Timeline



Aug 12, 2008:

9:35 PM Changeset in webkit [35704] by weinig@apple.com
  • 2 edits in trunk/LayoutTests

2008-08-12 Sam Weinig <sam@webkit.org>

Reviewed by Anders Carlsson.

  • platform/mac/Skipped: Add fast/dom/cssTarget-crash.html to the skipped list.
8:17 PM Changeset in webkit [35703] by oliver@apple.com
  • 3 edits in trunk/JavaScriptCore

Add peephole optimisation to 'op_not... jfalse...' (eg. if(!...) )

Reviewed by Geoff Garen

This is a very slight win in sunspider, and a fairly substantial win
in hot code that does if(!...), etc.

3:49 PM Changeset in webkit [35702] by Nikolas Zimmermann
  • 5 edits in trunk/LayoutTests

Not reviewed.

Revert the last SVGLineElement commit. It's not needed, the first run on buildbot showed them as failures
the next build doesn't complain anymore (this was before adjusting the offsets.) No idea what's going on,
but if it's fixed I'm happy.

While I'm at it, regenerate struct-image-02-b.svg (invisible differences) to have a clean baseline.
All SVG pixel test results are accurate on Leopard now.

3:26 PM Changeset in webkit [35701] by Nikolas Zimmermann
  • 3 edits in trunk/LayoutTests

Not reviewed. The last SVGLineElement* fix attempt, fixed 4/8 failing tests.
Trying two variations in two tests to find out the right offset to make the other four tests work on Tiger.

3:18 PM Changeset in webkit [35700] by Nikolas Zimmermann
  • 1 edit
    40 adds in trunk/LayoutTests

Rubber stamped by Eric.
Add new dynamic-update layout tests covering SVGRectElement.

3:13 PM Changeset in webkit [35699] by Nikolas Zimmermann
  • 9 edits in trunk/LayoutTests

Not reviewed. Trying to modify mouseevent target location, to work around known CG/Tiger hit test problem.

2:59 PM Changeset in webkit [35698] by Nikolas Zimmermann
  • 5 edits in trunk/LayoutTests

Reviewed by Sam.
Forgot to include these in my last commit.

2:49 PM Changeset in webkit [35697] by mitz@apple.com
  • 4 edits
    2 adds in trunk

WebCore:

2008-08-12 Mihnea Ovidenie <mihnea@adobe.com>

Fix for https://bugs.webkit.org/show_bug.cgi?id=19891
Broken HTML object elements cause de-reference of pointer to freed memory.
If we fail to load an image for an object tag and we no longer believe the object tag points at
an image, then clear m_imageLoader in the HTMLObjectElement so that we attempt to render the
fall back content.

Reviewed by Dave Hyatt and Alexey Proskuryakov.

Test: http/tests/misc/object-image-error-with-onload.html

  • html/HTMLObjectElement.cpp: (WebCore::HTMLObjectElement::renderFallbackContent):
  • page/Frame.cpp: (WebCore::Frame::Frame):

LayoutTests:

2008-08-12 Mihnea Ovidenie <mihnea@adobe.com>

Test for https://bugs.webkit.org/show_bug.cgi?id=19891
Broken HTML object elements cause de-reference of pointer to freed memory.

Reviewed by Dave Hyatt and Alexey Proskuryakov.

  • http/tests/misc/object-image-error-with-onload-expected.txt: Added.
  • http/tests/misc/object-image-error-with-onload.html: Added.
2:46 PM Changeset in webkit [35696] by Nikolas Zimmermann
  • 84 edits in trunk/LayoutTests

Reviewed by Eric.

Fix typo in most files: "mesage" -> "message".
Remove unused code in SVGTestCase.js

2:18 PM Changeset in webkit [35695] by Nikolas Zimmermann
  • 1 edit
    70 adds in trunk/LayoutTests

Rubber stamped by Eric.

Add new dynamic-update layout tests covering SVGPatternElement.

2:11 PM Changeset in webkit [35694] by Nikolas Zimmermann
  • 13 edits in trunk

Reviewed by Dave.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=19798
Masks are translated, and the mask images are swapped on the y-axis.

Turned out that http://trac.webkit.org/changeset/31830/trunk/WebCore/svg/graphics/cg/SVGResourceMaskerCg.mm
is guilty. GraphicsContext::clipToImageBuffer() does some extra transformations that SVGResourcesMaskerCg does not want.

Long term goal is to remove the SVGResource*/SVGPaintServer* classes anyway, so it's okay to duplicate
the "clip to image buffer" functionality, in the CG specific SVGResourceMaskerCg class - as it was before.

2:05 PM Changeset in webkit [35693] by Nikolas Zimmermann
  • 2 edits in trunk/LayoutTests

Reviewed by Sam.

Speculative fix for hit testing problems in dynamic-updates/SVGLineElement* on Tiger.
Supposed to fix the hanging tests on the buildbot.

2:00 PM Changeset in webkit [35692] by weinig@apple.com
  • 2 edits in trunk/LayoutTests

2008-08-12 Sam Weinig <sam@webkit.org>

  • fast/dom/Window/window-properties-expected.txt: Update results.
1:54 PM Changeset in webkit [35691] by mitz@apple.com
  • 11 edits in trunk

JavaScriptCore:

  • JavaScriptCore part of <rdar://problem/6121636> Make fast*alloc() abort() on failure and add "try" variants that return NULL on failure.

Reviewed by Darin Adler.

