16 years agoLayoutTests:
darin [Thu, 3 Aug 2006 15:22:52 +0000 (15:22 +0000)]

        Reviewed by Anders.

        - test for http://bugzilla.opendarwin.org/show_bug.cgi?id=10202
          REGRESSION: Repro crash when loading an empty image document

        * fast/tokenizer/image-empty-crash-expected.txt: Added.
        * fast/tokenizer/image-empty-crash.html: Added.


        Reviewed by Anders.

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=10202
          REGRESSION: Repro crash when loading an empty image document

        Test: fast/tokenizer/image-empty-crash.html

        * loader/ImageDocument.cpp:
        (WebCore::ImageTokenizer::stopParsing): Added null check for m_imageElement.
        (WebCore::ImageTokenizer::finish): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Anders.
darin [Thu, 3 Aug 2006 15:21:10 +0000 (15:21 +0000)]
    Reviewed by Anders.

        - clean up "flip"-related code a tiny bit

        * platform/Screen.h: Removed redundant parameter names.
        * platform/mac/ScreenMac.mm:
        (WebCore::flipScreenRect): Changed to call flipScreenPoint so we only have one
        copy of the flipping code.
        (WebCore::flipScreenPoint): Fixed indentation.

        * bridge/mac/WebCoreFrameBridge.mm: (globalPoint): Removed type casts and
        simplified a bit.

        * platform/mac/PlatformMouseEventMac.mm: (WebCore::globalPositionForEvent):
        * platform/mac/WheelEventMac.mm: (WebCore::globalPositionForEvent):
        Removed unneeded local variable.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Eric Seidel.
darin [Thu, 3 Aug 2006 15:17:54 +0000 (15:17 +0000)]
    Reviewed by Eric Seidel.

        - fix storage leak

        * WebView/WebFrame.m: (-[WebFramePrivate dealloc]): Release the frame loader.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Rubber stamped by Maciej.
thatcher [Thu, 3 Aug 2006 06:08:04 +0000 (06:08 +0000)]
    Rubber stamped by Maciej.

        Adding back resultsWithXpathQuery, removed by Darin's earlier change.
        This function is called from ObjC, but not used from JavaScript.

        * WebInspector/webInspector/inspector.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
thatcher [Wed, 2 Aug 2006 23:18:00 +0000 (23:18 +0000)]
    Reviewed by Darin.

        Bug 9632: [Drosera] syntax highlighting is slow (reproducible 40 second hang)

        Do not change the file source when normalizing the line endings.
        We use this file source to compare against new versions of the source
        as it comes in, so we can skip re-syntax highlighting if they are the same.
        The problem is apparent on yahoo.com since they have mixed line endings and
        once we normalize them the source will always be different. This was
        compounded by the fact that yahoo has around 40 inline scripts. Each
        inline script causes us to check if the main document has more loaded,
        that is when we compare the source strings. Since they are always different
        we would syntax highlight yahoo.com 40 times! We do check source length before
        doing a more expensive string comparison, but the lengths were the same.

        * Drosera/debugger.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
thatcher [Wed, 2 Aug 2006 23:15:27 +0000 (23:15 +0000)]
    Reviewed by Darin.

        Bug 10200: [Drosera] Deadlock between Drosera and Safari while loading page

        Prevent reentrancy in our debugger callbacks. This was causing a deadlock in Drosera because
        suspendProcessIfPaused was being called during a DO call into Safari.

        Preventing reentrancy also prevents scripts that Drosera injects and evaluates from showing
        up in rare cases (such as a iframe loading about:blank). I thought this would prevent cases
        where you call a function from the console and expect it to break on a breakpoint in them, but
        this appears to never have worked even without this change. When that is figured out we can
        reconsider a better solution to reentrancy. I have filed that as bug 10214.

        I also removed the NSRunLoop runMode:beforeDate: calls since DO handles this for us since
        we don't use "onway void" as the return type for the callbacks. Note: using onway void for
        the listener callbacks causes bad synchronization issues and obscure crashes.

        * DefaultDelegates/WebScriptDebugServer.m:
        (-[WebScriptDebugServer webView:didLoadMainResourceForDataSource:]):
        (-[WebScriptDebugServer webView:didParseSource:baseLineNumber:fromURL:sourceId:forWebFrame:]):
        (-[WebScriptDebugServer webView:failedToParseSource:baseLineNumber:fromURL:withError:forWebFrame:]):
        (-[WebScriptDebugServer webView:didEnterCallFrame:sourceId:line:forWebFrame:]):
        (-[WebScriptDebugServer webView:willExecuteStatement:sourceId:line:forWebFrame:]):
        (-[WebScriptDebugServer webView:willLeaveCallFrame:sourceId:line:forWebFrame:]):
        (-[WebScriptDebugServer webView:exceptionWasRaised:sourceId:line:forWebFrame:]):
        * DefaultDelegates/WebScriptDebugServerPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Abstract RenderLayer scrollbar creation and destruction so that there
hyatt [Wed, 2 Aug 2006 22:59:51 +0000 (22:59 +0000)]
    Abstract RenderLayer scrollbar creation and destruction so that there
        is a clean spot in which to drop in engine scrollbar stuff.

        Reviewed by timo

        * platform/ScrollBar.h:
        * rendering/RenderLayer.cpp:
        * rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by TimO.
ggaren [Wed, 2 Aug 2006 22:58:23 +0000 (22:58 +0000)]
    Reviewed by TimO.

        Tweaked the build instructions a bit.

        * building/build.html:
        * building/debug.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Fix by Mitz Pettel, reviewed by Darin.
ap [Wed, 2 Aug 2006 20:32:45 +0000 (20:32 +0000)]
    Fix by Mitz Pettel, reviewed by Darin.

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=3438
          incorrect display of transparent 1x1 PNGs

        Test: fast/replaced/image-solid-color-with-alpha.html

        * platform/mac/ImageMac.mm:
        (WebCore::Image::checkForSolidColor): Changed to divide each color component
        by the alpha component. You need to do that since the bitmap context has
        premultiplied alpha but m_solidColor should be non-premultiplied.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
aroben [Wed, 2 Aug 2006 20:26:11 +0000 (20:26 +0000)]
    Reviewed by Darin.

        - Fixed build

        * kjs/date_object.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by John.
mjs [Wed, 2 Aug 2006 19:06:45 +0000 (19:06 +0000)]
    Reviewed by John.

        - fix assertion failure on layout tests by stopping plugins from loading at a clearly defined time
        - add more assertions for safety

        * WebView/WebDataSource.m:
        (-[WebDataSourcePrivate dealloc]): Removed obsolete comment.
        (-[WebDataSource _updateLoading]): Add assertion ensuring this method is only called
        at a time when this data source is the one that might be loading for a frame.
        (-[WebDataSource _stopLoading]): Stop loading plugins as a FIXME suggests we should.
        * WebView/WebFrameLoader.m:
        (-[WebFrameLoader isLoadingPlugIns]): New helper method.
        (-[WebFrameLoader isLoading]): Consider plugin loads too - otherwise we won't stop them
        at stopLoading time.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago * building/checkout.html: Fix a typo: "OpenSource" instead of "Open Source".
darin [Wed, 2 Aug 2006 18:35:45 +0000 (18:35 +0000)]
    * building/checkout.html: Fix a typo: "OpenSource" instead of "Open Source".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoWebCore:
aroben [Wed, 2 Aug 2006 18:14:06 +0000 (18:14 +0000)]
        Reviewed by Brady.

        - Remove redundant #include

        * loader/loader.cpp:

        Reviewed by Brady.

        - Rename TransferJob to ResourceLoader (this file was forgotten in an
        earlier change by Maciej)

        * COM/WebView.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Timothy.
thatcher [Wed, 2 Aug 2006 18:13:46 +0000 (18:13 +0000)]
    Reviewed by Timothy.

        Bug 9931: [Drosera] Needs a cool icon

        * Drosera/Drosera.icns: Added.
        * Drosera/Drosera.xcodeproj/project.pbxproj:
        * Drosera/Info.plist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Adele.
mjs [Wed, 2 Aug 2006 08:11:39 +0000 (08:11 +0000)]
    Reviewed by Adele.

        - Change things around so WebFrameLoader tracks the main and provisional data source,
        as well as the frame load state, pulling much code out of WebFrame along the way.

        The most significant aspects of this change are:

        - management of WebDataSources and WebFrameState was moved into WebFrameLoader
        - there is now just one WebFrameLoader shared between the primary and provisional data source

        * WebKit.xcodeproj/project.pbxproj:
        * WebView/WebDataSource.m:
        (-[WebDataSourcePrivate dealloc]):
        (-[WebDataSource _updateLoading]):
        (-[WebDataSource _loadIcon]):
        (-[WebDataSource _setPrimaryLoadComplete:]):
        (-[WebDataSource _stopLoading]):
        (-[WebDataSource _startLoading]):
        (-[WebDataSource _addSubresourceLoader:]):
        (-[WebDataSource _removeSubresourceLoader:]):
        (-[WebDataSource _addPlugInStreamLoader:]):
        (-[WebDataSource _removePlugInStreamLoader:]):
        (-[WebDataSource _defersCallbacksChanged]):
        (-[WebDataSource _stopLoadingWithError:]):
        (-[WebDataSource _revertToProvisionalState]):
        (-[WebDataSource _setupForReplaceByMIMEType:]):
        (-[WebDataSource initWithRequest:]):
        (-[WebDataSource data]):
        (-[WebDataSource isLoading]):
        * WebView/WebFrame.m:
        (-[WebFramePrivate init]):
        (-[WebFramePrivate dealloc]):
        (-[WebFrame _closeOldDataSources]):
        (-[WebFrame _detachFromParent]):
        (-[WebFrame _makeDocumentView]):
        (-[WebFrame _receivedMainResourceError:]):
        (-[WebFrame _transitionToCommitted:]):
        (+[WebFrame _timeOfLastCompletedLoad]):
        (-[WebFrame _checkLoadCompleteForThisFrame]):
        (-[WebFrame _loadItem:withLoadType:]):
        (-[WebFrame _continueAfterWillSubmitForm:]):
        (-[WebFrame _continueLoadRequestAfterNavigationPolicy:formState:]):
        (-[WebFrame _initWithWebFrameView:webView:bridge:]):
        (-[WebFrame _frameLoader]):
        (-[WebFrame _provisionalLoadStarted]):
        (-[WebFrame _prepareForDataSourceReplacement]):
        (-[WebFrame _frameLoadCompleted]):
        (-[WebFrame provisionalDataSource]):
        (-[WebFrame dataSource]):
        (-[WebFrame stopLoading]):
        * WebView/WebFrameInternal.h:
        * WebView/WebFrameLoader.h:
        * WebView/WebFrameLoader.m:
        (-[WebFrameLoader initWithWebFrame:]):
        (-[WebFrameLoader dealloc]):
        (-[WebFrameLoader dataSource]):
        (-[WebFrameLoader _setDataSource:]):
        (-[WebFrameLoader clearDataSource]):
        (-[WebFrameLoader provisionalDataSource]):
        (-[WebFrameLoader _setProvisionalDataSource:]):
        (-[WebFrameLoader _clearProvisionalDataSource]):
        (-[WebFrameLoader state]):
        (+[WebFrameLoader timeOfLastCompletedLoad]):
        (-[WebFrameLoader _setState:]):
        (-[WebFrameLoader clearProvisionalLoad]):
        (-[WebFrameLoader markLoadComplete]):
        (-[WebFrameLoader commitProvisionalLoad]):
        (-[WebFrameLoader stopLoading]):
        (-[WebFrameLoader startLoading]):
        (-[WebFrameLoader startProvisionalLoad:]):
        (-[WebFrameLoader setupForReplace]):
        * WebView/WebFramePrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by John Sullivan.
tomernic [Tue, 1 Aug 2006 23:06:00 +0000 (23:06 +0000)]
    Reviewed by John Sullivan.

