1. 24 Jul, 2009 14 commits
    • abarth@webkit.org's avatar
      2009-07-24 Mike Fenton <mike.fenton@torchmobile.com> · 2def5eb9
      abarth@webkit.org authored
              Reviewed by Eric Seidel.
      
              Update WebCore/page/DOMTimer.cpp/h to conform to WebKit
              Style Guidelines as identified by cpplint.py.
              https://bugs.webkit.org/show_bug.cgi?id=27624
      
              * page/DragController.cpp:
              (WebCore::DragController::~DragController):
              (WebCore::documentFragmentFromDragData):
              (WebCore::DragController::dragEnded):
              (WebCore::DragController::dragEntered):
              (WebCore::DragController::dragExited):
              (WebCore::DragController::dragUpdated):
              (WebCore::DragController::performDrag):
              (WebCore::asFileInput):
              (WebCore::DragController::tryDocumentDrag):
              (WebCore::DragController::delegateDragSourceAction):
              (WebCore::DragController::concludeEditDrag):
              (WebCore::DragController::canProcessDrag):
              (WebCore::DragController::tryDHTMLDrag):
              (WebCore::DragController::mayStartDragAtEventLocation):
              (WebCore::getCachedImage):
              (WebCore::getImage):
              (WebCore::prepareClipboardForImageDrag):
              (WebCore::dragLocForDHTMLDrag):
              (WebCore::DragController::startDrag):
              (WebCore::DragController::doImageDrag):
              (WebCore::DragController::doSystemDrag):
              (WebCore::DragController::placeDragCaret):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46330 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2def5eb9
    • abarth@webkit.org's avatar
      2009-07-24 Mike Fenton <mike.fenton@torchmobile.com> · 97cc408e
      abarth@webkit.org authored
              Reviewed by Eric Seidel.
      
              Update WebCore/page/Chrome.cpp to conform to WebKit
              Style Guidelines as identified by cpplint.py.
              https://bugs.webkit.org/show_bug.cgi?id=27608
      
              * page/Chrome.cpp:
              (WebCore::Chrome::runBeforeUnloadConfirmPanel):
              (WebCore::Chrome::runJavaScriptAlert):
              (WebCore::Chrome::runJavaScriptConfirm):
              (WebCore::Chrome::runJavaScriptPrompt):
              (WebCore::Chrome::shouldInterruptJavaScript):
              (WebCore::Chrome::setToolTip):
              (WebCore::Chrome::requestGeolocationPermissionForFrame):
              (WebCore::ChromeClient::generateReplacementFile):
              (WebCore::ChromeClient::paintCustomScrollbar):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46329 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      97cc408e
    • abarth@webkit.org's avatar
      2009-07-24 Mike Fenton <mike.fenton@torchmobile.com> · bbf4d094
      abarth@webkit.org authored
              Reviewed by Eric Seidel.
      
              Update WebCore/page/Coordinates.cpp to conform to WebKit
              Style Guidelines as identified by cpplint.py.
              https://bugs.webkit.org/show_bug.cgi?id=27614
      
              * page/Coordinates.cpp:
              (WebCore::Coordinates::toString):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46328 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      bbf4d094
    • abarth@webkit.org's avatar
      2009-07-24 Mike Fenton <mike.fenton@torchmobile.com> · 807a99d3
      abarth@webkit.org authored
              Reviewed by Eric Seidel.
      
              Update WebCore/page/DOMSelection.cpp/h to conform to WebKit
              Style Guidelines as identified by cpplint.py.
              https://bugs.webkit.org/show_bug.cgi?id=27614
      
              * page/DOMSelection.cpp:
              (WebCore::DOMSelection::setBaseAndExtent):
              (WebCore::DOMSelection::modify):
              (WebCore::DOMSelection::addRange):
              (WebCore::DOMSelection::deleteFromDocument):
              * page/DOMSelection.h:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46327 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      807a99d3
    • eric@webkit.org's avatar
      2009-07-24 Eric Seidel <eric@webkit.org> · a4f0bad6
      eric@webkit.org authored
              Reviewed by Adam Barth.
      
              x-frame constructor calls are instantiated with the wrong prototype chain
              https://bugs.webkit.org/show_bug.cgi?id=27566
      
              Add the test.  Soon to be fixed by bug 27634.
      
              * fast/dom/constructed-objects-prototypes-expected.txt: Added.
              * fast/dom/constructed-objects-prototypes.html: Added.
              * fast/dom/resources/constructed-objects-prototypes.js: Added.
              (classNameForObject):
              (constructorPropertiesOnWindow):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46326 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a4f0bad6
    • mrowe@apple.com's avatar
      Roll out r46321 as it broke the Windows build. · 01af504c
      mrowe@apple.com authored
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46325 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      01af504c
    • abarth@webkit.org's avatar
      2009-07-24 Mike Fenton <mike.fenton@torchmobile.com> · 33c74813
      abarth@webkit.org authored
              Reviewed by Eric Seidel.
      
