Timeline



Oct 31, 2009:

4:22 PM Changeset in webkit [50383] by oliver@apple.com
  • 2 edits in trunk/WebCore

Build fix

4:10 PM Changeset in webkit [50382] by oliver@apple.com
  • 2 edits in trunk/WebCore

Remove obsolete null checks from CanvasRenderingContext3DMac
https://bugs.webkit.org/show_bug.cgi?id=30983

Reviewed by Darin Adler

3:19 PM Changeset in webkit [50381] by oliver@apple.com
  • 6 edits
    6 adds in trunk

WebGL allows objects to be used with the wrong context
https://bugs.webkit.org/show_bug.cgi?id=30981

Reviewed by Jon Honeycutt.

Simply add null checks and a few context guards to ensure we don't
deref null or attempt to use an object from a different context.

Tests: fast/canvas/webgl/incorrect-context-object-behaviour.html

fast/canvas/webgl/null-object-behaviour.html

2:54 PM BuildingOnWindows edited by jberlin@webkit.org
(diff)

Oct 30, 2009:

8:48 PM Changeset in webkit [50380] by mrowe@apple.com
  • 4 edits in branches/safari-4-branch

Versioning.

8:48 PM Changeset in webkit [50379] by mrowe@apple.com
  • 1 copy in tags/Safari-6531.21.8

New tag.

8:47 PM Changeset in webkit [50378] by mrowe@apple.com
  • 2 edits in branches/safari-4-branch/WebCore

Fix the build.

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::positionalDescriptionForMSAA):

7:11 PM Changeset in webkit [50377] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

Fixed failing layout test: restore a special case I accidentally deleted.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-30
Reviewed by Oliver Hunt.

  • runtime/DatePrototype.cpp:

(JSC::setNewValueFromDateArgs): In the case of applying a change to a date
that is NaN, reset the date to 0 *and* then apply the change; don't just
reset the date to 0.

5:51 PM Changeset in webkit [50376] by jhoneycutt@apple.com
  • 2 edits in trunk/WebKitTools

Fix an issue that Adam noticed in DRT.

Reviewed by Darin Adler.

  • DumpRenderTree/win/AccessibilityUIElementWin.cpp:

(AccessibilityUIElement::role):
Get the length of the role text, and create a buffer dynamically.

5:26 PM Changeset in webkit [50375] by mrowe@apple.com
  • 4 edits in branches/safari-4-branch

Versioning.

5:25 PM Changeset in webkit [50374] by mrowe@apple.com
  • 1 copy in tags/Safari-6531.21.7

New tag.

5:23 PM Changeset in webkit [50373] by mrowe@apple.com
  • 3 edits
    2 adds in branches/safari-4-branch

Merge r50356.

5:23 PM Changeset in webkit [50372] by mrowe@apple.com
  • 6 edits
    2 adds in branches/safari-4-branch

Merge r50355.

5:23 PM Changeset in webkit [50371] by mrowe@apple.com
  • 7 edits in branches/safari-4-branch

Merge r50354.

5:23 PM Changeset in webkit [50370] by mrowe@apple.com
  • 8 edits in branches/safari-4-branch

Merge r50353.

4:18 PM Changeset in webkit [50369] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's MediaQueryResult
https://bugs.webkit.org/show_bug.cgi?id=30857

Reviewed by Darin Adler.

Inherits MediaQueryResult class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:5984 and
it is no need to be copyable.

  • css/CSSStyleSelector.h:
4:16 PM Changeset in webkit [50368] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's MediaQueryEvaluator
https://bugs.webkit.org/show_bug.cgi?id=30854

Reviewed by Darin Adler.

Inherits MediaQueryEvaluator class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:401 and
it is no need to be copyable.

  • css/MediaQueryEvaluator.h:
4:10 PM Changeset in webkit [50367] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's MediaQueryEvaluator
https://bugs.webkit.org/show_bug.cgi?id=30854

Reviewed by Darin Adler.

Inherits MediaQueryEvaluator class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:401 and
it is no need to be copyable.

  • css/MediaQueryEvaluator.h:
4:08 PM Changeset in webkit [50366] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's CSSRuleSet
https://bugs.webkit.org/show_bug.cgi?id=30852

Reviewed by Darin Adler.

Inherits CSSRuleSet class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:455 and
it is no need to be copyable.

  • css/CSSStyleSelector.cpp:
4:06 PM Changeset in webkit [50365] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's CSSRuleData
https://bugs.webkit.org/show_bug.cgi?id=30851

Reviewed by Darin Adler.

Inherits CSSRuleData class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.h:320 and
it is no need to be copyable.

  • css/CSSStyleSelector.h:
4:02 PM Changeset in webkit [50364] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's CSSRuleDataList
https://bugs.webkit.org/show_bug.cgi?id=30850

Reviewed by Darin Adler.

Inherits CSSRuleDataList class from Noncopyable because it has been
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:2715 and
it is no need to be copyable.

  • css/CSSStyleSelector.h:
3:59 PM Changeset in webkit [50363] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's CSSNamespace
https://bugs.webkit.org/show_bug.cgi?id=30849

Reviewed by Darin Adler.

Inherits CSSNamespace struct from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSheet.cpp:141 and
it is no need to be copyable.

  • css/CSSNamespace.h:
3:41 PM Changeset in webkit [50362] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-10-30 Eric Seidel <eric@webkit.org>

No review, rolling out r50105.
http://trac.webkit.org/changeset/50105

This commit was causing:
https://bugs.webkit.org/show_bug.cgi?id=30869
We'll re-implement the feature a different way.

  • Scripts/bugzilla-tool:
3:35 PM Changeset in webkit [50361] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

Windows build fix: update for object-to-pointer change.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-30

  • runtime/DatePrototype.cpp:

(JSC::formatLocaleDate):

3:26 PM Changeset in webkit [50360] by dimich@chromium.org
  • 5 edits in trunk/WebCore

Refactor DatabaseTask in preparation for removing threadsafe refcounting from it.
Move the synchronizer object out of the DatabaseTask so there is no need to keep
the pointer to Databasetask around after passing it to MessageQueue.
Also pass the references to return parameters to the task so it can update them.
https://bugs.webkit.org/show_bug.cgi?id=30941

Reviewed by David Levin.

No new tests, since this is just moving the code around, no change in functionality.

  • storage/Database.cpp:

(WebCore::Database::Database):
(WebCore::Database::openAndVerifyVersion): Use new DatabaseTaskSynchronizer to wait for task completion.
(WebCore::Database::markAsDeletedAndClose): Ditto.
(WebCore::Database::tableNames): Ditto.
(WebCore::Database::stop): Use the boolean flag rather then 'killed' flag built into MessageQueue.
(WebCore::Database::scheduleTransaction): Transaction queue is a Deque now, change the way to fetch the transaction.

  • storage/Database.h: Change the SQLTransaction queue to be a Deque rather then a MessageQueue.
  • storage/DatabaseTask.cpp:

(WebCore::DatabaseTaskSynchronizer::DatabaseTaskSynchronizer):
(WebCore::DatabaseTaskSynchronizer::waitForTaskCompletion):
(WebCore::DatabaseTaskSynchronizer::taskCompleted):
(WebCore::DatabaseTask::DatabaseTask): Ctor takes DatabaseTaskSynchronizer which can be 0.
(WebCore::DatabaseTask::performTask): Signal completion. m_synchronizer should still be around since main thread is waiting on it.
(WebCore::DatabaseOpenTask::DatabaseOpenTask): Pass synchronizer and return parameters via constructor.
(WebCore::DatabaseCloseTask::DatabaseCloseTask): Ditto.
(WebCore::DatabaseTransactionTask::DatabaseTransactionTask): Ditto.
(WebCore::DatabaseTableNamesTask::DatabaseTableNamesTask): Ditto.

  • storage/DatabaseTask.h:

(WebCore::DatabaseOpenTask::create):
(WebCore::DatabaseCloseTask::create):
(WebCore::DatabaseTransactionTask::create):
(WebCore::DatabaseTableNamesTask::create):

3:23 PM Changeset in webkit [50359] by ggaren@apple.com
  • 4 edits in trunk/JavaScriptCore

https://bugs.webkit.org/show_bug.cgi?id=30942
Use pointers instead of copies to pass GregorianDateTime objects around.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-29
Reviewed by Darin Adler.

SunSpider reports a shocking 4.5% speedup on date-format-xparb, and 1.3%
speedup on date-format-tofte.

  • runtime/DateInstance.cpp:

(JSC::DateInstance::gregorianDateTime):

  • runtime/DateInstance.h:
  • runtime/DatePrototype.cpp:

(JSC::formatLocaleDate):
(JSC::dateProtoFuncToString):
(JSC::dateProtoFuncToUTCString):
(JSC::dateProtoFuncToISOString):
(JSC::dateProtoFuncToDateString):
(JSC::dateProtoFuncToTimeString):
(JSC::dateProtoFuncGetFullYear):
(JSC::dateProtoFuncGetUTCFullYear):
(JSC::dateProtoFuncToGMTString):
(JSC::dateProtoFuncGetMonth):
(JSC::dateProtoFuncGetUTCMonth):
(JSC::dateProtoFuncGetDate):
(JSC::dateProtoFuncGetUTCDate):
(JSC::dateProtoFuncGetDay):
(JSC::dateProtoFuncGetUTCDay):
(JSC::dateProtoFuncGetHours):
(JSC::dateProtoFuncGetUTCHours):
(JSC::dateProtoFuncGetMinutes):
(JSC::dateProtoFuncGetUTCMinutes):
(JSC::dateProtoFuncGetSeconds):
(JSC::dateProtoFuncGetUTCSeconds):
(JSC::dateProtoFuncGetTimezoneOffset):
(JSC::setNewValueFromTimeArgs):
(JSC::setNewValueFromDateArgs):
(JSC::dateProtoFuncSetYear):
(JSC::dateProtoFuncGetYear): Renamed getGregorianDateTime to gregorianDateTime,
since it no longer has an out parameter. Uses 0 to indicate invalid dates.

3:09 PM Changeset in webkit [50358] by adele@apple.com
  • 3 edits
    2 adds in trunk

REGRESSION: In Mail, Undo does not restore some characters I have deleted at the end of a line
https://bugs.webkit.org/show_bug.cgi?id=30955
<rdar://problem/7067033>

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-30
Reviewed by Darin Adler.

WebCore:

When the command is deleteWordBackward or deleteWordForward
we should not add to the open typing command, but
create a new one.

Test: editing/undo/undo-deleteWord.html

  • editing/TypingCommand.cpp:

(WebCore::TypingCommand::deleteKeyPressed): Always start a new command if the granularity is
not CharacterGranularity.
(WebCore::TypingCommand::forwardDeleteKeyPressed): Always start a new command if the granularity is
not CharacterGranularity.

LayoutTests:

  • editing/undo/undo-deleteWord-expected.txt: Added.
  • editing/undo/undo-deleteWord.html: Added.
3:03 PM Changeset in webkit [50357] by eric.carlson@apple.com
  • 3 edits in trunk/WebCore

2009-10-30 Eric Carlson <eric.carlson@apple.com>

Reviewed by Darin Adler.

Make MediaPlayer constructor private
https://bugs.webkit.org/show_bug.cgi?id=30965

  • html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::loadResource): (WebCore::HTMLMediaElement::finishParsingChildren): Use MediaPlayer::create.
  • platform/graphics/MediaPlayer.h: (WebCore::MediaPlayer::create): New.
2:38 PM Changeset in webkit [50356] by jhoneycutt@apple.com
  • 3 edits
    2 adds in trunk

Test for MSAA: Accessibility of headings is not correct

https://bugs.webkit.org/show_bug.cgi?id=30937

Reviewed by Adam Roben.

WebKitTools:

  • DumpRenderTree/win/AccessibilityUIElementWin.cpp:

(AccessibilityUIElement::role):
Allow the role returned to be a BSTR.
(AccessibilityUIElement::description):
Fix a copy/paste error.

LayoutTests:

  • platform/win/accessibility/heading-elements-expected.txt: Added.
  • platform/win/accessibility/heading-elements.html: Added.
2:38 PM Changeset in webkit [50355] by jhoneycutt@apple.com
  • 7 edits
    2 adds in trunk

Test for MSAA: Accessibility of links is wrong

https://bugs.webkit.org/show_bug.cgi?id=30928

Reviewed by Darin Adler.

WebKitTools:

  • DumpRenderTree/AccessibilityUIElement.cpp:

(getAccessibilityValueCallback):
Return the accessibility value.
(AccessibilityUIElement::getJSClass):
Added "accessibilityValue" value.

  • DumpRenderTree/AccessibilityUIElement.h:
  • DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:

(AccessibilityUIElement::accessibilityValue):
Stubbed.

  • DumpRenderTree/mac/AccessibilityUIElementMac.mm:

(AccessibilityUIElement::accessibilityValue):
Stubbed.

  • DumpRenderTree/win/AccessibilityUIElementWin.cpp:

(AccessibilityUIElement::accessibilityValue):
Get the object's value, and return it as a JS string.

LayoutTests:

  • platform/win/accessibility/linked-elements-expected.txt: Added.
  • platform/win/accessibility/linked-elements.html: Added.
2:38 PM Changeset in webkit [50354] by jhoneycutt@apple.com
  • 7 edits in trunk

MSAA: Accessibility of headings is not correct

https://bugs.webkit.org/show_bug.cgi?id=30937

Reviewed by Alice Liu.

WebCore:

  • accessibility/AccessibilityObject.h:

(WebCore::AccessibilityObject::stringRoleForMSAA):
(WebCore::AccessibilityObject::descriptionForMSAA):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::shouldReturnTagNameAsRoleForMSAA):
If the element's tag name is one of h1, h2, h3, h4, h5, h6, return
true.
(WebCore::AccessibilityRenderObject::stringRoleForMSAA):
If the element should return its tag name as the role, return the tag
name.
(WebCore::AccessibilityRenderObject::positionalDescriptionForMSAA):
If the object is a heading, return the string "L" followed by the
heading level.
(WebCore::AccessibilityRenderObject::descriptionForMSAA):
If the object has a positional description, return it. Otherwise, get
the accessibility description, and prefix it with "Description" so that
MSAA clients know that it's not a positional description.

  • accessibility/AccessibilityRenderObject.h:

WebKit/win:

  • AccessibleBase.cpp:

(AccessibleBase::get_accDescription):
Call the object's descriptionForMSAA(). Moved the comment to the
WebCore file.
(AccessibleBase::get_accRole):
If the object has a string role, return that. Otherwise, return the
integer role.

  • AccessibleBase.h:

Removed description(), as this was moved to WebCore.

2:37 PM Changeset in webkit [50353] by jhoneycutt@apple.com
  • 8 edits in trunk

MSAA: Accessibility of links is wrong

https://bugs.webkit.org/show_bug.cgi?id=30928

Reviewed by Darin Adler.

WebCore:

  • accessibility/AccessibilityImageMapLink.cpp:

(WebCore::AccessibilityImageMapLink::stringValueForMSAA):
Return the URL.
(WebCore::AccessibilityImageMapLink::nameForMSAA):
Return the alt text.

  • accessibility/AccessibilityImageMapLink.h:

(WebCore::AccessibilityImageMapLink::isLinked):
Return true.

  • accessibility/AccessibilityObject.h:

(WebCore::AccessibilityObject::isLinked):
(WebCore::AccessibilityObject::stringValueForMSAA):
(WebCore::AccessibilityObject::nameForMSAA):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::isLinkable):
Return true if the element is considered "linkable" with respect to
accessibility.
(WebCore::AccessibilityRenderObject::stringValueForMSAA):
If the element is linkable, check whether it has a parent anchor
element. If so, return the anchor element's href.
(WebCore::AccessibilityRenderObject::isLinked):
Return true if the element is linkable and if it's parent anchor tag's
href is non-empty.
(WebCore::AccessibilityRenderObject::nameForMSAA):
For text nodes, return the text.

  • accessibility/AccessibilityRenderObject.h:

WebKit/win:

  • AccessibleBase.cpp:

(AccessibleBase::get_accState):
Check whether the element is linked, instead of whether the node is an
anchor, so the child "linkable" elements of anchor elements will also
be "linked".
(AccessibleBase::name):
Return the name for MSAA.
(AccessibleBase::value):
Return the string value for MSAA.

2:20 PM Changeset in webkit [50352] by bweinstein@apple.com
  • 1 edit in trunk/WebKit/win/WebKitPrefix.cpp

Touch WebKitPrefix to force rebuild of WebKit.

1:59 PM Changeset in webkit [50351] by levin@chromium.org
  • 24 edits in trunk

Notify the chrome when the focused node has changed.
https://bugs.webkit.org/show_bug.cgi?id=30832

Patch by Evan Stade <estade@chromium.org> on 2009-10-30
Reviewed by David Levin.

WebCore:

This is similar to AX code that is already in place, except that this also informs the
chrome when there stops being a focused node. This is needed for a browser to show the
anchor for links that have keyboard focus.

  • dom/Document.cpp:

(WebCore::Document::setFocusedNode):

  • loader/EmptyClients.h:

(WebCore::EmptyChromeClient::focusedNodeChanged):

  • page/Chrome.cpp:

(WebCore::Chrome::focusedNodeChanged):

  • page/Chrome.h:
  • page/ChromeClient.h:

WebKit/gtk:

Added stub implementation for new ChromeClient function.

  • WebCoreSupport/ChromeClientGtk.cpp:

(WebKit::ChromeClient::focusedNodeChanged):

  • WebCoreSupport/ChromeClientGtk.h:

WebKit/haiku:

Added stub implementation for new ChromeClient function.

  • WebCoreSupport/ChromeClientHaiku.cpp:

(WebCore::ChromeClientHaiku::focusedNodeChanged):

  • WebCoreSupport/ChromeClientHaiku.h:

WebKit/mac:

Added stub implementation for new ChromeClient function.

  • WebCoreSupport/WebChromeClient.h:
  • WebCoreSupport/WebChromeClient.mm:

(WebChromeClient::focusedNodeChanged):

WebKit/qt:

Added stub implementation for new ChromeClient function.

  • WebCoreSupport/ChromeClientQt.cpp:

(WebCore::ChromeClientQt::focusedNodeChanged):

  • WebCoreSupport/ChromeClientQt.h:

WebKit/win:

Added stub implementation for new ChromeClient function.

  • WebCoreSupport/WebChromeClient.cpp:

(WebChromeClient::focusedNodeChanged):

  • WebCoreSupport/WebChromeClient.h:

WebKit/wx:

Added stub implementation for new ChromeClient function.

  • WebKitSupport/ChromeClientWx.cpp:

(WebCore::ChromeClientWx::focusedNodeChanged):

  • WebKitSupport/ChromeClientWx.h:
1:20 PM Changeset in webkit [50350] by eric@webkit.org
  • 3 edits
    2 adds in trunk

2009-10-30 Ben Murdoch <benm@google.com>

Reviewed by David Kilzer.

openDatabase() with empty version sets db version up incorrectly
https://bugs.webkit.org/show_bug.cgi?id=28417

  • storage/open-database-set-empty-version-expected.txt: Added.
  • storage/open-database-set-empty-version.html: Added.

2009-10-30 Ben Murdoch <benm@google.com>

Reviewed by David Kilzer.

openDatabase() with empty version sets db version up incorrectly
https://bugs.webkit.org/show_bug.cgi?id=28417

Test: storage/open-database-set-empty-version.html

  • storage/Database.cpp: (WebCore::Database::performOpenAndVerify): Raise an exception if the current database version does not match the expected version when the current version is the empty string.
12:52 PM Changeset in webkit [50349] by kevino@webkit.org
  • 2 edits in trunk/WebKitTools

Fix typo in command name used by wx build system.

