1. 30 Jan, 2007 4 commits
  2. 29 Jan, 2007 33 commits
    • ggaren's avatar
      I am the chosen one! I have seen the faces of the final five! I will · 0309d898
      ggaren authored
              save the WebKit build!
              * loader/ResourceLoader.cpp:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19247 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • ap's avatar
      Reviewed by Darin. · 1f0642c1
      ap authored
              Fix XPath test suite conversion problems
              * dom/svg/level3/xpath/svgunit.js: Fixed createXPathEvaluator() to work in Firefox (whose
              document.getFeature returns null for XPath). Changed load() to use XMLHttpRequest, instead of
              merely cloning the current document.
              * dom/svg/level3/xpath/resources/internaldtd.xml: Added.
              * dom/svg/level3/xpath/resources/staff.dtd: Added.
              * dom/svg/level3/xpath/resources/staff.xml: Added.
              * dom/svg/level3/xpath/resources/staffNS.dtd: Added.
              * dom/svg/level3/xpath/resources/staffNS.xml: Added.
              Added external test documents.
              * dom/svg/level3/xpath/XPathEvaluator_evaluate_WRONG_DOCUMENT_ERR-expected.txt:
              * dom/svg/level3/xpath/XPathExpression_evaluate_WRONG_DOCUMENT_ERR-expected.txt:
              These tests now run, but fail. They also fail in Firefox, but pass in Opera.
              * dom/svg/level3/xpath/XPath...
    • ggaren's avatar
      JavaScriptCore: · b76a6c20
      ggaren authored
              Reviewed by Maciej Stachowiak.
              Fixed <rdar://problem/4485644> REGRESSION: JavaScriptCore has init routines
              The TCMalloc module now initializes, if needed, inside GetCache() and 
              fastMallocSetIsMultiThreaded(). We leverage the same synchronization 
              technique used for enabling / disabling the single-threaded optimization 
              to synchronize initialization of the library without requiring a lock 
              for every malloc.
              1,251 runs of tcmalloc_unittest, 2 runs of a custom, massively multi-threaded 
              tcmalloc_unittest, and my custom version of the PLT show no regressions.
              Super-accurate JS iBench reports a .24% regression, which is right at the
              limit of its error range, so I'm declaring victory.
              * wtf/FastMalloc.cpp:
              (WTF::fastMallocSetIsMultiThreaded): Initialize, if needed. (InitModule()
              checks the "if needed" part.)
              (WTF::TCMalloc_ThreadCache::GetCache): Restored o...
    • gdennis's avatar
      LayoutTests: · 1b1375a2
      gdennis authored
              Reviewed by Maciej.
              - test for http://bugs.webkit.org/show_bug.cgi?id=10725
                Image data in from RTFD clipboard data thrown away
              * editing/pasteboard/paste-RTFD-expected.checksum: Added.
              * editing/pasteboard/paste-RTFD-expected.png: Added.
              * editing/pasteboard/paste-RTFD-expected.txt: Added.
              * editing/pasteboard/paste-RTFD.html: Added.
              * editing/pasteboard/paste-TIFF-expected.checksum: Added.
              * editing/pasteboard/paste-TIFF-expected.png: Added.
              * editing/pasteboard/paste-TIFF-expected.txt: Added.
              * editing/pasteboard/paste-TIFF.html: Added.
              * editing/resources/RTFD-pasteboard-data.dat: Added.
              * editing/resources/TIFF-pasteboard-data.dat: Added.
              Reviewed by Maciej.
              Part of fix for http://bugs.webkit.org/show_bug.cgi?id=10725
              Image data in from RTFD clipboard data thrown away
              ResourceHandles no longer need to be able to defer load...
    • bdash's avatar
      2007-01-29 Jim Correia <jim.correia@pobox.com> · 78827b4a
      bdash authored
              Reviewed by Mark.
              Added support for -allowsUndo/-setAllowsUndo: to allow editable WebView
              clients to completely disable undo registration. This is functionally
              equivalent to the methods with the same names on NSTextView.