  • JavaScriptCore.exp: Exported tryFastCalloc().
  • VM/RegisterFile.h: (KJS::RegisterFile::RegisterFile): Removed an ASSERT().
  • kjs/JSArray.cpp: (KJS::JSArray::putSlowCase): Changed to use tryFastRealloc(). (KJS::JSArray::increaseVectorLength): Ditto.
  • kjs/ustring.cpp: (KJS::allocChars): Changed to use tryFastMalloc(). (KJS::reallocChars): Changed to use tryFastRealloc().
  • wtf/FastMalloc.cpp: (WTF::fastZeroedMalloc): Removed null checking of fastMalloc()'s result and removed extra call to InvokeNewHook(). (WTF::tryFastZeroedMalloc): Added. Uses tryFastMalloc(). (WTF::tryFastMalloc): Renamed fastMalloc() to this. (WTF::fastMalloc): Added. This version abort()s if allocation fails. (WTF::tryFastCalloc): Renamed fastCalloc() to this. (WTF::fastCalloc): Added. This version abort()s if allocation fails. (WTF::tryFastRealloc): Renamed fastRealloc() to this. (WTF::fastRealloc): Added. This version abort()s if allocation fails. (WTF::do_malloc): Made this a function template. When the abortOnFailure template parameter is set, the function abort()s on failure to allocate. Otherwise, it sets errno to ENOMEM and returns zero. (WTF::TCMallocStats::fastMalloc): Defined to abort() on failure. (WTF::TCMallocStats::tryFastMalloc): Added. Does not abort() on failure. (WTF::TCMallocStats::fastCalloc): Defined to abort() on failure. (WTF::TCMallocStats::tryFastCalloc): Added. Does not abort() on failure. (WTF::TCMallocStats::fastRealloc): Defined to abort() on failure. (WTF::TCMallocStats::tryFastRealloc): Added. Does not abort() on failure.
  • wtf/FastMalloc.h: Declared the "try" variants.

WebCore:

  • WebCore part of <rdar://problem/6121636> Make fast*alloc() abort() on failure and add "try" variants that return NULL on failure.

Reviewed by Darin Adler.

  • platform/Arena.cpp: (WebCore::ArenaAllocate): Removed null checking of fastMalloc()'s result.
  • platform/graphics/cg/ImageBufferCG.cpp: (WebCore::ImageBuffer::create): Changed to use tryFastCalloc().
11:27 AM Changeset in webkit [35690] by mitz@apple.com
  • 3 edits
    4 adds in trunk

WebCore:

Reviewed by Darin Adler.

Test: fast/frames/frameset-style-recalc.html

  • html/HTMLFrameSetElement.cpp: (WebCore::HTMLFrameSetElement::recalcStyle): Changed to call the base class implementation after marking for layout.

LayoutTests:

Reviewed by Darin Adler.

  • fast/frames/frameset-style-recalc.html: Added.
  • platform/mac/fast/frames/frameset-style-recalc-expected.checksum: Added.
  • platform/mac/fast/frames/frameset-style-recalc-expected.png: Added.
  • platform/mac/fast/frames/frameset-style-recalc-expected.txt: Added.
10:45 AM Changeset in webkit [35689] by timothy@apple.com
  • 4 edits in trunk/WebKit/mac

Remove unneeded header imports from some Web Inspector files.

Reviewed by Adam Roben.

  • WebCoreSupport/WebInspectorClient.mm:
  • WebInspector/WebInspector.mm:
  • WebInspector/WebNodeHighlightView.m:
10:45 AM Changeset in webkit [35688] by timothy@apple.com
  • 2 edits in trunk/WebKit/mac

Remove the call to WKNSWindowMakeBottomCornersSquare on the
Web Inspector's window. This isn't needed anymore since the
window style masks used always have square bottom corners.

Reviewed by Adam Roben.

  • WebCoreSupport/WebInspectorClient.mm: (-[WebInspectorWindowController window]): Remove the call to WKNSWindowMakeBottomCornersSquare.
10:45 AM Changeset in webkit [35687] by timothy@apple.com
  • 2 edits in trunk/WebKit/mac

Make attaching and detaching the Web Inspector instantaneous.
This also preserves the current view, other state, and keeps
the script debugger attached.

https://bugs.webkit.org/show_bug.cgi?id=19301

Reviewed by Adam Roben.

  • WebCoreSupport/WebInspectorClient.mm: (-[WebInspectorWindowController close]): Only call setWindowVisible(false) when not moving windows.
10:13 AM Changeset in webkit [35686] by Adam Roben
  • 2 edits in trunk/WebKit/win

Export WTFLog in Debug builds

Reviewed by John Sullivan.

  • WebKit.vcproj/WebKit_debug.def:
10:04 AM Changeset in webkit [35685] by mitz@apple.com
  • 19 edits in trunk

WebCore:

Reviewed by John Sullivan.

  • move shouldUpdateWhenOffscreen from Settings to FrameView and rename it shouldUpdateWhileHidden
  • WebCore.base.exp:
  • page/FrameView.cpp: (WebCore::FrameViewPrivate::FrameViewPrivate): (WebCore::FrameView::shouldUpdateWhileHidden): (WebCore::FrameView::setShouldUpdateWhileHidden):
  • page/FrameView.h:
  • page/Settings.cpp:
  • page/Settings.h:

WebKit/mac:

Reviewed by John Sullivan.