              Update WebCore/page/DOMTimer.cpp/h to conform to WebKit
              Style Guidelines as identified by cpplint.py.
              https://bugs.webkit.org/show_bug.cgi?id=27618
      
              * page/DOMTimer.cpp:
              (WebCore::DOMTimer::DOMTimer):
              (WebCore::DOMTimer::~DOMTimer):
              (WebCore::DOMTimer::fired):
              (WebCore::DOMTimer::suspend):
              (WebCore::DOMTimer::resume):
              (WebCore::DOMTimer::canSuspend):
              * page/DOMTimer.h:
              (WebCore::DOMTimer::minTimerInterval):
              (WebCore::DOMTimer::setMinTimerInterval):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46324 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      33c74813
    • mrowe@apple.com's avatar
      Add a command line option to run-javascriptcore-tests to explicitly opt out of... · 77962a87
      mrowe@apple.com authored
      Add a command line option to run-javascriptcore-tests to explicitly opt out of attempting to rebuild JavaScriptCore.
      
      Reviewed by Simon Hausmann.
      
      This is primarily useful to speed up the build bots where one machine builds JavaScriptCore and a second runs the tests,
      leading to situations where the build system unnecessarily feels that the need to recompile.
      
      * BuildSlaveSupport/build.webkit.org-config/master.cfg:
      * Scripts/run-javascriptcore-tests:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46323 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      77962a87
    • eric@webkit.org's avatar
      2009-07-24 Eric Seidel <eric@webkit.org> · f67d5b65
      eric@webkit.org authored
              Reviewed by Adam Barth.
      
              Classes call DOMObject::mark() explicitly, should call DOMObjectWithGlobal::mark() instead
              https://bugs.webkit.org/show_bug.cgi?id=27641
      
              Nothing uses globalObject() yet, but this was causing crashes
              in the patch for bug 27634.  This is covered by fast/dom/gc-6.html.
      
              I decided to change these to Base:: instead of DOMObjectWithGlobal::
              for future-proofing.  All autogenerated classes use a typedef Base
              to avoid bugs like these.  Sadly C++ does not have a built-in super:: we could use.
      
              * WebCore.xcodeproj/project.pbxproj:
              * bindings/js/JSAbstractWorkerCustom.cpp:
              (WebCore::JSAbstractWorker::mark):
              * bindings/js/JSDOMApplicationCacheCustom.cpp:
              (WebCore::JSDOMApplicationCache::mark):
              * bindings/js/JSMessageChannelCustom.cpp:
              (WebCore::JSMessageChannel::mark):
              * bindings/js/JSMessagePortCustom.cpp:
              (WebCore::JSMessagePort::mark):
              * bindings/js/JSNamedNodesCollection.cpp:
              (WebCore::JSNamedNodesCollection::getOwnPropertySlot):
              * bindings/js/JSNodeCustom.cpp:
              (WebCore::JSNode::mark):
              * bindings/js/JSNodeFilterCustom.cpp:
              (WebCore::JSNodeFilter::mark):
              * bindings/js/JSNodeIteratorCustom.cpp:
              (WebCore::JSNodeIterator::mark):
              * bindings/js/JSSVGElementInstanceCustom.cpp:
              (WebCore::JSSVGElementInstance::mark):
              * bindings/js/JSTreeWalkerCustom.cpp:
              (WebCore::JSTreeWalker::mark):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46322 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f67d5b65
    • bfulgham@webkit.org's avatar
      WebCore: · b740c154
      bfulgham@webkit.org authored
      2009-07-23  Brent Fulgham  <bfulgham@webkit.org>
      
              Reviewed by Eric Seidel.
      
              Clean up dependencies on Apple support libraries for non-Apple build.
              http://bugs.webkit.org/show_bug.cgi?id=27532.
      
              * platform/graphics/win/SimpleFontDataWin.cpp: Conditionalize references
                to ApplicationServices.h and WebKitSystemInterface.h
      
      WebKit/win:
      
      2009-07-23  Brent Fulgham  <bfulgham@webkit.org>
      
              Reviewed by Eric Seidel.
      
              Clean up dependencies on Apple support libraries for non-Apple build.
              http://bugs.webkit.org/show_bug.cgi?id=27532.
      
              * DefaultDownloadDelegate.h: Conditionalize include of CFNetwork.h
              * WebDownload.h: Conditionalize include of CFNetwork.h
              * WebMutableURLRequest.cpp: Conditionalize include of CFNetwork.h
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46321 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b740c154
    • eric@webkit.org's avatar
      2009-07-22 Eric Seidel <eric@webkit.org> · 0560e061
      eric@webkit.org authored
              Reviewed by Adam Barth.
      