              * WebView/WebView.mm:
              (-[WebViewPrivate init]):
              (-[WebView initWithCoder:]):
              (-[WebView encodeWithCoder:]):
              (-[WebView allowsUndo]):
              (-[WebView setAllowsUndo:]):
              (-[WebView undoManager]):
              * WebView/WebViewPrivate.h:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19243 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • beidson's avatar
      Use a RetainPtr instead of a raw pointer · ded5c60b
      beidson authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19242 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • justing's avatar
      LayoutTests: · bce32c2f
      justing authored
              Reviewed by harrison
              ASSERTION FAILURE calling SelectionController::setBaseAndExtent (11833)
              * editing/selection/4960137-expected.checksum: Added.
              * editing/selection/4960137-expected.png: Added.
              * editing/selection/4960137-expected.txt: Added.
              * editing/selection/4960137.html: Added.
              Reviewed by harrison
              ASSERTION FAILURE calling SelectionController::setBaseAndExtent (11833)
              * editing/SelectionController.cpp:
              (WebCore::SelectionController::setSelection): Hand off the selection
              to the selectionController responsible for it, instead of asserting.
              A selection object can't be constructed with endpoints in different 
              documents, so this code is only necessary for one endpoint.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19239 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • adachan's avatar
      2007-01-29 Ada Chan <adachan@apple.com> · 942b524d
      adachan authored
              Reviewed by Brady.
              Moved the update of the title of the current entry in the backforward list to WebCore.
              * loader/FrameLoader.cpp:
              * WebCoreSupport/WebFrameLoaderClient.mm:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19238 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • adele's avatar
      WebCore: · 91175f31
      adele authored
              Reviewed by Darin.
              More preparation for event handling fixes.
              * bridge/EditorClient.h: Removed EventTargetNode parameter from handleKeyPress since we can just get that from the event.
              * dom/EventTargetNode.cpp: (WebCore::EventTargetNode::defaultEventHandler): ditto.
              * page/EventHandler.cpp: (WebCore::EventHandler::defaultKeyboardEventHandler): ditto.
              * page/EventHandler.h: ditto.
              * platform/graphics/svg/SVGImageEmptyClients.h: (WebCore::SVGEmptyEditorClient::handleKeyPress): ditto.
              * editing/Editor.h:
              * editing/Editor.cpp:
              (WebCore::Editor::handleKeyPress): Removed EventTargetNode parameter.
              (WebCore::Editor::execCommand): Made commandMap use AtomicStringImpls instead of Strings.
              (WebCore::Editor::insertText): Moved from the bridge.
              * page/mac/WebCoreFrameBridge.h: Moved insertText method to Editor.
              * page/mac/WebCoreFrameBridge.mm:
              * WebCore.exp: Updated exports.
              Reviewed by Darin.
              More preparation for event handling fixes.
              * WebCoreSupport/WebEditorClient.h: Removed EventTargetNode parameter, since you can
                just get this from the KeyboardEvent.
              * WebCoreSupport/WebEditorClient.mm: (WebEditorClient::handleKeyPress): ditto.
              * WebView/WebHTMLViewInternal.h: Added interpretKeyEventsParameters struct.
              * WebView/WebViewInternal.h: Changed parameter from NSEvent to WebCoreKeyboardEvent in _interceptEditingKeyEvent.
              * WebView/WebHTMLView.mm:
              (-[WebHTMLView yankAndSelect:]): Updated for new triggeringEvent parameter.
              (-[WebHTMLView _interceptEditingKeyEvent:]): Set the WebHTMLViewInterpretKeyEventsParameters.
              (-[WebHTMLView doCommandBySelector:]): Access WebHTMLViewInterpretKeyEventsParameters.
              (-[WebHTMLView insertText:]): ditto.
              (-[WebHTMLView _insertText:selectInsertedText:triggeringEvent:]): Added parameter for triggeringEvent.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19237 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • andersca's avatar
      LayoutTests: · ddd35425
      andersca authored
              Reviewed by Geoff.
              Assertion failure in -[WebCoreFrameBridge installInFrame:]
              * fast/parser/xhtml-document-with-html-object-expected.txt: Added.
              * fast/parser/xhtml-document-with-html-object.xhtml: Added.
              Reviewed by Geoff.
              Assertion failure in -[WebCoreFrameBridge installInFrame:]
              The problem was that the XML tokenizer didn't create elements with 
              createdByParser set to true.