12:37 PM Changeset in webkit [50348] by eric@webkit.org
  • 8 edits in trunk/WebCore

2009-10-30 John Gregg <johnnyg@google.com>

Reviewed by David Levin.

Need to turn off notifications properly at runtime
https://bugs.webkit.org/show_bug.cgi?id=30409

Moving the notificationsEnabled bit from NotificationCenter
to the new V8 RuntimeEnabledFeatures object.

Just moving a bit around, so no new tests.

  • bindings/v8/RuntimeEnabledFeatures.cpp:
  • bindings/v8/RuntimeEnabledFeatures.h: (WebCore::RuntimeEnabledFeatures::setNotificationsEnabled): (WebCore::RuntimeEnabledFeatures::notificationsEnabled):
  • bindings/v8/custom/V8DOMWindowCustom.cpp: (WebCore::ACCESSOR_RUNTIME_ENABLER):
  • bindings/v8/custom/V8WorkerContextCustom.cpp: (WebCore::ACCESSOR_RUNTIME_ENABLER):
  • notifications/NotificationCenter.cpp:
  • notifications/NotificationCenter.h:
11:31 AM Changeset in webkit [50347] by dimich@chromium.org
  • 14 edits in trunk/WebCore

[V8] More cleanup after r49949: remove ListenerGuard.
ListenerGuard is no longer needed since EventListeners do not depend on frame or v8 context.
https://bugs.webkit.org/show_bug.cgi?id=30943

Reviewed by Dimitri Glazkov.

Covered by fast/events/add-event-without-document.html which will now pass in Chromium.

  • bindings/v8/V8AbstractEventListener.cpp:

(WebCore::V8AbstractEventListener::V8AbstractEventListener):
(WebCore::V8AbstractEventListener::handleEvent):

  • bindings/v8/V8AbstractEventListener.h:
  • bindings/v8/V8DOMWrapper.cpp:

(WebCore::V8DOMWrapper::getEventListener):

  • bindings/v8/V8EventListenerList.h:

(WebCore::V8EventListenerList::findOrCreateWrapper):

  • bindings/v8/V8LazyEventListener.cpp:

(WebCore::V8LazyEventListener::V8LazyEventListener):

  • bindings/v8/V8Proxy.cpp:

(WebCore::V8Proxy::V8Proxy):
(WebCore::V8Proxy::disconnectFrame):
(WebCore::V8Proxy::clearForNavigation):

  • bindings/v8/V8Proxy.h:
  • bindings/v8/V8WorkerContextEventListener.cpp:

(WebCore::V8WorkerContextEventListener::V8WorkerContextEventListener):
(WebCore::V8WorkerContextEventListener::handleEvent):
(WebCore::V8WorkerContextEventListener::reportError):

  • bindings/v8/V8WorkerContextEventListener.h:

(WebCore::V8WorkerContextEventListener::create):

  • bindings/v8/WorkerContextExecutionProxy.cpp:

(WebCore::WorkerContextExecutionProxy::WorkerContextExecutionProxy):
(WebCore::WorkerContextExecutionProxy::dispose):
(WebCore::WorkerContextExecutionProxy::findOrCreateEventListener):

  • bindings/v8/WorkerContextExecutionProxy.h:
  • bindings/v8/custom/V8CustomEventListener.cpp:

(WebCore::V8EventListener::V8EventListener):

  • bindings/v8/custom/V8CustomEventListener.h:

(WebCore::V8EventListener::create):

11:19 AM Changeset in webkit [50346] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-10-30 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Patch v1 is a dumb default name for patches
https://bugs.webkit.org/show_bug.cgi?id=30952

Let's use "Patch" instead.

  • Scripts/bugzilla-tool:
10:57 AM Changeset in webkit [50345] by hamaji@chromium.org
  • 3 edits
    2 adds in trunk

2009-10-30 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

counterValueForElementById should return space-separated string for multiple counters
https://bugs.webkit.org/show_bug.cgi?id=30939

Test: fast/css/counters/counterValueForElementById.html

  • rendering/RenderTreeAsText.cpp: (WebCore::writeCounterValuesFromChildren): (WebCore::counterValueForElement):

2009-10-30 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

counterValueForElementById should return space-separated string for multiple counters
https://bugs.webkit.org/show_bug.cgi?id=30939

  • fast/css/counters/counterValueForElementById-expected.txt: Added.
  • fast/css/counters/counterValueForElementById.html: Added.
10:50 AM Changeset in webkit [50344] by eric@webkit.org
  • 4 edits in trunk/WebKit/qt

2009-10-30 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>

Reviewed by Tor Arne Vestbø.

[Qt] Remove the QWebInspector::windowTitleChanged signal,
QEvent::WindowTitleChange can be used to achieve the same.
https://bugs.webkit.org/show_bug.cgi?id=30927

  • Api/qwebinspector.cpp:
  • Api/qwebinspector.h:
  • WebCoreSupport/InspectorClientQt.cpp: (WebCore::InspectorClientQt::updateWindowTitle):
10:43 AM Changeset in webkit [50343] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-30 Sebastian Dröge <sebastian.droege@collabora.co.uk>

Reviewed by Gustavo Noronha.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivate::paint): Add some comments to explain what is happening here.
10:35 AM Changeset in webkit [50342] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-30 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Timothy Hatcher.

Fix Web Inspector crash on the errors/warnings counter click

RenderObject::createVisiblePosition(const Position& position)
understands "null Positions", so we can construct such a Position manually.

https://bugs.webkit.org/show_bug.cgi?id=30499

  • rendering/RenderBox.cpp: (WebCore::RenderBox::positionForPoint):
10:27 AM Changeset in webkit [50341] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-30 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Timothy Hatcher.

Fix Web Inspector crash on the errors/warnings counter click

RenderObject::createVisiblePosition(const Position& position)
understands "null Positions", so we can construct such a Position manually.

https://bugs.webkit.org/show_bug.cgi?id=30499

  • rendering/RenderBox.cpp: (WebCore::RenderBox::positionForPoint):
10:26 AM Changeset in webkit [50340] by hamaji@chromium.org
  • 3 edits in trunk/LayoutTests

2009-10-30 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Mark Rowe.

fast/css/counters/t1204* are not asynchronous anymore
https://bugs.webkit.org/show_bug.cgi?id=30940

  • fast/css/counters/t1204-increment-00-c-o.html:
  • fast/css/counters/t1204-increment-01-c-o.html:
10:19 AM Changeset in webkit [50339] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-30 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25534
[GTK] Objects of ROLE_TABLE should implement the accessible table interface

Second part of the implementation of AtkTable.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (cellAtIndex): (webkit_accessible_table_get_column_at_index): (webkit_accessible_table_get_row_at_index): (webkit_accessible_table_get_caption): (atk_table_interface_init):
9:40 AM Changeset in webkit [50338] by bweinstein@apple.com
  • 1 edit in trunk/WebCore/WebCorePrefix.cpp

Touch WebCorePrefix to try and force a clean build.

9:18 AM Changeset in webkit [50337] by pfeldman@chromium.org
  • 3 edits in trunk/WebCore

2009-10-30 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Add TimelinePanel into the panels enum.

https://bugs.webkit.org/show_bug.cgi?id=30915

  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::showPanel):
  • inspector/front-end/inspector.js: (WebInspector.showTimelinePanel):
8:48 AM Changeset in webkit [50336] by Csaba Osztrogonác
  • 2 edits in trunk/WebKitTools

Unreviewed trivial buildfix.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-30
[Qt] Buildfix for r50333.

  • DumpRenderTree/qt/DumpRenderTree.pro:
8:00 AM Changeset in webkit [50335] by kenneth@webkit.org
  • 7 edits in trunk/LayoutTests

Unreviewed layout test fixes.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
The existing excepted results were wrong, due to the visited links not
being reset between tests.

This was fixed in r50329, so updating the results.

  • platform/qt/fast/block/margin-collapse/block-inside-inline/010-expected.txt:
  • platform/qt/fast/block/margin-collapse/block-inside-inline/011-expected.txt:
  • platform/qt/fast/block/margin-collapse/block-inside-inline/012-expected.txt:
  • platform/qt/fast/block/margin-collapse/block-inside-inline/015-expected.txt:
  • platform/qt/fast/block/margin-collapse/block-inside-inline/016-expected.txt:
  • platform/qt/fast/block/margin-collapse/block-inside-inline/020-expected.txt:
7:13 AM Changeset in webkit [50334] by Csaba Osztrogonác
  • 1 edit
    9 adds in trunk/LayoutTests

[Qt] Add expected files for new svg tests we pass.
Compared to the Mac expected png files.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-30
Reviewed by Holger Hans Peter Freyther.

  • platform/qt/svg/css/arrow-with-shadow-expected.txt: Added.
  • platform/qt/svg/css/circle-in-mask-with-shadow-expected.txt: Added.
  • platform/qt/svg/css/clippath-with-shadow-expected.txt: Added.
  • platform/qt/svg/css/mask-with-shadow-expected.txt: Added.
  • platform/qt/svg/css/path-with-shadow-expected.txt: Added.
  • platform/qt/svg/css/stars-with-shadow-expected.txt: Added.
  • platform/qt/svg/filters/shadow-on-filter-expected.txt: Added.
  • platform/qt/svg/filters/shadow-on-rect-with-filter-expected.txt: Added.
7:02 AM Changeset in webkit [50333] by tonikitoo@webkit.org
  • 2 edits
    2 deletes in trunk/WebKitTools

[Qt] Remove qt/WorkQueue.cpp|h in favor of platform independent WorkQueue
https://bugs.webkit.org/show_bug.cgi?id=30953

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-30
Reviewed by Holger Freyther.

DumpRenderTree/WorkQueue and DumpRenderTree/qt/WorkQueue share mostly the
same implementation. Some Q_ASSERTs differ from ASSERTs basically. Patch
makes qt DRT to share this implementation (as gtk and mac ports do).

  • DumpRenderTree/qt/DumpRenderTree.pro:
  • DumpRenderTree/qt/WorkQueue.cpp: Removed.
  • DumpRenderTree/qt/WorkQueue.h: Removed.
6:51 AM Changeset in webkit [50332] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Unreviewed potential buildbot fix.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
Second try: Reset page history before running each test.

Apparently the QWebHistory::clear() keeps the current page
in history which is not what we want, so we not additionally
sets the history capacity to 0 (forces removing everything)
and then sets it back to its original value.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):

6:38 AM Changeset in webkit [50331] by yurys@chromium.org
  • 1 edit
    2 adds in trunk/LayoutTests

changelog date

6:29 AM Changeset in webkit [50330] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

If the owner widget of the page has a palette set, we
should use that one. This was only working when the
owner was a QWebView. This patch fixes that.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
Reviewed by Holger Hans Peter Freyther.

  • platform/qt/RenderThemeQt.cpp:

(WebCore::RenderThemeQt::applyTheme):

6:07 AM Changeset in webkit [50329] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Unreviewed potential buildbot fix.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
Reset page history before running each test.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):

5:12 AM QtWebKitPerformanceToolBackLog: edited by zecke@selfish.org
(diff)
3:50 AM Changeset in webkit [50328] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-10-30 Pavel Feldman <pfeldman@chromium.org>

Not reviewed: Wire CookieJarChromium to the cookies
backend. This is a final step of a 3-steps raw cookies
access implementation in Chromium.

  • platform/network/chromium/CookieJarChromium.cpp: (WebCore::getRawCookies): (WebCore::deleteCookie):
3:32 AM Changeset in webkit [50327] by yurys@chromium.org
  • 7 edits in trunk/WebCore

2009-10-30 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Fix Chromium crash in console.log in "deeply recursive" function

Check that result of 'frameSourceName' is not null handle
before casting it to String.


Allow V8Proxy::sourceName/sourceLineNumber() to report
that they have failed due to JavaScript stack overflow.

https://bugs.webkit.org/show_bug.cgi?id=30904

  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/ScriptCallStack.cpp: (WebCore::ScriptCallStack::create): (WebCore::ScriptCallStack::ScriptCallStack):
  • bindings/v8/ScriptCallStack.h:
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::sourceLineNumber): (WebCore::V8Proxy::sourceName):
  • bindings/v8/V8Proxy.h:
  • bindings/v8/custom/V8WorkerContextCustom.cpp: (WebCore::CALLBACK_FUNC_DECL):
2:27 AM Changeset in webkit [50326] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's MediaQueryExp
https://bugs.webkit.org/show_bug.cgi?id=30855

Reviewed by Darin Adler.

Inherits MediaQueryExp class from FastAllocBase because it is
instantiated by 'new' in WebCore/css/CSSParser.cpp:4874.

  • css/MediaQueryExp.h:
1:55 AM Changeset in webkit [50325] by abarth@webkit.org
  • 1 edit
    1 add in trunk

2009-10-30 Adam Barth <abarth@webkit.org>

Reviewed by Mark Rowe.

Teach git to ignore some files
https://bugs.webkit.org/show_bug.cgi?id=30951

Ignore WebKitBuild because we never want to version that directory.
Also, ignore the xcode project files so git clean doesn't blow away
your project settings. Finally, ignore the compiled python files in
WebKitTools/Script modules because they clutter up git status.

  • .gitignore: Added.
1:53 AM Changeset in webkit [50324] by zoltan@webkit.org
  • 2 edits in trunk/JavaScriptCore

Allow custom memory allocation control for JavaScriptCore's ListHashSet
https://bugs.webkit.org/show_bug.cgi?id=30853

Reviewed by Darin Adler.

Inherits ListHashSet class from FastAllocBase because it is
instantiated by 'new' in WebCore/rendering/RenderBlock.cpp:1813.

  • wtf/ListHashSet.h:
1:13 AM Changeset in webkit [50323] by oliver@apple.com
  • 5 edits in trunk

Regression: crash enumerating properties of an object with getters or setters
https://bugs.webkit.org/show_bug.cgi?id=30948

Reviewed by Gavin Barraclough

Add a guard to prevent us trying to cache property enumeration on
objects with getters or setters.

12:55 AM Changeset in webkit [50322] by rolandsteiner@chromium.org
  • 17 edits in trunk

Commit patch 1 of ruby implementation, reviewed by Eric Seidel. (see issue 28420)

12:24 AM Changeset in webkit [50321] by mitz@apple.com
  • 2 edits in trunk/WebCore

Removed test for an impossible condition (a glyph in a right-to-left run not having
the RTL flag 0x800)

Reviewed by Adele Peterson.

  • platform/graphics/mac/ComplexTextControllerATSUI.cpp:

(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation):

12:13 AM Changeset in webkit [50320] by oliver@apple.com
  • 2 edits in trunk/JavaScriptCore

REGRESSION (r50218-r50262): E*TRADE accounts page is missing content
https://bugs.webkit.org/show_bug.cgi?id=30947
<rdar://problem/7348833>

Reviewed by Maciej Stachowiak

The logic for flagging that a structure has non-enumerable properties
was in addPropertyWithoutTransition, rather than in the core Structure::put
method. Despite this I was unable to produce a testcase that caused
the failure that etrade was experiencing, but the new assertion in
getEnumerablePropertyNames triggers on numerous layout tests without
the fix, so in effecti all for..in enumeration in any test ends up
doing the required consistency check.

Oct 29, 2009:

11:51 PM Changeset in webkit [50319] by abarth@webkit.org
  • 4 edits in trunk/WebCore

2009-10-29 Adam Barth <abarth@webkit.org>

Reviewed by Dimitri Glazkov.

[V8] Out-of-memory crash in isolated worlds
https://bugs.webkit.org/show_bug.cgi?id=30906

We need to handle the fact that creating a V8:Context might fail. I
don't know how to test this change because creating a context usually
only fails when V8 decides it's using too much memory.

  • bindings/v8/V8IsolatedWorld.cpp: (WebCore::V8IsolatedWorld::V8IsolatedWorld):
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::evaluateInIsolatedWorld): (WebCore::V8Proxy::evaluateInNewContext):
11:29 PM QtBackLog edited by zecke@selfish.org
(diff)
11:26 PM QtBackLog edited by zecke@selfish.org
(diff)
11:12 PM Changeset in webkit [50318] by mitz@apple.com
  • 4 edits in trunk/WebKitTools

Fix “Undefined subroutine” errors in svn-*apply by moving the removeEOL subroutine
from the two scripts that define it but don’t use it to the script that uses it but doesn’t
define it.

Reviewed by Mark Rowe.

  • Scripts/VCSUtils.pm:
  • Scripts/svn-apply:
  • Scripts/svn-unapply:
10:57 PM Changeset in webkit [50317] by mrowe@apple.com
  • 2 edits in trunk/WebKit/mac

Validate the stopSpeaking: selector so that is not always enabled when a WebView is first responder.

Reviewed by Oliver Hunt.

  • WebView/WebHTMLView.mm:

(-[WebHTMLView validateUserInterfaceItemWithoutDelegate:]):

9:07 PM Changeset in webkit [50316] by mrowe@apple.com
  • 2 edits in trunk/WebKit/win

<http://webkit.org/b/30938> REGRESSION(r50233): Windows nightlies crash on launch due to changes
to IWebFramePrivate vtable ordering

Rubber-stamped by Dan Bernstein.

  • Interfaces/IWebFramePrivate.idl: Move the newly-added method to the end of the interface so that

it doesn't affect the ordering of the vtable.

7:39 PM Changeset in webkit [50315] by eric@webkit.org
  • 5 edits
    3 adds
    1 delete in trunk

2009-10-29 Erik Arvidsson <arv@chromium.org>

Reviewed by Eric Seidel.

Make links mouse focusable only on GTK and QT.

Links are now always mouse focusable on GTK and QT. On other platforms
the link needs a tabIndex or it needs to be contentEditable.

https://bugs.webkit.org/show_bug.cgi?id=26856

  • fast/events/click-focus-anchor-expected.txt:
  • fast/events/click-focus-anchor.html:
  • platform/gtk/fast/events/click-focus-anchor-expected.txt: Added.
  • platform/mac/fast/events/click-focus-anchor-expected.txt: Removed.
  • platform/qt/fast/events/click-focus-anchor-expected.txt: Added.

2009-10-29 Erik Arvidsson <arv@chromium.org>

Reviewed by Eric Seidel.

Make links mouse focusable only on GTK and QT.

Links are now always mouse focusable on GTK and QT. On other platforms
the link needs a tabIndex or it needs to be contentEditable.

https://bugs.webkit.org/show_bug.cgi?id=26856

  • html/HTMLAnchorElement.cpp: (WebCore::HTMLAnchorElement::isMouseFocusable):
7:30 PM Changeset in webkit [50314] by eric@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

2009-10-29 Jonathan Dixon <joth@chromium.org>

Reviewed by Eric Seidel.

Add expected files for GTK and QT builds, for test text-input-webkit-radius.html
Expected fils collected from buildbot:
http://build.webkit.org/results/Qt%20Linux%20Release/r50286%20(3257)/fast/css/
http://build.webkit.org/results/GTK%20Linux%20Release/r50286%20(5308)/fast/css/

  • platform/gtk/fast/css/text-input-with-webkit-border-radius-expected.txt: Added.
  • platform/qt/fast/css/text-input-with-webkit-border-radius-expected.txt: Added.
7:00 PM Changeset in webkit [50313] by mitz@apple.com
  • 2 edits in trunk/LayoutTests

Updated Tiger-specific result now that Tiger matches current behavior. A Tiger-specific
result is still needed because of the difference in Thai fonts.

  • platform/mac-tiger/fast/text/atsui-multiple-renderers-expected.txt:
5:49 PM Changeset in webkit [50312] by mitz@apple.com
  • 6 edits
    3 deletes in trunk

WebCore: Complete the fix for hit-testing and selection highlighting in ligatures for the ATSUI
code path.

Reviewed by Sam Weinig.

  • platform/graphics/mac/ComplexTextController.h: Added m_ltr member to ComplexTextRun.
  • platform/graphics/mac/ComplexTextControllerATSUI.cpp:

(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation): Skip over deleted
glyphs, but update indexes and advances correctly.
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): Initialize m_ltr.

