⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Jan 16, 2010:

11:18 PM Changeset in webkit [53371] by mjs@apple.com
  • 12 edits in trunk

2010-01-16 Maciej Stachowiak <​mjs@apple.com>

Reviewed by Oliver Hunt.

Cache JS string values made from DOM strings (Dromaeo speedup)
​https://bugs.webkit.org/show_bug.cgi?id=33768
<rdar://problem/7353576>

  • runtime/JSString.h: (JSC::jsStringWithFinalizer): Added new mechanism for a string to have an optional finalizer callback, for the benefit of weak-referencing caches. (JSC::): (JSC::Fiber::JSString): (JSC::Fiber::~JSString):
  • runtime/JSString.cpp: (JSC::JSString::resolveRope): Clear fibers so this doesn't look like a string with a finalizer.
  • runtime/WeakGCMap.h: Include "Collector.h" to make this header includable by itself.

2010-01-16 Maciej Stachowiak <​mjs@apple.com>

Reviewed by Oliver Hunt.

Cache JS string values made from DOM strings (Dromaeo speedup)
​https://bugs.webkit.org/show_bug.cgi?id=33768
<rdar://problem/7353576>


  • Plugins/Hosted/ProxyInstance.mm: (WebKit::ProxyInstance::stringValue): Explicitly make a String, since char* is now ambiguous.

2010-01-16 Maciej Stachowiak <​mjs@apple.com>

Reviewed by Oliver Hunt.

Cache JS string values made from DOM strings (Dromaeo speedup)
​https://bugs.webkit.org/show_bug.cgi?id=33768
<rdar://problem/7353576>

Added a new cache for JSString values that are created from Strings or AtomicStrings
in the DOM. It's common for the same string to be retrieved from the DOM repeatedly,
and it is wasteful to make a new JS-level string value every time.


The string cache is per-world, and thus thread-safe and not a
vector for accidental information exchange.


~30% speedup on Dromaeo Attributes test, also substantially helps other Dromaeo DOM tests.

  • bindings/js/JSDOMBinding.cpp: (WebCore::jsStringCache): Helper function to get the string cache for the current world. (WebCore::jsString): Some new overloads including the caching version. (WebCore::stringWrapperDestroyed): Finalizer callback - remove from relevant caches.
  • bindings/js/JSDOMBinding.h: (WebCore::jsString): Prototype new overloads (and define a few inline).
  • bindings/js/JSJavaScriptCallFrameCustom.cpp: (WebCore::JSJavaScriptCallFrame::type): Explicitly make a UString.
  • bindings/js/ScriptFunctionCall.cpp: (WebCore::ScriptFunctionCall::appendArgument): Ditto.
  • WebCore.base.exp: Add new JSString overloads that WebCore gets to see.
6:31 PM Changeset in webkit [53370] by oliver@apple.com
  • 2 edits in trunk/WebCore

2010-01-16 Oliver Hunt <​oliver@apple.com>

Reviewed by Nikolas Zimmermann.

SVG Markers are getting culled incorrectly
​https://bugs.webkit.org/show_bug.cgi?id=33767

Use applyTransformToPaintInfo to transform the paintInfo, otherwise
the paint rect is not updated leading to incorrect culling.

  • svg/graphics/SVGResourceMarker.cpp: (WebCore::SVGResourceMarker::draw):
5:42 PM Changeset in webkit [53369] by jhoneycutt@apple.com
  • 3 edits
    2 adds in trunk

MSAA: Screen rect for <option> elements is always the zero rect

​https://bugs.webkit.org/show_bug.cgi?id=33758

Reviewed by Oliver Hunt.

WebKit/win:

  • AccessibleBase.cpp:

(AccessibleBase::accLocation):
Use elementRect() rather than boundingBoxRect(), which
AccessibilityListBoxOption overrides.

LayoutTests:

  • platform/win/accessibility/option-element-position-and-size-expected.txt: Added.
  • platform/win/accessibility/option-element-position-and-size.html: Added.
5:42 PM Changeset in webkit [53368] by jhoneycutt@apple.com
  • 5 edits
    2 adds in trunk

get_accParent should try to retrieve parent AccessibilityObject, before
calling upon window

​https://bugs.webkit.org/show_bug.cgi?id=22893

Reviewed by Darin Adler.

WebKit/win:

  • AccessibleBase.cpp:

(AccessibleBase::get_accParent):
If the object has a parent object, return it. If not, return the
accessible for the WebView window.

WebKitTools:

  • DumpRenderTree/win/AccessibilityUIElementWin.cpp:

(AccessibilityUIElement::parentElement):
Get the object's parent. Query it for IAccessible, and return it.

LayoutTests:

  • platform/win/accessibility/parent-element-expected.txt: Added.
  • platform/win/accessibility/parent-element.html: Added.
5:03 PM Changeset in webkit [53367] by sullivan@apple.com
  • 3 edits in trunk/WebCore

​https://bugs.webkit.org/show_bug.cgi?id=33751 and <rdar://problem/7538330>
Zip code field is misidentified as street address because id attribute isn't checked.

Reviewed by Darin Adler

No new tests. I believe this code is used only by Safari AutoFill, but in any case
it does not affect page rendering or anything else at the WebCore/WebKit level.

  • page/Frame.cpp:

(WebCore::matchLabelsAgainstString):
New function, split out from matchLabelsAgainstElement.
(WebCore::Frame::matchLabelsAgainstElement):
Now calls matchLabelsAgainstString for the id attribute if no match is found for the name attribute.

  • page/mac/FrameMac.mm:

(WebCore::matchLabelsAgainstString):
Same as above. This is a parallel copy of the function using Mac-specific data structures.
(WebCore::Frame::matchLabelsAgainstElement):
Ditto.

3:30 PM Changeset in webkit [53366] by timothy@apple.com
  • 8 edits in trunk/WebCore

Use String.trim() instead of a regex in the Web Inspector.

​https://bugs.webkit.org/show_bug.cgi?id=33765

Reviewed by George Staikos.

  • inspector/front-end/ElementsPanel.js:

(WebInspector.ElementsPanel.prototype.performSearch): Use .trim().

  • inspector/front-end/ObjectPropertiesSection.js:

(WebInspector.ObjectPropertyTreeElement.prototype.applyExpression): Ditto.

  • inspector/front-end/ProfileView.js:
  • inspector/front-end/SourceFrame.js:

(WebInspector.SourceFrame.prototype._evalSelectionInCallFrame): Ditto.

  • inspector/front-end/StylesSidebarPane.js:

(WebInspector.StylePropertyTreeElement.prototype): Ditto.

  • inspector/front-end/utilities.js:

(String.prototype.trimLeadingWhitespace): Removed.
(String.prototype.trimTrailingWhitespace): Removed.
(String.prototype.trimWhitespace): Removed.