              * dom/Document.cpp:
              New method which takes a QName. This is used by both the XML tokenizer and createElementNS.
              Call new createElement method.
              * dom/XMLTokenizer.h:        
              * dom/XMLTokenizer.cpp:
              Use an OwnPtr for the pending callbacks object. Also, use String instead of StringImpl in the hash map.
              Call createElement.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19236 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • beidson's avatar
      Reviewed by Oliver · 65b67427
      beidson authored
              Only set the iconURL port for non-zero ports
              * loader/FrameLoader.cpp:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19235 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • kmccullo's avatar
      JavaScriptCore: · 7401f7ae
      kmccullo authored
              Reviewed by Geoff and Oliver.
              - rdar://problem/4955561
              - missusing JavaScript shouldn't crash webkit.  Now it doesn't, in this case.
              * bindings/objc/objc_runtime.mm:
              * bindings/runtime_method.cpp:
              * bindings/runtime_object.cpp:
              Reviewed by Geoff and Oliver.
              - rdar://problem/4955561
              - missusing JavaScript shouldn't crash webkit.  Now it doesn't in this case.
              * plugins/call-as-function-test-expected.txt: Added.
              * plugins/call-as-function-test.html: Added.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19234 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • oliver's avatar
      2007-01-29 Oliver Hunt <oliver@apple.com> · f4707497
      oliver authored
              build fix
              * WebView/WebHTMLView.mm:
              (-[WebHTMLView _startDraggingImage:at:operation:event:sourceIsDHTML:DHTMLWroteData:]):
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19233 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • sullivan's avatar
      * platform/network/mac/ResourceRequestMac.mm: · cd5f82ba
      sullivan authored
              Build fix -- removed #import of no-longer-extant WebDataProtocol.h (and also one of the
              two #imports for WebCoreSystemInterface.h, while I was at it)
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19232 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • andersca's avatar
      Reviewed by Oliver. · 67c57781
      andersca authored
              * page/DragController.cpp:
              Only try a drag operation if the drag client explicitly specified that it is OK
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19231 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • oliver's avatar
      2007-01-25 Oliver Hunt <oliver@apple.com> · e71315c1
      oliver authored
              Reviewed by Adam.
              Migrated drag state and logic to WebCore, removed superfluous methods
              * ChangeLog:
              * WebCoreSupport/WebDragClient.h:
              * WebCoreSupport/WebDragClient.mm:
              * WebCoreSupport/WebFrameBridge.mm:
                 allowDHTMLDrag move to WebCore::EventHandler
              * WebView/WebHTMLView.mm:
              (-[WebHTMLView _startDraggingImage:at:operation:event:sourceIsDHTML:DHTMLWroteData:]):
              (-[WebHTMLView draggedImage:movedTo:]):
              (-[WebHTMLView draggedImage:endedAt:operation:]):
                dragOffset and dragSourecAction is now stored in WebCore::DragController
                migrated _delegateDragSourceActionMask to WebCore::DragController
              * WebView/WebHTMLViewInternal.h:
                Removed dragOffset declaration, migrated to WebCore::DragController
              * WebView/WebView.mm:
                removed unnecessary method, _loadingDragOperationForDraggingInfo                       
              Reviewed by Adam.
              Stub for new DragClient method
              * WebCoreSupport/DragClientQt.cpp:
              * WebCoreSupport/DragClientQt.h:                  
              Reviewed by Adam.
              Migrated more drag logic and state to WebCore
              * page/DragClient.h:       
              * page/DragController.cpp:
                New forwarding method for the UI delegate
              * page/DragController.h:
                Now store the drag offset -- the offset of the cursor relative to the drag image)
                in DragController
              * page/EventHandler.cpp:
                Moved from WebFrameBridge
              * page/EventHandler.h:
              * page/mac/EventHandlerMac.mm:
                EventHandlerMac now uses EventHandler::allowDHTMLDrag rather than the 
                implementation in WebFrameBridge
              * page/mac/WebCoreFrameBridge.h:
                removed definition of allowDHTMLDrag from WebFrameBridge
              * platform/graphics/svg/SVGImageEmptyClients.h:
                Updated empty DragClient
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19230 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • ap's avatar
      Reviewed by Darin. · fd9f8312
      ap authored
              XPathEvaluator may return some nodes more than once in a result set
              Test: fast/xpath/nodeset-duplicates.html
              * xml/XPathPath.cpp:
              (WebCore::XPath::LocationPath::doEvaluate): Ensure uniqueness of elements
              in the node-set.