LayoutTests: Updated a test to account for font metric differences between Mac OS X versions. Removed
Leopard-specific results for another test now that Leopard matches current behavior.

Reviewed by Sam Weinig.

  • platform/mac-leopard/fast/text/atsui-multiple-renderers-expected.checksum: Removed.
  • platform/mac-leopard/fast/text/atsui-multiple-renderers-expected.png: Removed.
  • platform/mac-leopard/fast/text/atsui-multiple-renderers-expected.txt: Removed.
  • platform/mac/fast/text/ligature-subdivision-expected.txt:
  • platform/mac/fast/text/script-tests/ligature-subdivision.js:
5:26 PM Changeset in webkit [50311] by adele@apple.com
  • 2 edits in trunk

Updating Radar numbers

4:37 PM Changeset in webkit [50310] by adele@apple.com
  • 6 edits
    2 adds in trunk

REGRESSION(3.2.3 - 4.0.2): Message composing: when I undo a color change to text in Mail, undo/redo behaves strangely
<rdar://problem/7067033>
https://bugs.webkit.org/show_bug.cgi?id=30892

WebCore:

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-29
Reviewed by Darin Adler.

This problem shows in any scenario where it is necessary to split a text
node to apply a style. SplitElementCommand and WrapContentsInDummySpanCommand both
have member variables initialized in the constructor to keep reference to elements
they need to operate upon. These reference are not updated when reapplying the command.
For this reason it is necessary to guarantee that unapply doesn not delete the references
and that these commands implement doReapply to correctly reuse the existing
elements.

Test: editing/undo/redo-style.html

  • editing/SplitElementCommand.cpp:

(WebCore::SplitElementCommand::executeApply): Added.
(WebCore::SplitElementCommand::doApply): Modified to call executeApply.
(WebCore::SplitElementCommand::doUnapply): Doesn't release m_element1.
(WebCore::SplitElementCommand::doReapply): Added.

  • editing/SplitElementCommand.h: Added doReapply and executeApply.
  • editing/WrapContentsInDummySpanCommand.cpp:

(WebCore::WrapContentsInDummySpanCommand::executeApply): Added.
(WebCore::WrapContentsInDummySpanCommand::doApply): Modified to call executeApply.
(WebCore::WrapContentsInDummySpanCommand::doUnapply): Doesn't release m_dummySpan.
(WebCore::WrapContentsInDummySpanCommand::doReapply): Added.

  • editing/WrapContentsInDummySpanCommand.h: Added doReapply and executeApply.

LayoutTests:

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-29

  • editing/undo/redo-style-expected.txt: Added.
  • editing/undo/redo-style.html: Added.
4:22 PM Changeset in webkit [50309] by jorlow@chromium.org
  • 3 edits in trunk/WebCore

2009-10-29 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Darin Fisher.

DOM Storage's condition variable needs to handle spurious wakeups
https://bugs.webkit.org/show_bug.cgi?id=30920

Add a boolean to keep track of whether it's been terminated. Clean
up the locking code a tiny bit to make it easier to read. There's
no way to reproduce this reliably in a LayoutTest.

  • storage/LocalStorageThread.cpp: (WebCore::LocalStorageThread::LocalStorageThread): (WebCore::LocalStorageThread::terminate): (WebCore::LocalStorageThread::performTerminate):
  • storage/LocalStorageThread.h:
4:22 PM Changeset in webkit [50308] by andersca@apple.com
  • 2 edits in trunk/WebKit/mac

https://bugs.webkit.org/show_bug.cgi?id=30932
<rdar://problem/7350269>

Reviewed by Sam Weinig.

REGRESSION: Crash when turning on Private Browsing on site with flash.

Null check setValue.

  • Plugins/WebNetscapePluginView.mm:

(-[WebNetscapePluginView privateBrowsingModeDidChange]):

3:49 PM Changeset in webkit [50307] by kenneth@webkit.org
  • 4 edits in trunk

WebKitTools: Implement the Qt version of DRT dumpBackForwardList().

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Reviewed by Oliver Hunt.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::dumpHistoryItem):
(WebCore::DumpRenderTree::dumpBackForwardList):

LayoutTests: Removed test that passes after implementing the Qt DRT version
of dumpBackForwardList().

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Reviewed by Oliver Hunt.

  • platform/qt/Skipped:
3:43 PM Changeset in webkit [50306] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2009-10-29 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

[Chromium] Include ComplextTextController into Chromium Mac project.

  • WebCore.gyp/WebCore.gyp: Added include rule for ComplextText* files.
3:32 PM Changeset in webkit [50305] by eric.carlson@apple.com
  • 2 edits in trunk/WebKit/mac

2009-10-29 Eric Carlson <eric.carlson@apple.com>

Reviewed by Kevin Decker.

Refactor some duplicate plug-in clean up code into shared functions.

  • Plugins/WebPluginController.mm: (-[WebPluginController stopOnePlugin:]): New, stop a plug-in. (-[WebPluginController destroyOnePlugin:]): New, destroy plug-in. (-[WebPluginController stopAllPlugins]): Call stopOnePlugin. (-[WebPluginController destroyPlugin:]): Call stopOnePlugin and destroyOnePlugin. (-[WebPluginController destroyAllPlugins]): Call destroyOnePlugin.
3:32 PM Changeset in webkit [50304] by hamaji@chromium.org
  • 2 edits
    1 add
    3 deletes in trunk/LayoutTests

2009-10-29 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

fast/css/counters/invalidate-cached-counter-node can be dumpAsText
https://bugs.webkit.org/show_bug.cgi?id=30931

  • fast/css/counters/invalidate-cached-counter-node-expected.txt: Added.
  • fast/css/counters/invalidate-cached-counter-node.html:
  • platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.checksum: Removed.
  • platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.png: Removed.
  • platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.txt: Removed.
2:59 PM Changeset in webkit [50303] by timothy@apple.com
  • 2 edits in trunk/WebCore

Fix tabbing through element attributes in the Web Insector.

https://bugs.webkit.org/show_bug.cgi?id=30429

Reviewed by Pavel Feldman.

  • inspector/front-end/ElementsTreeOutline.js:

(WebInspector.ElementsTreeElement.prototype._attributeEditingCommitted): Don't call _updateTitle,
it is called for us when removeAttribute succeeds in the back-end.
(WebInspector.ElementsTreeElement.prototype._textNodeEditingCommitted): Ditto except for nodeValue.
(WebInspector.ElementsTreeElement.prototype._editingCancelled): Don't call _updateTitle, editing code reverts.
(WebInspector.ElementsTreeElement.prototype._updateTitle): Return early if we are editing.

2:54 PM Changeset in webkit [50302] by Csaba Osztrogonác
  • 2 edits in trunk/LayoutTests

[Qt] Remove some tests that now pass.

Patch by Csaba Osztrogonác <Csaba Osztrogonác> on 2009-10-29

  • platform/qt/Skipped:
2:25 PM Changeset in webkit [50301] by mitz@apple.com
  • 4 edits
    1 copy
    4 adds in trunk

Problem editing or selecting text containing ligatures
https://bugs.webkit.org/show_bug.cgi?id=30025

Reviewed by Sam Weinig.

WebCore:

Test: platform/mac/fast/text/ligature-subdivision.html

  • platform/graphics/mac/ComplexTextController.cpp:

(WebCore::ComplexTextController::ComplexTextController): Initialize
m_characterInCurrentGlyph.
(WebCore::ComplexTextController::offsetForPosition): If the hit glyph spans multiple
characters, compute the hit character based on dividing the glyph’s total advance into
a number of equal intervals equal to the number of characters and assigning the hit to the
character corresponding to the hit interval.
(WebCore::ComplexTextController::advance): If the final offset occurs mid-glyph, advance
by a fraction of the glyph’s total advance.

  • platform/graphics/mac/ComplexTextController.h: Added m_characterInCurrentGlyph.

LayoutTests:

  • platform/mac/fast/text/ligature-subdivision-expected.txt: Added.
  • platform/mac/fast/text/ligature-subdivision.html: Added.
  • platform/mac/fast/text/script-tests: Added.
  • platform/mac/fast/text/script-tests/TEMPLATE.html: Copied from LayoutTests/fast/js/script-tests/TEMPLATE.html.
  • platform/mac/fast/text/script-tests/ligature-subdivision.js: Added.
2:25 PM Changeset in webkit [50300] by bweinstein@apple.com
  • 2 edits in trunk/WebCore

Fixes <http://webkit.org/b/30918>.
Web Inspector: Datagrid Rows on Windows not properly aligned.

Reviewed by Timothy Hatcher.

Use line-height for the table rows to make sure the height
of our text and the height of the table rows are consistent.

  • inspector/front-end/inspector.css:
1:33 PM Changeset in webkit [50299] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2009-10-29 Adam Barth <abarth@webkit.org>

No review, rolling out r50296.
http://trac.webkit.org/changeset/50296

  • bindings/v8/V8IsolatedWorld.cpp: (WebCore::V8IsolatedWorld::V8IsolatedWorld):
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::evaluateInIsolatedWorld): (WebCore::V8Proxy::evaluateInNewContext):
1:25 PM Changeset in webkit [50298] by xan@webkit.org
  • 2 edits in trunk/WebKitTools

2009-10-29 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

Use the GTK+ main loop instead of rolling our own mini-version of
it.

  • DumpRenderTree/gtk/DumpRenderTree.cpp: (dump): (runTest): (webViewLoadFinished):
1:21 PM Changeset in webkit [50297] by kenneth@webkit.org
  • 2 edits in trunk/LayoutTests

Rubberstamped by Adam Roben.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Remove two tests that pass on the buildbot.

  • platform/qt/Skipped:
1:15 PM Changeset in webkit [50296] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-10-29 Adam Barth <abarth@webkit.org>

Reviewed by Dimitri Glazkov.

[V8] Out-of-memory crash in isolated worlds
https://bugs.webkit.org/show_bug.cgi?id=30906

We need to handle the fact that creating a V8:Context might fail. I
don't know how to test this change because creating a context usually
only fails when V8 decides it's using too much memory.

  • bindings/v8/V8IsolatedWorld.cpp: (WebCore::V8IsolatedWorld::V8IsolatedWorld):
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::evaluateInIsolatedWorld): (WebCore::V8Proxy::evaluateInNewContext):
1:05 PM Changeset in webkit [50295] by kov@webkit.org
  • 2 edits in trunk/WebCore

Reviewed by Xan Lopez.

Remove build warning introduced by r50284.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivate::duration):
1:05 PM Changeset in webkit [50294] by kenneth@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed Qt buildbot fix.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Add some files to skipped that depend on phonon support,
which is disabled on the Qt buildbot for now.

  • platform/qt/Skipped:
1:01 PM Changeset in webkit [50293] by Dimitri Glazkov
  • 4 edits in trunk/WebCore

2009-10-29 Dimitri Glazkov <Dimitri Glazkov>

Reviewed by Adam Barth.

[V8] Remove random crashes by removing retrieval of V8 context during garbage collection.

https://bugs.webkit.org/show_bug.cgi?id=30919

Unfortunately, I haven't been able to trigger this crash explicitly, so no test :(.

  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::jsWrapperForDOMObject): Added new "assume-it's-there" getter.
  • bindings/v8/V8DOMWrapper.h: Added getter decl.
  • bindings/v8/V8GCController.cpp: (WebCore::GCPrologueVisitor::visitDOMWrapper): Changed to use explicit getter.
12:43 PM Changeset in webkit [50292] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2009-10-29 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

[Chromium] Adjust the project files to sync up with
http://trac.webkit.org/changeset/50259

  • WebCore.gypi: Renamed and added files.
12:41 PM Changeset in webkit [50291] by mitz@apple.com
  • 2 edits in trunk/LayoutTests

Updated expected result for Tiger

  • platform/mac-tiger/fast/text/international/thai-line-breaks-expected.txt:
12:24 PM Changeset in webkit [50290] by kenneth@webkit.org
  • 2 edits in trunk/LayoutTests

Rubberstamped by Adam Roben.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Remove tests that pass after the change introduced in r50289.

  • platform/qt/Skipped:
12:06 PM Changeset in webkit [50289] by kenneth@webkit.org
  • 4 edits in trunk/WebKitTools

Implement DRT functionality for Qt introduced in
r28690 and r28705.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Reviewed by Adam Roben.

  • Implemented pathToLocalResource which exposes the functionality of

converting a given unix path to the correct location on Windows.

  • Implemented a way to remove machine-dependent information from paths

in layout test results.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::urlSuitableForTestResult):
(WebCore::WebPage::javaScriptConsoleMessage):

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp:

(LayoutTestController::pathToLocalResource):

  • DumpRenderTree/qt/LayoutTestControllerQt.h:
11:29 AM Changeset in webkit [50288] by eric@webkit.org
  • 4 edits in trunk

2009-10-29 Laszlo Gombos <Laszlo Gombos>

Reviewed by Tor Arne Vestbø.

[Qt] Implement DELETE HTTP method for XmlHttpRequest
https://bugs.webkit.org/show_bug.cgi?id=30894

No new tests as this functionality is already tested by the
xmlhttprequest LayoutTests. As this patch depends on an unreleased
version of the dependent QtNetwork library and the tests will be
enabled later once the dependent library is released (and the
buildbot is updated).

  • platform/network/qt/QNetworkReplyHandler.cpp: (WebCore::QNetworkReplyHandler::QNetworkReplyHandler): (WebCore::QNetworkReplyHandler::start):

2009-10-29 Laszlo Gombos <Laszlo Gombos>

Reviewed by Tor Arne Vestbø.

[Qt] Implement DELETE HTTP method for XmlHttpRequest
https://bugs.webkit.org/show_bug.cgi?id=30894

No new tests as this functionality is already tested by the
xmlhttprequest LayoutTests. As this patch depends on an unreleased
version of the dependent QtNetwork library and the tests will be
enabled later once the dependent library is released (and the
buildbot is updated).

  • Api/qwebframe.cpp: (QWebFrame::load):
11:15 AM Changeset in webkit [50287] by bweinstein@apple.com
  • 3 edits in trunk/WebCore

Part of <http://webkit.org/b/30483>.
Web Inspector: Always show the Local and Session Storage Views.

Reviewed by Timothy Hatcher.

Even if the length of the DOM Storage entry array is 0, still
generate the Datagrid because users can add things storage items
through the UI, so we should allow them to even if there isn't
anything there currently.

  • English.lproj/localizedStrings.js: Removed "This Storage is Empty".
  • inspector/front-end/DOMStorageItemsView.js:

(WebInspector.DOMStorageItemsView.prototype._showDOMStorageEntries):

10:58 AM Changeset in webkit [50286] by pfeldman@chromium.org
  • 3 edits in trunk/WebCore

2009-10-29 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Add TimelinePanel into the panels enum.

https://bugs.webkit.org/show_bug.cgi?id=30915

  • inspector/InspectorController.cpp: (WebCore::InspectorController::specialPanelForJSName):
  • inspector/InspectorController.h: (WebCore::InspectorController::):
10:53 AM Changeset in webkit [50285] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-10-29 Xan Lopez <xlopez@igalia.com>

Reviewed by Oliver Hunt.

[GTK] Threading problems with some of the tests
https://bugs.webkit.org/show_bug.cgi?id=30814

Create strings shared among threads with crossThreadString
constructor method.

  • storage/Database.cpp: (WebCore::Database::Database):
10:50 AM Changeset in webkit [50284] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-10-29 Sebastian Dröge <sebastian.droege@collabora.co.uk>

Reviewed by Gustavo Noronha.

https://bugs.webkit.org/show_bug.cgi?id=30308

Add support for ARGB videos.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivate::paint): Create the Cairo image surface for ARGB32 or RGB24 depending on the buffer's caps.
  • platform/graphics/gtk/VideoSinkGStreamer.cpp: (webkit_video_sink_timeout_func): (webkit_video_sink_render): Handle ARGB video and convert GStreamer's ARGB to Cairo's for displaying.
10:42 AM Changeset in webkit [50283] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-29 Anton Muhin <antonm@chromium.org>

Reviewed by David Levin.

[chromium] expose a method to access memory usage information in ChromiumBridge
Declare a static method to be implemented by http://codereview.chromium.org/332010/
https://bugs.webkit.org/show_bug.cgi?id=30829

  • platform/chromium/ChromiumBridge.h:
10:34 AM Changeset in webkit [50282] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-10-29 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25679
[Gtk] Improve accessibility of focusable lists

Implements the AtkSelection interface and enables the corresponding
(and expected) object:selection-changed event.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (optionFromList): (optionFromSelection): (atk_selection_interface_init): (webkit_accessible_selection_add_selection): (webkit_accessible_selection_clear_selection): (webkit_accessible_selection_ref_selection): (webkit_accessible_selection_get_selection_count): (webkit_accessible_selection_is_child_selected): (webkit_accessible_selection_remove_selection): (webkit_accessible_selection_select_all_selection): (GetAtkInterfaceTypeFromWAIType):
  • accessibility/gtk/AXObjectCacheAtk.cpp: (AXObjectCache::postPlatformNotification):
10:27 AM Changeset in webkit [50281] by sfalken@apple.com
  • 1 edit in trunk/WebKit/win/Interfaces/WebKit.idl

Touch WebKit.idl to force a rebuild.

10:26 AM Changeset in webkit [50280] by bweinstein@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Steve Falkenburg.

Windows Release bots are having problems in xmlhttprequest tests, Skipping them temporarily
to see if this eliminates the problem, and it is a test causing httpd problems.

  • platform/win/Skipped:
10:18 AM Changeset in webkit [50279] by jianli@chromium.org
  • 2 edits in trunk/WebCore

Bug 30655 - Only plain text should be copied to clipboard for TextArea.
https://bugs.webkit.org/show_bug.cgi?id=30655

Reviewed by Darin Adler.

  • editing/Editor.cpp:

(WebCore::nodeIsInTextFormControl):
(WebCore::Editor::cut):
(WebCore::Editor::copy):

9:58 AM Changeset in webkit [50278] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-10-29 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: hide timeline for now - not yet ready.

https://bugs.webkit.org/show_bug.cgi?id=30912

  • inspector/front-end/inspector.js: (WebInspector._createPanels):
9:46 AM Changeset in webkit [50277] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-10-29 Pavel Feldman <pfeldman@chromium.org>

Not reviewed: deploy Web Inspector's images in WebCore.gypi.

  • WebCore.gypi:
9:36 AM Changeset in webkit [50276] by pfeldman@chromium.org
  • 7 edits in trunk/WebCore

2009-10-29 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: get rid of timelineProfilerEnabled method.

https://bugs.webkit.org/show_bug.cgi?id=30911

  • inspector/InspectorBackend.cpp:
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::setWindowVisible):
  • inspector/InspectorController.h:
  • inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel.prototype._toggleTimelineButtonClicked):
9:12 AM Changeset in webkit [50275] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-10-29 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Darin Fisher.

Create stub methods for raw cookies access in ChromiumBridge.h

https://bugs.webkit.org/show_bug.cgi?id=30910

  • platform/chromium/ChromiumBridge.h:
8:58 AM QtWebKitContrib edited by vestbo@webkit.org
(diff)
8:55 AM QtWebKitContrib edited by vestbo@webkit.org
(diff)
8:44 AM Changeset in webkit [50274] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's RareData
https://bugs.webkit.org/show_bug.cgi?id=30858

Reviewed by Darin Adler.

Inherits RareData struct from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSSelector.h:259 and
it is no need to be copyable.

  • css/CSSSelector.h:
8:41 AM Changeset in webkit [50273] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's ShorthandScope
https://bugs.webkit.org/show_bug.cgi?id=30859

Reviewed by Darin Adler.

