
        * fast/loader/resources/external-script-URL-location.js: Added.

WebCore:

        Reviewed by Adam.

        - Added the test case: external-script-URL-location.html
        - Fix an issue where the url of a document is null after an open if the document
        has no parent.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        <rdar://problem/4960250>
        http://bugs.webkit.org/show_bug.cgi?id=11627
        REGRESSION: Reproducible crash at IMDb in WebCore::FrameLoader::stopLoadingSubframes

        In rare cases, we could end up calling checkLoadComplete twice for the same frame. This would cause the
        didFailProvisionalLoad delegate method to be called twice for the same frame, and also cause the provisional document loader
        to be reset to null when other code wasn't expecting it.

        This regressed in revision 10904 with the fix for <rdar://problem/4184719>. The fix is to only call stopLoading on the frame
        if either the document loader is loading, or the document is still being parsed. I've verified that the bug is still fixed and
        that no leaks occur.

        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::stopLoading):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        Fix ARM crash due to accessing non-4-byte-aligned memory
        as 32-bit values.

        * platform/AtomicString.cpp:
        (WebCore::UCharBufferTranslator::equal):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by harrison

        <rdar://problem/4838199>
        Integrate Mail and WebKit paste operations

        Provide subresources used to create the fragment as a
        convenience.

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _documentFragmentFromPasteboard:inContext:allowPlainText:chosePlainText:]):
        Update the calls to the changed method.
        (-[WebHTMLView _documentFragmentFromPasteboard:forType:inContext:subresources:]):
        Give the caller the subresources in the WebArchive and RTF cases.
        * WebView/WebHTMLViewPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin.

        <rdar://problem/5033905> Have the DOM secondary thread check raise an exception by default

        * platform/mac/ThreadCheck.mm:
        (WebCore::_WebCoreThreadViolationCheck):
        Initialize threadViolationIsException to true.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19928 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin

        <rdar://problem/5030628> - Crash opening a new window with the
        "New windows open to the same page" pref set

        * history/HistoryItem.cpp:
        (WebCore::HistoryItem::HistoryItem): Set the m_subItems vector capacity correctly

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej.

        http://bugs.webkit.org/show_bug.cgi?id=12801
        Assertion failure in createMarkup() (root) when doing Select All, Copy in an SVG document

        Test: editing/pasteboard/createMarkup-assert.xml

        * editing/markup.cpp:
        (WebCore::createMarkup): Removed the assertion, because it's wrong in a non-HTML world.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Oliver and Hyatt.

        Fix <rdar://problem/5024233> Crash while using Find on empty document

        No layout test possible, as this involves a null document.

        * dom/Range.cpp:
        (WebCore::rangeOfContents): Added an ASSERT.
        * page/Frame.cpp:
        (WebCore::Frame::findString): Added a null-check for document().
        (WebCore::Frame::markAllMatchesForText): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-02-28  Oliver Hunt  <oliver@apple.com>

        Reviewed by Maciej.

        Fixes <rdar://problem/5012009>

        When looking for a requested resource we should also check
        the set of manually added subresources if WebCore can't find it.

        * WebView/WebDataSource.mm:
        (-[WebDataSource subresourceForURL:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Tim H.

Fixes: http://bugs.webkit.org/show_bug.cgi?id=12887
        Bug#12887: [Drosera] Add ability to close loaded files

        * Drosera/DebuggerDocument.h:
        * Drosera/DebuggerDocument.m:
        (-[DebuggerDocument closeCurrentFile:]): Adds delegation to call a JS script to close files
        * Drosera/Drosera.xcodeproj/project.pbxproj:
        * Drosera/English.lproj/MainMenu.nib/classes.nib:
        * Drosera/English.lproj/MainMenu.nib/info.nib:
        * Drosera/English.lproj/MainMenu.nib/keyedobjects.nib: Adds Close Current File menu item
        * Drosera/debugger.html: Changed "no files loaded" to "<No files loaded>" to match Xcode style
        * Drosera/debugger.js: Adds implementation of closeFile() to unload currently loaded file

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Not reviewed - simple gdk build fix.

        Add stubs for newly introduced localized strings.