3:04 PM Changeset in webkit [53365] by oliver@apple.com
  • 15 edits in trunk/WebCore

2010-01-16 Oliver Hunt <​oliver@apple.com>

Reviewed by Nikolas Zimmermann.

Copying TransformationMatrix consumed a lot of cpu time while scroll with cursor over content
​https://bugs.webkit.org/show_bug.cgi?id=33766

Make localToParentTransform return by reference to avid copy overhead.
This is a little gnarly in places as it means we need to be able to
return temporary values in a few implementations, so we have to add
class fields to hold them, heppily the classes that these effect are

sufficiently uncommon for this to be okay.

  • rendering/RenderForeignObject.cpp: (WebCore::RenderForeignObject::localToParentTransform):
  • rendering/RenderForeignObject.h:
  • rendering/RenderObject.cpp: (WebCore::RenderObject::localTransform): (WebCore::RenderObject::localToParentTransform):
  • rendering/RenderObject.h:
  • rendering/RenderPath.cpp: (WebCore::RenderPath::localToParentTransform): (WebCore::RenderPath::nodeAtFloatPoint):
  • rendering/RenderPath.h:
  • rendering/RenderSVGImage.h: (WebCore::RenderSVGImage::localToParentTransform):
  • rendering/RenderSVGRoot.cpp: (WebCore::RenderSVGRoot::localToParentTransform):
  • rendering/RenderSVGRoot.h:
  • rendering/RenderSVGText.h: (WebCore::RenderSVGText::localToParentTransform):
  • rendering/RenderSVGTransformableContainer.cpp: (WebCore::RenderSVGTransformableContainer::localToParentTransform):
  • rendering/RenderSVGTransformableContainer.h:
  • rendering/RenderSVGViewportContainer.cpp: (WebCore::RenderSVGViewportContainer::localToParentTransform):
  • rendering/RenderSVGViewportContainer.h:
1:36 PM Changeset in webkit [53364] by Darin Adler
  • 5 edits
    2 adds in trunk

ValidityState can hold a stale pointer to control
​https://bugs.webkit.org/show_bug.cgi?id=33729
rdar://problem/7545114

Reviewed by Oliver Hunt and Alexey Proskuryakov.

WebCore:

Test: fast/forms/ValidityState-removed-control.html

  • html/HTMLFormControlElement.h: Use OwnPtr instead of

RefPtr to point to the ValidityState object.

  • html/ValidityState.cpp: Use a constant instead of a

macro for the email validation regular expression.
(WebCore::ValidityState::validationMessage): Use
m_control instead of control function; we don't need
a function for this. Also marked const.
(WebCore::ValidityState::typeMismatch): Ditto.
Fixed some minor style problems.
(WebCore::ValidityState::rangeUnderflow): Ditto.
(WebCore::ValidityState::rangeOverflow): Ditto.
(WebCore::ValidityState::stepMismatch): Ditto.
(WebCore::ValidityState::valid): Ditto.
(WebCore::ValidityState::isValidEmailAddress):
Changed local variable names for clarity. Got rid of
an unneeded global variable.

  • html/ValidityState.h: Removed RefCounted as a base

class, deriving from Noncopyable instead. Changed
creation to use PassOwnPtr instead of PassRefPtr.
Eliminated unneeded control function. Added ref and
deref functions that forward the reference counting
to the control. Moved constructor here and made it
inline.

LayoutTests:

  • fast/forms/ValidityState-removed-control-expected.txt: Added.
  • fast/forms/ValidityState-removed-control.html: Added.
12:53 PM Changeset in webkit [53363] by ap@apple.com
  • 5 edits
    5 adds in trunk

Reviewed by Darin Adler.

​https://bugs.webkit.org/show_bug.cgi?id=33752
Assertion failure when getting a href attribute with prefix

This was due to a temporary change made in 2005 (r9639) - checks in checkSetPrefix were
disabled during introduction of QualifiedName, but never re-enabled.

Tests: fast/dom/bad-href-attribute.html

fast/dom/node-prefix-setter-namespace-exception.html

  • dom/Attr.cpp: (WebCore::Attr::setPrefix):
  • dom/Element.cpp: (WebCore::Element::setPrefix):
  • dom/Node.cpp: (WebCore::Node::checkSetPrefix): Re-enabled the checks. Also, changed the prefix setter to treat "" as null, matching Firefox (DOM 3 Core spec says this behavior is implementation defined).
11:29 AM Changeset in webkit [53362] by mitz@apple.com
  • 2 edits in trunk/WebKitTools

<rdar://problem/7529519> DumpRenderTree crashes in makeLargeMallocFailSilently()

Reviewed by Mark Rowe.

  • DumpRenderTree/mac/CheckedMalloc.cpp:

(makeLargeMallocFailSilently): Temporarily allow writing to the default
malloc zone structure while modifying it.

11:18 AM Changeset in webkit [53361] by beidson@apple.com
  • 10 edits in trunk

<rdar://problem/7536748> and ​https://bugs.webkit.org/show_bug.cgi?id=33571
History traversals to a new document do not get the popstate event

Reviewed by Darin Adler.

WebCore:

State objects now live-on in their HistoryItem indefinitely.
This means any back/forward navigation might result in a popstate event, not just to
pre-existing documents as was previously the case.

  • history/HistoryItem.cpp:

(WebCore::HistoryItem::documentDetached): State objects are held beyond Document lifetime.

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::begin): If there is a pending state object for this Frame load, pass it

on to the Document after the Document is created.

(WebCore::FrameLoader::transitionToCommitted): If the current history item has a state object,

set it as the FrameLoad's pending state object.

  • loader/FrameLoader.h:


  • page/Page.cpp:

(WebCore::Page::goToItem): Remove a now-invalid ASSERT.

LayoutTests:

Updated existing tests to match new correct behavior.

  • fast/loader/stateobjects/document-destroyed-navigate-back-expected.txt:
  • fast/loader/stateobjects/document-destroyed-navigate-back-with-fragment-scroll-expected.txt:
  • fast/loader/stateobjects/document-destroyed-navigate-back-with-fragment-scroll.html:
  • fast/loader/stateobjects/document-destroyed-navigate-back.html:
7:20 AM Changeset in webkit [53360] by ddkilzer@apple.com
  • 3 edits in trunk/WebKitTools

Added back removed properties to iExploder input files

While the update-iexploder-cssproperties script works great to
add new properties, it has the side-effect of removing old
properties that once were parsed.

  • iExploder/htdocs/cssproperties.in: Added back removed

properties into their own section.

  • iExploder/htdocs/htmlattrs.in: Ditto.
6:58 AM Changeset in webkit [53359] by ddkilzer@apple.com
  • 1 edit in trunk/WebKitTools/ChangeLog

Added bug url to ChangeLog for r53358.