Inherits ShorthandScope class from FastAllocBase because it is
instantiated by 'new' in WebCore/css/CSSParser.cpp:902 and
it is no need to be copyable.

  • css/CSSParser.h:
8:38 AM Changeset in webkit [50272] by zoltan@webkit.org
  • 2 edits in trunk/WebCore

Allow custom memory allocation control for WebCore's AXObjectCache
https://bugs.webkit.org/show_bug.cgi?id=30848

Reviewed by Darin Adler.

Inherits AXObjectCache class from FastAllocBase because it is
instantiated by 'new' in WebCore/dom/Document.cpp:1537 and
it is no need to be copyable.

  • accessibility/AXObjectCache.h:
8:18 AM Changeset in webkit [50271] by abarth@webkit.org
  • 3 edits
    2 adds in trunk

2009-10-29 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Don't run JavaScript URLs in view source mode
https://bugs.webkit.org/show_bug.cgi?id=30881

Test that we don't run JavaScript URLs in view source mode.

This test actually passes before this change. I don't know how to
actually run JavaScript URLs in view source frames using DRT, but it's
possible in an abitrary embedding of WebKit.

  • http/tests/security/view-source-no-javascript-url-expected.txt: Added.
  • http/tests/security/view-source-no-javascript-url.html: Added.

2009-10-28 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Don't run JavaScript URLs in view source mode
https://bugs.webkit.org/show_bug.cgi?id=30881

Just say no.

Test: http/tests/security/view-source-no-javascript-url.html

  • bindings/ScriptControllerBase.cpp: (WebCore::ScriptController::executeIfJavaScriptURL):
7:43 AM Changeset in webkit [50270] by kenneth@webkit.org
  • 6 edits in trunk/WebKit/qt

Remove QWebView::guessUrlFromString() and replace its use
with the new QUrl::fromUserInput() if using Qt 4.6 or newer.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Reviewed by Tor Arne Vestbø.

  • Api/qwebview.cpp:
  • Api/qwebview.h:
  • QGVLauncher/main.cpp:

(urlFromUserInput):
(WebPage::applyProxy):
(MainWindow::load):

  • QtLauncher/main.cpp:

(urlFromUserInput):
(MainWindow::MainWindow):
(MainWindow::changeLocation):

  • tests/qwebview/tst_qwebview.cpp:
6:23 AM Changeset in webkit [50269] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

Serialize directly to the stream, and not first to an QByteArray,
that is later serialized. That is slower and also uses more bytes.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-28
Reviewed by Tor Arne Vestbø.

  • Api/qwebhistory.cpp:

(operator<<):
(operator>>):

5:48 AM Changeset in webkit [50268] by kov@webkit.org
  • 2 edits in trunk/WebCore

Unreviewed. Fixes style problems pointed out by Evan Martin.

5:46 AM Changeset in webkit [50267] by kov@webkit.org
  • 2 edits in trunk/WebKit/gtk

Unreviewed. Trivial fix - move public API declaration into the
public subsection of the doc.

  • docs/webkitgtk-sections.txt:
2:21 AM Changeset in webkit [50266] by mitz@apple.com
  • 2 edits in trunk/WebCore

Rubber-stamped by Mark Rowe.

64-bit Leopard build fix after r50259

  • platform/graphics/mac/ComplexTextControllerATSUI.cpp:

Declared ATSUTextInserted in 64-bit.
(WebCore::fontHasMirroringInfo): Use %d format and cast to int.
(WebCore::disableLigatures): Ditto.
(WebCore::initializeATSUStyle): Ditto.
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Ditto.

12:32 AM Changeset in webkit [50265] by zoltan@webkit.org
  • 2 edits in trunk/JavaScriptCore

Add cacheFlush support for Thumb-2 on Linux
https://bugs.webkit.org/show_bug.cgi?id=30865

Patch by Gabor Loki <loki@inf.u-szeged.hu> on 2009-10-29
Reviewed by Gavin Barraclough.

  • jit/ExecutableAllocator.h:

(JSC::ExecutableAllocator::cacheFlush):

12:23 AM Changeset in webkit [50264] by mitz@apple.com
  • 3 edits in trunk/WebCore

Tiger build fix after r50259

  • platform/graphics/mac/ComplexTextController.h:
  • platform/graphics/mac/ComplexTextControllerATSUI.cpp:
12:12 AM Changeset in webkit [50263] by mitz@apple.com
  • 2 edits in trunk/WebCore

Tiger build fix after r50259

  • platform/graphics/mac/ComplexTextControllerATSUI.cpp:

Oct 28, 2009:

11:07 PM Changeset in webkit [50262] by sfalken@apple.com
  • 1 edit in trunk/WebCore/ChangeLog

Fix changelog

11:06 PM Changeset in webkit [50261] by sfalken@apple.com
  • 2 edits
    1 add in trunk/WebCore

https://bugs.webkit.org/show_bug.cgi?id=30899
WebKit fails to build release on 32-bit Windows systems

Reviewed by Mark Rowe.

  • WebCore.vcproj/WebCore.vcproj: Excluded files from project.
  • bindings/js/JSBindingsAllInOne.cpp: Added.
11:04 PM Changeset in webkit [50260] by mitz@apple.com
  • 2 edits in trunk/WebCore

Fixed typos in color names.

Reviewed by Jon Honeycutt.

  • inspector/front-end/Color.js:
10:35 PM Changeset in webkit [50259] by mitz@apple.com
  • 3 edits
    1 copy
    3 moves
    1 add
    2 deletes in trunk

WebCore: Share code between the ATSUI- and Core Text-based Font implementations by doing the
following:

  • Generalize CoreTextController as ComplexTextController, keeping the Core Text-specific parts in ComplexTextControllerCoreText.cpp.
  • Generalize FontMacCoreText as FontComplexTextMac using ComplexTextController
  • Implement ATSUI-specific parts of ComplexTextController in ComplexTextControllerATSUI.
  • Remove FontMacATSUI.

Reviewed by Sam Weinig.

  • WebCore.xcodeproj/project.pbxproj: Removed CoreTextController.{cpp,h}, FontMacATSUI.mm,

and FontMacCoreText.cpp, and added ComplexTextController.{cpp,h},
ComplexTextControllerATSUI.cpp, ComplexTextControllerCoreText.cpp, and
FontComplexTextMac.cpp.

  • platform/graphics/mac/ComplexTextController.cpp: Copied from CoreTextController.cpp and

kept the non-Core Text-specific bits.
(WebCore::ComplexTextController::ComplexTextController): Updated for renames, including
its own.
(WebCore::ComplexTextController::offsetForPosition): Updated for renames and for
m_complexTextRuns holding references instead of objects.
(WebCore::ComplexTextController::collectComplexTextRuns): Updated for renames, including
its own.
(WebCore::ComplexTextController::advance): Updated for renames.
(WebCore::ComplexTextController::adjustGlyphsAndAdvances): Updated for renames and for
m_complexTextRuns holding references instead of objects, and changed to use the glyphs()
and advances() accessors.

  • platform/graphics/mac/ComplexTextController.h: Copied from CoreTextController.h and

renamed CoreTextController to ComplexTextController and CoreTextRun to ComplexTextRun. Made
the latter RefCounted, added ATSUI-specific members to it, and made some other members
Core Text-specific. Renamed m_coreTextRuns to m_complexTextRuns and made it hold references
rather than objects.
(WebCore::ComplexTextController::ComplexTextRun::create):
(WebCore::ComplexTextController::ComplexTextRun::glyphs):
(WebCore::ComplexTextController::ComplexTextRun::advances):

  • platform/graphics/mac/ComplexTextControllerATSUI.cpp: Added. Includes ATSUI-specific

parts of the ComplexTextController implementation.
(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation): This ATSUI
callback populates the ComplexTextRun’s glyphs, advances and indices vectors. It is invoked
when the ComplexTextRun constructor calls ATSUGetGlyphBounds().
(WebCore::isArabicLamWithAlefLigature): Helper function, copied from FontMacATSUI.mm.
(WebCore::shapeArabic): Helper function, adapted from FontMacATSUI.mm.
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): Sets up the
ATSUTextLayout, substituting the text buffer if necessary for things like shaping Arabic,
mirroring glyphs or directionality overrides, then calls ATSUGetGlyphBounds() in order to
get the glyphs, advances and indices vectors populated.
(WebCore::fontHasMirroringInfo): Helper function, copied from FontMacATSUI.mm.
(WebCore::disableLigatures): Ditto.
(WebCore::initializeATSUStyle): Ditto, somewhat cleaned up and simplified.
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Constructs
ComplexTextRuns, either missing-glyphs ones or ATSUTextLayout-based ones.

  • platform/graphics/mac/ComplexTextControllerCoreText.cpp: Copied from

CoreTextController.cpp and kept the Core Text-specific bits.
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): Updated for renames,
including its own, and moved the code to initialize m_glyphs and m_advances here. Previously
this was done in adjustGlyphsAndAdvances().
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Updated for renames,
including its own.

  • platform/graphics/mac/CoreTextController.cpp: Removed.
  • platform/graphics/mac/CoreTextController.h: Removed.
  • platform/graphics/mac/FontComplexTextMac.cpp: Renamed FontMacCoreText.cpp to this.

(WebCore::Font::selectionRectForComplexText): Changed to use ComplexTextController instead
of CoreTextController.
(WebCore::Font::drawComplexText): Ditto.
(WebCore::Font::floatWidthForComplexText): Ditto.
(WebCore::Font::offsetForPositionForComplexText): Ditto.

  • platform/graphics/mac/FontMacATSUI.mm: Removed.
  • platform/graphics/mac/FontMacCoreText.cpp: Removed.

LayoutTests: Removed Leopard-specific result after changing ATSUI missing-glyph behavior to match the
Core Text and fast code paths' behavior.

  • platform/mac-leopard/fast/text/international/khmer-selection-expected.txt: Removed.
10:23 PM Changeset in webkit [50258] by rolandsteiner@chromium.org
  • 2 edits in trunk/WebKitTools

2009-10-28 Roland Steiner <rolandsteiner@chromium.org>

Adding myself to the committers list.

  • Scripts/modules/committers.py:
9:20 PM Changeset in webkit [50257] by Chris Fleizach
  • 2 edits in trunk/WebKitTools

Adding myself to the committers list.

7:48 PM Changeset in webkit [50256] by Chris Fleizach
  • 6 edits
    2 adds in trunk

WebCore: WAI-ARIA: add support for 'option' role
https://bugs.webkit.org/show_bug.cgi?id=30843

Reviewed by Darin Adler.

Test: accessibility/aria-option-role.html

  • accessibility/AccessibilityListBoxOption.h:

(WebCore::AccessibilityListBoxOption::canHaveChildren):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::RoleEntry::):
(WebCore::AccessibilityRenderObject::canHaveChildren):

LayoutTests: Fix video-pause-immediately.html to avoid using setTimeout().

Patch by Andrew Scherkus <scherkus@chromium.org> on 2009-10-28
Reviewed by Eric Carlson.

This also addresses the flakiness on Windows, where 200ms was not enough time to advance the clock past zero.

https://bugs.webkit.org/show_bug.cgi?id=30889

  • media/video-pause-immediately.html: Replace setTimeout() with timeupdate events.
6:40 PM Changeset in webkit [50255] by barraclough@apple.com
  • 5 edits in trunk/JavaScriptCore

JSC JIT on ARMv7 cannot link jumps >16Mb range
https://bugs.webkit.org/show_bug.cgi?id=30891

Patch by Gavin Barraclough <barraclough@apple.com> on 2009-10-28
Reviewed by Oliver Hunt.

Start planing all relative jumps as move-32-bit-immediate-to-register-BX.
In the cases where the jump would fall within a relative jump range, use a relative jump.

  • JavaScriptCore.xcodeproj/project.pbxproj:
  • assembler/ARMv7Assembler.h:

(JSC::ARMv7Assembler::~ARMv7Assembler):
(JSC::ARMv7Assembler::LinkRecord::LinkRecord):
(JSC::ARMv7Assembler::):
(JSC::ARMv7Assembler::executableCopy):
(JSC::ARMv7Assembler::linkJump):
(JSC::ARMv7Assembler::relinkJump):
(JSC::ARMv7Assembler::setInt32):
(JSC::ARMv7Assembler::isB):
(JSC::ARMv7Assembler::isBX):
(JSC::ARMv7Assembler::isMOV_imm_T3):
(JSC::ARMv7Assembler::isMOVT):
(JSC::ARMv7Assembler::isNOP_T1):
(JSC::ARMv7Assembler::isNOP_T2):
(JSC::ARMv7Assembler::linkJumpAbsolute):
(JSC::ARMv7Assembler::twoWordOp5i6Imm4Reg4EncodedImmFirst):
(JSC::ARMv7Assembler::twoWordOp5i6Imm4Reg4EncodedImmSecond):
(JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp5i6Imm4Reg4EncodedImm):

  • assembler/MacroAssemblerARMv7.h:

(JSC::MacroAssemblerARMv7::makeJump):
(JSC::MacroAssemblerARMv7::makeBranch):

  • jit/JIT.h:
  • wtf/Platform.h:
6:25 PM Changeset in webkit [50254] by oliver@apple.com
  • 15 edits in trunk/JavaScriptCore

Improve for..in enumeration performance
https://bugs.webkit.org/show_bug.cgi?id=30887

Reviewed by Geoff Garen.

Improve indexing of an object with a for..in iterator by
identifying cases where get_by_val is being used with a iterator
as the subscript and replace it with a new get_by_pname
bytecode. get_by_pname then optimizes lookups that directly access
the base object.

5:54 PM Changeset in webkit [50253] by eric@webkit.org
  • 12 edits in trunk/WebCore

2009-10-28 Jens Alfke <snej@chromium.org>

Reviewed by Eric Seidel.

Fix GCC compiler warnings in WebCore, and enable -Wall and -Werror for Chromium build.
https://bugs.webkit.org/show_bug.cgi?id=30716

  • WebCore.gyp/WebCore.gyp: Enable "chromium_code" flag, just on Mac build for now.
  • accessibility/AccessibilityRenderObject.cpp: (WebCore::createARIARoleMap): Fix struct visibiity warning.
  • bindings/v8/ScriptCallStack.h: Fix out-of-order member initialization warning.
  • bindings/v8/V8Collection.h: (WebCore::getV8Object): Function in header should not be 'static' (fixes unused-static warning.)
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::convertNewNodeToV8Object): Fix signed/unsigned comparison warning.
  • bindings/v8/V8GCController.cpp: (WebCore::ObjectGrouperVisitor::applyGrouping): Fix unused-variable warning.
  • css/CSSPrimitiveValueMappings.h: (WebCore::CSSPrimitiveValue::CSSPrimitiveValue): Enable ListButtonPart case to avoid

warning about missing cases in 'switch' statement.

  • editing/EditorCommand.cpp: (WebCore::createCommandMap): Fix struct visibiity warning.
  • platform/graphics/skia/PlatformContextSkia.cpp: (PlatformContextSkia::State::State): Fix out-of-order member initialization warning.
  • rendering/RenderMediaControlsChromium.cpp: (WebCore::RenderMediaControlsChromium::shouldRenderMediaControlPart): Add empty 'default' case in

'switch' statement to avoid missing-case warning.

(WebCore::RenderMediaControlsChromium::paintMediaControlsPart): Ditto.

  • xml/XPathFunctions.cpp: (WebCore::XPath::createFunctionMap): Fix struct visibiity warning.
5:45 PM Changeset in webkit [50252] by eric@webkit.org
  • 2 edits in trunk/LayoutTests

2009-10-28 Andrew Scherkus <scherkus@chromium.org>

Reviewed by Eric Carlson.

Fix video-pause-immediately.html to avoid using setTimeout().

This also addresses the flakiness on Windows, where 200ms was not enough time to advance the clock past zero.

https://bugs.webkit.org/show_bug.cgi?id=30889

  • media/video-pause-immediately.html: Replace setTimeout() with timeupdate events.
5:22 PM Changeset in webkit [50251] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-28 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

REGRESSION: crashes in WebCore::RedirectScheduler::timerFired(WebCore::Timer<WebCore::RedirectScheduler>*)
https://bugs.webkit.org/show_bug.cgi?id=30839

Added null check for the case when the frame is detached from the page.

  • loader/RedirectScheduler.cpp: (WebCore::RedirectScheduler::timerFired):
5:12 PM Changeset in webkit [50250] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-28 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=30817
Use parentObjectUnignored instead of parentObject in webkit_accessible_get_parent

Also removes the hack I had originally added to solve bug 25411, because
the fix here is what I should have done in the first place.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (webkit_accessible_get_parent):
5:08 PM Changeset in webkit [50249] by bweinstein@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Ada Chan.

Add accessibility test to the list of Windows Skipped tests.

  • platform/win/Skipped:
4:37 PM Changeset in webkit [50248] by dimich@chromium.org
  • 2 edits in trunk/JavaScriptCore

Not reviewed, attemp to fix Windows build.

Touch the cpp file to cause recompile.

  • wtf/Threading.cpp:

(WTF::threadEntryPoint):

4:21 PM Changeset in webkit [50247] by dimich@chromium.org
  • 4 edits in trunk

JavaScriptCore: Add MessageQueue::removeIf(Predicate&) to remove certain tasks without pulling them from the queue.
Existing Database tests cover this since Database removes tasks when it is stopped.
https://bugs.webkit.org/show_bug.cgi?id=30805

Reviewed by David Levin.

  • wtf/MessageQueue.h:

(WTF::::removeIf):

WebCore: Add MessageQueue::removeIf(Predicate&) to remove certain tasks without pulling them from the queue.
Existing Database tests cover this, no change in functionality.
https://bugs.webkit.org/show_bug.cgi?id=30805

Reviewed by David Levin.

  • storage/DatabaseThread.cpp:

(WebCore::SameDatabasePredicate::SameDatabasePredicate): Added predicate that flags the tasks belonging to a specified database.
(WebCore::SameDatabasePredicate::operator()):
(WebCore::DatabaseThread::unscheduleDatabaseTasks): changed to use the new removeIf method.

4:19 PM Changeset in webkit [50246] by pfeldman@chromium.org
  • 4 edits in trunk/WebCore

2009-10-28 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Glue subsequent timeline records with same category
and title together. Enable Timeline Panel!!!

https://bugs.webkit.org/show_bug.cgi?id=30885

  • English.lproj/localizedStrings.js:
  • inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel.prototype.addRecordToTimeline): (WebInspector.TimelinePanel.prototype._formatRecord): (WebInspector.TimelineRecordTreeElement.prototype.onattach): (WebInspector.TimelineRecordTreeElement.prototype.refresh):
3:56 PM Changeset in webkit [50245] by kov@webkit.org
  • 10 edits in trunk

LayoutTests

Reviewed by Jan Alonzo.

[GTK] API to start inspector for a WebView
https://bugs.webkit.org/show_bug.cgi?id=22551

Unskip inspector tests that we are now able to run.

  • platform/gtk/Skipped:

WebKit/gtk

Reviewed by Jan Alonzo.

[GTK] API to start inspector for a WebView
https://bugs.webkit.org/show_bug.cgi?id=22551

Provide a simple, coordinates-based API to start the inspector.

  • WebCoreSupport/InspectorClientGtk.cpp: (WebKit::InspectorClient::createPage): Use files from the source tree when running from the top of the source directory. (WebKit::InspectorClient::localizedStringsURL): Ditto.
  • webkit/webkitprivate.h:
  • webkit/webkitwebinspector.cpp: (webkit_web_inspector_inspect_coordinates): (webkit_web_inspector_close): (webkit_web_inspector_execute_script):
  • webkit/webkitwebinspector.h:

WebKitTools

Reviewed by Jan Alonzo.

[GTK] API to start inspector for a WebView
https://bugs.webkit.org/show_bug.cgi?id=22551

Use the new inspector API to implement the LayoutTestController
interfaces used to test the inspector.