              Make most DOMObjects hold onto a JSDOMGlobalObject*
              https://bugs.webkit.org/show_bug.cgi?id=27588
      
              inner.document.constructor is fixed because all properties on the window
              object are created with the correct globalObject (instead of the lexical one).
              Since all objects now carry a globalObject pointer, when document creates
              HTMLDocumentConstructor it now has the right globalObject to use.
      
              * fast/dom/prototype-inheritance-2-expected.txt:
              * fast/dom/prototype-inheritance-expected.txt:
      
      2009-07-22  Eric Seidel  <eric@webkit.org>
      
              Reviewed by Adam Barth.
      
              Make most DOMObjects hold onto a JSDOMGlobalObject*
              https://bugs.webkit.org/show_bug.cgi?id=27588
      
              This change is almost entirely plumbing.  Only one functional
              change as part of this all (window.document.constructor has the correct prototype)
              Changes are detailed below.
      
              inner.document.constructor is fixed because all properties on the window
              object are created with the correct globalObject (instead of the lexical one).
              Since all objects now carry a globalObject pointer, when document creates
              HTMLDocumentConstructor it now has the right globalObject to use.
      
              Tests:
                fast/dom/prototype-inheritance.html
                fast/dom/prototype-inheritance-2.html
      
              * bindings/js/DOMObjectWithSVGContext.h:
              (WebCore::DOMObjectWithSVGContext::DOMObjectWithSVGContext):
                  Update the comment and add an ignored globalObject argument.
              * bindings/js/JSDOMBinding.h:
                  Pass a globalObject to all DOMObjects during creation.  Currently it's the wrong global object.
                  Once toJS is passed a globalObject it will be the right one.
              (WebCore::createDOMObjectWrapper):
              (WebCore::createDOMNodeWrapper):
              * bindings/js/JSDOMGlobalObject.h:
              (WebCore::JSDOMGlobalObject::globalObject): Makes binding generation easier.
              * bindings/js/JSDOMWindowCustom.cpp:
              (WebCore::JSDOMWindow::history): JSHistory is now passed a globalObject, but since it has no custom constructor it doesn't use it.
              (WebCore::JSDOMWindow::location): JSLocation is now passed a globalObject, but since it has no custom constructor it doesn't use it.
              * bindings/js/JSDocumentCustom.cpp:
              (WebCore::JSDocument::location): JSLocation is now passed a globalObject, but since it has no custom constructor it doesn't use it.
              * bindings/js/JSHTMLAllCollection.h:
              (WebCore::JSHTMLAllCollection::JSHTMLAllCollection):
              * bindings/js/JSHTMLCollectionCustom.cpp: Re-factoring needed to pass globalObject to JSNamedNodesCollection constructor.
              (WebCore::getNamedItems):
              (WebCore::callHTMLCollection):
              (WebCore::JSHTMLCollection::canGetItemsForName):
              (WebCore::JSHTMLCollection::nameGetter):
              (WebCore::JSHTMLCollection::item):
              (WebCore::JSHTMLCollection::namedItem):
              * bindings/js/JSHTMLFormElementCustom.cpp:
              (WebCore::JSHTMLFormElement::nameGetter):
              * bindings/js/JSNamedNodesCollection.cpp:
                   Now passed globalObject.  This is tested by inner.document.forms.testForm.
                   The passed globalObject is still wrong until toJS is fixed.
              (WebCore::JSNamedNodesCollection::JSNamedNodesCollection):
              * bindings/js/JSNamedNodesCollection.h:
              * bindings/js/JSSharedWorkerConstructor.cpp:
                   Add DOMConstructorObject missed by http://trac.webkit.org/changeset/45938
                   This class is not compiled by default, so not testable.
              (WebCore::JSSharedWorkerConstructor::JSSharedWorkerConstructor):
              * bindings/js/JSSharedWorkerConstructor.h:
              * bindings/scripts/CodeGeneratorJS.pm:
                   Make all bindings objects carry a globalObject pointer using DOMObjectWithGlobalPointer.
                   SVG bindings which need a context() pointer do not have enough space for globalObject() too.
                   WorkerContext does not need a globalObject (it is one), so special case it.
                   Make all .constructor calls use the stored globalObject().  This is what fixes window.document.constructor.
                   Make all constructors inherit from DOMConstructorObject for consistency.  Since the auto-bound constructors
                   override createStructure anyway, there is no functional change here.  Just completing work started in r45938.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46320 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0560e061
    • jmalonzo@webkit.org's avatar
      2009-07-23 Jan Michael Alonzo <jmalonzo@webkit.org> · e91f3954
      jmalonzo@webkit.org authored
              Reviewed by Eric Seidel.
      
              [Gtk] Add implementation of GCController for DRT
              https://bugs.webkit.org/show_bug.cgi?id=27636
      
              * DumpRenderTree/gtk/DumpRenderTree.cpp:
              (webViewWindowObjectCleared):
              (main):
              * DumpRenderTree/gtk/GCControllerGtk.cpp:
              (GCController::collect):
              (GCController::collectOnAlternateThread):
      
              Enable tests that are now passing.
      