6:44 AM Changeset in webkit [53358] by eric@webkit.org
  • 4 edits in trunk/WebKitTools

2010-01-16 Holger Hans Peter Freyther <​zecke@selfish.org>

Reviewed by David Kilzer.

Update iExploder/htdocs/*.in by running update-iexploder-cssproperties

  • iExploder/htdocs/cssproperties.in: New CSS attributes
  • iExploder/htdocs/htmlattrs.in: New HTML attributes
  • iExploder/htdocs/htmltags.in: New HTML tags
12:51 AM QtWebKitMirrorGuide edited by zecke@selfish.org
Mention the manipulate-content.py script for post-processing. (diff)
12:48 AM QtWebKitPerformanceUtilities edited by zecke@selfish.org
Add an example to post-process the db. (diff)
12:40 AM QtWebKitPerformanceToolBackLog: edited by zecke@selfish.org
Add some new commits (diff)
12:35 AM QtWebKitReleases edited by zecke@selfish.org
(diff)

Jan 15, 2010:

11:38 PM QtWebKitReleases edited by zecke@selfish.org
add some ideas on release engineering (diff)
9:07 PM Changeset in webkit [53357] by Simon Fraser
  • 13 edits
    72 adds in trunk

2010-01-15 Simon Fraser <Simon Fraser>

Reviewed by Dan Bernstein and Adele Peterson.

Support reflections on composited layers
​https://bugs.webkit.org/show_bug.cgi?id=31885

Implement reflections (via -webkit-box-reflect:) on compositing layers.

We add to the GraphicsLayer the notion of having a replica, and being a replicated layer.
The replica layer is not parented in the tree, but referenced by another layer.
RenderLayerBacking sets this up when it finds RenderLayers for reflections.

GraphicsLayerCA implements rendering of replica layers by cloning CA layers,
and copying their properties, including animations and contents. Deep reflections
are supported by a hash of clone layers on each GraphicsLayerCA, indexed by
the path down the tree to each replica instance.

When GraphicsLayerCA properties are changed, in most cases the clones must also
be updated.

Tests: compositing/masks/direct-image-mask.html

compositing/reflections/animation-inside-reflection.html
compositing/reflections/compositing-change-inside-reflection.html
compositing/reflections/deeply-nested-reflections.html
compositing/reflections/masked-reflection-on-composited.html
compositing/reflections/nested-reflection-anchor-point.html
compositing/reflections/nested-reflection-animated.html
compositing/reflections/nested-reflection-mask-change.html
compositing/reflections/nested-reflection-on-overflow.html
compositing/reflections/nested-reflection-opacity.html
compositing/reflections/nested-reflection-size-change.html
compositing/reflections/nested-reflection-transformed.html
compositing/reflections/nested-reflection-transition.html
compositing/reflections/nested-reflection.html
compositing/reflections/reflection-opacity.html
compositing/reflections/reflection-ordering.html
compositing/reflections/reflection-positioning.html
compositing/reflections/transform-inside-reflection.html

  • platform/graphics/GraphicsLayer.h: (WebCore::GraphicsLayer::isReplicated): Returns true when this layer has a replicated layer. (WebCore::GraphicsLayer::replicatedLayerPosition): (WebCore::GraphicsLayer::setReplicatedLayerPosition): The position of the replica layer must be special-cased; we cannot just copy the position of the original. (WebCore::GraphicsLayer::didDisplay): Method that indicates that the contents of the layer changed, which gives us a chance to update clone layers. (WebCore::GraphicsLayer::replicaLayer): reference to the replica layer. (WebCore::GraphicsLayer::replicatedLayer): reference to the layer that this (replica) layer is replicating. (WebCore::GraphicsLayer::setReplicatedLayer):
  • platform/graphics/GraphicsLayer.cpp: (WebCore::GraphicsLayer::GraphicsLayer): (WebCore::GraphicsLayer::setReplicatedByLayer): Hook up a replica with its replicated layer. (WebCore::GraphicsLayer::dumpProperties):
  • platform/graphics/mac/GraphicsLayerCA.h: (WebCore::GraphicsLayerCA::primaryLayer): Returns a CALayer, since structural layers may not be WebLayers. (WebCore::GraphicsLayerCA::isReplicatedRootClone): Given a cloneID (string representation of the path to a clone down the tree, which is a bitstring of 1 (replica), or 0 (non-replica)), returns true if this ID represents the root of a replica tree. (WebCore::GraphicsLayerCA::primaryLayerClones): Returns a pointer to the hash map of clones of the primary layers.

(WebCore::GraphicsLayerCA::ReplicaState::ReplicaState): Small struct used to track original/clone branching
down the tree during recursion, to build cloneID paths.

(WebCore::GraphicsLayerCA::hasCloneLayers): returns true if this layer has clone layers.

  • platform/graphics/mac/GraphicsLayerCA.mm: (WebCore::GraphicsLayerCA::~GraphicsLayerCA): remove the clone layers. (WebCore::GraphicsLayerCA::setChildren): call noteSublayersChanged() since we may have to update replicas too. (WebCore::GraphicsLayerCA::addChild): ditto (WebCore::GraphicsLayerCA::addChildAtIndex): ditto (WebCore::GraphicsLayerCA::addChildBelow): ditto (WebCore::GraphicsLayerCA::addChildAbove): ditto (WebCore::GraphicsLayerCA::replaceChild): ditto (WebCore::GraphicsLayerCA::removeFromParent): ditto (WebCore::GraphicsLayerCA::setMaskLayer): call propagateLayerChangeToReplicas() (WebCore::GraphicsLayerCA::setReplicatedLayer): note replica changed. (WebCore::GraphicsLayerCA::setReplicatedByLayer): ditto (WebCore::GraphicsLayerCA::moveOrCopyAllAnimationsForProperty): Enhanced to allow moving or copying animations. (WebCore::GraphicsLayerCA::moveOrCopyAnimationsForProperty): Ditto. (WebCore::GraphicsLayerCA::setContentsToImage): call noteSublayersChanged() (WebCore::GraphicsLayerCA::setContentsToVideo): call noteSublayersChanged() (WebCore::GraphicsLayerCA::didDisplay): here is our chance to copy updated contents to clone layers. (WebCore::GraphicsLayerCA::recursiveCommitChanges): (WebCore::GraphicsLayerCA::commitLayerChangesBeforeSublayers): pre-order commit, for things that need to be

committed before we recurse on children.

(WebCore::GraphicsLayerCA::commitLayerChangesAfterSublayers): post-order commit, for things that need to be

committed after we recurse on children, like clones.