  • DumpRenderTree/gtk/DumpRenderTree.cpp: (webInspectorInspectWebView): (createWebView):
  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::showWebInspector): (LayoutTestController::closeWebInspector): (LayoutTestController::evaluateInWebInspector):
3:42 PM Changeset in webkit [50244] by kov@webkit.org
  • 4 edits in trunk

LayoutTests

Reviewed by Xan Lopez.

[GTK] Fails new test fast/js/navigator-language.html
https://bugs.webkit.org/show_bug.cgi?id=30440

Unskip the test that now passes.

  • platform/gtk/Skipped:

WebCore

Reviewed by Xan Lopez.

[GTK] Fails new test fast/js/navigator-language.html
https://bugs.webkit.org/show_bug.cgi?id=30440

Reimplement WebCore::defaultLanguage to account for changes in
locale done by setLocale.

Already existing test: fast/js/navigator-language.html

  • platform/gtk/Language.cpp: (WebCore::defaultLanguage):
3:40 PM Changeset in webkit [50243] by sfalken@apple.com
  • 5 edits
    1 delete in trunk/WebKit/win

Should not generate/build COM DOM bindings
https://bugs.webkit.org/show_bug.cgi?id=30880

Reviewed by Sam "Horatio" Weinig.

  • Interfaces/WebKit.idl: Remove generated interfaces.
  • WebKit.vcproj/InterfacesGenerated.vcproj: Removed.
  • WebKit.vcproj/WebKit.sln: Remove InterfacesGenerated.vcproj.
  • WebKit.vcproj/WebKit.submit.sln: Remove InterfacesGenerated.vcproj.
  • WebKit.vcproj/WebKit.vcproj: Remove generated interfaces.
3:36 PM Changeset in webkit [50242] by eric.carlson@apple.com
  • 10 edits in trunk

2009-10-28 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

<rdar://problem/7303145>
Can't exit full screen mode or restart movie after pressing command -R.

  • html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::removedFromDocument): (WebCore::HTMLMediaElement::documentWillBecomeInactive):

Exit from fullscreen if necessary.

  • html/HTMLMediaElement.h:

2009-10-28 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

<rdar://problem/7303145>
Can't exit full screen mode or restart movie after pressing command -R.

  • WebView/WebVideoFullscreenController.mm: (-[WebVideoFullscreenController _requestExit]):

New, call media element's exitFullscreen method.

(-[WebVideoFullscreenController requestExitFullscreenWithAnimation:]):

Call mediaElement->exitFullscreen() after a delay as it turns around and calls
WebVideoFullscreenController:exitFullscreen.

  • WebView/WebVideoFullscreenHUDWindowController.h:
  • WebView/WebVideoFullscreenHUDWindowController.mm: (-[WebVideoFullscreenHUDWindow resignFirstResponder]):

New, don't give up responder status.

(-[WebVideoFullscreenHUDWindow performKeyEquivalent:]):

New, block all command key events.

(-[WebVideoFullscreenHUDWindowController exitFullscreen:]):

Remember when we are in the process of exiting fullscreen, don't allow recursion.

  • WebView/WebView.mm: (-[WebView _exitFullscreen]):

Early return if fullscreen controller is nil.

  • WebView/WebWindowAnimation.h:

Correct comment about _window retain status.

  • WebView/WebWindowAnimation.m: (-[WebWindowScaleAnimation dealloc]):

Don't release _window, we never reatained it.

3:19 PM Changeset in webkit [50241] by hamaji@chromium.org
  • 5 edits in trunk

2009-10-28 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Eric Seidel.

[Qt] WebFrame::counterValueForElementById must not be exposed
https://bugs.webkit.org/show_bug.cgi?id=30882

  • Api/qwebframe.cpp: (qt_drt_counterValueForElementById):
  • Api/qwebframe.h:

2009-10-28 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Eric Seidel.

[Qt] WebFrame::counterValueForElementById must not be exposed
https://bugs.webkit.org/show_bug.cgi?id=30882

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::counterValueForElementById):
2:45 PM Changeset in webkit [50240] by ap@apple.com
  • 2 edits in trunk/WebCore

Unreviewed - a trivial fix to get Windows bots running.

https://bugs.webkit.org/show_bug.cgi?id=30841
<rdar://problem/7342730> WebKit should not pass Referer header through a redirect to a non-secure site

  • platform/network/cf/ResourceRequestCFNet.cpp: (WebCore::setHeaderFields): Don't try to access empty vector's data.
2:44 PM Changeset in webkit [50239] by eric@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-10-28 Afonso R. Costa Jr. <afonso.costa@openbossa.org>

Reviewed by Oliver Hunt.

[Qt] Enable YARR when YARR_JIT is enabled
https://bugs.webkit.org/show_bug.cgi?id=30730

When enabling or disabling JIT using JAVASCRIPTCORE_JIT, the ENABLE_YARR should
be toggled also.

2:36 PM Changeset in webkit [50238] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-28 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25897
[Gtk] Extraneous object of ROLE_PANEL in hierarchy for entries

Remove the extraneous object of ROLE_PANEL.

  • accessibility/gtk/AccessibilityObjectAtk.cpp: (AccessibilityObject::accessibilityPlatformIncludesObject):
2:29 PM Changeset in webkit [50237] by eric@webkit.org
  • 3 edits
    4 adds in trunk

2009-10-28 Jonathan Dixon <joth@chromium.org>

Reviewed by Eric Seidel.

Bug 30547: (Chromium) searchbox not rendered properly due to the css property -webkit-border-radius
https://bugs.webkit.org/show_bug.cgi?id=30547

  • fast/css/text-input-with-webkit-border-radius-expected.checksum: Added.
  • fast/css/text-input-with-webkit-border-radius-expected.png: Added.
  • fast/css/text-input-with-webkit-border-radius-expected.txt: Added.
  • fast/css/text-input-with-webkit-border-radius.html: Added.

2009-10-28 Jonathan Dixon <joth@chromium.org>

Reviewed by Eric Seidel.

Bug 30547: (Chromium) searchbox not rendered properly due to the css property -webkit-border-radius
https://bugs.webkit.org/show_bug.cgi?id=30547

Test: fast/css/text-input-with-webkit-border-radius.html

  • rendering/RenderThemeChromiumWin.cpp: (WebCore::RenderThemeChromiumWin::paintTextFieldInternal): Implemented rounded border rendering in Chromium Windows theme renderer.
2:17 PM Changeset in webkit [50236] by pfeldman@chromium.org
  • 7 edits in trunk/WebCore

2009-10-28 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Pull items collections from resources panel and
timeline panel into AbstractTimelinePanel.

https://bugs.webkit.org/show_bug.cgi?id=30875

  • inspector/front-end/AbstractTimelinePanel.js: (WebInspector.AbstractTimelinePanel): (WebInspector.AbstractTimelinePanel.prototype.populateSidebar): (WebInspector.AbstractTimelinePanel.prototype.createItemTreeElement): (WebInspector.AbstractTimelinePanel.prototype.createItemGraph): (WebInspector.AbstractTimelinePanel.prototype._showCategory): (WebInspector.AbstractTimelinePanel.prototype._hideCategory): (WebInspector.AbstractTimelinePanel.prototype.filter): (WebInspector.AbstractTimelinePanel.prototype._createGraph): (WebInspector.AbstractTimelinePanel.prototype.updateMainViewWidth): (WebInspector.AbstractTimelinePanel.prototype.refresh): (WebInspector.AbstractTimelinePanel.prototype.reset): (WebInspector.AbstractTimelinePanel.prototype.get calculator): (WebInspector.AbstractTimelinePanel.prototype.set calculator): (WebInspector.AbstractTimelinePanel.prototype.addItem): (WebInspector.AbstractTimelinePanel.prototype.removeItem): (WebInspector.AbstractTimelinePanel.prototype.refreshItem): (WebInspector.AbstractTimelinePanel.prototype.revealAndSelectItem): (WebInspector.AbstractTimelinePanel.prototype.sortItems): (WebInspector.AbstractTimelinePanel.prototype.adjustScrollPosition): (WebInspector.AbstractTimelineCategory): (WebInspector.AbstractTimelineCategory.prototype.toString):
  • inspector/front-end/ResourceCategory.js: (WebInspector.ResourceCategory):
  • inspector/front-end/ResourcesPanel.js: (WebInspector.ResourcesPanel): (WebInspector.ResourcesPanel.prototype.createItemTreeElement): (WebInspector.ResourcesPanel.prototype.createItemGraph): (WebInspector.ResourcesPanel.prototype.isCategoryVisible): (WebInspector.ResourcesPanel.prototype.populateSidebar): (WebInspector.ResourcesPanel.prototype.get searchableViews): (WebInspector.ResourcesPanel.prototype.get searchResultsSortFunction.sortFuction): (WebInspector.ResourcesPanel.prototype.get searchResultsSortFunction): (WebInspector.ResourcesPanel.prototype.searchMatchFound): (WebInspector.ResourcesPanel.prototype.searchCanceled): (WebInspector.ResourcesPanel.prototype.performSearch): (WebInspector.ResourcesPanel.prototype.refresh): (WebInspector.ResourcesPanel.prototype.reset): (WebInspector.ResourcesPanel.prototype.removeResource): (WebInspector.ResourcesPanel.prototype.addMessageToResource): (WebInspector.ResourcesPanel.prototype.clearMessages): (WebInspector.ResourcesPanel.prototype.refreshResource): (WebInspector.ResourcesPanel.prototype.recreateViewForResourceIfNeeded): (WebInspector.ResourcesPanel.prototype.showResource): (WebInspector.ResourcesPanel.prototype._sortResourcesIfNeeded): (WebInspector.ResourcesPanel.prototype._toggleLargerResources): (WebInspector.ResourcesPanel.prototype._toggleResourceTracking): (WebInspector.ResourcesPanel.prototype.get _resources): (WebInspector.ResourceTimeCalculator.prototype._upperBound):
  • inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel): (WebInspector.TimelinePanel.prototype.get categories): (WebInspector.TimelinePanel.prototype.populateSidebar): (WebInspector.TimelinePanel.prototype.addRecordToTimeline): (WebInspector.TimelinePanel.prototype.createItemTreeElement): (WebInspector.TimelinePanel.prototype.createItemGraph): (WebInspector.TimelinePanel.prototype._formatRecord): (WebInspector.TimelineCategory):
  • inspector/front-end/inspector.css:
  • inspector/front-end/inspector.html:
2:05 PM Changeset in webkit [50235] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Rubberstamped by Oliver Hunt.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-28
Fix the warning:

"warning: ignoring return value of 'char* getcwd(char*, size_t)',
declared with attribute warn_unused_result".

by actually checking the result. In the case it is null, an
error has occoured, so treat it as the other fatal errors.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::DumpRenderTree::initializeFonts):

1:45 PM Changeset in webkit [50234] by eric@webkit.org
  • 4 edits in trunk/WebCore

2009-10-28 Kelly Norton <knorton@google.com>

Reviewed by Pavel Feldman.

Resets InspectorFrontend in InspectorTimelineAgent instead of removing it so
that it remains active on refreshs and page transitions.
https://bugs.webkit.org/show_bug.cgi?id=30874

  • inspector/InspectorController.cpp: (WebCore::InspectorController::setFrontendProxyObject):
  • inspector/InspectorTimelineAgent.cpp: (WebCore::InspectorTimelineAgent::resetFrontendProxyObject):
  • inspector/InspectorTimelineAgent.h:
1:14 PM Changeset in webkit [50233] by hamaji@chromium.org
  • 28 edits
    2 adds
    8 deletes in trunk

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

  • Api/qwebframe.cpp: (QWebFrame::counterValueForElementById): (QWebHitTestResult::frame):
  • Api/qwebframe.h:

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

  • webkit/webkitprivate.h:
  • webkit/webkitwebframe.cpp: (webkit_web_frame_counter_value_for_element_by_id):

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

  • Interfaces/IWebFramePrivate.idl:
  • WebFrame.cpp: (WebFrame::counterValueForElementById):
  • WebFrame.h:

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

Add counterValueForElement in WebFrame to call WebCore::counterValueForElement.

  • Misc/WebCoreStatistics.h:
  • Misc/WebCoreStatistics.mm: (-[WebFrame counterValueForElement:]):

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

Expose WebCore::counterValueForElement as a WebCore API.

  • WebCore.base.exp:
  • rendering/RenderTreeAsText.cpp: (WebCore::writeCounterValuesFromChildren): (WebCore::counterValueForElement):
  • rendering/RenderTreeAsText.h:

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

Convert two pixel tests to dumpAsText tests using counterValueForElementById.

  • fast/css/counters/t1204-increment-00-c-o-expected.txt: Added.
  • fast/css/counters/t1204-increment-00-c-o.html:
  • fast/css/counters/t1204-increment-01-c-o-expected.txt: Added.
  • fast/css/counters/t1204-increment-01-c-o.html:
  • platform/mac/fast/css/counters/t1204-increment-00-c-o-expected.checksum: Removed.
  • platform/mac/fast/css/counters/t1204-increment-00-c-o-expected.png: Removed.
  • platform/mac/fast/css/counters/t1204-increment-00-c-o-expected.txt: Removed.
  • platform/mac/fast/css/counters/t1204-increment-01-c-o-expected.checksum: Removed.
  • platform/mac/fast/css/counters/t1204-increment-01-c-o-expected.png: Removed.
  • platform/mac/fast/css/counters/t1204-increment-01-c-o-expected.txt: Removed.
  • platform/qt/fast/css/counters/t1204-increment-00-c-o-expected.txt: Removed.
  • platform/qt/fast/css/counters/t1204-increment-01-c-o-expected.txt: Removed.

2009-10-27 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Provide a way to get counter values with layoutTestContoller
https://bugs.webkit.org/show_bug.cgi?id=30555

Define layoutTestContoller.counterValueForElementById.

  • DumpRenderTree/LayoutTestController.cpp: (counterValueForElementByIdCallback): (LayoutTestController::staticFunctions):
  • DumpRenderTree/LayoutTestController.h:
  • DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::counterValueForElementById):
1:07 PM Changeset in webkit [50232] by oliver@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-10-24 Martin Robinson <martin.james.robinson@gmail.com>

Reviewed by Oliver Hunt.

Fix strict aliasing warning by switching reinterpret_cast to bitwise_cast.

strict-aliasing warnings in JSFunction.h
https://bugs.webkit.org/show_bug.cgi?id=27869

  • runtime/JSFunction.h: (JSC::JSFunction::nativeFunction): (JSC::JSFunction::scopeChain): (JSC::JSFunction::setScopeChain): (JSC::JSFunction::setNativeFunction):
12:57 PM Changeset in webkit [50231] by Nate Chapin
  • 2 edits in trunk/WebCore

2009-10-28 Nate Chapin <Nate Chapin>

Unreviewed, Chromium build fix for r50225.

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::defaultObjectContentType):
12:52 PM Changeset in webkit [50230] by eric@webkit.org
  • 8 edits in trunk/WebCore

2009-10-28 Eric Z. Ayers <zundel@google.com>

Reviewed by Pavel Feldman.

Adds InspectorTimelineAgent instrumentation for encountering a
<SCRIPT> tag when parsing an HTML document.

https://bugs.webkit.org/show_bug.cgi?id=30861

  • bindings/js/ScriptSourceCode.h: (WebCore::ScriptSourceCode::ScriptSourceCode): (WebCore::ScriptSourceCode::startLine): (WebCore::ScriptSourceCode::url):
  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::scriptHandler): (WebCore::HTMLTokenizer::scriptExecution):
  • inspector/InspectorTimelineAgent.cpp: (WebCore::InspectorTimelineAgent::willLoadXHR): (WebCore::InspectorTimelineAgent::willScriptTag): (WebCore::InspectorTimelineAgent::didScriptTag):
  • inspector/InspectorTimelineAgent.h: (WebCore::):
  • inspector/TimelineRecordFactory.cpp: (WebCore::TimelineRecordFactory::createScriptTagTimelineRecord):
  • inspector/TimelineRecordFactory.h:
  • inspector/front-end/TimelineAgent.js:
12:23 PM Changeset in webkit [50229] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-28 Steve Block <steveblock@google.com>

Reviewed by Darin Adler.

Makes sure that Geolocation watch IDs remain positive on overflow.
https://bugs.webkit.org/show_bug.cgi?id=30122

No new tests possible.

  • page/Geolocation.cpp: Modified. (WebCore::Geolocation::watchPosition): Modified. Reset the watch ID to 1 on overflow.
12:03 PM Changeset in webkit [50228] by staikos@webkit.org
  • 2 edits in trunk/WebCore

2009-10-28 George Staikos <george.staikos@torchmobile.com>

Attempt to fix the Mac debug build after 50225.

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::defaultObjectContentType):
11:35 AM Changeset in webkit [50227] by bweinstein@apple.com
  • 1 edit in trunk/WebKit/win/WebFrame.cpp

Windows build fix, add missing #include.

11:15 AM Changeset in webkit [50226] by ap@apple.com
  • 8 edits
    8 adds in trunk

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=30841
<rdar://problem/7342730> WebKit should not pass Referer header through a redirect to a non-secure site

Tests: http/tests/ssl/referer-301.html

http/tests/ssl/referer-303.html

  • platform/network/mac/ResourceHandleMac.mm: (-[WebCoreResourceHandleAsDelegate connection:willSendRequest:redirectResponse:]): Remove Referer header if redirecting from https to another protocol.
  • platform/network/ResourceRequestBase.cpp: (WebCore::ResourceRequestBase::clearHTTPReferrer): Update request counterparts, as it is always done when changing or adding header fields. (WebCore::ResourceRequestBase::clearHTTPOrigin): Ditto.
  • platform/network/ResourceRequestBase.h: clearHTTPReferrer() and clearHTTPOrigin() are no longer inline, since they have non-trivial implementations.
  • platform/network/mac/ResourceRequestMac.mm: (WebCore::ResourceRequest::doUpdatePlatformRequest): Fixed to synchronize header field removals. (WebCore::ResourceRequest::doUpdateResourceRequest): Ditto.
  • platform/network/cf/ResourceHandleCFNet.cpp: (WebCore::willSendRequest):
  • platform/network/cf/ResourceRequestCFNet.cpp: (WebCore::setHeaderFields): (WebCore::ResourceRequest::doUpdatePlatformRequest): (WebCore::ResourceRequest::doUpdateResourceRequest): Match Mac changes.
11:12 AM Changeset in webkit [50225] by staikos@webkit.org
  • 7 edits in trunk

2009-10-28 Joe Mason <jmason@rim.com>

Reviewed by Adam Treat.

Add FrameLoader::defaultObjectContentType, containing common code for
implementing FrameLoaderClient::objectContentType. (Currently the gtk
and win ports have copied this code, and the qt port uses similar code
with a few extra clauses. Moving this to a utility function cuts down
on copied code.) This causes no behavioural change.

https://bugs.webkit.org/show_bug.cgi?id=30868

11:00 AM Changeset in webkit [50224] by kenneth@webkit.org
  • 5 edits in trunk

[Qt] QWebHistory::saveState() is inconsistent with the Qt API
https://bugs.webkit.org/show_bug.cgi?id=30710

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-28
Reviewed by Tor Arne Vestbø.

WebCore:

Enforce the versioning, by ignoring any version different
from 1.

  • history/qt/HistoryItemQt.cpp:

(WebCore::HistoryItem::restoreState):
(WebCore::HistoryItem::saveState):

WebKit/qt:

Make the versioning internal and enforce it in the WebCore
part. Adjust the comments, as well as remove now dead code.

  • Api/qwebhistory.cpp:

(operator<<):
(operator>>):

  • Api/qwebhistory.h:
10:10 AM Changeset in webkit [50223] by eric@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

2009-10-28 Shu Chang <Chang.Shu@nokia.com>

Reviewed by Eric Seidel.