              * platform/gtk/Skipped:
      
              Add SPI for GCController to use.
      
              * webkit/webkitprivate.h:
              * webkit/webkitwebframe.cpp:
              (webkit_gc_collect_javascript_objects):
              (webkit_gc_collect_javascript_objects_on_alternate_thread):
              (webkit_gc_count_javascript_objects):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46319 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      e91f3954
    • eric@webkit.org's avatar
      2009-07-23 Eric Seidel <eric@webkit.org> · d72bb631
      eric@webkit.org authored
              Reviewed by Darin Adler.
      
              Add new test to cover more prototype/constructor inheritance cases
              https://bugs.webkit.org/show_bug.cgi?id=27622
      
              This new test tries hard not to duplicate testing done
              by other tests and instead test new objects and prototype chains
              which will be fixed as soon as toJS is passed through the globalObject.
      
              This test crawls the object tree under the inner window, finding
              one of each type of JavaScript object.  It depends on all objects having
              a valid prototype chain with a unique name (which is not always the case).
              The test tries to stress the cases where the objects would be lazily
              creating other objects or their prototypes or constructors which would
              require these objects to have a back-pointer to the globalObject to be
              able to correctly wire up the prototype chains.
      
              * fast/dom/prototype-inheritance-2-expected.txt: Added.
              * fast/dom/prototype-inheritance-2.html: Added.
              * fast/dom/resources/prototype-inheritance-2.js: Added.
              (propertiesOnObject):
              ():
              (constructorNamesForWindow):
              (makeCrawlObject):
              (evalToCrawlObject):
              (pushPropertyValuesWithUnseenTypes):
              (crawl):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46318 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      d72bb631
    • jmalonzo@webkit.org's avatar
      2009-07-23 Jan Michael Alonzo <jmalonzo@webkit.org> · 2a9b6562
      jmalonzo@webkit.org authored
              Skip transitions/change-values-during-transition.html which
              pass/fails intermittently
      
              * platform/gtk/Skipped:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46317 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2a9b6562
  2. 23 Jul, 2009 26 commits
    • jmalonzo@webkit.org's avatar
      2009-07-23 Jan Michael Alonzo <jmalonzo@webkit.org> · ceaa977f
      jmalonzo@webkit.org authored
              Skip tests that require expected outputs. Also skip failing tests,
              with bugs raised against the failures.
      
              * platform/gtk/Skipped:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46316 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ceaa977f
    • mrowe@apple.com's avatar
      Skip two new tests for GTK that use eventSender. · 877f51d0
      mrowe@apple.com authored
      * platform/gtk/Skipped:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46315 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      877f51d0
    • mrowe@apple.com's avatar
      Fix layout tests that incorrectly attempt to detect whether they're running... · b5cf863a
      mrowe@apple.com authored
      Fix layout tests that incorrectly attempt to detect whether they're running inside DumpRenderTree using "if (layoutTestController)".
      
      Rubber-stamped by Brady Eidson.
      
      * editing/execCommand/25256.html:
      * editing/execCommand/25320.html:
      * fast/canvas/arc-crash.html:
      * fast/css/emptyStyleTag.html:
      * fast/events/js-keyboard-event-creation.html:
      * fast/events/scrollbar-double-click.html:
      * fast/forms/select-no-name.html:
      * fast/history/window-open.html:
      * platform/mac/fast/events/objc-event-api.html:
      * platform/mac/fast/events/objc-keyboard-event-creation.html:
      * scrollbars/scrollbar-miss-mousemove-disabled.html:
      * scrollbars/scrollbar-miss-mousemove.html:
      * storage/domstorage/localstorage/clear.html:
      * storage/domstorage/localstorage/delete-removal.html:
      * storage/domstorage/localstorage/enumerate-storage.html:
      * storage/domstorage/localstorage/index-get-and-set.html:
      * storage/domstorage/localstorage/onstorage-attribute-markup.html:
      * storage/domstorage/localstorage/onstorage-attribute-setattribute.html:
      * storage/domstorage/localstorage/private-browsing-affects-storage.html:
      * storage/domstorage/localstorage/resources/window-open-second.html:
      * storage/domstorage/localstorage/simple-events.html:
      * storage/domstorage/localstorage/simple-usage.html:
      * storage/domstorage/localstorage/string-conversion.html:
      * storage/domstorage/localstorage/window-open.html:
      * storage/domstorage/sessionstorage/clear.html:
      * storage/domstorage/sessionstorage/delete-removal.html:
      * storage/domstorage/sessionstorage/enumerate-storage.html:
      * storage/domstorage/sessionstorage/index-get-and-set.html:
      * storage/domstorage/sessionstorage/onstorage-attribute-markup.html:
      * storage/domstorage/sessionstorage/onstorage-attribute-setattribute.html:
      * storage/domstorage/sessionstorage/private-browsing-affects-storage.html:
      * storage/domstorage/sessionstorage/resources/window-open-second.html:
      * storage/domstorage/sessionstorage/simple-events.html:
      * storage/domstorage/sessionstorage/simple-usage.html:
      * storage/domstorage/sessionstorage/string-conversion.html:
      * storage/domstorage/sessionstorage/window-open.html:
      * storage/domstorage/window-attributes-exist.html:
      * svg/custom/stop-crash.svg:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46314 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b5cf863a
    • mrowe@apple.com's avatar
      And now with some valid JavaScript syntax! · 0e1ee81c
      mrowe@apple.com authored
      * http/tests/misc/redirect-to-about-blank.html:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46313 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0e1ee81c
    • mrowe@apple.com's avatar
      Remove unnecessary check for eventSender in an attempt to please the GTK build bot. · 8d61c1f4
      mrowe@apple.com authored
      * fast/css/getComputedStyle/getComputedStyle-text-overflow.html:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46312 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      8d61c1f4
    • mrowe@apple.com's avatar
      Update build.webkit.org config for new machine. · 6b3df756
      mrowe@apple.com authored
      * BuildSlaveSupport/build.webkit.org-config/config.json:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46311 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      6b3df756
    • mrowe@apple.com's avatar
      Add a workaround for <rdar://problem/6925892> to the test that it affects. · 13d187ef
      mrowe@apple.com authored
      Reviewed by Brady Eidson.
      