(WebCore::GraphicsLayerCA::updateLayerNames): New method to match the other 'update' methods.
(WebCore::GraphicsLayerCA::updateSublayerList): Insert replica layers into the hierarchy.
(WebCore::GraphicsLayerCA::updateLayerPosition): update clones.
(WebCore::GraphicsLayerCA::updateLayerSize): ditto
(WebCore::GraphicsLayerCA::updateAnchorPoint): ditto
(WebCore::GraphicsLayerCA::updateTransform): ditto
(WebCore::GraphicsLayerCA::updateChildrenTransform): ditto
(WebCore::GraphicsLayerCA::updateMasksToBounds): ditto
(WebCore::GraphicsLayerCA::updateContentsOpaque): ditto
(WebCore::GraphicsLayerCA::updateBackfaceVisibility): ditto
(WebCore::GraphicsLayerCA::updateStructuralLayer): call ensureStructuralLayer()
(WebCore::moveAnimation): utility to move a CAAnimation from one layer to another.
(WebCore::GraphicsLayerCA::ensureStructuralLayer): refactored code which creates enclosing CALayers for reflection
flattening, or CATransformLayers for preserve-3d.
(WebCore::GraphicsLayerCA::structuralLayerPurpose): indicates why we need a structural layer.
(WebCore::GraphicsLayerCA::updateLayerDrawsContent): update clones
(WebCore::GraphicsLayerCA::updateContentsImage): ditto
(WebCore::GraphicsLayerCA::updateContentsRect): ditto
(WebCore::GraphicsLayerCA::updateMaskLayer): ditto
(WebCore::GraphicsLayerCA::updateReplicatedLayers): This is where we ask for the tree of layers for the replica
and its children, and attach them as sublayers.
(WebCore::GraphicsLayerCA::ReplicaState::cloneID): Build a bitstring from the array of original/clone values; this
string serves to identify clones in the hash map.
(WebCore::GraphicsLayerCA::replicatedLayerRoot): Request the tree of clone layers, set its position and transform,
and return it.
(WebCore::GraphicsLayerCA::setAnimationOnLayer): update clones
(WebCore::GraphicsLayerCA::removeAnimationFromLayer): ditto
(WebCore::GraphicsLayerCA::pauseAnimationOnLayer): ditto
(WebCore::GraphicsLayerCA::setContentsToGraphicsContext3D): udpate sublayers.
(WebCore::GraphicsLayerCA::suspendAnimations): update clones.
(WebCore::GraphicsLayerCA::resumeAnimations): ditto
(WebCore::GraphicsLayerCA::animatedLayerClones): return the hash map for clones of the appropriate layer for the given property.
(WebCore::GraphicsLayerCA::ensureCloneLayers): create and return clones for the CALayers for this layer.
(WebCore::GraphicsLayerCA::removeCloneLayers): clear out the clone layers.
(WebCore::GraphicsLayerCA::positionForCloneRootLayer): the root of a clonal subtree needs its position and transform to be special-cased,
since it doesn't just copy those properties from the original.
(WebCore::GraphicsLayerCA::propagateLayerChangeToReplicas): push the change flags onto the replica.
(WebCore::GraphicsLayerCA::fetchCloneLayers): recurse down sublayers, creating clones of the CALayers along the way, and returning
the root of the clone tree.
(WebCore::copyAnimation): utility to copy an animation from one layer to another. Animations can be shared between layers.
(WebCore::GraphicsLayerCA::cloneLayer): utility to clone a CALayer, copying those properties which GraphicsLayerCA makes use of
(WebCore::GraphicsLayerCA::setOpacityInternal): push opacity changes to clones.
(WebCore::GraphicsLayerCA::updateOpacityOnLayer): ditto
(WebCore::GraphicsLayerCA::noteSublayersChanged): set the ChildrenChanged flag, and proprate changes to the replica, if any.

  • platform/graphics/mac/WebLayer.mm: (-[WebLayer display]): override -display so we know when to update the contents of clone layers
  • platform/graphics/mac/WebTiledLayer.mm: (-[WebTiledLayer display]): ditto.
  • rendering/RenderLayer.h: (WebCore::RenderLayer::isReflection): New method that returns true if the renderer is a replica.
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::RenderLayer): initialize m_isReflection (WebCore::RenderLayer::updateReflectionStyle): call setIsReflection
  • rendering/RenderLayerBacking.cpp: (WebCore::RenderLayerBacking::createGraphicsLayer): Put a name on the reflection layer. (WebCore::RenderLayerBacking::updateGraphicsLayerConfiguration): Hook up the GraphicsLayers for the reflection. (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Reflection overrides preserve-3d (you have to flatten to reflect). Also hook up updating the reflection layer geometry, and the relica position.

(WebCore::RenderLayerBacking::paintIntoLayer): We no longer paint the reflection in software.

  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::computeCompositingRequirements): Hook reflection layers into the compositing logic. (WebCore::RenderLayerCompositor::canAccelerateVideoRendering): No longer have to push video into software if it's reflected. (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree): Update the bounds of the reflection layer. (WebCore::RenderLayerCompositor::updateCompositingDescendantGeometry): ditto (WebCore::RenderLayerCompositor::requiresCompositingWhenDescendantsAreCompositing): a compositing descendant forces a reflection ancestor to composite now. (WebCore::RenderLayerBacking::containsPaintedContent): Reflection layers don't paint anything. (WebCore::RenderLayerBacking::isDirectlyCompositedImage): No need to fall out of direct compositing mode for masks or reflections any more. (WebCore::RenderLayerBacking::paintIntoLayer): No need to paint the reflection manually now.
  • rendering/RenderObject.h: (WebCore::RenderObject::isReplica):
  • rendering/RenderReplica.h: (WebCore::RenderReplica::isReplica): New method used to determine if a render is a replica.
9:05 PM Changeset in webkit [53356] by eric@webkit.org
  • 4 edits in trunk/LayoutTests

2010-01-15 Doug Muir <​dmuir@google.com>

Fix some Layout Test timeouts when run under valgrind.

  • http/tests/history/redirect-js-form-submit-0-seconds.html:
  • http/tests/history/redirect-js-form-submit-2-seconds.html:
  • http/tests/misc/timer-vs-loading.html:
6:14 PM Changeset in webkit [53355] by eric@webkit.org
  • 6 edits
    2 adds in trunk

2010-01-15 Carol Szabo <​carol.szabo@nokia.com>

Reviewed by Darin Adler.

CSS2.1 Counters not updated when new elements are inserted in the DOM.
​https://bugs.webkit.org/show_bug.cgi?id=32884

  • fast/css/counters/adding-nodes-expected.txt: Added.
  • fast/css/counters/adding-nodes.html: Added.

2010-01-15 Carol Szabo <​carol.szabo@nokia.com>

Reviewed by Darin Adler.