              * xml/XPathPredicate.cpp:
              (WebCore::XPath::Union::doEvaluate): Fixed a uniqueness algorithm that was
              already present here. Added a FIXME about incorrect result ordering.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19227 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • ap's avatar
      Reviewed by Darin. · af0e1445
      ap authored
              XPathEvaluator behavior does not match Firefox - name() and attribute nodes
              * xml/XPathFunctions.cpp:
              (WebCore::XPath::FunName::doEvaluate): Do not try to append a prefix if it doesn't exist.
              * fast/xpath/name-null-namespace-expected.txt: Added.
              * fast/xpath/name-null-namespace.html: Added.
              Test for the bug.
              * fast/dom/xpath-empty-string-expected.txt: Removed.
              * fast/dom/xpath-empty-string.html: Removed.
              Moved these to fast/xpath.
              * fast/xpath/xpath-namespaces.html: Fixed an inline document to be well-formed, so the test now passes in Firefox.
              This is really a quirk of our DOMParser that the test passed in WebKit: on error, a descriptive message is appended to
              the partially parsed document, instead of replacing it.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19226 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • ap's avatar
      Reviewed by Darin. · 6eb5af47
      ap authored
              XPathEvaluator behavior does not match Firefox - substring() and empty element
              Test: fast/xpath/empty-string-substring.html
              * xml/XPathValue.cpp:
              * xml/XPathValue.h:
              Added a constructor taking const char*, so that string literals and arrays don't 
              get converted to boolean values.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19225 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • zack's avatar
      Cosmetic fixes - don't resize the widget in the constructor · 6d5f76ba
      zack authored
      (it was a hack from the initial port) and fix the paint
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19224 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • mjs's avatar
      - change from struct to class to make picky compilers happy · d9116de7
      mjs authored
              * loader/SubstituteData.h:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19223 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • bdash's avatar
      2007-01-30 Mark Rowe <mrowe@apple.com> · 638e3d0a
      bdash authored
              Reviewed by Maciej.
              Bug 12458: Crash in W3C-SVG-1.1/animate-elem-09-t.svg running layout tests under guard malloc
              * ksvg2/svg/SVGAnimationElement.cpp: 
              (WebCore::parseValues): ptr either points to a ';' or off the end of the buffer.  We're
              interested in what comes before the ';', so start there instead.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19222 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • zack's avatar
      Take the scrollbars into considation when computing · 3b817005
      zack authored
      the contents size of the main frame
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19221 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • mjs's avatar
      - fix includes · e52bab7a
      mjs authored
              * loader/SubstituteData.h:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19220 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • mjs's avatar
      - added new file that I forgot in the last commit · 32bef010
      mjs authored
              * loader/SubstituteData.h: Added.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19219 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • mjs's avatar
      WebCore: · 2d326f5c
      mjs authored
              Reviewed by Mark.
              - made data loading cross-platform
              - fixed <rdar://problem/4910106>
              * WebCore.exp:
              * WebCore.xcodeproj/project.pbxproj:
              * loader/CachedResource.cpp:
              * loader/CachedResource.h:
              * loader/DocLoader.cpp:
              * loader/DocumentLoader.cpp:
              * loader/DocumentLoader.h:
              * loader/FrameLoader.cpp:
              * loader/FrameLoader.h:
              * loader/FrameLoaderClient.h:
              * loader/MainResourceLoader.cpp:
              * loader/MainResourceLoader.h:
              * loader/mac/DocumentLoaderMac.mm:
              * loader/mac/WebDataProtocol.h: Removed.
              * loader/mac/WebDataProtocol.mm: Removed.
              * loader/qt/DocumentLoaderQt.cpp:
              * platform/graphics/svg/SVGImageEmptyClients.h:
              Reviewed by Mark.