        <rdar://problem/4480737> Flash crashes after it replaces itself via a document.write()

        I kind of hate to do this, but this is the best way to work around buggy plug-ins like Flash that assume that
        NPP_Destroy() cannot be called while the browser is calling one of its other plug-in functions.  The classic
        situation is a plug-in that replaces itself via an NPN_Invoke() that executes a document.write().

        * Plugins/WebBaseNetscapePluginView.h:
        * Plugins/WebBaseNetscapePluginView.m:
        (-[WebBaseNetscapePluginView sendEvent:]):
        Call -willCallPlugInFunction and -didCallPlugInFunction around calls to the NPP_* functions.
        (-[WebBaseNetscapePluginView setWindowIfNecessary]):
        (-[WebBaseNetscapePluginView start]):
        It should not be possible to start a plug-in instance while we are calling into it (one of those chicken/egg
        problems).  Added a sanity-checking assertion.
        (-[WebBaseNetscapePluginView stop]):
        If we're already calling a plug-in function, do not call NPP_Destroy().  The plug-in function we are calling
        may assume that its instance->pdata, or other memory freed by NPP_Destroy(), is valid and unchanged until said
        plugin-function returns.
        (-[WebBaseNetscapePluginView pluginScriptableObject]):
        Call -willCallPlugInFunction and -didCallPlugInFunction around calls to the NPP_* functions.
        (-[WebBaseNetscapePluginView willCallPlugInFunction]):
        Increment plug-in function call depth.
        (-[WebBaseNetscapePluginView didCallPlugInFunction]):
        Decrement plug-in function call depth.  Stop if we're supposed to stop.
        (-[WebBaseNetscapePluginView evaluateJavaScriptPluginRequest:]):
        Call -willCallPlugInFunction and -didCallPlugInFunction around calls to the NPP_* functions.
        (-[WebBaseNetscapePluginView webFrame:didFinishLoadWithReason:]):
        (-[WebBaseNetscapePluginView _printedPluginBitmap]):

        * Plugins/WebBaseNetscapePluginStream.m:
        (-[WebBaseNetscapePluginStream startStreamResponseURL:expectedContentLength:lastModifiedDate:MIMEType:]):
        Call -willCallPlugInFunction and -didCallPlugInFunction around calls to the NPP_* functions.
        (-[WebBaseNetscapePluginStream _destroyStream]):
        (-[WebBaseNetscapePluginStream _deliverData]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago - fix build after last change
mjs [Tue, 1 Aug 2006 22:53:40 +0000 (22:53 +0000)]
    - fix build after last change

        * WebView/WebFrame.m:
        (-[WebFrame _checkLoadCompleteForThisFrame]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Beth.
mjs [Tue, 1 Aug 2006 22:05:57 +0000 (22:05 +0000)]
    Reviewed by Beth.

        - revert part of my last fix that broke the Safari bookmarks view

        * WebView/WebFrame.m:
        (-[WebFrame _checkLoadCompleteForThisFrame]): still send layout message for non-HTML views

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Anders.
tomernic [Tue, 1 Aug 2006 20:20:42 +0000 (20:20 +0000)]
    Reviewed by Anders.

        Fixed an assertion failure I ran into while debugging <rdar://problem/4652683>.

        * Plugins/WebNetscapePluginEmbeddedView.m:
        (-[WebNetscapePluginEmbeddedView redeliverStream]):
        Don't clear the "instance" ivar here.  This code was refactored here from the old WebNetscapePluginRepresentation,
        which also had an "instance" ivar.  It is never appropriate to clear a plug-in view's instance.  That is done when
        the plug-in is destroyed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Beth.
ggaren [Tue, 1 Aug 2006 18:41:04 +0000 (18:41 +0000)]
    Reviewed by Beth.

        - Added instructions for debugging in Xcode

        * building/build-window.png: Added.
        * building/build.html:
        * building/custom-executable.png: Added.
        * building/info-tab.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Geoff.
mjs [Tue, 1 Aug 2006 18:38:33 +0000 (18:38 +0000)]
    Reviewed by Geoff.

        - some refactoring in preparation for moving more stuff to WebFrameLoader.

        * WebView/WebFrame.m:
        (-[WebFrame _clearDataSource]):
        (-[WebFrame _detachFromParent]):
        (-[WebFrame _commitProvisionalLoad]):
        (-[WebFrame _transitionToCommitted:]):
        (-[WebFrame _clearProvisionalLoad]):
        (-[WebFrame _markLoadComplete]):
        (-[WebFrame _checkLoadCompleteForThisFrame]):
        (-[WebFrame _startProvisionalLoad:]):
        (-[WebFrame _continueLoadRequestAfterNavigationPolicy:formState:]):
        (-[WebFrame stopLoading]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Tue, 1 Aug 2006 16:40:22 +0000 (16:40 +0000)]

        - updated a test affected by changes to the DumpRenderTree tool
          to return a current mouse position

        * fast/inline/dirtyLinesForInline-expected.txt: Updated.


        - fix a bug in my recent change where the mouse position at the end of
          the last test would affect the results of the next test

        * DumpRenderTree/DumpRenderTree.m: (dumpRenderTree): Set lastMousePosition to 0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Tue, 1 Aug 2006 06:15:38 +0000 (06:15 +0000)]

        Reviewed by Darin.

        - tests changed by http://bugzilla.opendarwin.org/show_bug.cgi?id=9955
        Colors scaled from [0, 1] range to [0, 255] incorrectly

        * fast/css/hsl-color-expected.checksum:
        * fast/css/hsl-color-expected.png:
        * fast/css/hsl-color-expected.txt:
        * fast/forms/input-disabled-color-expected.checksum:
        * fast/forms/input-disabled-color-expected.png:
        * fast/forms/input-disabled-color-expected.txt:


        Reviewed by Darin.

        - patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=9955
        Colors scaled from [0, 1] range to [0, 255] incorrectly

        Fixes scaling issue and fixes Color to follow style guidelines.

        * css/cssstyleselector.cpp:
        * platform/Color.cpp:
        * platform/Color.h:
        * rendering/RenderObject.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Tue, 1 Aug 2006 06:13:20 +0000 (06:13 +0000)]

        - test for http://bugzilla.opendarwin.org/show_bug.cgi?id=10158
          REGRESSION: Selection highlight is dark and opaque when solid-color images are used

        Reviewed by Darin.

        * fast/backgrounds/solid-color-context-restore-expected.checksum: Added.
        * fast/backgrounds/solid-color-context-restore-expected.png: Added.
        * fast/backgrounds/solid-color-context-restore-expected.txt: Added.
        * fast/backgrounds/solid-color-context-restore.html: Added.


        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=10158
          REGRESSION: Selection highlight is dark and opaque when solid-color images are used

        Reviewed by Darin.

        Pixel test: fast/backgrounds/solid-color-context-restore.html

        * platform/mac/ImageMac.mm:
        (WebCore::Image::draw): Added calls to preserve the graphics context
        around the painting of a solid color image.
        (WebCore::Image::drawTiled): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Maciej.
darin [Tue, 1 Aug 2006 06:10:04 +0000 (06:10 +0000)]
    Reviewed by Maciej.

        Bug 10182: [Drosera] Evaluating expressions in Console is slow when current frame has many variables

        * Drosera/console.js: String.indexOf returns -1 when the string is not found.
        Correct the logic to not reload local variable list unless an '=' character is in the expression.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Tue, 1 Aug 2006 06:00:32 +0000 (06:00 +0000)]

        Reviewed by Maciej.

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=10171
          REGRESSION: failing layout test: fast/events/objc-event-api.html

        * fast/events/objc-event-api-expected.txt: Updated test results for changes in the
        DumpRenderTree tool.


        Reviewed by Maciej.

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=10171
          REGRESSION: failing layout test: fast/events/objc-event-api.html

        * DumpRenderTree/DumpRenderTree.m:
        (main): Add pose so we can implement +[NSEvent mouseLocation]. Put the window at a predictable
        location in flipped coordinates, since those are the coordinates that we use.
        (+[DumpRenderTreeEvent mouseLocation]): Implement this, since it's used for mouse event handling
        inside WebHTMLView.

        * DumpRenderTree/EventSendingController.m:
        (+[EventSendingController initialize]): Removed unused webkitDomEventProperties array.
        (-[EventSendingController init]): Removed, since the whole thing was a no-op.
        (-[EventSendingController mouseMoveToX:Y:]): Convert the x,y pair to window coordinates. The old code
        probably worked OK, but this is needed to be correct.
        (-[EventSendingController handleEvent:]): Put clientX/Y before screenX/Y and removed the flipping
        code from screenY. We didn't really need to flip screenY -- what was actually happening was that the
        position that DumpRenderTree chose was in non-flipped coordinates, and with that fixed we can just
        dump the screenY as-is.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoWebCore:
mjs [Tue, 1 Aug 2006 05:15:40 +0000 (05:15 +0000)]

        Rubber-stamped by Tim Hatcher.

        - renamed TransferJob to ResourceLoader

        * Projects/gdk/webcore-gdk.bkl:
        * WebCore.vcproj/WebCore/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * WebCoreSources.bkl:
        * bridge/mac/FrameMac.mm:
        * bridge/mac/WebCoreResourceLoaderImp.h:
        * bridge/mac/WebCoreResourceLoaderImp.mm:
        (-[WebCoreResourceLoaderImp redirectedToURL:]):
        (-[WebCoreResourceLoaderImp addData:]):
        (-[WebCoreResourceLoaderImp finishJobAndHandle:]):
        * bridge/win/FrameWin.cpp:
        * dom/XMLTokenizer.cpp:
        * loader/Cache.cpp:
        * loader/LoaderFunctions.h:
        * loader/loader.cpp:
        * loader/loader.h:
        * loader/mac/LoaderFunctionsMac.mm:
        * platform/ResourceLoader.cpp: Added.
        * platform/ResourceLoader.h: Added.
        * platform/ResourceLoaderClient.h: Added.
        * platform/ResourceLoaderInternal.h: Added.
        * platform/TransferJob.cpp: Removed.
        * platform/TransferJob.h: Removed.
        * platform/TransferJobClient.h: Removed.
        * platform/TransferJobInternal.h: Removed.
        * platform/gdk/FrameGdk.cpp:
        * platform/gdk/FrameGdk.h:
        * platform/gdk/ResourceLoaderCurl.cpp: Added.
        * platform/gdk/ResourceLoaderManager.cpp: Added.
        * platform/gdk/ResourceLoaderManager.h: Added.
        * platform/gdk/TemporaryLinkStubs.cpp:
        * platform/gdk/TransferJobCurl.cpp: Removed.
        * platform/gdk/TransferJobManager.cpp: Removed.
        * platform/gdk/TransferJobManager.h: Removed.
        * platform/mac/ResourceLoaderMac.mm: Added.
        * platform/mac/TransferJobMac.mm: Removed.
        * platform/win/ResourceLoaderWin.cpp: Added.
        * platform/win/ResourceLoaderWin.h: Added.
        * platform/win/TemporaryLinkStubs.cpp:
        * platform/win/TransferJobWin.cpp: Removed.
        * platform/win/TransferJobWin.h: Removed.
        * xml/XSLTProcessor.cpp:
        * xml/xmlhttprequest.cpp:
        * xml/xmlhttprequest.h:


        Reviewed by Tim Hatcher.