CSS2.1 Counters not updated when new elements are inserted in the DOM.
​https://bugs.webkit.org/show_bug.cgi?id=32884

Test: fast/css/counters/adding-nodes.html

  • rendering/CounterNode.cpp: (WebCore::CounterNode::insertAfter): Modified to handle the addition of nodes with children. Needed when formerly root nodes become descendants of a new node.
  • rendering/RenderCounter.cpp: (WebCore::makeCounterNode): Changed to handle the case when root counter nodes lose their root status as a result of a new root counter node creation. (WebCore::destroyCounterNodeWithoutMapRemoval): Refactored more code into destroyCounterNodeChildren and renamed the function according to its new action. (WebCore::RenderCounter::destroyCounterNodes): Simplified to share more code with the new destroyCounterNode. (WebCore::RenderCounter::destroyCounterNode): Added to allow for selective counterNode destruction. (WebCore::RenderCounter::rendererSubtreeAttached): Added to refresh counter values in response to DOM changes. For renderers with no attached counters the execution time of this function cannot be discerned in comparison with the time needed to add a node or change the style of a node. (WebCore::updateCounters): Helper function for rendererSubtreeAttached. Updates the counters attached to a Renderer in response to the renderer or its ancestors being attached to the renderer tree.
  • rendering/RenderCounter.h:
  • rendering/RenderObject.cpp: (WebCore::RenderObject::addChild): Changed to update counter values if needed.
5:56 PM Changeset in webkit [53354] by weinig@apple.com
  • 2 edits in trunk/JavaScriptCore

Fix for <rdar://problem/7548432>
Add ALWAYS_INLINE to jsLess for a 1% speedup on llvm-gcc.

Reviewed by Maciej Stachowiak.

  • runtime/Operations.h:

(JSC::jsLess):

5:24 PM Changeset in webkit [53353] by kevino@webkit.org
  • 3 edits in trunk/WebKit/wx

Reviewed by Alexey Proskuryakov.

Move the code adding the child frame to the FrameTree into wxWebFrame constructor
so that it happens before we call init() on the new frame. Fixes asserts checking
frame count consistency.

5:18 PM Changeset in webkit [53352] by darin@chromium.org
  • 2 edits in trunk/LayoutTests

2010-01-15 Darin Fisher <​darin@chromium.org>

Fix flakey test.

Add a call to clearBackForwardList.

  • fast/history/history-length.html:
5:13 PM Changeset in webkit [53351] by eric@webkit.org
  • 7 edits in trunk

2010-01-15 Alejandro G. Castro <​alex@igalia.com>

Reviewed by Xan Lopez.

Original patch by David Ronis <​david.ronis@mcgill.ca>

Upgrade 1.1.17->1.1.18 fails: GTK_WIDGET_TOPLEVEL' was not declared in this scope
​https://bugs.webkit.org/show_bug.cgi?id=33486

Deprecated symbols replaced for gtk+ versions over 2.18.

  • platform/gtk/PlatformScreenGtk.cpp: (WebCore::getVisual): (WebCore::screenRect):

2010-01-15 Alejandro G. Castro <​alex@igalia.com>

Reviewed by Xan Lopez.

Original patch by David Ronis <​david.ronis@mcgill.ca>

Upgrade 1.1.17->1.1.18 fails: GTK_WIDGET_TOPLEVEL' was not declared in this scope
​https://bugs.webkit.org/show_bug.cgi?id=33486

Deprecated symbols replaced for gtk+ versions over 2.18.

  • WebCoreSupport/ChromeClientGtk.cpp: (WebKit::ChromeClient::windowRect):
  • webkit/webkitprivate.cpp: (currentToplevelCallback):
  • webkit/webkitwebframe.cpp: (webkit_web_frame_print_full): (webkit_web_frame_print):
  • webkit/webkitwebview.cpp: (webkit_web_view_focus_in_event):
4:58 PM Changeset in webkit [53350] by darin@chromium.org
  • 2 edits
    1 move in trunk/LayoutTests

2010-01-15 Darin Fisher <​darin@chromium.org>

Reviewed by Dimitri Glazkov.

Disable page cache and add diagnostic output in the failure case.
​https://bugs.webkit.org/show_bug.cgi?id=33749

  • fast/history/history-length-expected.txt:
  • fast/history/history-length.html: Copied from fast/history/history-length.html-disabled.
  • fast/history/history-length.html-disabled: Removed.
4:16 PM Changeset in webkit [53349] by oliver@apple.com
  • 2 edits in trunk/WebCore

2010-01-15 Oliver Hunt <​oliver@apple.com>

Reviewed by Sam Weinig.

Bad DOM performance in large SVG files
​https://bugs.webkit.org/show_bug.cgi?id=30055

Improve the performance of the early culling check by avoiding save/restore
of graphics state in culled case.

  • rendering/RenderPath.cpp: (WebCore::RenderPath::paint):
3:21 PM Changeset in webkit [53348] by darin@chromium.org
  • 1 edit
    1 move in trunk/LayoutTests

2010-01-15 Darin Fisher <​darin@chromium.org>

Disable this test temporarily since it is failing on all of the
builders.

  • fast/history/history-length.html: Removed.
  • fast/history/history-length.html-disabled: Copied from fast/history/history-length.html.
2:37 PM Changeset in webkit [53347] by eric@webkit.org
  • 16 edits
    1 delete in trunk

2010-01-15 Eric Seidel <​eric@webkit.org>

No review, rolling out r53344.
​http://trac.webkit.org/changeset/53344
​https://bugs.webkit.org/show_bug.cgi?id=32920

Broke the Chromium Mac builder.

  • WebCore.gypi:
  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/SerializedScriptValue.cpp: Removed.
  • bindings/v8/SerializedScriptValue.h: (WebCore::SerializedScriptValue::create): (WebCore::SerializedScriptValue::release): (WebCore::SerializedScriptValue::toString): (WebCore::SerializedScriptValue::SerializedScriptValue):
  • bindings/v8/custom/V8DOMWindowCustom.cpp: (WebCore::V8DOMWindow::postMessageCallback):
  • bindings/v8/custom/V8DedicatedWorkerContextCustom.cpp: (WebCore::V8DedicatedWorkerContext::postMessageCallback):
  • bindings/v8/custom/V8HistoryCustom.cpp: (WebCore::V8History::pushStateCallback): (WebCore::V8History::replaceStateCallback):
  • bindings/v8/custom/V8MessageEventCustom.cpp: (WebCore::V8MessageEvent::initMessageEventCallback):
  • bindings/v8/custom/V8MessagePortCustom.cpp: (WebCore::V8MessagePort::postMessageCallback):
  • bindings/v8/custom/V8PopStateEventCustom.cpp: (WebCore::V8PopStateEvent::initPopStateEventCallback): (WebCore::V8PopStateEvent::stateAccessorGetter):
  • bindings/v8/custom/V8WorkerCustom.cpp: (WebCore::V8Worker::postMessageCallback):

2010-01-15 Eric Seidel <​eric@webkit.org>

No review, rolling out r53344.
​http://trac.webkit.org/changeset/53344
​https://bugs.webkit.org/show_bug.cgi?id=32920

Broke the Chromium Mac builder.