  • replace -[WebPreferences updatesWhenOffscreen] with -[WebView shouldUpdateWhileHidden]
  • WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::transitionToCommittedForNewPage):
  • WebView/WebFrame.mm: (-[WebFrame _updateBackgroundAndUpdatesWhileHidden]):
  • WebView/WebFrameInternal.h:
  • WebView/WebHTMLView.mm: (-[WebHTMLView windowWillOrderOnScreen:]):
  • WebView/WebPreferenceKeysPrivate.h:
  • WebView/WebPreferences.mm: (+[WebPreferences initialize]):
  • WebView/WebPreferencesPrivate.h:
  • WebView/WebView.mm: (-[WebViewPrivate init]): (-[WebView _preferencesChangedNotification:]): (-[WebView setBackgroundColor:]): (-[WebView setDrawsBackground:]): (-[WebView shouldUpdateWhileHidden]): (-[WebView setShouldUpdateWhileHidden:]):
  • WebView/WebViewPrivate.h:
7:58 AM Changeset in webkit [35684] by Adam Roben
  • 2 edits in trunk/WebCore

Windows build fix

  • bindings/js/JSSVGPODTypeWrapper.h: Align JSSVGDynamicPODTypeWrapper's and JSSVGStaticPODTypeWrapperWithParent's members on 16-byte boundaries to avoid an alignment warning.
7:40 AM Changeset in webkit [35683] by Nikolas Zimmermann
  • 3 edits
    20 adds in trunk/LayoutTests

Rubber stamped by Anders.

Add new dynamic-update layout tests covering SVGCursorElement.

Scale svg/webarchive/resources/green-checker.png to a value < 128x128,
as this is the maximum hardcoded cursor size (fixes manual testing of svg-cursor-subresources.png).

5:01 AM Changeset in webkit [35682] by Nikolas Zimmermann
  • 1 edit
    4 moves
    60 adds in trunk/LayoutTests

Rubber stamped by Oliver.

Add new dynamice-update layout tests covering SVGMaskElement.
Move pixel tests results of SVGMarkerElement* to the right positions, while I'm at it.

4:59 AM Changeset in webkit [35681] by Simon Hausmann
  • 3 edits in trunk/WebKit/qt

2008-08-12 Urs Wolfer <uwolfer@kde.org>

Reviewed by Simon.

https://bugs.webkit.org/show_bug.cgi?id=20357

Fix crash in QWebPage in case contextMenuEvent has been overwritten
because context menu has the view as parent and thus is deleted too early.

Add testcase for this crash.

3:35 AM Changeset in webkit [35680] by Nikolas Zimmermann
  • 3 edits
    75 adds in trunk

Reviewed by Oliver.

Add new dynamice-update layout tests covering SVGMarkerElement.
Fix bug: SVGMarkerElement's SVG DOM function calls don't update rendering.
Fix orientAngle/orientType confusion: "auto" orient should always return "0" as angle.

3:27 AM Changeset in webkit [35679] by Nikolas Zimmermann
  • 3 edits
    50 adds in trunk

Reviewed by Oliver.

Add new dynamic-update layout tests covering SVGImageElement.
Fix bug: SVGImageElement doesn't react on 'preserveAspectRatio' changes.

Aug 11, 2008:

9:28 PM Changeset in webkit [35678] by timothy@apple.com
  • 2 edits in trunk/WebCore

Changed Option/Alt-Up or Down in CSS editing when the value is
near zero to jump to the next integer.

Reviewed by Tim Hatcher.

https://bugs.webkit.org/show_bug.cgi?id=20326

  • page/inspector/StylesSidebarPane.js:
9:20 PM Changeset in webkit [35677] by timothy@apple.com
  • 2 edits in trunk/WebCore

Changed the line highlight transition for an easier animation.

Reviewed by Tim Hatcher.

  • page/inspector/SourceFrame.js:
8:57 PM Changeset in webkit [35676] by timothy@apple.com
  • 2 edits in trunk/WebCore

Added support for some Firebug Command Line APIs.

Reviewed by Tim Hatcher.

https://bugs.webkit.org/show_bug.cgi?id=19867
https://bugs.webkit.org/show_bug.cgi?id=19868
https://bugs.webkit.org/show_bug.cgi?id=19869
https://bugs.webkit.org/show_bug.cgi?id=19875
https://bugs.webkit.org/show_bug.cgi?id=19876
https://bugs.webkit.org/show_bug.cgi?id=19880

  • page/inspector/Console.js: (WebInspector.Console.prototype._evalInInspectedWindow): Create an object on the inspected window that holds the console command line API functions. This object is used in a with statement around the typed expression.
7:55 PM Changeset in webkit [35675] by Nikolas Zimmermann
  • 1 edit
    110 adds in trunk/LayoutTests

Rubber stamped by Oliver.

Add a bunch of new dynamic-updates tests covering SVGLineElement & SVGRadialGradientElement.
No functional changes needed, all work out of the box.

7:22 PM Changeset in webkit [35674] by Nikolas Zimmermann
  • 11 edits
    15 adds in trunk

Reviewed by Antti.

Fixes: http://bugs.webkit.org/show_bug.cgi?id=17736

JS wrapper objects around SVG POD types, that contain other SVG POD types with writable properties
failed to update. Modification of the values were completly ignored (ie. transform.matrix.a = 50, didn't take any effect)

Added tests: svg/custom/svg-modify-currentTranslate.html

svg/custom/tearoffs-with-tearoffs.html
svg/custom/immutable-properties.html

Fixed tests: svg/dynamic-updates/SVGLinearGradientElement-svgdom-gradientTransform-prop.html

5:34 PM Changeset in webkit [35673] by beidson@apple.com
  • 3 edits in trunk/WebCore

2008-08-11 Brady Eidson <beidson@apple.com>

Reviewed by John and Anders

Fix for <rdar://problem/6141797>

When WebArchives were entirely a WebKit concept, there was a guarantee that a WebResource
would never have nil data.