        * platform/gdk/TemporaryLinkStubs.cpp:
        (WebCore::searchMenuNoRecentSearchesText):
        (WebCore::searchMenuRecentSearchesText):
        (WebCore::searchMenuClearRecentSearchesText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

        Reviewed by Beth

        Move the Thread Safety Check functions into their own header for export to WebKit

        * WebCore.exp: Export the function
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/scripts/CodeGeneratorObjC.pm: Include the new header
        * platform/Logging.h: Move stuff to ThreadCheck.h
        * platform/ThreadCheck.h: Added.
        * platform/mac/LoggingMac.mm: Move stuff to ThreadCheck.mm
        * platform/mac/ThreadCheck.mm: Added.
        (WebCore::_WebCoreThreadViolationCheck):
        (WebCoreReportThreadViolation):

WebKit:

        Reviewed by Beth

        Start using the Thread Safety Check implemented in WebCore for the DOM bindings in the rest of
        the WebKit API instead of the ASSERT_MAIN_THREAD() hack

        * History/WebBackForwardList.mm:
        (-[WebBackForwardList initWithWebCoreBackForwardList:]):
        (-[WebBackForwardList init]):
        (-[WebBackForwardList dealloc]):
        (-[WebBackForwardList finalize]):
        * History/WebHistoryItem.mm:
        (-[WebHistoryItem initWithURLString:title:lastVisitedTimeInterval:]):
        (-[WebHistoryItem dealloc]):
        (-[WebHistoryItem finalize]):
        (-[WebHistoryItem copyWithZone:]):
        (-[WebHistoryItem initWithWebCoreHistoryItem:]):
        * Misc/WebIconDatabase.mm:
        (-[WebIconDatabase init]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Brady.

        Fix for http://bugs.webkit.org/show_bug.cgi?id=12923 REGRESSION:
        Assertion failure copying standalone image

        * platform/mac/PasteboardMac.mm:
        (WebCore::Pasteboard::writeImage): Use the CachedImage as the
        resource. This makes more sense anyway. There is no need to null-
        check the renderer or the CachedImage since we return early if
        there is no Image* and the HitTestResult::image() function checks
        for these things.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore:

        Reviewed by Beth.

        Fix for <rdar://problem/4887423> REGRESSION: search results popup menu strings are not localized
        and <rdar://problem/3517227> accessibility-related strings in WebCore are not localized

        Use localized strings from WebKit instead of hard coded strings.

        * bridge/mac/WebCoreAXObject.mm: (-[WebCoreAXObject roleDescription]):
        * page/mac/WebCoreViewFactory.h:
        * platform/LocalizedStrings.h:
        * platform/mac/LocalizedStringsMac.mm:
        (WebCore::searchMenuNoRecentSearchesText):
        (WebCore::searchMenuRecentSearchesText):
        (WebCore::searchMenuClearRecentSearchesText):
        (WebCore::AXWebAreaText):
        (WebCore::AXLinkText):
        (WebCore::AXListMarkerText):
        (WebCore::AXImageMapText):
        (WebCore::AXHeadingText):
        * rendering/RenderTextControl.cpp: (WebCore::RenderTextControl::itemText):

WebKit:

        Reviewed by Beth.

        Fix for <rdar://problem/4887423> REGRESSION: search results popup menu strings are not localized
        and <rdar://problem/3517227> accessibility-related strings in WebCore are not localized

        * WebCoreSupport/WebViewFactory.mm:
        (-[WebViewFactory searchMenuNoRecentSearchesText]):
        (-[WebViewFactory searchMenuRecentSearchesText]):
        (-[WebViewFactory searchMenuClearRecentSearchesText]):
        (-[WebViewFactory AXWebAreaText]):
        (-[WebViewFactory AXLinkText]):
        (-[WebViewFactory AXListMarkerText]):
        (-[WebViewFactory AXImageMapText]):
        (-[WebViewFactory AXHeadingText]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Bradeee Eidson.

        Disabling this test for now because it fails on the buildbot but not on
        my Intel or PPC machines. I've filed http://bugs.webkit.org/show_bug.cgi?id=12930
        to track this issue.

        * fast/leaks/003.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Maciej.