Added special expected test result for Tiger to fix regression tests.
https://bugs.webkit.org/show_bug.cgi?id=30811

  • platform/mac-tiger/fast/js/navigator-language-expected.txt: Added.
10:09 AM Changeset in webkit [50222] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-10-28 Pavel Feldman <pfeldman@chromium.org>

Not reviewed: follow up fix to InspectorControllerStub.
Define it after inspector.js due to namespace

https://bugs.webkit.org/show_bug.cgi?id=30866

  • inspector/front-end/inspector.html:
10:03 AM Changeset in webkit [50221] by pfeldman@chromium.org
  • 5 edits
    1 add in trunk/WebCore

2009-10-28 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Add InspectorController stub in order to
support opening in standalone mode.

https://bugs.webkit.org/show_bug.cgi?id=30866

  • WebCore.gypi:
  • WebCore.vcproj/WebCore.vcproj:
  • inspector/front-end/InspectorControllerStub.js: Added. (.WebInspector.InspectorControllerStub):
10:02 AM Changeset in webkit [50220] by eric@webkit.org
  • 9 edits in trunk/WebCore

2009-10-28 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Jan Alonzo.

https://bugs.webkit.org/show_bug.cgi?id=25897
[Gtk] Extraneous object of ROLE_PANEL in hierarchy for entries

Expands upon the new funtionality which allows platforms to indicate
that a particular object should not be ignored, so that they can
specify that an object should be ignored or that it should be left up
to the default handling to decide.

  • accessibility/AccessibilityObject.h:
  • accessibility/chromium/AccessibilityObjectChromium.cpp:
  • accessibility/gtk/AccessibilityObjectAtk.cpp:
  • accessibility/mac/AccessibilityObjectMac.mm:
  • accessibility/qt/AccessibilityObjectQt.cpp:
  • accessibility/win/AccessibilityObjectWin.cpp:
  • accessibility/wx/AccessibilityObjectWx.cpp: (AccessibilityObject::accessibilityPlatformIncludesObject):
  • accessibility/AccessibilityRenderObject.cpp: (AccessibilityRenderObject::accessibilityIsIgnored):
9:53 AM Changeset in webkit [50219] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-10-28 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Jan Alonzo.

https://bugs.webkit.org/show_bug.cgi?id=25534
[GTK] Objects of ROLE_TABLE should implement the accessible table interface

Expose the table summary as the accessible description because there is no summary object.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (webkit_accessible_get_description):
9:14 AM Changeset in webkit [50218] by vestbo@webkit.org
  • 2 edits in trunk/JavaScriptCore

Use JSVALUE64 on 64-bit Windows

Patch by Jan-Arve Sæther <jan-arve.saether@nokia.com> on 2009-10-28
Reviewed by Tor Arne Vestbø.

  • wtf/Platform.h: Make sure WTF_USE_JSVALUE64 is defined
8:10 AM Changeset in webkit [50217] by tonikitoo@webkit.org
  • 1 edit
    2 adds in trunk/WebKit/qt

Pushing missing WebKit/qt/tests/qwebframe/resources/ dir from bug 29248.

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-28
[Qt] [API] Make it possible to have 'invisible' loads
https://bugs.webkit.org/show_bug.cgi?id=29248

  • tests/qwebframe/resources/image2.png: Copied from WebKit/qt/tests/qwebelement/image.png.
6:54 AM Changeset in webkit [50216] by Csaba Osztrogonác
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Holger Freyther.

Patch by Csaba Osztrogonác <Csaba Osztrogonác> on 2009-10-28
[Qt] Remove http/tests/security tests from
skiplist, and add only now failing tests.

  • platform/qt/Skipped:
6:54 AM Changeset in webkit [50215] by steveblock@google.com
  • 7 edits in trunk/WebCore

Adds ScriptController::intializeThreading(), with both JSC and V8 implementations.
https://bugs.webkit.org/show_bug.cgi?id=30678

Reviewed by Eric Seidel.

This new method is used from common code, rather than calling JSC::initializeThreading() (or the V8 equivalent) directly.

Build fix, no new tests required.

  • bindings/js/ScriptController.cpp: Modified.

(WebCore::ScriptController::initializeThreading): Added.

  • bindings/js/ScriptController.h: Modified. Adds ScriptController::initializeThreading().
  • bindings/v8/ScriptController.cpp: Modified.

(WebCore::ScriptController::initializeThreading): Added.

  • bindings/v8/ScriptController.h: Modified. Adds ScriptController::initializeThreading().
  • loader/icon/IconDatabase.cpp: Modified.

(WebCore::iconDatabase): Call ScriptController::initializeThreading(), rather than JSC::initializeThreading().

  • storage/Database.cpp: Modified.

(WebCore::Database::Database): Call ScriptController::initializeThreading(), rather than JSC::initializeThreading().

6:18 AM Changeset in webkit [50214] by kenneth@webkit.org
  • 4 edits in trunk/WebKit/qt

[Qt] QWebHistory::saveState() is inconsistent with the Qt API
https://bugs.webkit.org/show_bug.cgi?id=30710

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-28
Reviewed by Holger Freyther.

Remove the QWebHistory::saveState() and ::restoreState() as
they are inconsistent with the Qt API.

Update unittests to reflect the change.

  • Api/qwebhistory.cpp:

(operator<<):
(operator>>):

  • Api/qwebhistory.h:
  • tests/qwebhistory/tst_qwebhistory.cpp:

(saveHistory):
(restoreHistory):
(tst_QWebHistory::saveAndRestore_crash_1):
(tst_QWebHistory::saveAndRestore_crash_2):
(tst_QWebHistory::saveAndRestore_crash_3):
(tst_QWebHistory::clear):

4:44 AM Changeset in webkit [50213] by zecke@webkit.org
  • 3 edits in trunk/WebCore

Document a feature of the m_liveDecodedResources list.

https://bugs.webkit.org/show_bug.cgi?id=30209
Reviewed by Darin Adler.

Document a feature of the m_liveDecodedResources list.
https://bugs.webkit.org/show_bug.cgi?id=30209

The code made the assumption that the list is sorted by
the m_lastDecodedAccessTime property of the CachedResource.
The above is not true when CachedResource::setDecodedSize
is called and the item is inserted the first time. In this
case the m_lastDecodedAccessTime is still zero and the
m_liveDecodedResources list becomes unsorted.

It is impossible that Cache::pruneLiveResources will
stop to process the list too early due this feature and
the alternatives of updating m_lastDecodedAccessTime in
CachedResource::setDecodedSize or changing the insert
to search the right position have a negative impact on
performance. The best solution for now is to document
this feature.

  • loader/Cache.cpp:

(WebCore::Cache::pruneLiveResources):

  • loader/CachedResource.cpp:

(WebCore::CachedResource::setDecodedSize):

4:24 AM Changeset in webkit [50212] by zecke@webkit.org
  • 2 edits in trunk/WebKit/win

Speculative build-fix for WebElementPropertyBag.cpp

  • WebElementPropertyBag.cpp: Include RenderObject.h
4:06 AM Changeset in webkit [50211] by barraclough@apple.com
  • 2 edits in trunk/JavaScriptCore
  • jit/JIT.h:

Reviewed by NOBODY (build fix!).

4:05 AM Changeset in webkit [50210] by Csaba Osztrogonác
  • 2 edits in trunk/LayoutTests

[Qt] Remove http/tests/misc tests from
skiplist, and add only now failing tests.

Patch by Csaba Osztrogonác <Csaba Osztrogonác> on 2009-10-28

  • platform/qt/Skipped:
3:23 AM Changeset in webkit [50209] by tonikitoo@webkit.org
  • 2 edits in trunk/WebKit/qt

Complementary fix to bug 30779.

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-27
Reviewed by Holger Freyther.

By mistake I used QWeakPointer's toStrongRef() method which docs
explicitly say to not be used in this situation (when the tracked
pointer is devired from QObject). Instead QWeakPointer's data()
is recommended.

  • Api/qwebpage.cpp:

(QWebPage::view):

3:18 AM Changeset in webkit [50208] by xan@webkit.org
  • 4 edits in trunk

WebCore:

2009-10-28 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

[Gtk] Extraneous object of ROLE_PANEL in hierarchy for entries
https://bugs.webkit.org/show_bug.cgi?id=25897

Make text controls always implement the text interface, and the
editable text interface when they are not read only. This is what
ATK-users expect.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (getInterfaceMaskFromObject):

WebKit/gtk:

2009-10-28 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

[Gtk] Extraneous object of ROLE_PANEL in hierarchy for entries
https://bugs.webkit.org/show_bug.cgi?id=25897

Update test to also check that entries implement the AtkText
interface.

  • tests/testatk.c: (run_get_text_tests): (test_webkit_atk_get_text_at_offset_forms): (test_webkit_atk_get_text_at_offset): (main):
3:16 AM Changeset in webkit [50207] by zecke@webkit.org
  • 7 edits in trunk

Fix assertion in SVGRenderBase::mapLocalToContainer resulting from HitTestResult::boundinBox()

https://bugs.webkit.org/show_bug.cgi?id=27347

The current HitTestResult methods are not using the (3d)
transformation aware routines. This can lead to an assertion
SVGRenderBase::mapLocalToContainer method.

Remove HitTestResult::boundingBox() as it is only used in
two places and conceptually doesn't belong into a HitTest
which is operating on points. Update the Qt and Win code
to determine the rect themselves in a (3d) transformation
fashion.

Change HitTestResult::imageRect to use the (3d) transformation
aware RenderBox::absoluteContentQuad to avoid running into
an assertion with SVG content.

  • rendering/HitTestResult.cpp:

(WebCore::HitTestResult::imageRect):

  • rendering/HitTestResult.h:
  • WebElementPropertyBag.cpp:

(WebElementPropertyBag::Read):

  • Api/qwebframe.cpp:

(QWebHitTestResultPrivate::QWebHitTestResultPrivate):

3:16 AM Changeset in webkit [50206] by zecke@webkit.org
  • 2 edits in trunk/WebCore

[Qt] Custom Cursor doesn't use hotspot.

Cursors loaded from a QPixmap didn't specifiy the
hotspot. Take the hotspot values from the mac
implementation.

  • platform/qt/CursorQt.cpp:
3:16 AM Changeset in webkit [50205] by zecke@webkit.org
  • 2 edits in trunk

Export fastMalloc, fastCalloc, fastRealloc and fastFree

Export the FastMalloc functions outside of the libwebkit library
to be able to instrument memory allocations. These are C++ symbols
but do not require the C++ runtime to be useful and should be of
no harm to plain C code.

  • autotools/symbols.filter:
3:14 AM Changeset in webkit [50204] by zecke@webkit.org
  • 2 edits in trunk/JavaScriptCore

Export fastMalloc, fastCalloc, fastRealloc and fastFree on GCC/Unix

https://bugs.webkit.org/show_bug.cgi?id=30769

When using -fvisibility=hidden to hide all internal symbols by default
the malloc symbols will be hidden as well. For memory instrumentation
it is needed to provide an instrumented version of these symbols and
override the normal routines and by changing the visibility back to
default this becomes possible.

The only other solution would be to use system malloc instead of the
TCmalloc implementation but this will not allow to analyze memory
behavior with the default allocator.

  • wtf/FastMalloc.h: Define WTF_FAST_MALLOC_EXPORT for GCC and !darwin
2:59 AM Changeset in webkit [50203] by pfeldman@chromium.org
  • 8 edits
    9 adds in trunk/WebCore

2009-10-27 Pavel Feldman <pfeldman@chromium.org>

Reviewed by NOBODY (OOPS!).

Web Inspector: Initial revision of the Timeline grid.

https://bugs.webkit.org/show_bug.cgi?id=30834

  • English.lproj/localizedStrings.js:
  • WebCore.gypi:
  • inspector/front-end/AbstractTimelinePanel.js: (WebInspector.AbstractTimelinePanel.prototype.populateSidebar): (WebInspector.AbstractTimelinePanel.prototype.refresh): (WebInspector.AbstractTimelineCalculator):
  • inspector/front-end/Images/timelineBarBlue.png: Added.
  • inspector/front-end/Images/timelineBarGray.png: Added.
  • inspector/front-end/Images/timelineBarGreen.png: Added.
  • inspector/front-end/Images/timelineBarOrange.png: Added.
  • inspector/front-end/Images/timelineBarPurple.png: Added.
  • inspector/front-end/Images/timelineBarRed.png: Added.
  • inspector/front-end/Images/timelineBarYellow.png: Added.
  • inspector/front-end/Images/timelineCheckmarks.png: Added.
  • inspector/front-end/Images/timelineDots.png: Added.
  • inspector/front-end/ResourcesPanel.js: (WebInspector.ResourceTimeCalculator): (WebInspector.ResourceTransferSizeCalculator):
  • inspector/front-end/TimelineAgent.js:
  • inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel): (WebInspector.TimelinePanel.prototype.get statusBarItems): (WebInspector.TimelinePanel.prototype.get categories): (WebInspector.TimelinePanel.prototype.populateSidebar): (WebInspector.TimelinePanel.prototype._createStatusbarButtons): (WebInspector.TimelinePanel.prototype.addItemToTimeline): (WebInspector.TimelinePanel.prototype.refresh): (WebInspector.TimelinePanel.prototype._toggleTimelineButtonClick): (WebInspector.TimelinePanel.prototype.reset): (WebInspector.TimelinePanel.prototype._formatRecord): (WebInspector.TimelinePanel.prototype.showCategory): (WebInspector.TimelinePanel.prototype.hideCategory): (WebInspector.TimelineRecordTreeElement): (WebInspector.TimelineRecordTreeElement.prototype.onattach): (WebInspector.TimelineCalculator): (WebInspector.TimelineCalculator.prototype.computeBarGraphPercentages): (WebInspector.TimelineCalculator.prototype.computePercentageFromEventTime): (WebInspector.TimelineCalculator.prototype.computeBarGraphLabels): (WebInspector.TimelineCalculator.prototype.updateBoundaries): (WebInspector.TimelineCalculator.prototype.formatValue): (WebInspector.TimelineGraph): (WebInspector.TimelineGraph.prototype.get graphElement): (WebInspector.TimelineGraph.prototype.refreshLabelPositions): (WebInspector.TimelineGraph.prototype.refresh):
  • inspector/front-end/inspector.css:
12:48 AM Changeset in webkit [50202] by Csaba Osztrogonác
  • 2 edits in trunk/LayoutTests

Put tests into skiplist added in r50168 and r50162 because
layoutTestController.showWebInspector() is unimplemented now.

Patch by Csaba Osztrogonác <Csaba Osztrogonác> on 2009-10-28
Reviewed by Holger Freyther.

  • platform/qt/Skipped:

inspector/console-format-collections.html added
inspector/javascript-syntax-highlight.html added

12:13 AM Changeset in webkit [50201] by barraclough@apple.com
  • 4 edits in trunk/JavaScriptCore

Rubber Stamped by Samuel Q. Weinig.

Patch by Gavin Barraclough <barraclough@apple.com> on 2009-10-27
Make the asserts protecting the offsets in the JIT more descriptive.

  • jit/JIT.h:
  • jit/JITCall.cpp:

(JSC::JIT::compileOpCall):

  • jit/JITPropertyAccess.cpp:

(JSC::JIT::emit_op_method_check):
(JSC::JIT::compileGetByIdHotPath):
(JSC::JIT::compileGetByIdSlowCase):
(JSC::JIT::emit_op_put_by_id):

Oct 27, 2009:

11:50 PM Changeset in webkit [50200] by mitz@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Sam Weinig.

Changed this test to use -webkit-box-shadow instead of box-shadow after r49003.

  • fast/box-shadow/inset.html:
11:43 PM Changeset in webkit [50199] by Joseph Pecoraro
  • 5 edits in trunk

2009-10-27 Joseph Pecoraro <Joseph Pecoraro>

Reviewed by Pavel Feldman.

Web Inspector: Move runAfterPendingDispatches to TestController for Clarity
https://bugs.webkit.org/show_bug.cgi?id=30844

  • inspector/front-end/TestController.js: (WebInspector.TestController.prototype.runAfterPendingDispatches):
  • inspector/front-end/inspector.js: moved runAfterPendingDispatches
10:19 PM Changeset in webkit [50198] by darin@chromium.org
  • 2 edits in trunk/WebCore

2009-10-27 Darin Fisher <darin@chromium.org>

Reviewed by Dimitri Glazkov.

Add missing forward declaration of FramelessScrollView.
https://bugs.webkit.org/show_bug.cgi?id=30824

  • platform/chromium/FramelessScrollViewClient.h:
8:44 PM Changeset in webkit [50197] by Chris Fleizach
  • 2 edits in trunk/LayoutTests

No review.

ARIA menu/menu item need AXRoleDescription
https://bugs.webkit.org/show_bug.cgi?id=30804

Layout test fix for Tiger/Leopard.

  • platform/mac-leopard/accessibility/table-with-aria-role-expected.txt:
8:39 PM Changeset in webkit [50196] by Chris Fleizach
  • 4 edits
    2 adds in trunk

WAI-ARIA: add support for 'presentation' role
https://bugs.webkit.org/show_bug.cgi?id=30806

Reviewed by Adele Peterson.

WebCore:

Test: accessibility/aria-presentational-role.html

  • accessibility/AccessibilityObject.h:

(WebCore::):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::accessibilityIsIgnored):
(WebCore::RoleEntry::):

LayoutTests:

  • accessibility/aria-presentational-role-expected.txt: Added.
  • accessibility/aria-presentational-role.html: Added.
7:56 PM Changeset in webkit [50195] by Chris Fleizach
  • 4 edits
    2 adds in trunk

ARIA menu/menu item need AXRoleDescription
https://bugs.webkit.org/show_bug.cgi?id=30804

Reviewed by Darin Adler.

WebCore:

Make sure that role descriptions will default to the system's first, before
returning an unknown role.

Test: platform/mac/accessibility/aria-menu-role-descriptions.html

  • accessibility/mac/AccessibilityObjectWrapper.mm:

(-[AccessibilityObjectWrapper roleDescription]):

LayoutTests:

  • platform/mac/accessibility/aria-menu-role-descriptions-expected.txt: Added.
  • platform/mac/accessibility/aria-menu-role-descriptions.html: Added.
  • platform/mac/accessibility/table-with-aria-role-expected.txt:
7:20 PM Changeset in webkit [50194] by ggaren@apple.com
  • 2 edits in trunk/WebCore

Qt build fix: migrated away from API that no longer exists.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27

  • bridge/qt/qt_runtime.cpp:

(JSC::Bindings::convertQVariantToValue):

6:37 PM Changeset in webkit [50193] by ggaren@apple.com
  • 2 edits in trunk/WebCore

Qt build fix: migrated away from API that no longer exists.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27

  • bridge/qt/qt_runtime.cpp:

(JSC::Bindings::convertValueToQVariant):

6:30 PM Changeset in webkit [50192] by oliver@apple.com
  • 2 edits in trunk/WebCore

Fix wording in comment.

6:26 PM Changeset in webkit [50191] by oliver@apple.com
  • 2 edits in trunk/WebCore

Crash occurs after launching Newsfire 1.6 for the first time
https://bugs.webkit.org/show_bug.cgi?id=30807

Reviewed by Darin Adler.

We allocate an autorelease pool but then store it off the stack. In a
GC environment this led to it being collected, and thus caused badness
to ensue. To work around this we simply avoid using a pool at all in
a GC environment as it would be a no-op anyway.

6:07 PM Changeset in webkit [50190] by eric@webkit.org
  • 9 edits
    6 adds in trunk

2009-10-27 Steve Block <steveblock@google.com>

Reviewed by Darin Adler.

When a Geolocation method is called, immediately calls the error calback asynchronously if permissions have already been denied.
https://bugs.webkit.org/show_bug.cgi?id=27944.