When they were pushed down into WebCore, that guarantee was lost, subtly changing a few
semantics with some WebKit applications.

The guarantee was a good one and should be restored.

Note that ApplicationCacheResource doesn't need any updates to follow this rule as it already
creates an empty data object in the case of null data for its own purposes.

  • loader/SubstituteResource.h: (WebCore::SubstituteResource::SubstituteResource): ASSERT that the data is not null. This well help any future subclassers not make this mistake.
  • loader/archive/ArchiveResource.cpp: (WebCore::ArchiveResource::create): Return 0 if the data is null.
3:44 PM Changeset in webkit [35672] by Simon Fraser
  • 3 edits
    2 adds in trunk

2008-08-11 Simon Fraser <Simon Fraser>

Reviewed by Dave Hyatt

https://bugs.webkit.org/show_bug.cgi?id=20328
Fix a problem when an 'all' transition transition with more than
one property changing is interrupted, and did some AnimationController
cleanup.

Test: transitions/interrupted-all-transition.html

  • page/AnimationController.cpp: (WebCore::ImplicitAnimation::ImplicitAnimation): (WebCore::AnimationControllerPrivate::blendProperties): (WebCore::CompositeAnimation::updateTransitions): (WebCore::CompositeAnimation::cleanupFinishedAnimations): (WebCore::CompositeAnimation::setTransitionStartTime): (WebCore::CompositeAnimation::overrideImplicitAnimations): (WebCore::CompositeAnimation::resumeOverriddenImplicitAnimations): (WebCore::ImplicitAnimation::animate): (WebCore::ImplicitAnimation::onAnimationEnd): (WebCore::ImplicitAnimation::sendTransitionEvent): (WebCore::ImplicitAnimation::affectsProperty): (WebCore::KeyframeAnimation::endAnimation): (WebCore::KeyframeAnimation::onAnimationEnd):
1:45 PM Changeset in webkit [35671] by kmccullough@apple.com
  • 3 edits in trunk/WebCore

2008-08-11 Kevin McCullough <kmccullough@apple.com>

Reviewed by Tim.

  • Because console messages have group levels now, newly created messages that do not specify the level lose their message since the number of arguments is wrong.


  • page/inspector/Console.js:
  • page/inspector/Resource.js:
12:52 PM Changeset in webkit [35670] by alp@webkit.org
  • 2 edits in trunk/WebCore

2008-08-11 Alp Toker <alp@nuanti.com>

Build fix. Add new files from r35666 (WebKitAnimationEvent). Also take
the opportunity to sort the sources lists.

  • GNUmakefile.am:
11:48 AM Changeset in webkit [35669] by timothy@apple.com
  • 2 edits in trunk/WebCore

Speed up the the JavaScript syntax highlighter by generating
the finders only once per script instead of per line.

https://bugs.webkit.org/show_bug.cgi?id=20346

Reviewed by Adam Roben.

  • page/inspector/SourceFrame.js: (WebInspector.SourceFrame.prototype._syntaxHighlightJavascriptLine): Removed, factored into syntaxHighlightJavascript as an inline function. (WebInspector.SourceFrame.prototype.syntaxHighlightJavascript): Pulled in the _syntaxHighlightJavascriptLine so it will create a closure. Generate the finders before iterating the lines.
11:31 AM Changeset in webkit [35668] by Adam Roben
  • 2 edits in trunk/WebCore

Windows build fix

  • WebCore.vcproj/WebCore.vcproj: Added JSWebKitAnimationEvent.cpp and JSWebKitTransitionEvent.cpp to the project. Let VS reformat the file, too.
11:04 AM Changeset in webkit [35667] by mitz@apple.com
  • 7 edits
    4 adds in trunk

WebCore:

Reviewed by Darin Adler.

  • fix <rdar://problem/6131096> Reproducible crash in CounterNode::isReset under guard malloc

Test: fast/css/counters/invalidate-cached-counter-node.html

  • rendering/RenderContainer.cpp: (WebCore::RenderContainer::invalidateCounters): Added. Invalidates all RenderCounters in :before and :after content.
  • rendering/RenderContainer.h:
  • rendering/RenderCounter.cpp: (WebCore::RenderCounter::isCounter): Renamed isRenderCounter() to this to match the RenderObject method. (WebCore::RenderCounter::invalidate): Added. Resets the cached CounterNode and invalidates the object's layout and preferred widths. (WebCore::destroyCounterNodeChildren): Added a call to invalidateCounters().
  • rendering/RenderCounter.h:
  • rendering/RenderObject.h: (WebCore::RenderObject::invalidateCounters):

LayoutTests:

Reviewed by Darin Adler.

  • test for <rdar://problem/6131096> Reproducible crash in CounterNode::isReset under guard malloc
  • fast/css/counters/invalidate-cached-counter-node.html: Added.
  • platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.checksum: Added.
  • platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.png: Added.
  • platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.txt: Added.
10:24 AM Changeset in webkit [35666] by dino@apple.com
  • 19 edits
    7 adds in trunk/WebCore

Implement CSS Animation and Transition Events
https://bugs.webkit.org/show_bug.cgi?id=20337

Implement the events defined in the CSS Animations
and Transitions specifications so code can react
to animations and transitions.