      * http/tests/misc/redirect-to-about-blank.html:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46310 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      13d187ef
    • beidson@apple.com's avatar
      2009-07-23 Brady Eidson <beidson@apple.com> · 8bb49f2d
      beidson@apple.com authored
              Remove accidentally checked in code.
      
              * DumpRenderTree/mac/LayoutTestControllerMac.mm:
              (waitUntilDoneWatchdogFired):
              (LayoutTestController::setWaitToDump):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46309 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      8bb49f2d
    • beidson@apple.com's avatar
      WebCore: · df488b50
      beidson@apple.com authored
      2009-07-23  Brady Eidson  <beidson@apple.com>
      
              Reviewed by Geoff Garen.
      
              WebCore has a few places that don't gracefully handle a null request returned from willSendRequest.
              https://bugs.webkit.org/show_bug.cgi?id=27595
      
              Test: http/tests/misc/will-send-request-returns-null-on-redirect.html
              
              * inspector/InspectorController.cpp:
              (WebCore::InspectorController::removeResource): Null-check the request URL.
              
              * platform/network/cf/ResourceHandleCFNet.cpp: Ditto, and return null instead of creating an empty one.
              (WebCore::willSendRequest):
      
      WebKit/win:
      
      2009-07-23  Brady Eidson  <beidson@apple.com>
      
              Reviewed by Geoff Garen.
      
              WebFrameLoaderClient on Windows doesn't gracefully handle a null request returned from willSendRequest.
              https://bugs.webkit.org/show_bug.cgi?id=27595
      
              * WebCoreSupport/WebFrameLoaderClient.cpp:
              (WebFrameLoaderClient::dispatchWillSendRequest):
      
      WebKitTools:
      
      2009-07-23  Brady Eidson  <beidson@apple.com>
      
              Reviewed by Geoff Garen.
      
              WebCore has a few places that don't gracefully handle a null request returned from willSendRequest.
              https://bugs.webkit.org/show_bug.cgi?id=27595
      
              Add the ability to tell the LayoutTestController to return a null request for willSendRequest
              redirect callbacks.
              
              * DumpRenderTree/LayoutTestController.cpp:
              (LayoutTestController::LayoutTestController):
              (setWillSendRequestReturnsNullOnRedirectCallback):
              (LayoutTestController::staticFunctions):
              * DumpRenderTree/LayoutTestController.h:
              (LayoutTestController::willSendRequestReturnsNullOnRedirect):
              (LayoutTestController::setWillSendRequestReturnsNullOnRedirect):
      
              * DumpRenderTree/mac/ResourceLoadDelegate.mm:
              (-[ResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
              * DumpRenderTree/win/ResourceLoadDelegate.cpp:
              (ResourceLoadDelegate::willSendRequest):
      
      LayoutTests:
      
      2009-07-23  Brady Eidson  <beidson@apple.com>
      
              Reviewed by Geoff Garen.
      
              WebCore has a few places that don't gracefully handle a null request returned from willSendRequest.
              https://bugs.webkit.org/show_bug.cgi?id=27595
      
              * http/tests/misc/resources/redirect-to-http-url.php: Added.
              * http/tests/misc/will-send-request-returns-null-on-redirect-expected.txt: Added.
              * http/tests/misc/will-send-request-returns-null-on-redirect.html: Added.
              * platform/gtk/Skipped:
              * platform/qt/Skipped:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46308 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      df488b50
    • pdherbemont@apple.com's avatar
      2009-07-23 Pierre d'Herbemont <pdherbemont@apple.com> · 02ff2d94
      pdherbemont@apple.com authored
              Fix layout tests, no review.
      