        - test for http://bugs.webkit.org/show_bug.cgi?id=12911
        GoogleDocs: Ordered lists don't update immediately when start attribute changed

        * fast/lists/ol-start-dynamic-expected.checksum: Added.
        * fast/lists/ol-start-dynamic-expected.png: Added.
        * fast/lists/ol-start-dynamic-expected.txt: Added.
        * fast/lists/ol-start-dynamic.html: Added.

WebCore:

        Reviewed by Maciej.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12911
        GoogleDocs: Ordered lists don't update immediately when start attribute changed

        Update list marker value when start attribute changes.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reviewed by Maciej.

Corrected the use of the new ENABLE() macros in some ksvg2/svg files.
Add new build-webkit flags "--(no-)xpath" / "--(no-)xslt", to be able
to switch off build features easily. Also add "--(no-)svg-experimental-features"
flag, to be able to test filters/animations/use/foreignObject easily.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-02-28  Mark Rowe  <mrowe@apple.com>

        Reviewed by Maciej.

        <rdar://problem/5028473> WebKit allocates a huge number of NSCalendarDates while loading history file

        * History/WebHistory.mm:
        (-[WebHistoryPrivate insertItem:atDateIndex:]): Use lastVisitedTimeInterval rather than _lastVisitedDate to avoid allocating NSCalendarDates.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej.

        http://bugs.webkit.org/show_bug.cgi?id=12609
        Any SVG element will create renderers even when children of HTML elements

        Allow creation of svg renderers only when parent is SVG, except for
        the <svg> element.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej.

        http://bugs.webkit.org/show_bug.cgi?id=12913
        Markers do not render in webkit when it misses markerWidth or markerHeight attribute

        Set defaults for markerWidth/markerHeight so markers that do not specify them render.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Extending the timeout on this test to see if it fixes the buildbot failure.
        (I don't get a failure locally.)

        * fast/leaks/003.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    * loader/ResourceLoader.cpp:
        (WebCore::ResourceLoader::~ResourceLoader):
        Remove comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by oliver

        <rdar://problem/5027300>
        REGRESSION: Images inserted with align left/right are lost

        * editing/inserting/12882-expected.checksum: Added.
        * editing/inserting/12882-expected.png: Added.
        * editing/inserting/12882-expected.txt: Added.
        * editing/inserting/12882.html: Added.

WebCore:

        Reviewed by oliver

        <rdar://problem/5027300>
        REGRESSION: Images inserted with align left/right are lost

        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::shouldMerge):
        Don't attempt to merge to or from a position before
        or after a block because it will be a no-op and
        lead to infinite recursion.
        In this case it instead resulted in content loss because
        of bugs in start/endOfParagraph (5027702).
        * editing/visible_units.cpp: Added two FIXMEs for
        the problems with start/endOfParagraph.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19903 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Geoff.

        Make resource load delegate methods pass the right document loader.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::willSendRequest):
        (WebCore::FrameLoader::didReceiveResponse):
        (WebCore::FrameLoader::didReceiveData):
        (WebCore::FrameLoader::didFailToLoad):
        (WebCore::FrameLoader::didFinishLoad):
        (WebCore::FrameLoader::didReceiveAuthenticationChallenge):
        (WebCore::FrameLoader::didCancelAuthenticationChallenge):
        Use the resource loader's document loader instead of the active one.

        * loader/MainResourceLoader.cpp:
        (WebCore::MainResourceLoader::receivedError):
        Make it so we send the frame load delegate method before the resource load delegate method.
        This was a regression from 2.0 and was caused by the fix to rdar://problem/4609195. Because the way the loader
        now works, both delegate methods will be called.

        * loader/ResourceLoader.cpp:
        (WebCore::ResourceLoader::ResourceLoader):
        * loader/ResourceLoader.h:
        (WebCore::ResourceLoader::documentLoader):
        Add document loader pointer to ResourceLoader.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin Adler.

        Follow-up to fixing http://bugs.webkit.org/show_bug.cgi?id=12659 | <rdar://problem/4954306>
        JS objects not collected after closing window @ ebay.com/maps.google.com

        Changed Interpreter cache of global constructors and prototypes from
        ProtectedPtrs to bare, marked pointers. ProtectedPtrs are inefficient,
        and they increase the risk of reference cycles. Also, Darin said something
        about ProtectedPtrs giving him warts.

        Also changed data members to precise types from generic JSObject*'s.

        Layout tests and JS tests pass.