Reviewed by Tim Hatcher and Dave Hyatt.

  • DerivedSources.make:
  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:

Build configs for new files

  • bindings/js/JSDOMWindowBase.cpp:
  • bindings/js/JSDOMWindowBase.h:
  • bindings/js/JSEventCustom.cpp:
  • dom/Document.h:
  • dom/Event.cpp:
  • dom/Event.h:
  • dom/EventTarget.cpp:
  • dom/EventTargetNode.cpp:
  • dom/EventTargetNode.h:
  • html/HTMLElement.cpp:
  • page/AnimationController.cpp:

do all the new event stuff

  • html/HTMLAttributeNames.in:

the onwebkitanimation* and onwebkittransitionend attrs

  • dom/WebKitAnimationEvent.cpp: Added.
  • dom/WebKitAnimationEvent.h: Added.
  • dom/WebKitAnimationEvent.idl: Added.
  • dom/WebKitTransitionEvent.cpp: Added.
  • dom/WebKitTransitionEvent.h: Added.
  • dom/WebKitTransitionEvent.idl: Added.

New files for the events

  • manual-tests/transition-events.html: Added.

New testfile

10:21 AM Changeset in webkit [35665] by Adam Roben
  • 5 edits
    2 adds in trunk

Move WTF::notFound into its own header so that it can be used independently of Vector

JavaScriptCore:

Move WTF::notFound into its own header so that it can be used
independently of Vector

Rubberstamped by Darin Adler.

  • JavaScriptCore.vcproj/WTF/WTF.vcproj:
  • JavaScriptCore.xcodeproj/project.pbxproj: Added NotFound.h to the project.
  • wtf/NotFound.h: Added. Moved the notFound constant here...
  • wtf/Vector.h: ...from here.

WebCore:

Add a ForwardingHeader for wtf/NotFound.h

Rubberstamped by Darin Adler.

  • ForwardingHeaders/wtf/NotFound.h: Added.
9:53 AM Changeset in webkit [35664] by timothy@apple.com
  • 2 edits
    2 adds in trunk/WebCore

Fixes a bug where error bubbles in JavaScript resources would
be clobbered by the syntax highlighter.

https://bugs.webkit.org/show_bug.cgi?id=20345

Reviewed by Adam Roben.

  • manual-tests/inspector/resources/script-error.js: Added.
  • manual-tests/inspector/styled-error-bubbles-in-scripts.html: Added.
  • page/inspector/SourceFrame.js: (WebInspector.SourceFrame.prototype._addMessageToSource): Check the nodeType and not the nodeName, this is less fragile. (WebInspector.SourceFrame.prototype._syntaxHighlightJavascriptLine): Check if the lastChild is an error bubble, if so remove it before getting the line's textContent. Add the error bubble back at the end.
5:01 AM Changeset in webkit [35663] by ap@webkit.org
  • 3 edits in trunk/JavaScriptCore

Reviewed by Mark Rowe.

<rdar://problem/6130393> REGRESSION: PhotoBooth hangs after launching under TOT Webkit

  • API/JSContextRef.cpp: (JSGlobalContextRelease): Corrected a comment.
  • kjs/collector.cpp: (KJS::Heap::~Heap): Ensure that JSGlobalData is not deleted while sweeping the heap.

Aug 10, 2008:

10:07 PM Changeset in webkit [35662] by mrowe@apple.com
  • 3 edits
    3 copies in trunk

Roll over three large ChangeLogs.

9:51 PM Changeset in webkit [35661] by mitz@apple.com
  • 2 edits in trunk/WebCore

Reviewed by Eric Seidel.

  • dom/make_names.pl: Removed "using namespace WebCore::*Names" from *ElementFactory.cpp files.
7:17 PM Changeset in webkit [35660] by mrowe@apple.com
  • 4 edits in trunk/WebCore

Fix crash in Acid3 in Windows and Gtk ports, introduced in r35590.

Reviewed by Cameron Zwarich.

  • bindings/js/ScriptControllerGtk.cpp: Fix the logic.
  • bindings/js/ScriptControllerWin.cpp: Ditto.
  • bindings/js/ScriptControllerQt.cpp: Switch to isPluginView for consistency with Gtk and Windows.
3:36 PM Changeset in webkit [35659] by mitz@apple.com
  • 3 edits
    2 adds in trunk

WebCore:

Reviewed by Darin Adler.

  • fix <rdar://problem/6074587> REGRESSION (34722): Unable to view AT&T wireless bill - window title contains markup, window is empty

Test: fast/tokenizer/ampersand-in-special-tag.html

  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::parseSpecial): Changed to only advance lastDecodedEntityPosition if an entity was found and decoded.

LayoutTests:

Reviewed by Darin Adler.

  • test for <rdar://problem/6074587> REGRESSION (34722): Unable to view AT&T wireless bill - window title contains markup, window is empty
  • fast/tokenizer/ampersand-in-special-tag-expected.txt: Added.
  • fast/tokenizer/ampersand-in-special-tag.html: Added.
11:21 AM Changeset in webkit [35658] by alp@webkit.org
  • 9 edits
    3 deletes in trunk

2008-08-10 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed (and updated) by Alp Toker.

https://bugs.webkit.org/show_bug.cgi?id=16620
[GTK] Autotools make dist and make check support

Get make dist working.

Note that not all possible configurations have been tested yet.