              Actually add the video-layer tests results for leopard, that were
              forgotten in r46302. Also add test results for windows.
      
              * platform/mac-leopard/fast/layers/video-layer-expected.png: Added.
              * platform/mac-leopard/fast/layers/video-layer-expected.txt: Added.
              * platform/win/fast/layers/video-layer-expected.txt: Added.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46307 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      02ff2d94
    • mrowe@apple.com's avatar
      Fix DumpRenderTree to not fail an assertion on launch on a new machine. · be919533
      mrowe@apple.com authored
      Reviewed Oliver Hunt.
      
      * DumpRenderTree/mac/DumpRenderTree.mm:
      (setDefaultsToConsistentValuesForTesting): Disable the XSS auditor.  It being enabled
      breaks all of the JavaScript tests within testStringByEvaluatingJavaScriptFromString,
      which causes us to exit due to an assertion failure.  It's not clear why the XSS auditor
      decides to interfere with these tests.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46304 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      be919533
    • pdherbemont@apple.com's avatar
      2009-07-23 Pierre d'Herbemont <pdherbemont@apple.com> · c2737dd4
      pdherbemont@apple.com authored
              Reviewed by Simon Fraser.
      
              Following changes on the video controls, video-layer tests results need to
              be updated in order to fix the layout tests.
      
              * platform/mac-leopard/fast/layers/video-layer-expected.checksum: Copied from LayoutTests/platform/mac/fast/layers/video-layer-expected.checksum.
              * platform/mac-leopard/fast/layers/video-layer-expected.png: Copied from LayoutTests/platform/mac/fast/layers/video-layer-expected.png.
              * platform/mac-leopard/fast/layers/video-layer-expected.txt: Copied from LayoutTests/platform/mac/fast/layers/video-layer-expected.txt.
              * platform/mac/fast/layers/video-layer-expected.checksum: Removed.
              * platform/mac/fast/layers/video-layer-expected.png: Removed.
              * platform/mac/fast/layers/video-layer-expected.txt: Updated.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46302 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c2737dd4
    • cfleizach@apple.com's avatar
    • darin@apple.com's avatar
      2009-07-23 Darin Adler <darin@apple.com> · 0395090d
      darin@apple.com authored
              * WebView/WebDataSource.mm:
              (-[WebDataSource _makeRepresentation]): Landed a comment I forgot last time.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46300 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0395090d
    • mrowe@apple.com's avatar
      Update layout test results for three tests related to text fields as their metrics · b43babdd
      mrowe@apple.com authored
      have changed slightly since the results were last updated.
      
      Rubber-stamped by Dan Bernstein.
      
      * platform/mac-leopard/fast/forms/text-control-intrinsic-widths-expected.txt: Copied from LayoutTests/fast/forms/text-control-intrinsic-widths-expected.txt.
      * platform/mac/fast/forms/input-text-maxlength-expected.txt:
      * platform/mac/fast/forms/input-text-paste-maxlength-expected.txt:
      * platform/mac/fast/forms/text-control-intrinsic-widths-expected.txt: Renamed from LayoutTests/fast/forms/text-control-intrinsic-widths-expected.txt.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46299 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b43babdd
    • darin@apple.com's avatar
      WebCore: · 719d29cb
      darin@apple.com authored
      2009-07-23  Darin Adler  <darin@apple.com>
      
              Reviewed by Brady Eidson.
      
              URL appears in back/forward button menu instead of title for items with custom representation
              https://bugs.webkit.org/show_bug.cgi?id=27586
              rdar://problem/5060337
      
              * WebCore.base.exp: Exported DocumentLoader::setTitle for use by Mac WebKit.
              * loader/FrameLoader.cpp:
              (WebCore::FrameLoader::didChangeTitle): Tightened code to check if the document
              loader is the correct one; previously this would never happen because we'd
              commit the load before any title changes could be registered, but now we can
              encounter a case where we get a title during a provisional load.
      
      WebKit/mac:
      
      2009-07-23  Darin Adler  <darin@apple.com>
      
              Reviewed by Brady Eidson.
      
              URL appears in back/forward button menu instead of title for items with custom representation
              https://bugs.webkit.org/show_bug.cgi?id=27586
              rdar://problem/5060337
      
              The problem is that DocumentLoader expects to store a title, but for custom representations
              it is never passed to the document loader.
      