        * kjs/SavedBuiltins.h:
        * kjs/interpreter.cpp:
        (KJS::Interpreter::init):
        (KJS::Interpreter::~Interpreter):
        (KJS::Interpreter::initGlobalObject): Moved Identifier::init() call to
        constructor, for clarity.
        (KJS::Interpreter::mark):
        * kjs/interpreter.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Oops! Forgot to check in this file.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Hyatt.

        - repaint test for http://bugs.webkit.org/show_bug.cgi?id=12910
          REGRESSION (r18756-r18765): list-bullet doesn't redraw properly when changing the list's content using JavaScript

        * fast/repaint/list-marker-expected.checksum: Added.
        * fast/repaint/list-marker-expected.png: Added.
        * fast/repaint/list-marker-expected.txt: Added.
        * fast/repaint/list-marker.html: Added.

WebCore:

        Reviewed by Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12910
          REGRESSION (r18756-r18765): list-bullet doesn't redraw properly when changing the list's content using JavaScript

        Test: fast/repaint/list-marker.html

        * rendering/RenderListItem.cpp:
        (WebCore::RenderListItem::positionListMarker): Add the marker to the visual
        overflow of all its ancestor blocks up to the list item. This needs to be
        done here since the marker is positioned only after those blocks have been
        laid out.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-02-28  Mark Rowe  <mrowe@apple.com>

        Reviewed by Tim Hatcher.

        <rdar://problem/4985524> Problem with Blot and ToT WebKit (decoding WebCoreScrollView)

        References to WebCoreScrollView as a subview of a WebHTMLView may be present in some NIB
        files, so NSUnarchiver must be still able to look up the WebCoreScrollView class.

        * WebKit.exp: Export WebCoreScrollView symbol.
        * WebView/WebHTMLView.mm: Add empty WebCoreScrollView class.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests:

        Reviewed by Hyatt.

        - test for http://bugs.webkit.org/show_bug.cgi?id=12918
        REGRESSION: Google Finance dropdown Flickers
        <rdar://4988039>

        * fast/events/mouseout-dead-node-expected.checksum: Added.
        * fast/events/mouseout-dead-node-expected.png: Added.
        * fast/events/mouseout-dead-node-expected.txt: Added.
        * fast/events/mouseout-dead-node.html: Added.

WebCore:

        Reviewed by Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12918
        REGRESSION: Google Finance dropdown Flickers
        <rdar://4988039>

        If a node dies while mouse is over it, it is still supposed to receive
        mouseout event (wasn't case in Tiger webkit). However this event should
        not propagate to any other nodes. This patch matches Firefox behavior in
        this respect.

        * dom/EventTargetNode.cpp:
        (WebCore::EventTargetNode::dispatchGenericEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Bumped up the wiggle room on this test because it's failing on the buildbot.
        (We're still well below the ~100 objects you would see in a real failure.)

        * fast/leaks/003.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Oops! Forgot to commit this file

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

JavaScriptCore:

        Reviewed by Maciej Stachowiak.

        Fixed http://bugs.webkit.org/show_bug.cgi?id=12659 | <rdar://problem/4954306>
        JS objects not collected after closing window @ ebay.com/maps.google.com

        Don't GC in the Interpreter destructor. For that to work, the Interpreter
        would have to NULL out all of its ProtectedPtrs before calling collect(). But
        we've decided that we don't want things to work that way, anyway. We want the
        client to be in charge of manual GC so that it can optimize cases when
        it will be destroying many interpreters at once
        (e.g., http://bugs.webkit.org/show_bug.cgi?id=12900).

        Also removed Interpreter::collect() because it was redundant with
        Collector::collect().

        * JavaScriptCore.exp:
        * kjs/interpreter.cpp:
        (KJS::Interpreter::~Interpreter):
        * kjs/testkjs.cpp:
        (TestFunctionImp::callAsFunction):

LayoutTests:

        Reviewed by Maciej Stachowiak.

        Test for http://bugs.webkit.org/show_bug.cgi?id=12659 | <rdar://problem/4954306>
        JS objects not collected after closing window @ ebay.com/maps.google.com

        * fast/leaks/003-expected.txt: Added.
        * fast/leaks/003.html: Added.