              - updated for cross-platform data loading support
              * WebCoreSupport/WebFrameLoaderClient.h:
              * WebCoreSupport/WebFrameLoaderClient.mm:
              * WebView/WebDataSource.mm:
              (-[WebDataSource initWithRequest:]):
              * WebView/WebDocumentLoaderMac.h:
              * WebView/WebDocumentLoaderMac.mm:
              * WebView/WebFrame.mm:
              (-[WebFrame _loadData:MIMEType:textEncodingName:baseURL:unreachableURL:]):
              (-[WebFrame loadData:MIMEType:textEncodingName:baseURL:]):
              (-[WebFrame _loadHTMLString:baseURL:unreachableURL:]):
              (-[WebFrame loadArchive:]):
              * WebView/WebFrameInternal.h:
              * WebView/WebHTMLView.mm:
              (-[WebHTMLView _documentFragmentFromPasteboard:inContext:allowPlainText:chosePlainText:]):
              Reviewed by Mark.
              - updated for cross-platform data loading support
              * WebCoreSupport/FrameLoaderClientQt.cpp:
              * WebCoreSupport/FrameLoaderClientQt.h:
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19218 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • zack's avatar
      Order the build files alphabetically instead of · d5d0c607
      zack authored
      the random mess we had before
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19217 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • zack's avatar
      Use the platform independent scrollbar code. · 2fb3bc52
      zack authored
      Remove the redundant empty stubs and forward
      few missing platform scrollbar calls
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19216 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • staikos's avatar
      Make window resizing work · de769d58
      staikos authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19215 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • staikos's avatar
      Make popup windows work · 019150f8
      staikos authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19214 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • lars's avatar
      Reviewed by Zack · 80fc94e3
      lars authored
              Fix the way we handle native widgets (QWidget) inside
              WebCore. Now WebCore::Widget owns the QWidget in all
              cases. This is the only way to get well defined semantics
              for all teh native widgets we have (currently ScrollView
              and ScrollBar, but plugins will follow).
              This has the side effect that one cannot rely on a defined 
              lifetime of a QWebFrame when using the API. 
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19213 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • staikos's avatar
      Missed saving this file before the previous checkin. · 93b48bf8
      staikos authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19212 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • staikos's avatar
      Implement more functions for the Qt build · 089ddb4c
      staikos authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19211 268f45cc-cd09-0410-ab3c-d52691b4dbfc
  3. 28 Jan, 2007 3 commits
    • staikos's avatar
      Fix redirects with the Qt resource implementation. · 059e5aeb
      staikos authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19210 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • ggaren's avatar
      Reviewed by Maciej Stachowiak. · fda60011
      ggaren authored
              First step in fixing <rdar://problem/4485644> REGRESSION: JavaScriptCore 
              has init routines
              Don't rely on a static initializer to store the main thread's ID (which
              we would use to detect allocations on secondary threads). Instead, require 
              the caller to notify fastMalloc if it might allocate on a secondary thread.
              Also fixed what seemed like a race condition in do_malloc.
              tcmalloc_unittest and my custom versions of JS iBench and PLT show no
              * wtf/FastMalloc.cpp:
              (1) Renamed from "fastMallocRegisterThread", which was a misleading name because 
              not all threads need to register with fastMalloc -- only secondary threads 
              need to, and only for the purpose of disabling its single-threaded optimization. 
              (2) Use the pageheap_lock instead of a custom one, since we need to synchronize
              with the read of isMultiThreaded inside CreateCacheIfNecessary. This is a new
              requirement, now that we can't guarantee that the first call to CreateCacheIfNecessary
              will occur on the main thread at init time, before any other threads have been created.
              (WTF::do_malloc): Reverted WTF change only to call GetCache() if size <= kMaxSize.
              The WTF code would read phinited without holding the pageheap_lock, which
              seemed like a race condition. Regardless, calling GetCache reduces the number 
              of code paths to module initialization, which will help in writing the 
              final fix for this bug.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19209 268f45cc-cd09-0410-ab3c-d52691b4dbfc
    • bdash's avatar
      2007-01-29 Mark Rowe <mrowe@apple.com> · e1d0d467
      bdash authored
              Reviewed by Brady.
              Bug 11085: REGRESSION: favicon.ico always looked for on port 80
              * loader/FrameLoader.cpp:
              (WebCore::FrameLoader::iconURL): Copy the port over to the favicon URL.
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@19208 268f45cc-cd09-0410-ab3c-d52691b4dbfc