        - renamed TransferJob to ResourceLoader in WebCore

        * COM/WebFrame.cpp:
        * COM/WebFrame.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Maciej.
darin [Tue, 1 Aug 2006 04:54:55 +0000 (04:54 +0000)]
    Reviewed by Maciej.

        - omit the margin and padding boxes for display types where they are ignored
        - use CSS instead of properties for table spacing and padding as suggested by Tim H.

        * WebInspector/webInspector/inspector.css: Added rules for spacing and padding.
        Added rules that hide the margin and padding boxes (borders and all but the center cell)
        when the hide attribute is present.
        * WebInspector/webInspector/inspector.html: Added classes for the rules above.
        Removed cellpadding and cellspacing attributes.
        * WebInspector/webInspector/inspector.js: Added code to hide/show the margin and
        padding boxes based on the display type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Hyatt.
darin [Tue, 1 Aug 2006 04:14:20 +0000 (04:14 +0000)]
    Reviewed by Hyatt.

        - removed obsolete cell margin hack

        * rendering/RenderBox.cpp: (WebCore::RenderBox::calcVerticalMargins):
        Set top and bottom margins to 0 for table cells instead of -16384.
        No effect on layout (including layout tests).

        * css/CSSComputedStyleDeclaration.cpp:
        Added a few comments about differences between this class and the
        what the CSS specification says about computed style.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
thatcher [Tue, 1 Aug 2006 04:00:57 +0000 (04:00 +0000)]

        Reviewed by Maciej.

        Bug 9738: Unqualified :hover selector ignored in strict parsing mode

        * fast/selectors/unqualified-hover-quirks-expected.checksum: Added.
        * fast/selectors/unqualified-hover-quirks-expected.png: Added.
        * fast/selectors/unqualified-hover-quirks-expected.txt: Added.
        * fast/selectors/unqualified-hover-quirks.html: Added.
        * fast/selectors/unqualified-hover-strict-expected.checksum: Added.
        * fast/selectors/unqualified-hover-strict-expected.png: Added.
        * fast/selectors/unqualified-hover-strict-expected.txt: Added.
        * fast/selectors/unqualified-hover-strict.html: Added.


        Reviewed by Maciej.

        Bug 9738: Unqualified :hover selector ignored in strict parsing mode

        * css/cssstyleselector.cpp:
        (WebCore::CSSStyleSelector::checkSelector): Restrict the :hover and :active
        exclusion based on onlyHoverActive to quirks mode.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
thatcher [Tue, 1 Aug 2006 03:54:30 +0000 (03:54 +0000)]
    Reviewed by Darin.

        Fixes <http://bugzilla.opendarwin.org/show_bug.cgi?id=10159>
        "REGRESSION: delegate returning no menu elements crashes webkit"

        No automated test, because there's no way to programmatically open a context menu,
        no manual test because there's no way to customize the context menu delegate.

        * WebView/WebView.m:
        (-[WebView _menuForElement:defaultItems:]): Make sure the context menu returned
        some menu items before accessing the first one.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
thatcher [Tue, 1 Aug 2006 03:51:45 +0000 (03:51 +0000)]
    Reviewed by Darin.

        Bug 10178: [Drosera] Selecting a frame in the stack list should take you to that location in the source

        * Drosera/debugger.js:  Keep a stack that contains the source file and line number references for outer
        frames.  Use this stack to determine which file and line to highlight when a stack frame is selected.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
thatcher [Tue, 1 Aug 2006 03:48:39 +0000 (03:48 +0000)]
    Reviewed by Darin.

        Bug 10175: [Drosera] Anonymous functions show up as "(global scope)" in stack list

        * Drosera/DebuggerDocument.m:
        (-[WebScriptObject currentFunctionStack]): If the frame has a caller it isn't the global scope.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoAdd ScrollBar.cpp to fix the build.
hyatt [Tue, 1 Aug 2006 02:27:57 +0000 (02:27 +0000)]
Add ScrollBar.cpp to fix the build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Fix for bug 10179, digg.com scrolls slowly. Improve fixed positioning
hyatt [Tue, 1 Aug 2006 02:25:43 +0000 (02:25 +0000)]
    Fix for bug 10179, digg.com scrolls slowly.  Improve fixed positioning
        and fixed backgrounds so that a count of them is kept on the FrameView.
        This allows us to switch slow scrolling on and off as these objects come
        and go.

        * css/cssstyleselector.cpp:
        * page/FrameView.cpp:
        * page/FrameView.h:
        * rendering/RenderBox.cpp:
        * rendering/RenderObject.cpp:
        * rendering/RenderView.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Begin disentangling of scrollbar logic in preparation for engine-implemented
hyatt [Tue, 1 Aug 2006 01:50:45 +0000 (01:50 +0000)]
    Begin disentangling of scrollbar logic in preparation for engine-implemented
        scrollbars.  Split ScrollBar into two classes, an abstract base (still named
        ScrollBar) and a new PlatformScrollBar subclass.  This subclass is used only
        by platforms that want to continue to use a platform scrollbar (rather than
        the engine one).

        Reviewed by darin

        * WebCore.xcodeproj/project.pbxproj:
        * html/HTMLCanvasElement.h:
        * page/Frame.cpp:
        * platform/PlatformScrollBar.h: Added.
        * platform/ScrollBar.h:
        * platform/mac/PlatformScrollBarMac.mm: Added.
        (-[WebCoreScrollBar initWithPlatformScrollBar:]):
        (-[WebCoreScrollBar detachPlatformScrollBar]):
        (-[WebCoreScrollBar scroll:]):
        (-[WebCoreScrollBar widget]):
        (-[WebCoreScrollBar mouseDown:]):
        * platform/mac/ScrollBarMac.mm: Removed.
        * platform/mac/WebCoreTextArea.mm:
        * rendering/RenderBlock.cpp:
        * rendering/RenderFormElement.cpp:
        * rendering/RenderLayer.cpp:
        * rendering/RenderLayer.h:
        * rendering/RenderWidget.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
adele [Tue, 1 Aug 2006 00:05:30 +0000 (00:05 +0000)]

        Reviewed by John and Anders.

        - Test for <rdar://problem/4380576> onChange does not get called when field changed by return key

        * fast/forms/onchange-enter-submit-expected.txt: Added.
        * fast/forms/onchange-enter-submit.html: Added.


        Reviewed by John and Anders.

        - Fix for <rdar://problem/4380576> onChange does not get called when field changed by return key

        Test: fast/forms/onchange-enter-submit.html

        Blur (which will fire onChange appropriately) before submitting a form when pressing Enter.
        This matches the sequence of events fired in WinIE.

        * html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::defaultEventHandler):
        * html/HTMLSelectElement.cpp: (WebCore::HTMLSelectElement::defaultEventHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by John.
thatcher [Mon, 31 Jul 2006 23:55:15 +0000 (23:55 +0000)]
    Reviewed by John.

        <rdar://problem/4658194> REGRESSION: "Search in Google"
        and "Search in Spotlight" fail to work on text selected in a frame

        Use selectedFrame to get the frame with the text selection.

        * WebView/WebView.m:
        (-[WebView _searchWithGoogleFromMenu:]):
        (-[WebView _searchWithSpotlightFromMenu:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Kevin.
ggaren [Mon, 31 Jul 2006 22:27:46 +0000 (22:27 +0000)]
    Reviewed by Kevin.

        Moved string debug function outside of the WebCore namespace so that it's
        easier to call in the debugger. I will never doubt Darin again (x100).

        * platform/String.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoGet rid of the -50000 nonsense on scrollbars when they're first added to overflow...
hyatt [Mon, 31 Jul 2006 21:06:15 +0000 (21:06 +0000)]
Get rid of the -50000 nonsense on scrollbars when they're first added to overflow blocks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Partial fix for 5453. Improve overflow so that when it changes
hyatt [Mon, 31 Jul 2006 20:59:48 +0000 (20:59 +0000)]
    Partial fix for 5453.  Improve overflow so that when it changes
        dynamically from scroll/auto to hidden that we properly hide the
        scrollbars.  Also refine the dirty dashboard region updating so that
        it only happens if the scrollbars genuinely come and go.

        Also eliminate the extra repaint of the entire block.  This should not be
        necessary, since the scrollbars repaint themselves already and children
        will repaint already if they move.

        Reviewed by darin

        Adding test case from bug into fast/overflow/dynamic-hidden.html

        * rendering/RenderLayer.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoAdd layout test for bug 5453.
hyatt [Mon, 31 Jul 2006 20:58:28 +0000 (20:58 +0000)]
Add layout test for bug 5453.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Mon, 31 Jul 2006 19:49:10 +0000 (19:49 +0000)]

        Reviewed by justin

        Empty style spans created in applyInlineStyle
        REGRESSION: Blot and Mail both do a very poor job of pasting the main www.apple.com page

        * editing/pasteboard/paste-4039777-fix-expected.txt:
        * editing/pasteboard/testcase-9507-expected.checksum: Added.
        * editing/pasteboard/testcase-9507-expected.png: Added.
        * editing/pasteboard/testcase-9507-expected.txt: Added.
        * editing/pasteboard/testcase-9507.html: Added.
        * editing/unsupported-content/table-delete-001-expected.txt:
        * editing/selection/selectNode-expected.png:
        * editing/selection/selectNode-expected.checksum:
        * editing/selection/selectNode-expected.txt:


        Reviewed by justin

        Empty style spans created in applyInlineStyle

        Improves paste fidelity because some of these empty font/style spans had a non-zero
        size and were messing up the layout of pasted content:
        REGRESSION: Blot and Mail both do a very poor job of pasting the main www.apple.com page

        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::applyInlineStyle): Use the adjusted start node instead
        of start.node().  Don't do any application if the endpoints are swapped.  Adjust
        endNode if the start node is a descendant of it, so that the pre-order traversal will
        terminate properly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Tim Hatcher.
darin [Mon, 31 Jul 2006 19:14:26 +0000 (19:14 +0000)]
    Reviewed by Tim Hatcher.

        - http://bugzilla.opendarwin.org/show_bug.cgi?id=10168
          add a first cut at a Metrics pane to the inspector

        * WebInspector/webInspector/inspector.css: Add styles for the new metrics pane.
        * WebInspector/webInspector/inspector.html: Add the new metrics pane, starting with
        the table to show the box model.
        * WebInspector/webInspector/inspector.js: Add the new metrics pane. Add back some
        "title" attributes so we have more tooltips. Removed the optional parameter to

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
ggaren [Mon, 31 Jul 2006 18:44:35 +0000 (18:44 +0000)]
    Reviewed by Darin.