        * fast/dom/gc-10.html: Bumped the wiggle room threshold on this test by 2
        because it's failing on my machine with the patch applied but, according
        to the test notes, we're still well within the margin of error.

WebCore:

        Reviewed by Maciej Stachowiak.

        Fixed http://bugs.webkit.org/show_bug.cgi?id=12659 | <rdar://problem/4954306>
        JS objects not collected after closing window @ ebay.com/maps.google.com

        Garbage collect in the KJSProxy destructor, after clearing our reference
        to the interpreter, because that's when the interpreter has torn down fully.

        (Technically speaking, we can't *prove* that we have the only reference to
        our interpreter, but that's how it works in practice, and manual garbage
        collection is just an opportunistic optimization, so it's OK for it to
        work in practice even if it can't be proven in theory.)

        Layout tests pass. No leaks reported.

        * bindings/js/kjs_proxy.cpp:
        (WebCore::KJSProxy::~KJSProxy):
        * bindings/js/kjs_proxy.h:
        * page/Page.cpp:
        (WebCore::Page::~Page): Merged pageDestroyed() calls. Moved debug-only
        code to the bottom.
        * page/Frame.cpp:
        (WebCore::Frame::~Frame): Don't call getObject() because globalObject()
        returns a JSObject* already, and the call can leave a pointer to the Window
        object on the stack. Don't check for NULL because it is an invariant of
        JavaScriptCore that no JSObject* can be NULL. Do use a volatile pointer
        for w because the 'w = 0' assignment just screams to the compiler, "Don't
        generate any code for me!"

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Beth.

        Fix <rdar://problem/5011905> REGRESSION: "Open Link" contextual menu
        item appears twice

        * WebCoreSupport/WebContextMenuClient.mm:
        (fixMenusToSendToOldClients): Remove the "Open Link" item from the
        default menu items array before sending it off to Tiger Mail.
        (WebContextMenuClient::getCustomMenuFromDefaultItems): Set the
        representedObject on every NSMenuItem to match our old (correct) API
        behavior.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by David Hyatt.

        http://bugs.webkit.org/show_bug.cgi?id=4128
        !important is ignored in inline styling.

        Handle properties with !important flag better in inline
        style declarations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Bug 11435.  Make sure RenderViews always paintBoxDecorations.  They used to before I removed RenderView's
        paint method.  This fix restores the original behavior.

        Reviewed by mitz

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::setStyle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Fix for bug 12094, make sure setMinMaxKnown starts off false for the weird/rare case of empty
        plaintext documents (which really are buggy and should be fixed to generate root elements).

        Reviewed by mitz

        * rendering/RenderView.cpp:
        (WebCore::RenderView::RenderView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-02-27  Dex Deacon  <occupant4@gmail.com>

        Reviewed by Darin.

        Fixed the case where a BackForwardList of capacity==1 would grow without bound.

        * history/BackForwardList.cpp:
        (WebCore::BackForwardList::addItem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-02-22  Lars Naesbye Christensen  <lars@naesbye.dk>

        Reviewed by Maciej.

        http://bugs.webkit.org/show_bug.cgi?id=12848
        Help cursor should have a white outline

        * Resources/helpCursor.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2007-02-27  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Darin.

        - http://bugs.webkit.org/show_bug.cgi?id=12906
          REGRESSION: Canvas is pixelated when the page is opened in a background tab

        * WebCoreSupport/WebChromeClient.mm:
        (WebChromeClient::scaleFactor): If the view is not in a window, use the main
        screen's scale factor as a best guess.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej.

        http://bugs.webkit.org/show_bug.cgi?id=12594
        REGRESSION: Strange highlight in active input area

        No automated test possible.

        * rendering/InlineTextBox.cpp:
        (WebCore::InlineTextBox::paint): Restore a check lost in r12792.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Small tweak to run-webkit-tests.

        * Scripts/run-webkit-tests: Allow people with lots of RAM to run more than
        1000 MallocStackLogging tests at a time.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Mitz.

        - fix http://bugs.webkit.org/show_bug.cgi?id=12908
          crash in http/tests/incremental/frame-focus-before-load.html

        * page/mac/EventHandlerMac.mm: (WebCore::EventHandler::focusDocumentView):
        Add null checks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@19884 268f45cc-cd09-0410-ab3c-d52691b4dbfc