  • src/PlatformMessagePortChannel.cpp: (WebCore::PlatformMessagePortChannel::postMessageToRemote): (WebCore::PlatformMessagePortChannel::tryGetMessageFromRemote):
  • src/WebWorkerBase.cpp: (WebKit::WebWorkerBase::postMessageToWorkerObject):
  • src/WebWorkerClientImpl.cpp: (WebKit::WebWorkerClientImpl::postMessageToWorkerContext): (WebKit::WebWorkerClientImpl::postMessageToWorkerObjectTask):
  • src/WebWorkerImpl.cpp: (WebKit::WebWorkerImpl::postMessageToWorkerContextTask):
2:08 PM Changeset in webkit [53346] by darin@chromium.org
  • 5 edits
    2 adds in trunk

2010-01-15 Darin Fisher <​darin@chromium.org>

Reviewed by Brady Eidson.

history.length does not return number of elements in history list
​https://bugs.webkit.org/show_bug.cgi?id=24472

Test: fast/history/history-length.html

  • page/Page.cpp: (WebCore::Page::getHistoryLength):
1:47 PM Changeset in webkit [53345] by dimich@chromium.org
  • 3 edits in trunk/WebCore

Need to ensure that Document::postTask does not provide the Task with a dangling pointer to destroyed Document
​https://bugs.webkit.org/show_bug.cgi?id=31633

Reviewed by Darin Adler.

Don't see a way to add test for it, we don't have a way to reproduce the issue currently.

  • dom/Document.cpp:

(WebCore::DocumentWeakReference::DocumentWeakReference):
(WebCore::DocumentWeakReference::document):
(WebCore::DocumentWeakReference::clear):
(WebCore::Document::Document): Create a weak reference to this Document.
(WebCore::Document::~Document): Clear the weak pointer, preventing further execution of tasks.
(WebCore::PerformTaskContext::PerformTaskContext):
(WebCore::performTask): Check if the documentWeakReference is cleared by Document destructor - in this case do not run the task.
(WebCore::Document::postTask):

  • dom/Document.h:

(WebCore::DocumentWeakReference::create):

1:28 PM Changeset in webkit [53344] by eric@webkit.org
  • 16 edits
    1 add in trunk

2010-01-15 Vitaly Repeshko <​vitalyr@chromium.org>

Reviewed by David Levin.

[V8] Support SerializedScriptValue.
​https://bugs.webkit.org/show_bug.cgi?id=32920
​http://crbug.com/30620

Initial implementation of SerializedScriptValue which is used to
to create a serialized representation of JavaScript objects. This
representation is needed for structured clones and worker messages.

  • WebCore.gypi: Added SerializedScriptValue.cpp.
  • bindings/scripts/CodeGeneratorV8.pm: Removed conversion to string before using SerializedScriptValue.
  • bindings/v8/SerializedScriptValue.cpp: Added. (WebCore::): (WebCore::ZigZag::encode): (WebCore::ZigZag::decode): (WebCore::Writer::Writer): (WebCore::Writer::writeUndefined): (WebCore::Writer::writeNull): (WebCore::Writer::writeTrue): (WebCore::Writer::writeFalse): (WebCore::Writer::writeString): (WebCore::Writer::writeInt32): (WebCore::Writer::writeNumber): (WebCore::Writer::endComposite): (WebCore::Writer::data): (WebCore::Writer::doWriteUint32): (WebCore::Writer::append): (WebCore::Writer::ensureSpace): (WebCore::Writer::fillHole): (WebCore::Writer::charAt): (WebCore::Serializer::Serializer): (WebCore::Serializer::serialize): (WebCore::Serializer::StateBase::~StateBase): (WebCore::Serializer::StateBase::nextState): (WebCore::Serializer::StateBase::setNextState): (WebCore::Serializer::StateBase::composite): (WebCore::Serializer::StateBase::StateBase): (WebCore::Serializer::State::composite): (WebCore::Serializer::State::tag): (WebCore::Serializer::State::State): (WebCore::Serializer::StackCleaner::StackCleaner): (WebCore::Serializer::StackCleaner::~StackCleaner): (WebCore::Serializer::ArrayState::ArrayState): (WebCore::Serializer::ArrayState::done): (WebCore::Serializer::ArrayState::advance): (WebCore::Serializer::ObjectState::ObjectState): (WebCore::Serializer::ObjectState::done): (WebCore::Serializer::ObjectState::advance): (WebCore::Serializer::ObjectState::nextProperty): (WebCore::Serializer::doSerialize): (WebCore::Serializer::push): (WebCore::Serializer::top): (WebCore::Serializer::pop): (WebCore::Serializer::checkComposite): (WebCore::Reader::Reader): (WebCore::Reader::isEof): (WebCore::Reader::read): (WebCore::Reader::readTag): (WebCore::Reader::readString): (WebCore::Reader::readInt32): (WebCore::Reader::readNumber): (WebCore::Reader::doReadUint32): (WebCore::Deserializer::Deserializer): (WebCore::Deserializer::deserialize): (WebCore::Deserializer::doDeserialize): (WebCore::Deserializer::push): (WebCore::Deserializer::pop): (WebCore::Deserializer::stackDepth): (WebCore::Deserializer::element): (WebCore::SerializedScriptValue::SerializedScriptValue): (WebCore::SerializedScriptValue::deserialize):
  • bindings/v8/SerializedScriptValue.h: (WebCore::SerializedScriptValue::create): (WebCore::SerializedScriptValue::createFromWire): (WebCore::SerializedScriptValue::release): (WebCore::SerializedScriptValue::toWireString):

Updated uses of SerializedScriptValue:

  • bindings/v8/custom/V8DOMWindowCustom.cpp: (WebCore::V8DOMWindow::postMessageCallback):
  • bindings/v8/custom/V8DedicatedWorkerContextCustom.cpp: (WebCore::V8DedicatedWorkerContext::postMessageCallback):
  • bindings/v8/custom/V8HistoryCustom.cpp: (WebCore::V8History::pushStateCallback): (WebCore::V8History::replaceStateCallback):
  • bindings/v8/custom/V8MessageEventCustom.cpp: (WebCore::V8MessageEvent::initMessageEventCallback):
  • bindings/v8/custom/V8MessagePortCustom.cpp: (WebCore::V8MessagePort::postMessageCallback):
  • bindings/v8/custom/V8WorkerCustom.cpp: (WebCore::V8Worker::postMessageCallback):

2010-01-15 Vitaly Repeshko <​vitalyr@chromium.org>

Reviewed by David Levin.