  • GNUmakefile.am:

2008-08-10 Alp Toker <alp@nuanti.com>

Remove leftover qmake/GTK+ build files.

  • DumpRenderTree/gtk/DumpRenderTree.pro: Removed.
  • GtkLauncher/GtkLauncher.pro: Removed.
3:06 AM Changeset in webkit [35657] by ap@webkit.org
  • 5 edits in trunk/JavaScriptCore

Reviewed by Sam Weinig.

Added same heap debug checks to more code paths.

  • kjs/JSActivation.cpp: (KJS::JSActivation::put): (KJS::JSActivation::putWithAttributes):
  • kjs/JSGlobalObject.cpp: (KJS::JSGlobalObject::putWithAttributes):
  • kjs/JSObject.h: (KJS::JSObject::putDirect):
  • kjs/JSVariableObject.h: (KJS::JSVariableObject::symbolTablePut): (KJS::JSVariableObject::symbolTablePutWithAttributes):

Aug 9, 2008:

10:32 PM CygwinInstallerPackages created by wilsong@gmail.com
A list of all the packages included in the cygwin installer for …
10:19 PM BuildingOnWindows edited by wilsong@gmail.com
Adding a link to a new list of all packages included in the cygwin … (diff)
8:11 PM Changeset in webkit [35656] by alp@webkit.org
  • 2 edits in trunk

2008-08-09 Marco Barisione <marco.barisione@collabora.co.uk>

Reviewed by Alp Toker.

http://bugs.webkit.org/show_bug.cgi?id=19045
[gtk] ./configure doesn't check for x toolkit availability

Some old versions of Xt do not provide xt.pc, in this case use
AC_CHECK_LIB to check if the library is installed and then fall back
to using just -lXt.

  • configure.ac:
8:00 PM Changeset in webkit [35655] by alp@webkit.org
  • 2 edits in trunk

2008-08-09 Alp Toker <alp@nuanti.com>

GTK+ build fixes.

Back out r20338 libxslt check (conditional check already exists and
this broke the build).

Make the flex version check a warning rather than fatal to fix the
build bot.

  • configure.ac:
7:03 PM Changeset in webkit [35654] by jmalonzo@webkit.org
  • 2 edits in trunk

2008-08-08 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed by Mark Rowe.

http://bugs.webkit.org/show_bug.cgi?id=20338
[Gtk] Autotools should check for libxslt

Autotools check for libxslt

  • configure.ac:
2:14 PM Changeset in webkit [35653] by cwzwarich@webkit.org
  • 2 edits in trunk/JavaScriptCore

2008-08-09 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Maciej.

Fix some style issues in the sampling tool.

  • VM/SamplingTool.cpp: (KJS::sleepForMicroseconds): (KJS::SamplingTool::dump):
4:00 AM Changeset in webkit [35652] by cwzwarich@webkit.org
  • 5 edits in trunk/JavaScriptCore

2008-08-09 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Oliver.

Revision 35651, despite being a rather trivial change, introduced a
large regression on the regexp-dna SunSpider test. This regression
stemmed from an increase in the size of CodeBlock::dump(). There is
no reason for this method (and several related methods) to be compiled
in non-debug builds with the sampling tool disabled. This patch
conditionally compiles them, reversing the regression on SunSpider.

  • JavaScriptCore.exp:
  • VM/CodeBlock.cpp:
  • VM/CodeBlock.h:
  • VM/Machine.cpp:

Aug 8, 2008:

8:56 PM Changeset in webkit [35651] by cwzwarich@webkit.org
  • 4 edits
    3 adds in trunk

2008-08-08 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Oliver.

Bug 20330: JSCore crash loading any filehurricane media page
<https://bugs.webkit.org/show_bug.cgi?id=20330>

Fix a typo in the constant loading patch. Also, add a case for
op_unexpected_load to CodeBlock::dump().

JavaScriptCore:

  • VM/CodeBlock.cpp: (KJS::CodeBlock::dump):
  • VM/CodeGenerator.cpp: (KJS::CodeGenerator::addUnexpectedConstant):

LayoutTests:

  • fast/js/resources/unexpected-constant-crash.js: Added.
  • fast/js/unexpected-constant-crash-expected.txt: Added.
  • fast/js/unexpected-constant-crash.html: Added.
8:37 PM Changeset in webkit [35650] by mitz@apple.com
  • 2 edits in trunk/WebKit/mac

Reviewed by John Sullivan.

  • fix <rdar://problem/6130216> Exception "windowRegionBeingDrawn != nil" in NSView when caching image of a subframe

This change reintroduces <https://bugs.webkit.org/show_bug.cgi?id=5195> on Leopard.

  • WebView/WebHTMLView.mm: (-[WebHTMLView _recursive:displayRectIgnoringOpacity:inContext:topView:]): Removed an unsuccessful workaround for <rdar://problem/5668489>, because invoking layout may change the view hierarchy during the drawing operation, which is not supported on Leopard.
6:05 PM Changeset in webkit [35649] by Beth Dakin
  • 4 edits in trunk/WebCore

2008-08-08 Beth Dakin <Beth Dakin>

Reviewed by Darin.

This patch makes the findString and markAllMatchesForText functions
work with disconnected frames that are contained within
overflow:hidden blocks.