              * WebView/WebDataSource.mm:
              (-[WebDataSource _makeRepresentation]): Added a call to DocumentLoader::setTitle.
              Works as long as the title does not change during the document's lifetime, which
              is good enough for the simple cases in Safari.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46298 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      719d29cb
    • mrowe@apple.com's avatar
      Update default state of the XSS auditor in GTK DumpRenderTree to match Mac and Windows · 062ce4a1
      mrowe@apple.com authored
      in hopes of fixing some layout test failures seen on the build bot.
      
      Reviewed by Oliver Hunt.
      
      * DumpRenderTree/gtk/DumpRenderTree.cpp:
      (resetWebViewToConsistentStateBeforeTesting):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46297 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      062ce4a1
    • mrowe@apple.com's avatar
      Windows appears to match the Leopard results for the CSS 2.1 counter tests that were failing. · 23c2cc53
      mrowe@apple.com authored
      * platform/win/css2.1/t1202-counter-04-b-expected.txt: Copied from LayoutTests/platform/mac-leopard/css2.1/t1202-counter-04-b-expected.txt.
      * platform/win/css2.1/t1202-counters-04-b-expected.txt: Copied from LayoutTests/platform/mac-leopard/css2.1/t1202-counters-04-b-expected.txt.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46296 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      23c2cc53
    • mitz@apple.com's avatar
      WebCore: · fde1b4b0
      mitz@apple.com authored
              Reviewed by Dave Hyatt.
      
              [CSS3 Backgrounds and Borders] Add support for inset box shadows
              https://bugs.webkit.org/show_bug.cgi?id=27582
      
              Test: fast/box-shadow/inset.html
      
              * css/CSSComputedStyleDeclaration.cpp:
              (WebCore::valueForShadow): Set the ShadowValue’s shadow style to 'inset'
                  as needed.
      
              * css/CSSParser.cpp:
              (WebCore::ShadowParseContext::ShadowParseContext): Added style and allowStyle
                  members. Initialize the allowStyle member.
              (WebCore::ShadowParseContext::commitValue): Pass the style value to the
                  ShadowValue constructor. Reset allowStyle.
              (WebCore::ShadowParseContext::commitLength): Update allowStyle.
              (WebCore::ShadowParseContext::commitColor): Ditto.
              (WebCore::ShadowParseContext::commitStyle): Added. Sets the style member and
                  updates the state.
              (WebCore::CSSParser::parseShadow): Parse the 'inset' keyword.
      
              * css/CSSStyleSelector.cpp:
              (WebCore::CSSStyleSelector::applyProperty): Get the style value from the
                  shadow value and pass it to the ShadowData constructor.
      
              * css/ShadowValue.cpp:
              (WebCore::ShadowValue::ShadowValue): Added style.
              (WebCore::ShadowValue::cssText): Added style.
      
              * css/ShadowValue.h:
              (WebCore::ShadowValue::create): Added style.
      
              * page/animation/AnimationBase.cpp:
              (WebCore::blendFunc): Added a blend function for shadow styles. When blending
                  between normal and inset shadows, all intermediate values map to normal.
              (WebCore::PropertyWrapperShadow::blend): Added normal style to the default
                  shadow.
                  
              * rendering/InlineFlowBox.cpp:
              (WebCore::InlineFlowBox::paintBoxShadow): Added a shadow style parameter,
                  which is passed through to RenderBoxModelObject::paintBoxShadow().
      
              (WebCore::InlineFlowBox::paintBoxDecorations): Paint inset shadows on top of
                  the background.
      
              * rendering/InlineFlowBox.h:
      
              * rendering/RenderBox.cpp:
              (WebCore::RenderBox::paintBoxDecorations): Paint inset shadows on top of the
                  background.
      
              * rendering/RenderBoxModelObject.cpp:
              (WebCore::RenderBoxModelObject::paintBoxShadow): Added a shadow style
                  parameter, and code to paint inset shadows.
      
              * rendering/RenderBoxModelObject.h:
      
              * rendering/RenderFieldset.cpp:
              (WebCore::RenderFieldset::paintBoxDecorations): Paint inset shadows on top of
                  the background.
      
              * rendering/RenderTable.cpp:
              (WebCore::RenderTable::paintBoxDecorations): Ditto.
      
              * rendering/RenderTableCell.cpp:
              (WebCore::RenderTableCell::paintBoxDecorations): Ditto.
      
              * rendering/style/ShadowData.h:
              Added a ShadowStyle enum.
              (WebCore::ShadowData::ShadowData): Add and initialize a style member.
      
      LayoutTests:
      
              Reviewed by Dave Hyatt.
      
              [CSS3 Backgrounds and Borders] Add support for inset box shadows
              https://bugs.webkit.org/show_bug.cgi?id=27582
      
              * fast/box-shadow/inset.html: Added.
              * platform/mac/fast/box-shadow/inset-expected.checksum: Added.
              * platform/mac/fast/box-shadow/inset-expected.png: Added.
              * platform/mac/fast/box-shadow/inset-expected.txt: Added.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46295 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      fde1b4b0
    • treat@webkit.org's avatar
      2009-07-23 Jakob Petsovits <jakob.petsovits@torchmobile.com> · 4629ee3a
      treat@webkit.org authored
               Reviewed by Adam Treat.
      