[V8] Support SerializedScriptValue.
​https://bugs.webkit.org/show_bug.cgi?id=32920
​http://crbug.com/30620

Updated uses of SerializedScriptValue:

  • src/PlatformMessagePortChannel.cpp: (WebCore::PlatformMessagePortChannel::postMessageToRemote): (WebCore::PlatformMessagePortChannel::tryGetMessageFromRemote):
  • src/WebWorkerBase.cpp: (WebKit::WebWorkerBase::postMessageToWorkerObject):
  • src/WebWorkerClientImpl.cpp: (WebKit::WebWorkerClientImpl::postMessageToWorkerContext): (WebKit::WebWorkerClientImpl::postMessageToWorkerObjectTask):
  • src/WebWorkerImpl.cpp: (WebKit::WebWorkerImpl::postMessageToWorkerContextTask):
1:12 PM WebKit Team edited by Chris Jerdonek
Added myself. (diff)
12:54 PM Changeset in webkit [53343] by oliver@apple.com
  • 3 edits in trunk/WebCore

2010-01-15 Oliver Hunt <​oliver@apple.com>

Reviewed by Dirk Schulze.

Bad DOM performance in large SVG files
​https://bugs.webkit.org/show_bug.cgi?id=30055

Add an early return when we go to paint a RenderPath that
isn't in the current clip.

  • rendering/RenderPath.cpp: (WebCore::RenderPath::paint):
  • svg/graphics/SVGImage.cpp: (WebCore::SVGImage::draw):
12:53 PM Changeset in webkit [53342] by eric@webkit.org
  • 3 edits in trunk/WebCore

2010-01-15 Steve Block <​steveblock@google.com>

Reviewed by Eric Seidel.

Make Geolocation::suspend/resume public.
​https://bugs.webkit.org/show_bug.cgi?id=33679

These were made private in Bug 32499, but are required by Android to be public.

No new tests, build fix only.

  • page/Geolocation.cpp: Modified (WebCore::Geolocation::suspend): Added back in (WebCore::Geolocation::resume): Added back in
  • page/Geolocation.h: Modified. Make suspend and resume public
12:48 PM Changeset in webkit [53341] by ggaren@apple.com
  • 5 edits
    2 adds in trunk

2010-01-14 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Oliver Hunt.

REGRESISON: Google maps buttons not working properly
​https://bugs.webkit.org/show_bug.cgi?id=31871

REGRESSION(r52948): JavaScript exceptions thrown on Google Maps when
getting directions for a second time
​https://bugs.webkit.org/show_bug.cgi?id=33446


SunSpider and v8 report no change.

  • interpreter/Interpreter.cpp: (JSC::Interpreter::tryCacheGetByID): Update our cached offset in case flattening the dictionary changed any of its offsets.
  • jit/JITStubs.cpp: (JSC::JITThunks::tryCacheGetByID): (JSC::DEFINE_STUB_FUNCTION):
  • runtime/Operations.h: (JSC::normalizePrototypeChain): ditto

2010-01-15 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Oliver Hunt.

REGRESISON: Google maps buttons not working properly
​https://bugs.webkit.org/show_bug.cgi?id=31871

REGRESSION(r52948): JavaScript exceptions thrown on Google Maps when
getting directions for a second time
​https://bugs.webkit.org/show_bug.cgi?id=33446


Added a test for these bugs.

  • fast/js/pic/undictionary-expected.txt: Added.
  • fast/js/pic/undictionary.html: Added.
12:36 PM Changeset in webkit [53340] by eric@webkit.org
  • 3 edits in trunk/WebKitTools

2010-01-15 Chris Jerdonek <​chris.jerdonek@gmail.com>

Reviewed by David Kilzer.

Altered parseDiffHeader() to skip unrecognized lines and
other minor clean-ups.

​https://bugs.webkit.org/show_bug.cgi?id=33476

  • Scripts/VCSUtils.pm:
    • Changed parseDiffHeader() as follows:
      • Skips over unrecognized lines.
      • Addressed FIXME to remove substitution for "diff" line.
      • Renamed "version" header hash key to "sourceRevision".
      • Eliminated "copiedFromVersion" header hash key.
      • Included "sourceRevision" also for copied files.
      • Checks that copy revision number matches "sourceRevision".
      • No longer returns $foundHeaderEnding.
      • Dies if header ending not found.
      • Diff header dividing line now always added.
  • Scripts/webkitperl/VCSUtils_unittest/parseDiffHeader.pl:
    • Made necessary changes in parseDiffHeader() unit tests.
    • Shortened the file paths in some test cases.
12:26 PM Changeset in webkit [53339] by wsiegrist@apple.com
  • 1 edit in trunk/WebKitTools/BuildSlaveSupport/build.webkit.org-config/public_html/index.html

2010-01-15 William Siegrist <​wsiegrist@apple.com>

Cleanup buildbot main menu

12:20 PM Changeset in webkit [53338] by eric@webkit.org
  • 2 edits in trunk/WebCore

2010-01-15 Alexander Pavlov <​apavlov@chromium.org>

Reviewed by Pavel Feldman.

Set pointer cursor when over a breakpoint in the BreakpointsSidebarPane

​https://bugs.webkit.org/show_bug.cgi?id=33713

  • inspector/front-end/inspector.css:
12:20 PM Changeset in webkit [53337] by ojan@chromium.org
  • 2 edits in trunk/LayoutTests

2010-01-15 Ojan Vafai <​ojan@chromium.org>

Reviewed by Eric Seidel.

svg/custom/marker-changes.svg is flaky
​https://bugs.webkit.org/show_bug.cgi?id=33728

removing gratuitous setTimeout.

  • svg/custom/marker-changes.svg:
10:56 AM Changeset in webkit [53336] by ojan@chromium.org
  • 2 edits in trunk/LayoutTests

2010-01-15 Ojan Vafai <​ojan@chromium.org>

Reviewed by Adam Barth.

transitions/shorthand-border-transitions.html is still flaky on slower bots
​https://bugs.webkit.org/show_bug.cgi?id=33721

Increase the tolerance. The test still tests what it was meant to.

  • transitions/shorthand-border-transitions.html:
10:38 AM Changeset in webkit [53335] by Nate Chapin
  • 6 edits in trunk/WebCore

2010-01-15 Nate Chapin <Nate Chapin>

Reviewed by Dimitri Glazkov.

[V8] Generate more of the custom behaviors that the v8 bindings attach to v8 FunctionTemplates.

  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/V8Binding.cpp: (WebCore::getToStringName): (WebCore::constructorToString): (WebCore::getToStringTemplate):
  • bindings/v8/V8Binding.h:
  • bindings/v8/V8DOMWindowShell.cpp: (WebCore::V8DOMWindowShell::createNewContext):
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate):
9:46 AM Changeset in webkit [53334] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2010-01-15 Dimitri Glazkov <Dimitri Glazkov>

No review, rolling out r53331.
​http://trac.webkit.org/changeset/53331
​https://bugs.webkit.org/show_bug.cgi?id=30055

Broke fast/borders/svg-as-border-image-3.html. Don't give up,
Oliver! You are really, really close.