  • editing/Editor.cpp: (WebCore::Editor::rangeVisibility): (WebCore::Editor::firstVisibleRange): (WebCore::Editor::lastVisibleRange):
  • editing/Editor.h: (WebCore::Editor::):
  • page/Frame.cpp: (WebCore::Frame::findString): (WebCore::Frame::markAllMatchesForText):
5:15 PM Changeset in webkit [35648] by britto@apple.com
  • 15 edits
    2 adds in trunk

2008-08-08 Maxime Britto <britto@apple.com>

Reviewed by Adele.

Test: fast/events/scroll-to-anchor-in-overflow-hidden.html
https://bugs.webkit.org/show_bug.cgi?id=20270
Jump to an anchor wasn't working when the overflow:hidden CSS attribute was set on the page.

  • WebCore.base.exp:
  • dom/Element.cpp: (WebCore::Element::scrollIntoView): (WebCore::Element::scrollIntoViewIfNeeded):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::gotoAnchor):
  • page/EventHandler.cpp: (WebCore::EventHandler::handleMousePressEvent): (WebCore::EventHandler::handleMouseDraggedEvent):
  • page/Frame.cpp: (WebCore::Frame::revealSelection): (WebCore::Frame::revealCaret):
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::scrollRectToVisible): Takes a new parameter "scrollToAnchor" since it's a special case of scroll. It passes this new bool to canBeProgrammaticallyScrolled. (WebCore::RenderLayer::autoscroll):
  • rendering/RenderLayer.h:
  • rendering/RenderListBox.h: (WebCore::RenderListBox::canBeProgramaticallyScrolled):
  • rendering/RenderObject.cpp: (WebCore::RenderObject::canBeProgramaticallyScrolled): If the call is coming from a scrollToAnchor we don't need scrollBars to accept.
  • rendering/RenderObject.h:
  • rendering/RenderTextControl.h: (WebCore::RenderTextControl::canBeProgramaticallyScrolled):

2008-08-08 Maxime Britto <britto@apple.com>

Reviewed by Adele.

  • WebView/WebFrame.mm: (-[WebFrame _scrollDOMRangeToVisible:]):

2008-08-08 Maxime Britto <britto@apple.com>

Reviewed by Adele.
https://bugs.webkit.org/show_bug.cgi?id=20270
This test is automatic and can also be run manually.

  • fast/events/scroll-to-anchor-in-overflow-hidden-expected.txt: Added.
  • fast/events/scroll-to-anchor-in-overflow-hidden.html: Added.
4:40 PM Changeset in webkit [35647] by britto@apple.com
  • 2 edits in trunk/WebCore

2008-08-08 Maxime Britto <britto@apple.com>

Reviewed by Adele.

<rdar://problem/6137022>
<https://bugs.webkit.org/show_bug.cgi?id=20331>
REGRESSION(r35177-r35203): Scrolling slowly over a scrollable field scrolls the page instead of the field (20331)

  • page/EventHandler.cpp: (WebCore::scrollAndAcceptEvent): fixed the wrong cast for the wheel delta which was truncated to 0 on slow wheel scrolls.
1:44 PM Changeset in webkit [35646] by dino@apple.com
  • 7 edits
    3 adds in trunk/WebCore

Execute CSS Animations using new AnimationController
https://bugs.webkit.org/show_bug.cgi?id=20119

I still need to file some follow-on bugs that EricS noticed.

Reviewed by Dave Hyatt

Some new tests:

  • manual-tests/animate-duration.html: Added.
  • manual-tests/animate-left.html: Added.
  • manual-tests/animate-none.html: Added.
  • css/CSSStyleSelector.cpp:
  • dom/Document.cpp:
    • don't redo style when page is place into cache
  • page/AnimationController.cpp:
  • page/AnimationController.h:
    • execute the animations
  • rendering/RenderObject.cpp:
  • rendering/style/RenderStyle.h:
    • make border radii const (for blending)
1:25 PM Changeset in webkit [35645] by kmccullough@apple.com
  • 5 edits in trunk/WebCore

2008-08-08 Kevin McCullough <kmccullough@apple.com>

Reviewed by Tim.

<rdar://problem/6136326> JSProfiler: Start/stop profiling button in the
Web Inspector does not match the menu.

  • page/InspectorController.cpp: (WebCore::InspectorController::toggleRecordButton): Toggle the record button in the webInspector. (WebCore::InspectorController::startUserInitiatedProfiling): (WebCore::InspectorController::stopUserInitiatedProfiling):
  • page/InspectorController.h:
  • page/inspector/ProfilesPanel.js: Move the UI part of the record button into its funtion so that it can be called from the Inspector controller.
  • page/inspector/inspector.js:
11:57 AM Changeset in webkit [35644] by pewtermoose@webkit.org
  • 2 edits in trunk/JavaScriptCore

2008-08-08 Matt Lilek <webkit@mattlilek.com>

Not reviewed, build fix.

11:48 AM Changeset in webkit [35643] by beidson@apple.com
  • 2 edits in trunk/WebKit/mac

2008-08-08 Brady Eidson <beidson@apple.com>

Reviewed by Sam Weinig

Fix for <rdar://problem/5865504>

This bug was actually *largely* fixed by http://trac.webkit.org/changeset/35538.
But with that same patch, it became possible for a WebResource to fail to initialize.
Therefore we were trying to add nil to an NSCFArray for certain situations, which is bad.
Lets fix that, shall we?

  • WebView/WebArchive.mm: (-[WebArchive subresources]):
11:21 AM Changeset in webkit [35642] by kmccullough@apple.com
  • 3 edits in trunk/WebCore

2008-08-08 Kevin McCullough <kmccullough@apple.com>

Reviewed by Tim.