  • fast/dom/Geolocation/resources/permission-denied-already-error.js: Added. Tests that when permission has already been denied, the error callback is invoked with code PERMISSION_DENIED, when the Geolocation service reports an error.
  • fast/dom/Geolocation/permission-denied-already-error.html: Added. Wrapper for above test.
  • fast/dom/Geolocation/permission-denied-already-error-expected.txt: Added. Expected result for above test.
  • fast/dom/Geolocation/resources/permission-denied-already-success.js: Added. Tests that when permission has already been denied, the error callback is invoked with code PERMISSION_DENIED, when the Geolocation service has a good position fix.
  • fast/dom/Geolocation/permission-denied-already-success.html: Added. Wrapper for above test.
  • fast/dom/Geolocation/permission-denied-already-success-expected.txt: Added. Expected result for above test.
  • fast/dom/Geolocation/resources/permission-denied.js: Modified. Updates error string.
  • fast/dom/Geolocation/permission-denied-expected.txt: Modified. Expected result for above test.
  • fast/dom/Geolocation/resources/permission-denied-stops-watches.js: Modified. Updates error string.
  • fast/dom/Geolocation/permission-denied-expected-stops-watches.txt: Modified. Expected result for above test.
  • platform/gtk/Skipped: Modified. Adds above tests to skipped list.

2009-10-27 Steve Block <steveblock@google.com>

Reviewed by Darin Adler.

When a Geolocation method is called, immediately calls the error calback asynchronously if permissions
have already been denied.
https://bugs.webkit.org/show_bug.cgi?id=27944.

Tests: fast/dom/Geolocation/permission-denied-already-error.html

fast/dom/Geolocation/permission-denied-already-success.html

  • page/Geolocation.cpp: Modified. (WebCore::Geolocation::GeoNotifier::GeoNotifier): Modified. GeoNotifier takes Geolocation object as constructor argument. (WebCore::Geolocation::GeoNotifier::setFatalError): Added. Sets a fatal error for this notifier, causing it to terminate immediately and call the error callback asynchronously. (WebCore::Geolocation::GeoNotifier::timerFired): Added. Used to call the error callback asynchronously on fatal error. (WebCore::Geolocation::getCurrentPosition): Modified. Calls startRequest. (WebCore::Geolocation::watchPosition): Modified. Calls startRequest. (WebCore::Geolocation::startRequest): Added. Common functionality for starting a one-shot or watch request. Sets a fatal error on the notifier if permissions have already been denied. (WebCore::Geolocation::fatalErrorOccurred): Added. Registers that a notifier has encountered a fatal error and should be destroyed. (WebCore::Geolocation::setIsAllowed): Modified. Uses a standard error message for the error callback when permissions are denied.
  • page/Geolocation.h: Modified. (WebCore::Geolocation::isDenied): Added. Determines whether permissions have been denied.
5:58 PM Changeset in webkit [50189] by mrowe@apple.com
  • 4 edits in branches/safari-4-branch

Versioning.

5:58 PM Changeset in webkit [50188] by mrowe@apple.com
  • 1 copy in tags/Safari-6531.21.6

New tag.

5:57 PM Changeset in webkit [50187] by mrowe@apple.com
  • 3 edits
    2 adds in branches/safari-4-branch

Merge r50165.

5:21 PM Changeset in webkit [50186] by Simon Fraser
  • 2 edits in trunk/WebCore

2009-10-27 Chris Marrin <cmarrin@apple.com>

Reviewed by Oliver Hunt.

Fixed a bug where I was using the wrong Canvas pointer to do a cross-domain check
https://bugs.webkit.org/show_bug.cgi?id=30840

  • html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::drawImage):
5:18 PM Changeset in webkit [50185] by eric@webkit.org
  • 3 edits
    2 adds in trunk

2009-10-27 Eric Seidel <eric@webkit.org>

Reviewed by Nikolas Zimmermann.

SVGStyledElement::getPresentationAttribute() can return a shared CSSValue (some SVG tests randomly fail on the bot, and in release builds)
https://bugs.webkit.org/show_bug.cgi?id=29620

  • svg/dom/getPresentationAttribute-cache-corruption-expected.txt: Added.
  • svg/dom/getPresentationAttribute-cache-corruption.svg: Added.

2009-10-27 Eric Seidel <eric@webkit.org>

Reviewed by Nikolas Zimmermann.

SVGStyledElement::getPresentationAttribute() can return a shared CSSValue (some SVG tests randomly fail on the bot, and in release builds)
https://bugs.webkit.org/show_bug.cgi?id=29620

SVG was modifying CSSValues cached of off CSSMappedAttributeDeclarations.
This patch fixes the check to make sure that a new CSSMappedAttributeDeclaration is
created before returning a CSSValue that JavaScript can modify.

Test: svg/dom/getPresentationAttribute-cache-corruption.svg

  • svg/SVGStyledElement.cpp: (WebCore::SVGStyledElement::getPresentationAttribute):
4:32 PM Changeset in webkit [50184] by mitz@apple.com
  • 5 edits
    4 adds in trunk

Incomplete repaint of text field in relative positioned inline at imdb.com
https://bugs.webkit.org/show_bug.cgi?id=30047

Reviewed by Simon Fraser.

WebCore:

Test: fast/repaint/inline-relative-positioned.html

Implemented offsetFromContainer(), mapLocalToContainer() and
mapAbsoluteToLocalPoint() in RenderInline.

  • rendering/RenderBox.h: Fixed argument names in the declaration of

mapLocalToContainer().

  • rendering/RenderInline.cpp:

(WebCore::RenderInline::offsetFromContainer):
(WebCore::RenderInline::mapLocalToContainer):
(WebCore::RenderInline::mapAbsoluteToLocalPoint):

  • rendering/RenderInline.h:

LayoutTests:

  • fast/repaint/inline-relative-positioned-expected.checksum: Added.
  • fast/repaint/inline-relative-positioned-expected.png: Added.
  • fast/repaint/inline-relative-positioned-expected.txt: Added.
  • fast/repaint/inline-relative-positioned.html: Added.
4:29 PM Changeset in webkit [50183] by ggaren@apple.com
  • 9 edits in trunk

JavaScriptCore: A little bit of refactoring in the date code.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27
Reviewed by Sam Weinig.

  • runtime/DateConstructor.cpp:

(JSC::constructDate):

  • runtime/DateInstance.cpp:

(JSC::DateInstance::DateInstance):

  • runtime/DateInstance.h: Removed some unused functions. Changed the default

constructor to ensure that a DateInstance is always initialized.

  • runtime/DatePrototype.cpp:

(JSC::DatePrototype::DatePrototype): Pass an initializer to our constructor,
since it now requires one.

  • wtf/DateMath.cpp:

(WTF::msToGregorianDateTime): Only compute our offset from UTC if our
output will require it. Otherwise, our offset is 0.

WebKit/mac: Updated for refactoring in the date code.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27
Reviewed by Sam Weinig.

  • WebView/WebView.mm:

(aeDescFromJSValue): Since we just want a number of milliseconds, do that
instead of something more complicated.

4:22 PM Changeset in webkit [50182] by jorlow@chromium.org
  • 2 edits in trunk/WebCore

Speculative build fix for Chromium.

Patch by Jeremy Orlow <jorlow@chromium.org> on 2009-10-27

  • platform/graphics/GraphicsContext3D.h:
3:41 PM Changeset in webkit [50181] by cmarrin@apple.com
  • 9 edits in trunk/WebCore

Make WebGL context failure more robust and make it succeed in more cases
https://bugs.webkit.org/show_bug.cgi?id=30349

Bubble a failure to get a CGLContext up to HTMLContextElement so it can
return null from getContext. Also added a more robust pixel format choosing
mechanism and can render correctly when a SW renderer is chosen.

3:20 PM Changeset in webkit [50180] by ggaren@apple.com
  • 1 edit
    1 add in trunk/WebCore

Mac build fix: a forwarding header.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27

  • ForwardingHeaders/wtf/DateInstanceCache.h: Added.
3:12 PM Changeset in webkit [50179] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

Build fix: Mark DateInstaceCache.h private, so other frameworks can see it.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27

3:09 PM Changeset in webkit [50178] by ggaren@apple.com
  • 1 edit
    1 add in trunk/JavaScriptGlue

Build fix: Updated for JavaScriptCore export changes.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27

  • ForwardingHeaders/wtf/DateInstanceCache.h: Added.
3:09 PM Changeset in webkit [50177] by rniwa@webkit.org
  • 2 edits in trunk/WebCore

No Review.

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-10-27
Making fontWeightIsBold static to fix Snow Leopard build.

  • editing/ApplyStyleCommand.cpp:

(WebCore::fontWeightIsBold):

3:04 PM Changeset in webkit [50176] by eric@webkit.org
  • 3 edits in trunk/LayoutTests

2009-10-27 Kinuko Yasuda <kinuko@google.com>

Reviewed by Darin Adler.

Fixing a test case that yields FAIL results. Also updating
the corresponding expected results.
https://bugs.webkit.org/show_bug.cgi?id=30821

  • storage/domstorage/complex-values-expected.txt:
  • storage/domstorage/script-tests/complex-values.js:
3:04 PM Changeset in webkit [50175] by ggaren@apple.com
  • 1 edit
    1 add in trunk/JavaScriptCore

Build fix: re-readded this file.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27

  • runtime/DateInstanceCache.h: Added.

(JSC::DateInstanceData::create):
(JSC::DateInstanceData::DateInstanceData):
(JSC::DateInstanceCache::DateInstanceCache):
(JSC::DateInstanceCache::add):
(JSC::DateInstanceCache::lookup):

3:01 PM Changeset in webkit [50174] by ggaren@apple.com
  • 10 edits
    1 add in trunk

JavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=30800
Cache recently computed date data.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27
Reviewed by Darin Adler and Oliver Hunt.

SunSpider reports a ~0.5% speedup, mostly from date-format-tofte.js.

  • runtime/DateInstance.cpp:

(JSC::DateInstance::DateInstance):
(JSC::DateInstance::getGregorianDateTime): Use the shared cache.

  • runtime/DateInstance.h: Renamed m_cache to m_data, to avoid the confusion

of a "cache cache".

  • runtime/DatePrototype.cpp:

(JSC::formatLocaleDate):
(JSC::dateProtoFuncToString):
(JSC::dateProtoFuncToUTCString):
(JSC::dateProtoFuncToISOString):
(JSC::dateProtoFuncToDateString):
(JSC::dateProtoFuncToTimeString):
(JSC::dateProtoFuncGetFullYear):
(JSC::dateProtoFuncGetUTCFullYear):
(JSC::dateProtoFuncToGMTString):
(JSC::dateProtoFuncGetMonth):
(JSC::dateProtoFuncGetUTCMonth):
(JSC::dateProtoFuncGetDate):
(JSC::dateProtoFuncGetUTCDate):
(JSC::dateProtoFuncGetDay):
(JSC::dateProtoFuncGetUTCDay):
(JSC::dateProtoFuncGetHours):
(JSC::dateProtoFuncGetUTCHours):
(JSC::dateProtoFuncGetMinutes):
(JSC::dateProtoFuncGetUTCMinutes):
(JSC::dateProtoFuncGetSeconds):
(JSC::dateProtoFuncGetUTCSeconds):
(JSC::dateProtoFuncGetTimezoneOffset):
(JSC::setNewValueFromTimeArgs):
(JSC::setNewValueFromDateArgs):
(JSC::dateProtoFuncSetYear):
(JSC::dateProtoFuncGetYear): Pass an ExecState to these functions, so they
can access the DateInstanceCache.

  • runtime/JSGlobalData.h: Keep a DateInstanceCache.

JavaScriptGlue: Updated for JavaScriptCore export changes.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-27
Reviewed by Oliver Hunt.

  • ForwardingHeaders/wtf/DateMath.h: Added.
2:51 PM Changeset in webkit [50173] by darin@chromium.org
  • 3 edits in trunk/JavaScriptCore

2009-10-27 James Robinson <jamesr@chromium.org>

Reviewed by Darin Fisher.

Ensures that JavaScriptCore/wtf/CurrentTime.cpp is not built in PLATFORM(CHROMIUM) builds.

Chromium uses a different method to calculate the current time than is used in
JavaScriptCore/wtf/CurrentTime.cpp. This can lead to time skew when calls to currentTime() and Chromium's time
function are mixed. In particular, timers can get scheduled in the past which leads to 100% CPU use.
See http://code.google.com/p/chromium/issues/detail?id=25892 for an example.

https://bugs.webkit.org/show_bug.cgi?id=30833

2:50 PM Changeset in webkit [50172] by rniwa@webkit.org
  • 5 edits in trunk

WebKit cannot remove nested bold tags
https://bugs.webkit.org/show_bug.cgi?id=30784

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-10-27
Reviewed by Eric Seidel.

WebCore:

This patch modifies getPropertiesNotInComputedStyle so that it compares font-weight
more flexibly. Instead of comparing cssText, it uses fontWeightIsBold which converts
all font weights to either bold or unbold state.

  • editing/ApplyStyleCommand.cpp:

(WebCore::fontWeightIsBold):
(WebCore::getPropertiesNotInComputedStyle): Removes font-weight property
if two style have the same value under getFontWeightValue

LayoutTests:

Adds two more edge cases to remove-nested-inline-styles.js:
<b><b><b>12</b></b></b> and <b><b><b>1</b></b>2</b>
WebKit now passes the one existing test used to fail.

  • editing/style/remove-nested-inline-styles-expected.txt: Added two tests.
  • editing/style/script-tests/remove-nested-inline-styles.js: Added two tests.
2:04 PM Changeset in webkit [50171] by eric@webkit.org
  • 2 edits in trunk

2009-10-27 Carol Szabo <carol.szabo@nokia.com>

Reviewed by Darin Adler.

Give an ability to WebKit to free statically allocated pointers
before quit.
https://bugs.webkit.org/show_bug.cgi?id=27980

No new tests for this as the change does not affect WebKit
functionality, but only memory usage.

  • JavaScriptCore/wtf/StdLibExtras.h: Allowed for the override of DEFINE_STATIC_LOCAL via compiler commandline options or elsewhere in the code. Useful for implementing custom builds that free resources allocated via DEFINE_STATIC_LOCAL.
2:02 PM WebKit Team edited by Joseph Pecoraro
Added Myself (diff)
1:42 PM Changeset in webkit [50170] by dumi@chromium.org
  • 2 edits in trunk/WebCore

Build fix, no review needed.

Patch by Dumitru Daniliuc <dumi@chromium.org> on 2009-10-27
Fixing the BUILD. m_estimatedSize in Database.h should be of type
'unsigned long' instead of 'unsigned long long'.

No tests required.

  • storage/Database.h:
1:14 PM Changeset in webkit [50169] by dumi@chromium.org
  • 9 edits in trunk/WebCore

Refactoring the Database class to not depend on
OriginQuotaManager. Also, adding a SecurityOrigin copy to each
Database instance, that is safe to use on the file
thread. Finally, adding new simple fields and getters to the
Database object for storing/getting the display name and estimated
size specified by the user in the openDatabase() call.

Patch by Dumitru Daniliuc <dumi@chromium.org> on 2009-10-27
Reviewed by Adam Barth.

https://bugs.webkit.org/show_bug.cgi?id=30548

  • page/SecurityOrigin.h:
  • storage/Database.cpp:

(WebCore::Database::openDatabase): Storing the display name and
the estimated size in the Database object.
(WebCore::Database::Database): Storing the display name and the
estimated size in the Database object, as well as a SecurityOrigin
instance that is safe to use on the DB thread.
(WebCore::Database::maximumSize): Delegate the call to
DatabaseTracker::getMaxSizeForDatabase().
(WebCore::Database::databaseThreadSecurityOrigin): Return the
SecurityOrigin instance that's safe to use on the DB thread.
(WebCore::Database::threadSafeSecurityOrigin): Return the
SecurityOrigin instance that's safe to use on the current thread.
(WebCore::Database::displayName): Return the display name
specified by the user in the openDatabase() call.
(WebCore::Database::estimatedSize): Return the estimated size
specified by the user in the openDatabase() call.
(WebCore::Database::fileName): Return the name of the file where
the current Database is tored.

  • storage/Database.h:
  • storage/DatabaseTracker.cpp:

(WebCore::DatabaseTracker::getMaxSizeForDatabase): Returns the
maximum size for a DB file based on the current size of that file
and the space available for that origin.

  • storage/DatabaseTracker.h:
  • storage/SQLTransaction.cpp:
  • storage/SQLTransactionClient.cpp:

(WebCore::SQLTransactionClient::didCommitTransaction): Use the
correct SecurityOrigin instance.

12:37 PM Changeset in webkit [50168] by Joseph Pecoraro
  • 5 edits
    2 adds in trunk

2009-10-27 Joseph Pecoraro <Joseph Pecoraro>

Reviewed by Timothy Hatcher.

Web Inspector: Pretty Print all HTML Collection Types like we do for NodeList
https://bugs.webkit.org/show_bug.cgi?id=30709

Test: inspector/console-format-collections.html

  • inspector/front-end/InjectedScript.js: (Object.type): check for instances of HTMLCollection like we do for NodeList
  • inspector/front-end/inspector.js: added WebInspector.pendingDispatches counter (WebInspector.dispatch): increment and decrement dispatch counter (WebInspector.runAfterPendingDispatches): run when there are no more dispatches
12:30 PM Changeset in webkit [50167] by eric@webkit.org
  • 7 edits in trunk/WebCore

2009-10-27 Kelly Norton <knorton@google.com>

Reviewed by Timothy Hatcher.

Adds XMLHttpRequest support to InspectorTimelineAgent.
https://bugs.webkit.org/show_bug.cgi?id=30578

  • inspector/InspectorTimelineAgent.cpp: (WebCore::InspectorTimelineAgent::willChangeXHRReadyState): (WebCore::InspectorTimelineAgent::didChangeXHRReadyState): (WebCore::InspectorTimelineAgent::willLoadXHR): (WebCore::InspectorTimelineAgent::didLoadXHR):
  • inspector/InspectorTimelineAgent.h: (WebCore::):
  • inspector/TimelineRecordFactory.cpp: (WebCore::TimelineRecordFactory::createXHRReadyStateChangeTimelineRecord): (WebCore::TimelineRecordFactory::createXHRLoadTimelineRecord):
  • inspector/TimelineRecordFactory.h:
  • inspector/front-end/TimelineAgent.js:
  • xml/XMLHttpRequest.cpp: (WebCore::XMLHttpRequest::callReadyStateChangeListener):
12:23 PM Changeset in webkit [50166] by kenneth@webkit.org
  • 6 edits in trunk

Rubberstamped by Oliver Hunt.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-27
Change two methods to be internal for DRT use only.

Part of [Qt] Review all new API in Qt 4.6
https://bugs.webkit.org/show_bug.cgi?id=29843#c11

WebKit/qt:

  • Api/qwebsecurityorigin.cpp:

(qt_drt_whiteListAccessFromOrigin):
(qt_drt_resetOriginAccessWhiteLists):
(QWebSecurityOrigin::localSchemes):

  • Api/qwebsecurityorigin.h:

WebKitTools:

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp:

(LayoutTestController::whiteListAccessFromOrigin):

11:54 AM Changeset in webkit [50165] by adele@apple.com
  • 3 edits
    2 adds in trunk

WebCore: REGRESSION (Safari 4.0.3-ToT): After pressing Shift-PageDown, pressing Shift-Up
extends the top of the selection upwards (but should shrink the selection instead)
<rdar://problem/7269075>
https://bugs.webkit.org/show_bug.cgi?id=29981

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-27
Reviewed by Adele Peterson.

  • editing/SelectionController.cpp:

(WebCore::SelectionController::modify): Added setting m_lastChangeWasHorizontalExtension.