  • rendering/RenderPath.cpp: (WebCore::RenderPath::paint):
9:46 AM Changeset in webkit [53333] by ojan@chromium.org
  • 3 edits in trunk/LayoutTests

2010-01-14 Ojan Vafai <​ojan@chromium.org>

Reviewed by Eric Seidel.

animations/combo-transform-translate+scale.html failed on Snow Leopard Debug Bot
​https://bugs.webkit.org/show_bug.cgi?id=33461

Make test half as long. Remove a setTimeout and tweak tolerances to
reduce flakiness.

  • animations/combo-transform-translate+scale.html:
9:13 AM BuildingQtOnSymbian edited by Simon Hausmann
(diff)
9:13 AM BuildingQtOnSymbian edited by Simon Hausmann
(diff)
9:04 AM BuildingQtOnSymbian created by Simon Hausmann
8:46 AM Changeset in webkit [53332] by kov@webkit.org
  • 2 edits in trunk

Rubber-stamped by Xan Lopez.

Bump version to 1.1.19.

  • configure.ac:
8:43 AM QtWebKit edited by Simon Hausmann
(diff)
4:25 AM Changeset in webkit [53331] by oliver@apple.com
  • 2 edits in trunk/WebCore

2010-01-15 Oliver Hunt <​oliver@apple.com>

Reviewed by Nikolas Zimmermann.

Bad DOM performance in large SVG files
​https://bugs.webkit.org/show_bug.cgi?id=30055

Add an early return when we go to paint a RenderPath that
isn't in the current clip.

  • rendering/RenderPath.cpp: (WebCore::RenderPath::paint):
3:59 AM Changeset in webkit [53330] by vestbo@webkit.org
  • 2 edits in trunk/WebCore

2010-01-15 Tor Arne Vestbø <​tor.arne.vestbo@nokia.com>

Reviewed by Antti Koivisto.

Use OS(MAC_OS_X) instead of PLATFORM(MAC) when determining navigator.platform

​https://bugs.webkit.org/show_bug.cgi?id=33711

  • page/NavigatorBase.cpp:
3:54 AM Changeset in webkit [53329] by Philippe Normand
  • 2 edits in trunk/WebCore

2010-01-14 Philippe Normand <​pnormand@igalia.com>

Reviewed by Xan Lopez.

[Gtk] Compiler warning: comparisons like 'X<=Y<=Z' do not have their mathematical meaning
​https://bugs.webkit.org/show_bug.cgi?id=33575

Patch from Magnus Boman <​captain.magnus@gmail.com>

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::mimeTypeCache): fix compiler warnings.
3:20 AM Changeset in webkit [53328] by pfeldman@chromium.org
  • 9 edits
    1 add in trunk/WebCore

2010-01-14 Mikhail Naganov <​mnaganov@chromium.org>

Reviewed by Timothy Hatcher.

Add welcome screen to Profiles pane to provide some instructions for novices.

​https://bugs.webkit.org/show_bug.cgi?id=19268

  • English.lproj/localizedStrings.js:
  • WebCore.gypi:
  • WebCore.vcproj/WebCore.vcproj:
  • inspector/front-end/ProfileView.js: (WebInspector.CPUProfileType.prototype.get welcomeMessage):
  • inspector/front-end/ProfilesPanel.js: (WebInspector.ProfileType.prototype.get welcomeMessage): (WebInspector.ProfilesPanel): (WebInspector.ProfilesPanel.prototype.show): (WebInspector.ProfilesPanel.prototype.registerProfileType): (WebInspector.ProfilesPanel.prototype._addWelcomeMessage.messageButtonClicked): (WebInspector.ProfilesPanel.prototype._addWelcomeMessage): (WebInspector.ProfilesPanel.prototype.showProfile): (WebInspector.ProfilesPanel.prototype.closeVisibleView): (WebInspector.ProfilesPanel.prototype._updateInterface):
  • inspector/front-end/WebKit.qrc:
  • inspector/front-end/WelcomeView.js: Added. (WebInspector.WelcomeView): (WebInspector.WelcomeView.prototype._windowResized): (WebInspector.WelcomeView.prototype.addMessage):
  • inspector/front-end/inspector.css:
  • inspector/front-end/inspector.html:
2:00 AM QtWebKitBackportingFixes edited by Simon Hausmann
(diff)
1:54 AM QtWebKitBackportingFixes edited by Simon Hausmann
(diff)
1:45 AM QtWebKitBackportingFixes edited by Girish Ramakrishnan
(diff)
1:43 AM Changeset in webkit [53327] by Girish Ramakrishnan
  • 2 edits in trunk/WebCore

[Qt/Win] Flash in QGraphicsWebView does not process hover correctly.

Reviewed by Simon Hausmann.

Mouse hover does not work as expected with the flash in some sites.

The problem appears to be that Flash queries NPNVnetscapeWindow on every
mouse hover. I do not how flash uses this value but returning 0 when flash
is in windowless mode solves the problem (When using QGraphicsWebView we
inject wmode opaque, thereby putting the plugin in windowless mode).

​https://bugs.webkit.org/show_bug.cgi?id=33591

12:30 AM Changeset in webkit [53326] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

2010-01-15 Zoltan Horvath <​zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for Peer class
​https://bugs.webkit.org/show_bug.cgi?id=33670

Inherits the following class from Noncopyable because it is
instantiated by 'new' and no need to be copyable:

class name - instantiated at: WebCore/'location'
class Peer - websockets/WebSocketChannelClient.h:38

  • websockets/WorkerThreadableWebSocketChannel.h:
12:23 AM Changeset in webkit [53325] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

2010-01-15 Zoltan Horvath <​zoltan@webkit.org>

Reviewed by Oliver Hunt.

[Qt] Allow custom memory allocation control for GraphicsContextPlatformPrivate class
​https://bugs.webkit.org/show_bug.cgi?id=33669

Inherits the following class from Noncopyable because it is
instantiated by 'new' and no need to be copyable:

class name - instantiated at: WebCore/'location'
class GraphicsContextPlatformPrivate - platform/graphics/qt/GraphicsContextQt.cpp:254

  • platform/graphics/qt/GraphicsContextQt.cpp:
12:20 AM Changeset in webkit [53324] by darin@chromium.org
  • 4 edits
    2 adds in trunk

2010-01-14 Darin Fisher <​darin@chromium.org>

Reviewed by Brady Eidson.

history.pushState should clear the entire forward history
​https://bugs.webkit.org/show_bug.cgi?id=33160

Test: fast/loader/stateobjects/pushstate-clears-forward-history.html

  • history/BackForwardList.cpp: (WebCore::BackForwardList::addItem): (WebCore::BackForwardList::pushStateItem):
  • history/BackForwardList.h:
Note: See TracTimeline for information about the timeline view.