<rdar://problem/6131187> Profiles don't show up immediately when run
with the UI.

  • page/InspectorController.cpp: (WebCore::startProfiling): Added so that the inspector's JS code could call through to the same function used by the Develop menu. This way both can call didFinishAllExecution() which puts the profile in the Web Inspector. (WebCore::stopProfiling): Ditto (WebCore::InspectorController::windowScriptObjectAvailable): (WebCore::InspectorController::stopUserInitiatedProfiling): Now calls didFinishAllExecution() to put the profile in the Web Inspector. This should be safe since the UIs shouldn't work if JS is still running.
  • page/inspector/ProfilesPanel.js: Use the new functions.
10:38 AM Changeset in webkit [35641] by mitz@apple.com
  • 2 edits in trunk/WebCore

Reviewed by Darin Adler.

  • fix <rdar://problem/5961260> GDI text: transparent fill color text does not render if it it has a stroke or shadow
  • platform/graphics/win/FontCGWin.cpp: (WebCore::Font::drawGlyphs):
5:59 AM Changeset in webkit [35640] by jchaffraix@webkit.org
  • 11 edits
    7 adds in branches/XBL2

WebCore:

2008-08-08 Julien Chaffraix <jchaffraix@webkit.org>

Reviewed by Eric.

Tests: fast/xbl/xbl-crash-inline.xhtml

fast/xbl/xbl-document-sheets-modification.xhtml
fast/xbl/xbl-element-binding.xhtml

Bug 20212: [XBL] Add full support for element="" attachment

  • Add attributeChanged method to XBLBindingElement to catch "element" changes and update the bindings.
  • Translate "element" attribute to a stylesheet that is added to the bound document's sheets.
  • GNUmakefile.am: Add XBLBindingElement.cpp.
  • WebCore.pro: Ditto.
  • WebCore.xcodeproj/project.pbxproj: Ditto.
  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::CSSStyleSelector): Add binding sheets.
  • dom/Document.cpp: (WebCore::Document::~Document): Remove binding sheet information.

(WebCore::Document::recalcStyleSelector): When
recalculating the style selector, remove all the bindings
as the removed sheets would never remove theirs.

(WebCore::Document::addBindingSheet): Update binding manager
information.
(WebCore::Document::removeBindingSheet): Ditto.

  • dom/Document.h:
  • xbl/XBLBindingElement.cpp: Added. (WebCore::XBLBindingElement::XBLBindingElement): (WebCore::XBLBindingElement::attributeChanged): Check for "element" changes and update the cached sheet and the bound element's bindings style sheets.

(WebCore::XBLBindingElement::generateBindingSheet): Create
the style sheet from the "element" attribute.

  • xbl/XBLBindingElement.h:
  • xbl/XBLBindingManager.cpp: (WebCore::XBLBindingManager::addBindingSheet): (WebCore::XBLBindingManager::removeBindingSheet): (WebCore::XBLBindingManager::clearBindingSheets): (WebCore::XBLBindingManager::getBindingSheets): All 4 previous methods are used to manage the binding sheets.
  • xbl/XBLBindingManager.h:

LayoutTests:

2008-08-08 Julien Chaffraix <jchaffraix@webkit.org>

Reviewed by Eric.

Test case for bug 20212: [XBL] Add full support for element="" attachment


  • fast/xbl/xbl-crash-inline-expected.txt: Added.
  • fast/xbl/xbl-crash-inline.xhtml: Added.
  • fast/xbl/xbl-document-sheets-modification-expected.txt: Added.
  • fast/xbl/xbl-document-sheets-modification.xhtml: Added.
  • fast/xbl/xbl-element-binding-expected.txt: Added.
  • fast/xbl/xbl-element-binding.xhtml: Added.


2:48 AM Changeset in webkit [35639] by oliver@apple.com
  • 5 edits in trunk/JavaScriptCore

Improve performance of arithmetic operators

Reviewed by Cameron Zwarich

Added a fast (non-virtual) mechanism to determine if a non-immediate JSValue*
is a JSNumberCell. We then use this to allow improved specialisation in many
arithmetic operators. SunSpider reports a 2.5% progression overall, with greater
than 10% progressions on a number of arithmetic heavy tests.

2:32 AM Changeset in webkit [35638] by Simon Hausmann
  • 2 edits in trunk/WebCore

2008-08-08 Adam Treat <treat@kde.org>

Reviewed by Simon.

Return the canonical file path in Qt's listDirectory as WebCore::PluginDatabase
uses this method to retrieve a list of possible plugins. Many linux systems
have nested symbolic links scattered throughout the plugin directories which
resulted in the PluginDatabase loading the same plugin N times depending on
the number of symbolic links.

1:21 AM Changeset in webkit [35637] by jchaffraix@webkit.org
  • 5 edits in branches/XBL2/WebCore

008-08-08 Julien Chaffraix <jchaffraix@webkit.org>

Reviewed by Eric.

Re-enable last bits of XBLDocument loading
legacy code.

  • loader/DocLoader.cpp: (WebCore::DocLoader::requestXBLDocument): (WebCore::DocLoader::requestResource):
  • loader/DocLoader.h:
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::addLowBandwidthDisplayRequest):
  • xbl/XBLDocument.h: (WebCore::XBLDocument::create): Removed adoptRef to match other Document 'create' method as it would lead to refcount churn. Also added missing 'const' in isXBLDocument.
Note: See TracTimeline for information about the timeline view.