        * platform/String.cpp:
        (WebCore::string): Added a debugging function to create a String from a
        const char*.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoUpdate ChangeLog with bug link.
thatcher [Mon, 31 Jul 2006 18:21:03 +0000 (18:21 +0000)]
Update ChangeLog with bug link.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-31 Anders Carlsson <acarlsson@apple.com>
andersca [Mon, 31 Jul 2006 16:13:34 +0000 (16:13 +0000)]
2006-07-31  Anders Carlsson  <acarlsson@apple.com>

        Reviewed by John.

        * Plugins/WebPluginDatabase.m:
        (-[WebPluginDatabase refresh]):
        Create a mutable set instead of a mutable array.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-31 Anders Carlsson <acarlsson@apple.com>
andersca [Mon, 31 Jul 2006 15:43:14 +0000 (15:43 +0000)]
2006-07-31  Anders Carlsson  <acarlsson@apple.com>

        * css/CSSPrimitiveValue.cpp:
        Fix windows build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reinstating my last change. Everything builds now.
ggaren [Mon, 31 Jul 2006 08:44:46 +0000 (08:44 +0000)]
    Reinstating my last change. Everything builds now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Mon, 31 Jul 2006 08:41:52 +0000 (08:41 +0000)]

        Reviewed by Darin.

        Bug 10167: REGRESSION(r15688): ASSERTION FAILED: _private->mouseDownEvent != nil in layout tests

        * fast/events/objc-event-api-expected.txt: Update test results.


        Reviewed by Darin.

        Bug 10167: REGRESSION(r15688): ASSERTION FAILED: _private->mouseDownEvent != nil in layout tests

        * DumpRenderTree/EventSendingController.m: Don't set the mouse down flag
        in keyDown handler.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Backing out my last change because it causes a build failure in some
ggaren [Mon, 31 Jul 2006 08:31:37 +0000 (08:31 +0000)]
    Backing out my last change because it causes a build failure in some
        configurations. This seems to be the cleanest way to get svn to cooperate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago * editing/pasteboard/paste-table-002-expected.txt: Corrected one result
darin [Mon, 31 Jul 2006 07:26:14 +0000 (07:26 +0000)]
    * editing/pasteboard/paste-table-002-expected.txt: Corrected one result
        affected by my recent change to getComputedStyle.

        * editing/pasteboard/paste-table-002.html: Converted to a text test, since
        it dumps the serialized form of hte result.
        * editing/pasteboard/paste-table-002-expected.checksum: Removed.
        * editing/pasteboard/paste-table-002-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Maciej.
ggaren [Mon, 31 Jul 2006 07:17:51 +0000 (07:17 +0000)]
    Reviewed by Maciej.

        Moved files around to make WebCore's structure clearer
        and easier to explain.

        DerivedSources/JS* -> bindings/js/DerivedSources/ (so JS* files
        show up when you search inside bindings/js/, but are marked, at
        the same time, as files that you should not edit manually)

        WebCore+SVG/DOMList.h -> ksvg2/svg/ (only files inside ksvg2/svg/
        use DOMList.h)

        icon -> loader/icon

        I tested that everything still builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Tim Hatcher.
darin [Mon, 31 Jul 2006 07:07:12 +0000 (07:07 +0000)]
    Reviewed by Tim Hatcher.

        * WebInspector/webInspector/inspector.js: Fix bug where a null property value
        leads to an empty style pane.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Mon, 31 Jul 2006 06:52:06 +0000 (06:52 +0000)]

        - test that displays all computed style -- demonstrates bug fixes I just
          made to the computed style class and also acts as a regression test for
          the existence of all of them and the basic format for a default object

        * fast/css/computed-style-expected.txt: Added.
        * fast/css/computed-style.html: Added.


        Reviewed by Tim Hatcher.

        - some improvements for the benefit of the style pane of the inspector

        * css/CSSComputedStyleDeclaration.cpp: Removed background-position
        and border-spacing from the list of properties that show up in
        computed style, because of background-position-x, background-position-y,
        -webkit-border-horizontal-spacing and -webkit-border-vertical-spacing.
        (WebCore::valueForLength): Added handling for undefinedLength, intrinsic,
        and min-intrinsic.
        (WebCore::primitiveValueFromLength): Removed code that would add a
        space to the string for no good reason.
        (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue): Added a
        special case for a line clamp of -1, which should come back as "none"
        rather than an actual "-1".
        (WebCore::CSSComputedStyleDeclaration::length): Return 0 if the
        declaration has no corresponding node or no renderer.
        (WebCore::CSSComputedStyleDeclaration::item): Check against length()
        so that the two stay consistent rather than using a constant.

        * css/CSSPrimitiveValue.cpp:
        (WebCore::isCSSTokenizerIdentifier): Added.
        (WebCore::isCSSTokenizerURL): Added.
        (WebCore::quoteString): Added.
        (WebCore::quoteStringIfNeeded): Changed to quote strings in many more
        cases -- any cases where they would not parse in the CSS parser otherwise.
        The main case this affects is font names with spaces in them.
        (WebCore::quoteURLIfNeeded): Added.
        (WebCore::CSSPrimitiveValue::cssText): Use quoteURLIfNeeded in the case
        where we're making the text form of a URI.

        * css/tokenizer.flex: Whitespace tweaks to line things up better.


        Reviewed by Tim Hatcher.

        - some improvements for the inspector

        * WebInspector/WebInspector.m:
        (+[WebInspector sharedWebInspector:]): Fixed bug that could cause the inspector
        to be garbage collected if used in an application with GC enabled.
        (-[WebInspector dealloc]): Removed a call to a non-existent close method.
        (-[WebInspector window]): Added a custom WebPreferences object and called
        setPrivateBrowsingEnabled:YES so the inspector won't appear in the history menu.
        Also call setProhibitsMainFrameScrolling:YES to try to get rid of trouble where
        the inspector scrolls when dragging.

        * WebInspector/webInspector/inspector.css: Added style for the new color swatch,
        and JavaScript properties. More of the style should be shared between the panes,
        but this should be OK for now.

        * WebInspector/webInspector/inspector.html: Added a first cut at a JavaScript
        properties pane. Needs work, but better than nothing.

        * WebInspector/webInspector/inspector.js: Lots of improvements:
        - Omit "typical" property values from computed style display, making it much shorter.
        - Use the words "black", "white", and "transparent" when appropriate for color values.
        - Refactored the loaded() function to get rid of repetitive scrollbar setup.
        - Added a new scrollarea for the JavaScript properties pane.
        - Simplified refreshScrollbars() -- we now refresh all scrollbars every time, which does no harm.
        - Removed unused resultsWithXpathQuery().
        - Use [] instead of "new Array()" and {} instead of "new Object()".
        - Removed unused xpathForNode().
        - Changed style pane to display the style for a text node's parent instead of saying
          it can't display the style for text.
        - Fixed regression I caused a while back by checking the length of a computed style
          and not trying to display anything if its length is 0. Before this change and the
          corresponding change in WebCore, we'd see a complete list of all styles with the
          empty string as the value for each one.
        - Changed the name of the computedStyle flag on the style rules array to isComputedStyle
          to make it easier to understand it's a boolean.
        - Fixed an error in the code that does !important scanning where it was trying to
          do a special case for computed style, but was checking the computed style flag on
          the wrong object.
        - Added populateStyleListItem() function to factor out things in common between the
          items in the top level list and the expanded tree for shorthand properties.
        - Added code to make a color swatch next to the textual representation for any
          property that contains a color.
        - Implemented a first cut at a simple JavaScript properties pane.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Maciej.
thatcher [Mon, 31 Jul 2006 06:15:05 +0000 (06:15 +0000)]
    Reviewed by Maciej.

        Bug 9686: [Drosera] Need the ability to break into Drosera on Javascript exceptions

        * Drosera/DebuggerDocument.m:
        (-[WebScriptObject switchToServerNamed:]): Remove as listener before resuming to
        ensure that the server does not try and notify us of events related to the resumption.
        (-[WebScriptObject webView:exceptionWasRaised:sourceId:line:forWebFrame:]): Delegate
        call through to JavaScript.
        * Drosera/debugger.js: Pause debugger when exception is raised.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Beth.
ggaren [Sun, 30 Jul 2006 23:02:09 +0000 (23:02 +0000)]
    Reviewed by Beth.

        - Tweaked the checkout and build instructions to make them simpler and

        * building/build.html:
        * building/checkout.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-30 Eric Seidel <eric@eseidel.com>
eseidel [Sun, 30 Jul 2006 04:31:53 +0000 (04:31 +0000)]
2006-07-30  Eric Seidel  <eric@eseidel.com>

        Reviewed by gramps!

        Remove some unused SVG code.

        * ksvg2/ecma/Ecma.cpp: Removed.
        * ksvg2/ecma/Ecma.h: Removed.
        * ksvg2/ecma/GlobalObject.cpp: remove a couple dead functions.
        * ksvg2/ecma/GlobalObject.h: add a comment about this dead code.
        * ksvg2/ecma/SVGLookup.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoJavaScriptCore:
darin [Sun, 30 Jul 2006 00:04:22 +0000 (00:04 +0000)]

        - Removed tabs from these source files that still had them.
          We don't use them; that way source files look fine in editors
          that have tabs set to 8 spaces or to 4 spaces.
        - Removed allow-tabs Subversion property from the files too.

        * bindings/NP_jsobject.cpp:
        * bindings/c/c_utility.cpp:
        * bindings/jni/jni_runtime.cpp:
        * bindings/jni/jni_utility.cpp:
        * bindings/objc/objc_utility.mm:
        * bindings/runtime.cpp:
        * bindings/runtime_method.cpp:
        * bindings/testbindings.cpp:
        * bindings/testbindings.mm:
        * kjs/date_object.cpp:
        * kjs/function.cpp:
        * kjs/list.cpp:
        * kjs/nodes.cpp:
        * kjs/nodes.h:
        * kjs/string_object.cpp:
        * kjs/ustring.cpp:


        - Removed tabs from these source files that still had them.
          We don't use them; that way source files look fine in editors
          that have tabs set to 8 spaces or to 4 spaces.
        - Removed allow-tabs Subversion property from the files too.

        * bindings/objc/WebScriptObject.h:
        * bindings/objc/WebScriptObject.mm:
        * platform/FontData.cpp:


        - Removed tabs from these source files that still had them.
          We don't use them; that way source files look fine in editors
          that have tabs set to 8 spaces or to 4 spaces.
        - Removed allow-tabs Subversion property from the files too.