LayoutTests: Test for <rdar://problem/7269075> REGRESSION (Safari 4.0.3-ToT): After pressing Shift-PageDown, pressing Shift-Up
extends the top of the selection upwards (but should shrink the selection instead)
https://bugs.webkit.org/show_bug.cgi?id=29981

  • editing/selection/shrink-selection-after-shift-pagedown-expected.txt: Added.
  • editing/selection/shrink-selection-after-shift-pagedown.html: Added.
11:45 AM Changeset in webkit [50164] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

Make sure that initiating a rotation while rotating won't make
it end up at rotation positions that are not a multiply of
180 degrees.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-27
Reviewed by Tor Arne Vestbø.

  • QGVLauncher/main.cpp:

(MainView::animatedFlip):

11:37 AM Changeset in webkit [50163] by bweinstein@apple.com
  • 2 edits in trunk/WebCore

Rubber-stamped by Tim Hatcher.

Removed an extra console.log from ElementsTreeOutline.

  • inspector/front-end/ElementsTreeOutline.js:
11:26 AM Changeset in webkit [50162] by timothy@apple.com
  • 4 edits
    2 adds in trunk

WebCore: Rewrote JavaScriptSourceSyntaxHighlighter to be more accurate
https://bugs.webkit.org/show_bug.cgi?id=27147

Patch by Keishi Hattori <casey.hattori@gmail.com> on 2009-10-27
Reviewed by Timothy Hatcher.

Tests: LayoutTests/inspector/javascript-syntax-highlight.html

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

(WebInspector.JavaScriptSourceSyntaxHighlighter.process): Added. Processes 100 tokens at a time.
(WebInspector.JavaScriptSourceSyntaxHighlighter.lex): Added. Scans for a token.
(WebInspector.JavaScriptSourceSyntaxHighlighter.appendNonToken): Added. Appends the non-token characters that lex ignored.
(WebInspector.JavaScriptSourceSyntaxHighlighter.syntaxHighlightNode): Added. To syntax highlight node in ElementsTreeOutline.

LayoutTests: Tests for JavaScript syntax highlighter.
https://bugs.webkit.org/show_bug.cgi?id=27147

Patch by Keishi Hattori <casey.hattori@gmail.com> on 2009-10-27
Reviewed by Timothy Hatcher.

  • inspector/javascript-syntax-highlight-expected.txt: Added.
  • inspector/javascript-syntax-highlight.html: Added.
11:21 AM Changeset in webkit [50161] by xan@webkit.org
  • 2 edits in trunk/LayoutTests

2009-10-27 Xan Lopez <xlopez@igalia.com>

Skip a few more tests that are failing on and off.

  • platform/gtk/Skipped:
11:03 AM Changeset in webkit [50160] by eric@webkit.org
  • 3 edits
    2 adds in trunk

2009-10-27 Mads Ager <ager@chromium.org>

Reviewed by Adam Barth.

[V8] Missing null check after string conversion in error reporting
https://bugs.webkit.org/show_bug.cgi?id=30774

Test case for toString conversion failure on an error object in
stack overflow situations. Test passes if it doesn't crash.

  • fast/dom/error-to-string-stack-overflow-expected.txt: Added.
  • fast/dom/error-to-string-stack-overflow.html: Added.

2009-10-27 Mads Ager <ager@chromium.org>

Reviewed by Adam Barth.

[V8] Missing null check after string conversion in error reporting
https://bugs.webkit.org/show_bug.cgi?id=30774

Add null check after string conversion in error reporting code.
ToString conversion can fail for instance when an exception is
thrown during conversion.

Test: fast/dom/error-to-string-stack-overflow.html

  • bindings/v8/V8Utilities.cpp: (WebCore::reportException):
10:54 AM Changeset in webkit [50159] by steveblock@google.com
  • 3 edits in trunk/WebCore

Use a pair of maps to store Geolocation watchers.
https://bugs.webkit.org/show_bug.cgi?id=29178

Reviewed by Darin Adler.

The pair of maps allows us to look up a watcher either by its ID or by its GeoNotifier object.
The ability to look up by a watcher by its GeoNotifier object will be required when implementing
Geolocation::fatalErrorOccurred. See https://bugs.webkit.org/show_bug.cgi?id=27944

  • page/Geolocation.cpp: Modified.

(WebCore::Geolocation::Watchers::set): Added. Adds a watcher with the given ID.
(WebCore::Geolocation::Watchers::remove): Added. Removes a watcher by ID.
(WebCore::Geolocation::Watchers::remove): Added. Removes a watcher by GeoNotifier object.
(WebCore::Geolocation::Watchers::clear): Added. Removes all watchers.
(WebCore::Geolocation::Watchers::isEmpty): Added. Determines if there are no watchers.
(WebCore::Geolocation::Watchers::getNotifiersVector): Added. Gets a vector of the GeoNotifier objects.
(WebCore::Geolocation::watchPosition): Modified. Rename watcher identifier static variable.
(WebCore::Geolocation::stopTimersForWatchers): Modified. Use Watchers::getNotifiersVector.
(WebCore::Geolocation::handleError): Modified. Use Watchers::getNotifiersVector.
(WebCore::Geolocation::makeSuccessCallbacks): Modified. Use Watchers::getNotifiersVector.

  • page/Geolocation.h: Modified. Adds Geolocation::Watchers class.
10:45 AM Changeset in webkit [50158] by jorlow@chromium.org
  • 2 edits in trunk/WebCore

2009-10-27 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Dimitri Glazkov.

[V8] DOM Storage runtime flag should default to on
https://bugs.webkit.org/show_bug.cgi?id=30828

The DOM Storage runtime flag was supposed to default to on. I accidentally
changed the logic to default to false when I moved the flag.

  • bindings/v8/RuntimeEnabledFeatures.cpp:
10:38 AM Changeset in webkit [50157] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-10-27 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

REGRESSION: svn-apply exits(1) when applying a patch with a file add
https://bugs.webkit.org/show_bug.cgi?id=30826

  • Scripts/svn-apply:
    • Add () around all system() calls.
    • Use the correct system() == 0 or die instead of system() or die
    • Add descriptive messages to all die statements.
10:35 AM Changeset in webkit [50156] by steveblock@google.com
  • 2 edits in trunk/WebKitTools

Adds steveblock@google.com to list of committers.

Reviewed by NOBODY.

10:00 AM Changeset in webkit [50155] by eric@webkit.org
  • 4 edits in trunk/WebCore

2009-10-27 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Dimitri Glazkov.

Fix crash found in chromium test_shell.
https://bugs.webkit.org/show_bug.cgi?id=30808

When WebSocket is deleted without close, webkit would crash
when it handles didClose.

Check scriptExecutionContext before post task for event.
Use WebSocketChannel::disconnect() instead of close() in WebSocket
destructor, so that WebSocketChannel should not call deleted WebSocket
back in didClose().
To make sure WebSocketChannel alive while it is processing WebSocket
protocol over SocketStreamHandle, ref() in connect() and deref() in
didClose().

  • websockets/WebSocket.cpp: (WebCore::WebSocket::~WebSocket): (WebCore::WebSocket::didConnect): (WebCore::WebSocket::didReceiveMessage): (WebCore::WebSocket::didClose):
  • websockets/WebSocketChannel.cpp: (WebCore::WebSocketChannel::connect): (WebCore::WebSocketChannel::disconnect): (WebCore::WebSocketChannel::didClose): (WebCore::WebSocketChannel::didReceiveData):
  • websockets/WebSocketChannel.h:
9:49 AM Changeset in webkit [50154] by mitz@apple.com
  • 6 edits in trunk

JavaScriptGlue: Updated license

Reviewed by Darin Adler.

  • LICENSE:

WebCore: Updated Apple license

Reviewed by Darin Adler.

  • LICENSE-APPLE:

WebKit: Updated license

Reviewed by Darin Adler.

  • LICENSE:
9:43 AM Changeset in webkit [50153] by beidson@apple.com
  • 5 edits in trunk

Rubberstamped by Sam Weinig.

Re-enable the check that disallows pages with unload handlers in the page cache.

<rdar://problem/7196485> and http://webkit.org/b/29021

WebCore:

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::canCachePageContainingThisFrame):
(WebCore::FrameLoader::logCanCacheFrameDecision):

LayoutTests:

  • platform/mac/Skipped:
  • platform/win/Skipped:
9:30 AM Changeset in webkit [50152] by timothy@apple.com
  • 1 edit
    2 deletes in trunk/WebCore

Remove more empty files that didn't get removed earlier.

  • inspector/DOMDispatchTimelineItem.cpp: Removed.
  • inspector/DOMDispatchTimelineItem.h: Removed.
8:34 AM Changeset in webkit [50151] by Csaba Osztrogonác
  • 2 edits in trunk/WebCore

[Qt] Fix assertion crashes introduced by changes made in r50110.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-27
Reviewed by Tor Arne Vestbø.

  • dom/XMLTokenizerQt.cpp:

(WebCore::XMLTokenizer::parseEndElement): only call popCurrentNode() if there are nodes left in the stack.

8:27 AM Changeset in webkit [50150] by timothy@apple.com
  • 1 edit
    2 deletes in trunk/WebCore

Remove empty files that didn't get removed earlier.

  • inspector/TimelineItem.cpp: Removed.
  • inspector/TimelineItem.h: Removed.
7:44 AM Changeset in webkit [50149] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

Unreviewed Qt build fix.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-27
Update the tests as well to the new API change.

  • tests/qwebelement/tst_qwebelement.cpp:

(tst_QWebElement::clear):

7:28 AM Changeset in webkit [50148] by xan@webkit.org
  • 2 edits in trunk/LayoutTests

2009-10-27 Xan Lopez <xlopez@igalia.com>

Another typo.

  • platform/gtk/Skipped:
7:22 AM Changeset in webkit [50147] by kenneth@webkit.org
  • 3 edits in trunk/WebKit/qt

Rubberstamped by Tor Arne Vestbø.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-27
[Qt] QWebElement::removeChildren() should be
QWebElement::removeAllChildren()
https://bugs.webkit.org/show_bug.cgi?id=30630

  • Api/qwebelement.cpp:

(QWebElement::removeAllChildren):

  • Api/qwebelement.h:
7:18 AM Changeset in webkit [50146] by xan@webkit.org
  • 2 edits in trunk/LayoutTests

2009-10-27 Xan Lopez <xlopez@igalia.com>

Fix typo in test names.

  • platform/gtk/Skipped:
7:02 AM Changeset in webkit [50145] by xan@webkit.org
  • 2 edits in trunk/LayoutTests

2009-10-27 Xan Lopez <xlopez@igalia.com>

Skip a few tests with links to their tracking bugs.

  • platform/gtk/Skipped:
6:57 AM Changeset in webkit [50144] by tonikitoo@webkit.org
  • 1 edit in trunk/WebCore/ChangeLog

This reverts commit r50141 that wrongly messed up the WebCore/ChangeLog

6:51 AM Changeset in webkit [50143] by Csaba Osztrogonác
  • 3 edits in trunk/JavaScriptCore

Rubber-stamped by Tor Arne Vestbø.

Patch by Peter Varga <pvarga@inf.u-szeged.hu> on 2009-10-27
Fix typo in RegexInterpreter.cpp and RegexJIT.cpp alterantive to
alternative.

  • yarr/RegexInterpreter.cpp:

(JSC::Yarr::ByteCompiler::alternativeBodyDisjunction):
(JSC::Yarr::ByteCompiler::alternativeDisjunction):
(JSC::Yarr::ByteCompiler::emitDisjunction):

  • yarr/RegexJIT.cpp:

(JSC::Yarr::RegexGenerator::generateDisjunction):

6:38 AM Changeset in webkit [50142] by tonikitoo@webkit.org
  • 5 edits in trunk/WebKit/qt

Make QWebPagePrivate's (QWidget) view to be a QWeakPointer.
https://bugs.webkit.org/show_bug.cgi?id=30779

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-27
Reviewed by Antti Koivisto and Holger Freyther.

The fact that it was been set from external objects of qwebpage
and not being deleted internally can lead to dangling references.

  • Api/qgraphicswebview.cpp:

(QGraphicsWebView::~QGraphicsWebView):

  • Api/qwebpage.cpp:

(QWebPagePrivate::QWebPagePrivate):
(QWebPagePrivate::createContextMenu):
(QWebPagePrivate::handleSoftwareInputPanel):
(QWebPagePrivate::keyPressEvent):
(QWebPage::setView):
(QWebPage::view):
(QWebPage::javaScriptAlert):
(QWebPage::javaScriptConfirm):
(QWebPage::javaScriptPrompt):
(QWebPage::shouldInterruptJavaScript):
(QWebPage::createWindow):
(QWebPage::extension):
(QWebPage::chooseFile):
(QWebPage::userAgentForUrl):

  • Api/qwebpage_p.h:
  • Api/qwebview.cpp:

(QWebView::~QWebView):

6:37 AM Changeset in webkit [50141] by tonikitoo@webkit.org
  • 1 edit in trunk/WebCore/ChangeLog

2009-10-20 Mikhail Naganov <mnaganov@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: populate child nodes before sorting them.

https://bugs.webkit.org/show_bug.cgi?id=29673

  • inspector/front-end/BottomUpProfileDataGridTree.js: (WebInspector.BottomUpProfileDataGridNode): Swapped with BottomUpProfileDataGridTree to be consistent with TopDownProfileDataGridNode. (WebInspector.BottomUpProfileDataGridNode.prototype._takePropertiesFromProfileDataGridNode): (WebInspector.BottomUpProfileDataGridNode.prototype._keepOnlyChild): (WebInspector.BottomUpProfileDataGridNode.prototype._exclude): (WebInspector.BottomUpProfileDataGridNode.prototype._merge): (WebInspector.BottomUpProfileDataGridNode.prototype._sharedPopulate): (WebInspector.BottomUpProfileDataGridTree.prototype.exclude):
  • inspector/front-end/ProfileDataGridTree.js: (WebInspector.ProfileDataGridNode.prototype.sort): Added missing parentheses. (WebInspector.ProfileDataGridNode.prototype.get _parent): (WebInspector.ProfileDataGridNode.prototype._populate):
  • inspector/front-end/TopDownProfileDataGridTree.js: (WebInspector.TopDownProfileDataGridNode.prototype._sharedPopulate):
5:20 AM Changeset in webkit [50140] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-10-27 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25534
[GTK] Objects of ROLE_TABLE should implement the accessible table interface

First part of the implementation of AtkTable.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (getCell): (getCellIndex): (webkit_accessible_table_ref_at): (webkit_accessible_table_get_index_at): (webkit_accessible_table_get_n_columns): (webkit_accessible_table_get_n_rows): (webkit_accessible_table_get_column_extent_at): (webkit_accessible_table_get_row_extent_at): (webkit_accessible_table_get_row_header): (atk_table_interface_init): (AtkInterfacesInitFunctions): (GetAtkInterfaceTypeFromWAIType):
4:33 AM Changeset in webkit [50139] by vestbo@webkit.org
  • 2 edits in trunk/WebCore

[Qt] Don't lose remainder when computing wheel event deltas

Reviewed by Lars Knoll.

  • platform/qt/WheelEventQt.cpp:
3:17 AM Changeset in webkit [50138] by eric@webkit.org
  • 4 edits in trunk/WebCore

2009-10-27 Laszlo Gombos <Laszlo Gombos>

Reviewed by Antti Koivisto.

Fix QtWebKit build for WIN_OS if Netscape plug-in support
is turned off and refactor some related code
https://bugs.webkit.org/show_bug.cgi?id=30786

No new tests as there is no new functionality introduced.

  • platform/FileSystem.h: Refactor to make sure that each different type definition is only repeated once.
  • plugins/PluginPackage.cpp: (WebCore::PluginPackage::compareFileVersion): Move it out from the ENABLE_PLUGIN_PACKAGE_SIMPLE_HASH guard and combine it with the function body from PluginPackageWin.
  • plugins/win/PluginPackageWin.cpp: Remove compareFileVersion as it is now in PluginPackage.cpp.
3:09 AM Changeset in webkit [50137] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-10-27 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

svn-apply can exit(0) even on patch failure
https://bugs.webkit.org/show_bug.cgi?id=29622

  • Scripts/svn-apply:
    • Add a bunch of "or die" statements, hopefully catching all possible cases where failure could still exit(0).
3:01 AM Changeset in webkit [50136] by eric@webkit.org
  • 8 edits in trunk/WebKitTools

2009-10-27 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

svn-* scripts should share code through VCSUtils.pm
https://bugs.webkit.org/show_bug.cgi?id=30791

Just moving code into a shared location.

  • Scripts/VCSUtils.pm:
  • Scripts/prepare-ChangeLog:
  • Scripts/resolve-ChangeLogs:
  • Scripts/svn-apply:
  • Scripts/svn-create-patch:
  • Scripts/svn-unapply:
  • Scripts/update-webkit:
2:54 AM Changeset in webkit [50135] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-10-27 Vadim Zeitlin <vadim@wxwidgets.org>

Suppress a huge number of MSVC warnings when building wxWebKit.

  • wx/build/settings.py:
2:20 AM Changeset in webkit [50134] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-10-26 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

https://bugs.webkit.org/show_bug.cgi?id=27011
[Gtk] Implement support for get_index_in_parent

Implement atk_object_get_index_in_parent.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (webkit_accessible_get_index_in_parent):
2:07 AM Changeset in webkit [50133] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-10-27 Eric Seidel <eric@webkit.org>

No review, rolling out r50131.
http://trac.webkit.org/changeset/50131

  • bridge/objc/objc_instance.h:
  • bridge/objc/objc_instance.mm: (ObjcInstance::ObjcInstance): (ObjcInstance::virtualEnd):
12:12 AM Changeset in webkit [50132] by dbates@webkit.org
  • 4 edits
    6 adds in trunk

2009-10-26 Daniel Bates <dbates@webkit.org>

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=30491

Fixes an issue where pressing return/enter on the keyboard
in <isindex> does not submit it if is not within a <form>.

According to the HTML 2.0 thru HTML 4.01 spec
(http://www.w3.org/MarkUp/html-spec/html-spec_7.html#SEC7.5), the
<isindex> element does not need to be within a <form> element in
order to submit it. It can submitted on its own. Also, if present,
the href property of the <base> element will dictate where to submit
the value of the <isindex> element (this is analogous to the action
property of the <form> element).

Tests: http/tests/misc/isindex-with-no-form-base-href.html

http/tests/misc/isindex-with-no-form.html

  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::defaultEventHandler): Calls createTemporaryFormForIsIndex() to create a <form> if none is present and we are an <isindex> element. (WebCore::HTMLInputElement::createTemporaryFormForIsIndex): Added.
  • html/HTMLInputElement.h:

2009-10-26 Daniel Bates <dbates@webkit.org>

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=30491

Test that the <isindex> element can be submitted without being enclosed
within a <form> element. Also tests that the href property of the <base>
element can be used to direct where the submission is sent.

  • http/tests/misc/isindex-with-no-form-base-href-expected.txt: Added.
  • http/tests/misc/isindex-with-no-form-base-href.html: Added.
  • http/tests/misc/isindex-with-no-form-expected.txt: Added.
  • http/tests/misc/isindex-with-no-form.html: Added.
  • http/tests/misc/resources/isindex-with-no-form-base-href-submit.html: Added.
  • http/tests/misc/resources/isindex-with-no-form-base-href.html: Added.
12:10 AM Changeset in webkit [50131] by oliver@apple.com
  • 3 edits in trunk/WebCore

Crash occurs after launching Newsfire 1.6 for the first time
https://bugs.webkit.org/show_bug.cgi?id=30807

Reviewed by Maciej Stachowiak.

Use a RetainPtr in ObjcBinding to ensure that we adequately
protect the autorelease pool from GC in GC'd applications.

Note: See TracTimeline for information about the timeline view.