               Fix false positives for switch statement indentation check in cpplint.
               https://bugs.webkit.org/show_bug.cgi?id=27615
      
               Makes one-line case statements (e.g. "case foo: bar();") work.
               Also a few general improvements to the robustness and readability of
               the check, and more test cases.
      
               * Scripts/modules/cpplint.py:
               * Scripts/modules/cpplint_unittest.py:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46294 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      4629ee3a
    • simon.fraser@apple.com's avatar
      2009-07-23 Simon Fraser <simon.fraser@apple.com> · 69070f36
      simon.fraser@apple.com authored
              Fix the build with UNUSED_PARAM(frame) for when ENABLE(3D_RENDERING) is not defined.
      
              * css/MediaQueryEvaluator.cpp:
              (WebCore::transform_3dMediaFeatureEval):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46293 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      69070f36
    • treat@webkit.org's avatar
      2009-07-23 Jakob Petsovits <jakob.petsovits@torchmobile.com> · 2fe25536
      treat@webkit.org authored
               Reviewed by Adam Treat.
      
               Add check for line-breaking rule #3 to cpplint.
               https://bugs.webkit.org/show_bug.cgi?id=27610
      
               "An else if statement should be written as an if statement when
               the prior if concludes with a return statement."
      
               Implemented by a multi-line (kind of back-tracking) algorithm.
               Comes with loads of unit tests. Fixes the check for label indentation
               to be not so overzealous, as it didn't allow for completely unindented
               goto labels (at the very start of a line).
      
               * Scripts/modules/cpplint.py:
               * Scripts/modules/cpplint_unittest.py:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46292 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2fe25536
    • simon.fraser@apple.com's avatar
      2009-07-23 Simon Fraser <simon.fraser@apple.com> · ec4b4f9e
      simon.fraser@apple.com authored
              Reviewed by Adele Peterson.
      
              3d-transforms media query needs to look check that accelerated compositing is enabled
              https://bugs.webkit.org/show_bug.cgi?id=27621
      
              When evaluating a media query with '-webkit-transform-3d', we need to check the
              runtime switch that toggles accererated compositing, and therefore 3D.
      
              No test because we can't disable the pref dynamically in DRT.
      
              * css/MediaQueryEvaluator.cpp:
              (WebCore::transform_3dMediaFeatureEval):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46291 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ec4b4f9e
    • mrowe@apple.com's avatar
      Update layout test results for two international text tests that are failing... · 5752f6e1
      mrowe@apple.com authored
      Update layout test results for two international text tests that are failing due to font differences.
      
      Rubber-stamped by Dan Bernstein.
      
      * platform/mac-leopard/fast/text/international/thai-baht-space-expected.checksum: Renamed from LayoutTests/platform/mac/fast/text/international/thai-baht-space-expected.checksum.
      * platform/mac-leopard/fast/text/international/thai-baht-space-expected.png: Renamed from LayoutTests/platform/mac/fast/text/international/thai-baht-space-expected.png.
      * platform/mac-leopard/fast/text/international/thai-baht-space-expected.txt: Copied from LayoutTests/platform/mac/fast/text/international/thai-baht-space-expected.txt.
      * platform/mac/fast/text/international/bidi-neutral-directionality-paragraph-start-expected.txt:
      * platform/mac/fast/text/international/thai-baht-space-expected.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46290 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      5752f6e1
    • mrowe@apple.com's avatar
      Update layout test results for two CSS 2.1 counter tests that are failing due to font differences. · 9e962c0b
      mrowe@apple.com authored
      Rubber-stamped by Dan Bernstein.
      
      * platform/mac-leopard/css2.1/t1202-counter-04-b-expected.txt: Copied from LayoutTests/platform/mac/css2.1/t1202-counter-04-b-expected.txt.
      * platform/mac-leopard/css2.1/t1202-counters-04-b-expected.txt: Copied from LayoutTests/platform/mac/css2.1/t1202-counters-04-b-expected.txt.
      * platform/mac/css2.1/t1202-counter-04-b-expected.txt:
      * platform/mac/css2.1/t1202-counters-04-b-expected.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46289 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      9e962c0b
    • pdherbemont@apple.com's avatar
      2009-07-23 Pierre d'Herbemont <pdherbemont@apple.com> · b1a382bb
      pdherbemont@apple.com authored
              Layout test fix so no review.
      
              Add audio-controls-rendering layout test results for windows and leopard.
      
              * platform/mac-leopard/media/audio-controls-rendering-expected.txt: Added.
              * platform/win/media/audio-controls-rendering-expected.txt: Added.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@46288 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b1a382bb