        * DefaultDelegates/WebDefaultPolicyDelegate.m:
        * History/WebHistory.m:
        * Misc/WebDownload.m:
        * Misc/WebIconDatabase.m:
        * Misc/WebKitErrors.m:
        * Misc/WebKitLogging.m:
        * Misc/WebNSDataExtras.m:
        * Misc/WebNSFileManagerExtras.m:
        * Panels/WebPanelAuthenticationHandler.m:
        * Plugins/WebBaseNetscapePluginView.m:
        * Plugins/npfunctions.h:
        * WebCoreSupport/WebSubresourceLoader.m:
        * WebView/WebMainResourceLoader.m:
        * WebView/WebView.h:
        * WebView/WebView.m:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago * tests/mozilla/expected.html: Update test results now that regress-185165.js
darin [Sat, 29 Jul 2006 23:41:40 +0000 (23:41 +0000)]
    * tests/mozilla/expected.html: Update test results now that regress-185165.js
        is succeeding. I suspect Anders fix for bug 4620655 is the reason.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoJavaScriptCore:
weinig [Sat, 29 Jul 2006 23:15:25 +0000 (23:15 +0000)]

        Reviewed by Darin.

        - patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=10080
          Adopt pedantic changes from the Unity project to improve
          cross-compiler compatibility

          Changes include:
          * Removing trailing semicolon from namespace braces.
          * Removing trailing comma from last enum declaration.
          * Updating to match style guidelines.
          * Adding missing newline to the end of the file.
          * Turning on gcc warning for missing newline at the end of a source file
            (GCC_WARN_ABOUT_MISSING_NEWLINE in Xcode, -Wnewline in gcc).
          * Alphabetical sorting of Xcode source list files.
          * Replace use of non-portable variable-size array with Vector.
          * Use C-style comments instead of C++ comments in files that might
            be included by either C or C++ files.

        * API/JSCallbackConstructor.cpp:
        * API/JSCallbackFunction.cpp:
        * API/JSCallbackObject.cpp:
        * JavaScriptCore.xcodeproj/project.pbxproj:
        * JavaScriptCorePrefix.h:
        * bindings/jni/jni_class.cpp:
        * bindings/jni/jni_class.h:
        * bindings/jni/jni_instance.cpp:
        * bindings/jni/jni_objc.mm:
        * bindings/jni/jni_runtime.cpp:
        * bindings/jni/jni_runtime.h:
        * bindings/jni/jni_utility.h:
        * bindings/objc/objc_instance.h:
        * bindings/runtime_array.h:
        * kjs/collector.h:
        * kjs/config.h:
        * kjs/ustring.cpp:
        * wtf/Platform.h:


        Reviewed by Darin.

        - patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=10080
          Adopt pedantic changes from the Unity project to improve
          cross-compiler compatibility

          Changes include:
          * Turning on gcc warning for missing newline at the end of a source file
            (GCC_WARN_ABOUT_MISSING_NEWLINE in Xcode, -Wnewline in gcc).

        * JavaScriptGlue.xcodeproj/project.pbxproj:


        Reviewed by Darin.

        - patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=10080
          Adopt pedantic changes from the Unity project to improve
          cross-compiler compatibility

          Changes include:
          * Removing trailing semicolon from namespace braces.
          * Removing trailing comma from last enum declaration.
          * Updating to match style guidelines.
          * Adding missing newline to the end of the file.
          * Turning on gcc warning for missing newline at the end of a source file
            (GCC_WARN_ABOUT_MISSING_NEWLINE in Xcode, -Wnewline in gcc).
          * Alphabetical sorting of Xcode source list files.
          * Use abs() function from <math.h> instead of ABS() macro.
          * Use C-style comments instead of C++ comments in files that might
            be included by either C or C++ files.
          * Use -numeric_limits<double>::infinity() instead of -HUGE_VAL.

        * WebCore+SVG/DOMList.h:
        * WebCore.xcodeproj/project.pbxproj:
        * WebCorePrefix.h:
        * bindings/js/JSCanvasRenderingContext2DCustom.cpp:
        * bindings/js/JSXSLTProcessor.h:
        * bindings/js/kjs_domnode.h:
        * bindings/objc/DOMCSS.h:
        (-[DOMCSSValue enum]):
        * bindings/objc/DOMCore.h:
        (-[DOMImplementation createDocument:::]):
        * bindings/objc/DOMEvents.h:
        (-[DOMOverflowEvent enum]):
        * bindings/objc/DOMRange.h:
        * bindings/objc/DOMTraversal.h:
        * bindings/objc/DOMXPath.h:
        (-[DOMXPathNSResolver lookupNamespaceURI:]):
        * bridge/mac/WebCoreFrameBridge.h:
        * bridge/mac/WebCoreKeyboardAccess.h:
        * css/makeprop:
        * css/makevalues:
        * dom/ChildNodeList.h:
        * dom/DocPtr.h:
        * dom/Document.h:
        * dom/DocumentMarker.h:
        * dom/EventTargetNode.h:
        * dom/KeyboardEvent.h:
        * dom/NameNodeList.h:
        * dom/QualifiedName.cpp:
        * editing/TypingCommand.h:
        * editing/UnlinkCommand.h:
        * html/FormDataList.h:
        * html/HTMLBlockquoteElement.h:
        * html/HTMLDivElement.h:
        * html/HTMLFormElement.h:
        * html/HTMLHRElement.h:
        * html/HTMLHeadingElement.h:
        * html/HTMLMarqueeElement.h:
        * html/HTMLParagraphElement.h:
        * html/HTMLPlugInElement.h:
        * html/HTMLPreElement.h:
        * html/HTMLTokenizer.h:
        * icon/IconDatabase.cpp:
        * icon/SQLStatement.cpp:
        * kcanvas/KCanvasFilters.h:
        * kcanvas/RenderSVGImage.h:
        * kcanvas/device/quartz/KRenderingDeviceQuartz.h:
        * ksvg2/css/SVGRenderStyle.h:
        * ksvg2/css/SVGRenderStyleDefs.h:
        * ksvg2/events/SVGZoomEvent.h:
        * ksvg2/ksvg.h:
        * ksvg2/misc/KCanvasRenderingStyle.h:
        * ksvg2/misc/SVGImageLoader.h:
        * ksvg2/scripts/make_names.pl:
        * ksvg2/svg/SVGAElement.h:
        * ksvg2/svg/SVGAngle.h:
        * ksvg2/svg/SVGAnimateColorElement.h:
        * ksvg2/svg/SVGAnimateElement.h:
        * ksvg2/svg/SVGAnimateTransformElement.h:
        * ksvg2/svg/SVGAnimatedAngle.h:
        * ksvg2/svg/SVGAnimatedBoolean.h:
        * ksvg2/svg/SVGAnimatedColor.h:
        * ksvg2/svg/SVGAnimatedEnumeration.h:
        * ksvg2/svg/SVGAnimatedInteger.h:
        * ksvg2/svg/SVGAnimatedLength.h:
        * ksvg2/svg/SVGAnimatedLengthList.h:
        * ksvg2/svg/SVGAnimatedNumber.h:
        * ksvg2/svg/SVGAnimatedNumberList.h:
        * ksvg2/svg/SVGAnimatedPathData.h:
        * ksvg2/svg/SVGAnimatedPoints.h:
        * ksvg2/svg/SVGAnimatedPreserveAspectRatio.h:
        * ksvg2/svg/SVGAnimatedRect.h:
        * ksvg2/svg/SVGAnimatedString.h:
        * ksvg2/svg/SVGAnimatedTemplate.h:
        * ksvg2/svg/SVGAnimatedTransformList.h:
        * ksvg2/svg/SVGAnimationElement.cpp:
        * ksvg2/svg/SVGAnimationElement.h:
        * ksvg2/svg/SVGCircleElement.h:
        * ksvg2/svg/SVGClipPathElement.h:
        * ksvg2/svg/SVGColor.h:
        * ksvg2/svg/SVGComponentTransferFunctionElement.h:
        * ksvg2/svg/SVGCursorElement.h:
        * ksvg2/svg/SVGDOMImplementation.h:
        * ksvg2/svg/SVGDefsElement.h:
        * ksvg2/svg/SVGDescElement.h:
        * ksvg2/svg/SVGDocument.h:
        * ksvg2/svg/SVGElement.h:
        * ksvg2/svg/SVGElementInstance.h:
        * ksvg2/svg/SVGElementInstanceList.h:
        * ksvg2/svg/SVGEllipseElement.h:
        * ksvg2/svg/SVGExternalResourcesRequired.h:
        * ksvg2/svg/SVGFEBlendElement.h:
        * ksvg2/svg/SVGFEColorMatrixElement.h:
        * ksvg2/svg/SVGFEComponentTransferElement.h:
        * ksvg2/svg/SVGFECompositeElement.h:
        * ksvg2/svg/SVGFEDiffuseLightingElement.h:
        * ksvg2/svg/SVGFEDisplacementMapElement.h:
        * ksvg2/svg/SVGFEDistantLightElement.h:
        * ksvg2/svg/SVGFEFloodElement.h:
        * ksvg2/svg/SVGFEFuncAElement.h:
        * ksvg2/svg/SVGFEFuncBElement.h:
        * ksvg2/svg/SVGFEFuncGElement.h:
        * ksvg2/svg/SVGFEFuncRElement.h:
        * ksvg2/svg/SVGFEGaussianBlurElement.h:
        * ksvg2/svg/SVGFEImageElement.h:
        * ksvg2/svg/SVGFELightElement.h:
        * ksvg2/svg/SVGFEMergeElement.h:
        * ksvg2/svg/SVGFEMergeNodeElement.h:
        * ksvg2/svg/SVGFEOffsetElement.h:
        * ksvg2/svg/SVGFEPointLightElement.h:
        * ksvg2/svg/SVGFESpecularLightingElement.h:
        * ksvg2/svg/SVGFESpotLightElement.h:
        * ksvg2/svg/SVGFETileElement.h:
        * ksvg2/svg/SVGFETurbulenceElement.h:
        * ksvg2/svg/SVGFilterElement.h:
        * ksvg2/svg/SVGFilterPrimitiveStandardAttributes.h:
        * ksvg2/svg/SVGFitToViewBox.h:
        * ksvg2/svg/SVGForeignObjectElement.cpp:
        * ksvg2/svg/SVGForeignObjectElement.h:
        * ksvg2/svg/SVGGElement.h:
        * ksvg2/svg/SVGGradientElement.h:
        * ksvg2/svg/SVGHelper.h:
        * ksvg2/svg/SVGImageElement.h:
        * ksvg2/svg/SVGLangSpace.h:
        * ksvg2/svg/SVGLength.h:
        * ksvg2/svg/SVGLengthList.h:
        * ksvg2/svg/SVGLineElement.h:
        * ksvg2/svg/SVGLinearGradientElement.h:
        * ksvg2/svg/SVGList.h:
        * ksvg2/svg/SVGLocatable.h:
        * ksvg2/svg/SVGMarkerElement.h:
        * ksvg2/svg/SVGMaskElement.h:
        * ksvg2/svg/SVGMatrix.h:
        * ksvg2/svg/SVGNumber.h:
        * ksvg2/svg/SVGNumberList.h:
        * ksvg2/svg/SVGPaint.h:
        * ksvg2/svg/SVGPathElement.h:
        * ksvg2/svg/SVGPathSeg.h:
        * ksvg2/svg/SVGPathSegArc.h:
        * ksvg2/svg/SVGPathSegClosePath.h:
        * ksvg2/svg/SVGPathSegCurvetoCubic.h:
        * ksvg2/svg/SVGPathSegCurvetoCubicSmooth.h:
        * ksvg2/svg/SVGPathSegCurvetoQuadratic.h:
        * ksvg2/svg/SVGPathSegCurvetoQuadraticSmooth.h:
        * ksvg2/svg/SVGPathSegLineto.h:
        * ksvg2/svg/SVGPathSegLinetoHorizontal.h:
        * ksvg2/svg/SVGPathSegLinetoVertical.h:
        * ksvg2/svg/SVGPathSegList.h:
        * ksvg2/svg/SVGPathSegMoveto.h:
        * ksvg2/svg/SVGPatternElement.h:
        * ksvg2/svg/SVGPoint.h:
        * ksvg2/svg/SVGPointList.h:
        * ksvg2/svg/SVGPolyElement.h:
        * ksvg2/svg/SVGPolygonElement.h:
        * ksvg2/svg/SVGPolylineElement.h:
        * ksvg2/svg/SVGPreserveAspectRatio.h:
        * ksvg2/svg/SVGRadialGradientElement.h:
        * ksvg2/svg/SVGRect.h:
        * ksvg2/svg/SVGRectElement.h:
        * ksvg2/svg/SVGSVGElement.h:
        * ksvg2/svg/SVGScriptElement.h:
        * ksvg2/svg/SVGSetElement.h:
        * ksvg2/svg/SVGStopElement.h:
        * ksvg2/svg/SVGStringList.h:
        * ksvg2/svg/SVGStylable.h:
        * ksvg2/svg/SVGStyleElement.h:
        * ksvg2/svg/SVGStyledElement.h:
        * ksvg2/svg/SVGStyledLocatableElement.h:
        * ksvg2/svg/SVGStyledTransformableElement.h:
        * ksvg2/svg/SVGSwitchElement.h:
        * ksvg2/svg/SVGSymbolElement.h:
        * ksvg2/svg/SVGTRefElement.h:
        * ksvg2/svg/SVGTSpanElement.h:
        * ksvg2/svg/SVGTests.h:
        * ksvg2/svg/SVGTextContentElement.h:
        * ksvg2/svg/SVGTextElement.h:
        * ksvg2/svg/SVGTextPositioningElement.h:
        * ksvg2/svg/SVGTitleElement.h:
        * ksvg2/svg/SVGTransform.h:
        * ksvg2/svg/SVGTransformList.h:
        * ksvg2/svg/SVGTransformable.h:
        * ksvg2/svg/SVGURIReference.h:
        * ksvg2/svg/SVGUseElement.h:
        * ksvg2/svg/SVGViewElement.h:
        * ksvg2/svg/SVGZoomAndPan.h:
        * ksvg2/svg/svgpathparser.h:
        * page/Frame.h:
        * platform/AffineTransform.h:
        * platform/FontCache.cpp:
        * platform/FontData.cpp:
        * platform/FontData.h:
        * platform/TextBox.h:
        * platform/Timer.cpp:
        * platform/mac/FontCacheMac.mm:
        * platform/mac/GlyphMapMac.cpp:
        * platform/mac/WebFontCache.mm:
        * rendering/DeprecatedRenderSelect.cpp:
        * rendering/EllipsisBox.h:
        * rendering/RenderBR.h:
        * rendering/RenderBlock.h:
        * rendering/RenderFlexibleBox.h:
        * rendering/RenderFlow.h:
        * rendering/RenderFrame.cpp:
        * rendering/bidi.h:
        * rendering/break_lines.cpp:


        Reviewed by Darin.

        - patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=10080
          Adopt pedantic changes from the Unity project to improve
          cross-compiler compatibility

          Changes include:
          * Adding missing newline to the end of the file.
          * Turning on gcc warning for missing newline at the end of a source file
            (GCC_WARN_ABOUT_MISSING_NEWLINE in Xcode, -Wnewline in gcc).

        * WebKit.xcodeproj/project.pbxproj:
        * WebView/WebResourcePrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoClear stray executable bits.
darin [Sat, 29 Jul 2006 23:09:11 +0000 (23:09 +0000)]
Clear stray executable bits.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoClear some stray executable bits.
darin [Sat, 29 Jul 2006 22:50:14 +0000 (22:50 +0000)]
Clear some stray executable bits.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoRemove the output produced by an alert, so
rwlbuis [Sat, 29 Jul 2006 21:05:40 +0000 (21:05 +0000)]
Remove the output produced by an alert, so
hopefully buildbot will not complain anymore :)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Geoff.
rwlbuis [Sat, 29 Jul 2006 20:25:31 +0000 (20:25 +0000)]
    Reviewed by Geoff.

        ghostmap XHTML + SVG fails due to javascript undefined exception

        Make sure document.URL works when called from a non-HTML document.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoJavaScriptCore:
darin [Sat, 29 Jul 2006 15:44:20 +0000 (15:44 +0000)]

        Reviewed by Darin.

        - fixes for Linux build

        * JavaScriptCoreSources.bkl: Added new files to build, kjs/PropertyNameArray.cpp
        and kjs/testkjs.cpp, and removed old files.


        Reviewed and tweaked by Darin.

        - fixes for Linux build
          (still might not compile, since Darin tweaked but was unable to test).

        * WebCoreSources.bkl: Added some new sources, removed some obsolete ones.
        * Projects/gdk/webcore-gdk.bkl: Added RenderPopupMenuGdk.cpp

        * html/HTMLPlugInElement.h: Removed incorrect syntax with class name
        before ::.

        * platform/gdk/FrameGdk.h:
        * platform/gdk/FrameGdk.cpp: Updated parameters of constructor. Moved a couple stubs
        out of this file.
        * platform/gdk/RenderPopupMenuGdk.h: Added.
        * platform/gdk/RenderPopupMenuGdk.cpp: Added. Just a file full of empty stubs.
        * platform/gdk/RenderThemeGdk.h:
        * platform/gdk/RenderThemeGdk.cpp: (WebCore::RenderThemeGdk::createPopupMenu): Added.

        * platform/gdk/TemporaryLinkStubs.cpp: Add some new stubs.


        Reviewed by Darin.

        - fixes for Linux build

        * GdkLauncher/mk: Added call to bakefile_gen before calling make.
        * GdkLauncher/gdklauncher.bkl: Tweak comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago - changed this test to a plain-text one and added text explaining what it...
darin [Sat, 29 Jul 2006 15:43:11 +0000 (15:43 +0000)]
    - changed this test to a plain-text one and added text explaining what it tests

        * fast/forms/input-named-action-overrides-action-attribute.html: Added explanation text,
        got rid of some incorrect HTML syntax, added a call to dumpAsText(), and used a paragraph
        element instead of a span element for the result.
        * fast/forms/input-named-action-overrides-action-attribute-expected.txt: Updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
darin [Sat, 29 Jul 2006 15:21:41 +0000 (15:21 +0000)]
    Reviewed by Darin.

        Bug 10022: REGRESSION: Crash in WebCore::XMLTokenizer::characters

        * dom/XMLTokenizer.cpp:
        (WebCore::XMLTokenizer::resumeParsing): If the write() call resulted in new callbacks being
        added, don't call through to end() just yet.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoWebCore:
darin [Sat, 29 Jul 2006 15:19:13 +0000 (15:19 +0000)]

        Reviewed by John Sullivan.

        - manual test for http://bugzilla.opendarwin.org/show_bug.cgi?id=9984
          ASSERTION FAILURE: _private->mouseDownEvent != nil
          (WebKit/WebView/WebHTMLView.m:4863 -[WebHTMLView(WebInternal) _delegateDragSourceActionMask])

        * manual-tests/subview-click-assertion.html: Added.


        Reviewed by John Sullivan.

         - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9984
          ASSERTION FAILURE: _private->mouseDownEvent != nil
          (WebKit/WebView/WebHTMLView.m:4863 -[WebHTMLView(WebInternal) _delegateDragSourceActionMask])

        * WebView/WebHTMLView.m:
        (-[WebHTMLView _setMouseDownEvent:]): Moved into the WebHTMLViewFileInternal category and changed
        to accept nil.
        (-[WebHTMLView _startDraggingImage:at:operation:event:sourceIsDHTML:DHTMLWroteData:]):
        Copy the hit HTMLView's mouse down event to the top HTMLView.
        (-[WebHTMLView acceptsFirstMouse:]): Added a call to _setMouseDownEvent:nil before returning.
        (-[WebHTMLView shouldDelayWindowOrderingForEvent:]): Added a call to _setMouseDownEvent:nil
        before returning.
        (-[WebHTMLView mouseUp:]): Added a call to _setMouseDownEvent:nil to clear the event set in
        mouseDown: (and used during dragging).
        (-[WebHTMLView _delegateDragSourceActionMask]): Copy the hit HTMLView's mouse down event to
        the top HTMLView.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
darin [Sat, 29 Jul 2006 15:06:04 +0000 (15:06 +0000)]

        Reviewed by Darin.

        Test for http://bugzilla.opendarwin.org/show_bug.cgi?id=8587 / <rdar://problem/4631844>
        REGRESSION: {display:list-item} items outside an ol or ul element don't number correctly

        * fast/lists/ordered-list-with-no-ol-tag-expected.checksum: Added.
        * fast/lists/ordered-list-with-no-ol-tag-expected.png: Added.
        * fast/lists/ordered-list-with-no-ol-tag-expected.txt: Added.
        * fast/lists/ordered-list-with-no-ol-tag.html: Added.


        Reviewed by Darin.

        Fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=8587 / <rdar://problem/4631844>
        REGRESSION: {display:list-item} items outside an ol or ul element don't number correctly

        * rendering/RenderListItem.cpp:
        (WebCore::enclosingList): If no real list element is found, use parent instead.
        (WebCore::previousListItem): Check list elements to see if they are list items -- the
        old code assumed that a list would never be a list item but this is entirely possible
        with CSS and especially with the new rule that means any DOM element could act as a list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
ap [Sat, 29 Jul 2006 14:56:42 +0000 (14:56 +0000)]
    Reviewed by Darin.

        Fix http://bugzilla.opendarwin.org/show_bug.cgi?id=10147
        REGRESSION: custom attribute values set via javascript are not persistent

        Test: fast/dom/Element/setAttribute-case-insensitivity.html

        * dom/Element.cpp:
        (WebCore::Element::setAttribute): Use the lowercased localName for retrieving the old value, too.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago - rolled out the test for bug 9753, since we rolled out the fix
darin [Sat, 29 Jul 2006 14:53:40 +0000 (14:53 +0000)]
    - rolled out the test for bug 9753, since we rolled out the fix
          we'll bring back the test when we bring back the fix

        * svg/custom/width-full-percentage-expected.checksum: Removed.
        * svg/custom/width-full-percentage-expected.png: Removed.
        * svg/custom/width-full-percentage-expected.txt: Removed.
        * svg/custom/width-full-percentage.svg: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Suggested by Darin.
thatcher [Sat, 29 Jul 2006 00:25:56 +0000 (00:25 +0000)]
    Suggested by Darin.

        Moving the implementation of isContentEditable to the
        the WebPrivate category.

        * bindings/objc/DOM.mm:
        (-[DOMNode isContentEditable]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Sat, 29 Jul 2006 00:24:21 +0000 (00:24 +0000)]

        Reviewed by levi

        REGRESSION: Blot and Mail both do a very poor job of pasting the main www.apple.com page

        Removed incorrect nesting:
        * editing/pasteboard/paste-table-001-expected.checksum:
        * editing/pasteboard/paste-table-001-expected.png:
        * editing/pasteboard/paste-table-001-expected.txt:
        * editing/pasteboard/paste-table-003-expected.txt:


        Reviewed by levi

        First part of:
        REGRESSION: Blot and Mail both do a very poor job of pasting the main www.apple.com page

        Migrate to isBlock and enclosingBlock.
        Changed RendereringInfo::isBlock and ReplacementFragment::isBlock
        to wasBlock so that they don't conflict with isBlock and because
        it's more descriptive.

        * editing/ReplaceSelectionCommand.cpp:
        * editing/ReplaceSelectionCommand.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Fri, 28 Jul 2006 22:01:54 +0000 (22:01 +0000)]

        Reviewed by levi

        Poor paste fidelity of http://www.google.com/

        Removed one of the two extraneous newlines that are introduced on paste:
        * editing/pasteboard/paste-table-003-expected.checksum:
        * editing/pasteboard/paste-table-003-expected.png:
        * editing/pasteboard/paste-table-003-expected.txt:

        Fewer pastes remove nodes containing the selection b/c of
        the changes to interchange newlines:
        * editing/pasteboard/paste-line-endings-007-expected.txt:
        * editing/pasteboard/paste-line-endings-008-expected.txt:
        * editing/pasteboard/paste-line-endings-009-expected.txt:

        * editing/pasteboard/4242293-1-expected.checksum: Added.
        * editing/pasteboard/4242293-1-expected.png: Added.
        * editing/pasteboard/4242293-1-expected.txt: Added.
        * editing/pasteboard/4242293-1.html: Added.
        * editing/pasteboard/4242293-expected.checksum: Added.
        * editing/pasteboard/4242293-expected.png: Added.
        * editing/pasteboard/4242293-expected.txt: Added.
        * editing/pasteboard/4242293.html: Added.


        Reviewed by levi

        More changes for:
        Poor paste fidelity of http://www.google.com/

        * editing/markup.cpp:
        (WebCore::needInterchangeNewlineAfter): Added.
        Only add an interchangeNewline if we're annotating for interchange.  Can't test
        this until we expose createMarkup in non-annotate mode (I'd expose Range::toHTML).
        Remove the FIXME about prepending the interchangeNewline before creating the rest
        of the markup.  Its correct to surround the interchangeNewline in ancestors of the
        adjusted startNode.
        Fixed the check to see if an interchangeNewline is needed (!inSameBlock).  This would
        add an interchangeNewline for a selection that started before a table.
        Removed the workaround for the bug where markup for a table was not included when
        selecting the contents of a table, since this change fixes the general problem of
        markup for ancestors of the startNode being left out.
        Don't leave out ancestors of the startNode as we leave their subtrees.  Execute the
        code to include them even if we closed ancestors in the ancestorsToClose list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by John.
thatcher [Fri, 28 Jul 2006 19:59:37 +0000 (19:59 +0000)]
    Reviewed by John.

        <rdar://problem/4657473> REGRESSION: Spell check not available from contextual menu in Mail

        The context menu code should be checking isContentEditable
        on DOMNode not just DOMElement. This is needed because DOMText
        will be the node class of any text that is clicked.

        * DefaultDelegates/WebDefaultContextMenuDelegate.m:
        (-[WebDefaultUIDelegate webView:contextMenuItemsForElement:defaultMenuItems:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoWebCore:
thatcher [Fri, 28 Jul 2006 19:58:30 +0000 (19:58 +0000)]

        Reviewed by John.

        Adding isContentEditable to DOMNode's private category.

        * bindings/objc/DOMPrivate.h:


        Reviewed by John.

        The context menu code should be checking isContentEditable
        on DOMNode not just DOMElement. This is needed because DOMText
        will be the node class of any text that is clicked.

        * DefaultDelegates/WebDefaultContextMenuDelegate.m:
        (-[WebDefaultUIDelegate webView:contextMenuItemsForElement:defaultMenuItems:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Rolling out http://bugzilla.opendarwin.org/show_bug.cgi?id=6010 because of failin...
ap [Fri, 28 Jul 2006 19:19:07 +0000 (19:19 +0000)]
Rolling out bugzilla.opendarwin.org/show_bug.cgi?id=6010 because of failing layout tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Rolling out http://bugzilla.opendarwin.org/show_bug.cgi?id=9753 because of...
adele [Fri, 28 Jul 2006 19:01:18 +0000 (19:01 +0000)]
    Rolling out bugzilla.opendarwin.org/show_bug.cgi?id=9753 because of failing layout tests.
        SVG with width and height 100% leaves room for scrollbar on the right.

        * kcanvas/RenderPath.cpp:
        * kcanvas/RenderPath.h:
        * kcanvas/RenderSVGContainer.cpp:
        * ksvg2/svg/SVGCircleElement.cpp:
        * ksvg2/svg/SVGCircleElement.h:
        * ksvg2/svg/SVGEllipseElement.cpp:
        * ksvg2/svg/SVGEllipseElement.h:
        * ksvg2/svg/SVGLineElement.cpp:
        * ksvg2/svg/SVGLineElement.h:
        * ksvg2/svg/SVGRectElement.cpp:
        * ksvg2/svg/SVGRectElement.h:
        * ksvg2/svg/SVGStyledElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-27 Eric Seidel <eric@eseidel.com>
eseidel [Fri, 28 Jul 2006 18:07:35 +0000 (18:07 +0000)]
2006-07-27  Eric Seidel  <eric@eseidel.com>

        Reviewed by darin.

        Add better SVGLoad event support.
        There will still need to be additional support added for <link> and <script> elements.
        Not all of the error -> loaded -> error transition cases work yet.

        * dom/EventTargetNode.cpp:
        * html/HTMLImageLoader.cpp:
        * html/HTMLImageLoader.h:
        * ksvg2/misc/SVGImageLoader.cpp:
        * ksvg2/misc/SVGImageLoader.h:
        * ksvg2/svg/SVGAElement.cpp:
        * ksvg2/svg/SVGAnimatedBoolean.h:
        * ksvg2/svg/SVGElement.cpp:
        * ksvg2/svg/SVGElement.h:
        * ksvg2/svg/SVGExternalResourcesRequired.h:
        * ksvg2/svg/SVGImageElement.cpp:
        * ksvg2/svg/SVGImageElement.h:
        * ksvg2/svg/SVGSVGElement.cpp:
        * ksvg2/svg/SVGScriptElement.cpp:
        * xml/xmlhttprequest.cpp:
        (WebCore::XMLHttpRequest::callReadyStateChangeListener): use new Event() instead of createEvent

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Anders.
adele [Fri, 28 Jul 2006 05:13:24 +0000 (05:13 +0000)]
    Reviewed by Anders.

        - http://bugzilla.opendarwin.org/show_bug.cgi?id=10135
          Cleanup: remove useless parameters from RenderObject::position()

        No test possible (no functionality change)

        * rendering/RenderBox.cpp:
        * rendering/RenderBox.h:
        * rendering/RenderObject.h:
        * rendering/RenderTableRow.h:
        * rendering/RenderTableSection.h:
        * rendering/RenderText.cpp:
        * rendering/RenderText.h:
        * rendering/bidi.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
adele [Fri, 28 Jul 2006 04:53:11 +0000 (04:53 +0000)]

        Reviewed by Maciej.

        Test case for
        SVG with width and height 100% leaves room for scrollbar on the right.

        * svg/custom/width-full-percentage-expected.txt: Added.
        * svg/custom/width-full-percentage-expected.checksum: Added.
        * svg/custom/width-full-percentage-expected.png: Added.
        * svg/custom/width-full-percentage.svg: Added.


        Reviewed by Maciej.

        SVG with width and height 100% leaves room for scrollbar on the right.

        Relayout children of svg containers, except svg shapes that do not depend on percentages.

        * kcanvas/RenderPath.cpp: (WebCore::RenderPath::hasPercentageValues):
        * kcanvas/RenderPath.h:
        * kcanvas/RenderSVGContainer.cpp: (WebCore::RenderSVGContainer::layout):
        * ksvg2/svg/SVGCircleElement.cpp: (SVGCircleElement::hasPercentageValues):
        * ksvg2/svg/SVGCircleElement.h:
        * ksvg2/svg/SVGEllipseElement.cpp: (WebCore::SVGEllipseElement::hasPercentageValues):
        * ksvg2/svg/SVGEllipseElement.h:
        * ksvg2/svg/SVGLineElement.cpp: (SVGLineElement::hasPercentageValues):
        * ksvg2/svg/SVGLineElement.h:
        * ksvg2/svg/SVGRectElement.cpp: (WebCore::SVGRectElement::hasPercentageValues):
        * ksvg2/svg/SVGRectElement.h:
        * ksvg2/svg/SVGStyledElement.h: (WebCore::SVGStyledElement::hasPercentageValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Thu, 27 Jul 2006 23:50:51 +0000 (23:50 +0000)]

        Reviewed by levi

        Poor paste fidelity of http://www.google.com/

        * editing/pasteboard/styled-element-markup-expected.checksum: Added.
        * editing/pasteboard/styled-element-markup-expected.png: Added.
        * editing/pasteboard/styled-element-markup-expected.txt: Added.
        * editing/pasteboard/styled-element-markup.html: Added.


        Reviewed by levi

        Poor paste fidelity of http://www.google.com/

        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::applyInlineStyle):  Fixed crasher with nil-check.
        * editing/markup.cpp:
        (WebCore::startMarkup): Style information wasn't added to an element if it had
        attributes but no style attribute.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-27 Alexander Kellett <lypanov@kde.org>
lypanov [Thu, 27 Jul 2006 21:06:13 +0000 (21:06 +0000)]
2006-07-27  Alexander Kellett  <lypanov@kde.org>

        Reviewed by Darin.

        - fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=9867
          REGRESSION: code that modifies form element in Firefox modifies form attribute in WebKit

        * bindings/scripts/CodeGeneratorJS.pm: add new attribute - HasOverridingNameGetter
        * html/HTMLFormElement.idl:

2006-07-27  Alexander Kellett  <lypanov@kde.org>

        Reviewed by Darin.

        Test for: http://bugzilla.opendarwin.org/show_bug.cgi?id=9867
        REGRESSION: code that modifies form element in Firefox modifies form attribute in WebKit

        * fast/forms/input-named-action-overrides-action-attribute-expected.txt: Added.
        * fast/forms/input-named-action-overrides-action-attribute.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Anders.
rwlbuis [Thu, 27 Jul 2006 20:18:49 +0000 (20:18 +0000)]
    Reviewed by Anders.

        Some HTML DOM attributes have incorrect types

        Adjust idl and dom source code for HTMLImageElement and
        HTMLBaseFontElement to choose correct type for
        HTMLImageElement::border and HTMLBaseFontElement::size

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Thu, 27 Jul 2006 20:01:56 +0000 (20:01 +0000)]

        Reviewed by levi

        Safari crashes in CompositeEditCommand::removeBlockPlaceholder entering text in a web page form

        * fast/forms/4628409-expected.txt: Added.
        * fast/forms/4628409.html: Added.


        Reviewed by levi

        Safari crashes in CompositeEditCommand::removeBlockPlaceholder entering text in a web page form

        * editing/CompositeEditCommand.cpp:
        A script on the page shrinks the textarea down to zero rows just
        before insertion.  That gives the shadow div a zero height, removing
        [div, 0] as a candidate for a VisiblePosition.  Then
        InsertTextCommand::input() tries to create a VisiblePosition from
        [div, 0] and fails.  A nil check is a quick fix.  Another might
        be to make positions where there is a caret selection candidates.
        Another might be to make any [div, 0] where div is a shadow div inside
        a form element a candidate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Fix by Graham Dennis, reviewed by Darin.
ap [Thu, 27 Jul 2006 18:58:16 +0000 (18:58 +0000)]
    Fix by Graham Dennis, reviewed by Darin.

        - fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=10094
          Japanese characters improperly rendering in TOT

        Test: fast/encoding/denormalised-voiced-japanese-chars.html

        * platform/Font.cpp:
        (WebCore::WidthIterator::advance): pass currentCharacter to normalizeVoicingMarks
        (WebCore::WidthIterator::normalizeVoicingMarks): turn the currentCharacter variable
        into an argument as WidthIterator::advance doesn't update m_currentCharacter while

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-27 Eric Seidel <eric@eseidel.com>
eseidel [Thu, 27 Jul 2006 09:22:16 +0000 (09:22 +0000)]
2006-07-27  Eric Seidel  <eric@eseidel.com>

        Reviewed by ap.

        Complex entities seem to fail on TOT
        Use an alternative work-around to an XML Entity parsing bug.

        * dom/XMLTokenizer.cpp:
        (WebCore::hackAroundLibXMLEntityBug): added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
rwlbuis [Thu, 27 Jul 2006 07:46:00 +0000 (07:46 +0000)]
    Reviewed by Darin.

        Fix viewBox issues in Hixie's test

        Do a more accurate and quicker job of parsing the viewBox
        attribute. Handle properly on negative width/height and
        invalid syntax in general.

        * ksvg2/svg/SVGFitToViewBox.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago2006-07-27 Eric Seidel <eric@eseidel.com>
eseidel [Thu, 27 Jul 2006 05:26:00 +0000 (05:26 +0000)]
2006-07-27  Eric Seidel  <eric@eseidel.com>

        Reviewed by ap.

        Enables getSVGDocument() support for HTMLObjectElement and HTMLEmbedElement
        This patch does not try to autogenerate anything, but enables the SVG functionality.
        Also fix HTMLObjectElement vspace and hspace to be numbers instead of strings to match IE & Firefox.

        * bindings/js/kjs_dom.cpp:
        * bindings/js/kjs_html.cpp:
        * bindings/js/kjs_html.h:
        * dom/Document.h:
        * html/HTMLEmbedElement.cpp:
        * html/HTMLEmbedElement.h:
        * html/HTMLObjectElement.cpp:
        * html/HTMLObjectElement.h:
        * ksvg2/svg/SVGDocument.h:
        * rendering/RenderWidget.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Darin.
ap [Thu, 27 Jul 2006 04:31:29 +0000 (04:31 +0000)]
    Reviewed by Darin.

        Fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9901
        XSLTProcessor cannot parse text nodes for INPUT ELEMENTS in an XSLT stylesheet

        Test: fast/js/xhtml-serialize.html

        * editing/markup.cpp:
        (WebCore::endMarkup): If the element has child nodes, write a closing tag even if it
        is not permitted in HTML, to make it valid XML.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by John Sullivan (and Kevin and Timo).
aliceli1 [Thu, 27 Jul 2006 01:43:55 +0000 (01:43 +0000)]
    Reviewed by John Sullivan (and Kevin and Timo).

        landing this patch for John et al.
        Fixed <rdar://problem/4651380> REGRESSION (9A232->9A234): existing web clips don't render if you restart Dashboard

        * page/FrameView.cpp:
        moved the call to didFirstLayout to above the check for needsLayout
        because we need to call didFirstLayout for every webview, not just
        ones we think need layout.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Thu, 27 Jul 2006 00:42:44 +0000 (00:42 +0000)]

        Reviewed by darin

        REGRESSION: Can't change size or style of any text in message body

        * editing/style/designmode-expected.checksum: Added.
        * editing/style/designmode-expected.png: Added.
        * editing/style/designmode-expected.txt: Added.
        * editing/style/designmode.html: Added.


        Reviewed by darin

        REGRESSION: Can't change size or style of any text in message body

        * dom/Node.cpp:
        (WebCore::Node::isContentRichlyEditable): Mimic Node::isContentEditable.
        * html/HTMLElement.cpp:
        (WebCore::HTMLElement::isContentRichlyEditable): First check
        the frame for editability, then userModify, like what
        HTMLElement::isContentEditable() does.
        * html/HTMLElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Fix by Mitz Pettel, reviewed by ap.
ap [Wed, 26 Jul 2006 19:33:46 +0000 (19:33 +0000)]
    Fix by Mitz Pettel, reviewed by ap.

        - fix debug-only information() string for table cells

        * rendering/RenderObject.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by Maciej, inspired by John.
ggaren [Wed, 26 Jul 2006 03:04:51 +0000 (03:04 +0000)]
    Reviewed by Maciej, inspired by John.

        - Fixed <rdar://problem/4651931> 1% REGRESSION on iBench HTML due to
        repeated requests for non-existent favicon

        An optimization to avoid serializing favicon data for missing icons had stomped an
        optimization to avoid GETing a missing favicon more than once. The solution
        is a happy marriage of optimizations, ensuring that we *retain* the missing
        favicon's "i am missing" data without posting a notification or saving it to disk.

        * Misc/WebIconDatabase.m:
        (-[WebIconDatabase _setIconURL:forURL:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoWebCore:
harrison [Wed, 26 Jul 2006 00:30:06 +0000 (00:30 +0000)]

        Reviewed by timo and Darin.

        <rdar://problem/4618584> "Paste and Match Style" is not working in Mail (add SPI)

        * WebCore.xcodeproj/project.pbxproj:
        * bridge/mac/WebCoreFrameBridge.h:
        * bridge/mac/WebCoreFrameBridge.mm:
        (-[WebCoreFrameBridge replaceSelectionWithNode:selectReplacement:smartReplace:matchStyle:]):
        Add matchStyle parameter for use by WebKit.


        Reviewed by timo and Darin.

        <rdar://problem/4618584> "Paste and Match Style" is not working in Mail (add SPI)

        * WebKit.xcodeproj/project.pbxproj:
        * WebView/WebView.m:
        (-[WebView replaceSelectionWithNode:]):
        (-[WebView _replaceSelectionWithNode:matchStyle:]):
        * WebView/WebViewPrivate.h:
        (-[WebView _replaceSelectionWithNode:matchStyle::]):
        New SPI that is same as replaceSelectionWithNode: with added parameter whether to match existing style.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15633 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoLayoutTests:
justing [Tue, 25 Jul 2006 23:13:20 +0000 (23:13 +0000)]

        Reviewed by adele

        REGRESSION: Can't Select All when the body is contentEditable="true"

        * editing/selection/select-all-005-expected.checksum: Added.
        * editing/selection/select-all-005-expected.png: Added.
        * editing/selection/select-all-005-expected.txt: Added.
        * editing/selection/select-all-005.html: Added.
        * editing/selection/select-all-006-expected.checksum: Added.
        * editing/selection/select-all-006-expected.png: Added.
        * editing/selection/select-all-006-expected.txt: Added.
        * editing/selection/select-all-006.html: Added.


        Reviewed by adele

        REGRESSION: Can't Select All when the body is contentEditable="true"

        * editing/VisiblePosition.cpp:
        (WebCore::VisiblePosition::canonicalPosition): Allow descent from a non-editable html
        element into an editable body as a convenience.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15627 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years ago Reviewed by darin
justing [Tue, 25 Jul 2006 21:06:25 +0000 (21:06 +0000)]
    Reviewed by darin

        <rdar://problem/4468458> Incoming HTML message is editable <body contenteditable="true">

        * dom/Document.cpp:
        (WebCore::Document::inDesignMode): Removed some spaces.
        * page/Frame.cpp:
        (WebCore::Frame::applyEditingStyleToElement): Just add CSS properties needed for editing.
        (WebCore::Frame::removeEditingStyleFromElement): Made this a no-op, we'll remove
        it eventually.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15624 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoWebCore:
ddkilzer [Tue, 25 Jul 2006 15:35:24 +0000 (15:35 +0000)]

        Reviewed by Darin.

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9717
          Replace IDL ConvertUndefinedToTrue parameter attribute with Optional attribute

        No test added because there is no change in functionality.

        * bindings/js/kjs_binding.cpp: Removed valueToBooleanTreatUndefinedAsTrue() method.
        * bindings/js/kjs_binding.h: Ditto.
        * bindings/scripts/CodeGeneratorJS.pm: Removed support for ConvertUndefinedToTrue parameter attribute.
        * dom/Element.h: Set default value to true for scrollIntoView() and scrollIntoViewIfNeeded() arguments.
        * dom/Element.idl: Replaced ConvertUndefinedToTrue attribute with Optional attribute.
          Fixed name of scrollIntoViewIfNeeded() argument to match implementation.
        * page/DOMWindow.idl: Replaced ConvertUndefinedToTrue attribute with Optional attribute.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15619 268f45cc-cd09-0410-ab3c-d52691b4dbfc

16 years agoJavaScriptCore:
darin [Tue, 25 Jul 2006 05:05:09 +0000 (05:05 +0000)]

        Reviewed and tweaked a bit by Darin.

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9902
          jsNull and NSNull not properly converted between JS and ObjC

        * bindings/objc/objc_utility.mm: (KJS::Bindings::convertObjcValueToValue):
        Added case for converting NSNull to jsNull.


        Reviewed and tweaked a bit by Darin.

        - added some new tests that use "echo" including one for JavaScript null

        * plugins/bindings-test-expected.txt: Added results.
        * plugins/bindings-test.html: Added tests.


        Reviewed and tweaked a bunch by Darin.

        Test: plugins/bindings-test.html

        - fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9902
          jsNull and NSNull not properly converted between JS and ObjC

        * bindings/objc/WebScriptObject.mm:
        (+[WebScriptObject _convertValueToObjcValue:originExecutionContext:executionContext:]):
        Added case that converts jsNull to NSNull.


        Reviewed and tweaked a bit by Darin.

        * DumpRenderTree/ObjCPlugin.m:
        (+[ObjCPlugin isSelectorExcludedFromWebScript:]): Added "echo:".
        (+[ObjCPlugin webScriptNameForSelector:]): Use the name "echo" for
        "echo:" so it's nice to call from JavaScript.
        (-[ObjCPlugin echo:]): Just returns the same object -- can be used
        to test a round trip through Objective-C types.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15618 268f45cc-cd09-0410-ab3c-d52691b4dbfc