Timeline



Aug 11, 2009:

11:50 PM Changeset in webkit [47093] by oliver@apple.com
  • 3 edits in trunk/JavaScriptCore

Minor style fixes in previous patch and changelog correction.

Reviewed by Mark Rowe

11:22 PM Changeset in webkit [47092] by oliver@apple.com
  • 13 edits
    2 adds in trunk

Make it harder to misuse try* allocation routines
https://bugs.webkit.org/show_bug.cgi?id=27469

Reviewed by Gavin Barraclough

Jump through a few hoops to make it much harder to accidentally
miss null-checking of values returned by the try-* allocation
routines.

10:59 PM Changeset in webkit [47091] by barraclough@apple.com
  • 3 edits in trunk/JavaScriptCore

Reviewed by NOBODY (build fix part deux).

10:29 PM Changeset in webkit [47090] by barraclough@apple.com
  • 3 edits in trunk/JavaScriptCore

Reviewed by NOBODY (build fix).

10:22 PM Changeset in webkit [47089] by barraclough@apple.com
  • 18 edits in trunk

JavaScriptCore: Restrict use of FuncDeclNode & FuncExprNode to the parser.
https://bugs.webkit.org/show_bug.cgi?id=28209

Reviewed by Oliver Hunt.

These objects were also being referenced from the CodeBlock. By changing this
to just retain pointers to FunctionBodyNodes these classes can be restricted to
use during parsing.

No performance impact (or sub-percent progression).

Update symbols.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::mark):
(JSC::CodeBlock::reparseForExceptionInfoIfNecessary):
(JSC::CodeBlock::shrinkToFit):

  • bytecode/CodeBlock.h:

(JSC::CodeBlock::addFunction):
(JSC::CodeBlock::function):

Unify m_functions & m_functionExpressions into a single Vector<RefPtr<FuncExprNode> >.

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::addConstant):
(JSC::BytecodeGenerator::emitNewFunction):
(JSC::BytecodeGenerator::emitNewFunctionExpression):

  • bytecompiler/BytecodeGenerator.h:

FunctionStacks now contain FunctionBodyNodes not FuncDeclNodes.

  • interpreter/Interpreter.cpp:

(JSC::Interpreter::execute):
(JSC::Interpreter::privateExecute):

Update to reflect chnages in CodeBlock.

  • jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_new_func_exp):

  • jit/JITStubs.cpp:

(JSC::DEFINE_STUB_FUNCTION):

  • jit/JITStubs.h:

(JSC::):

Update to reflect chnages in CodeBlock.

  • parser/Grammar.y:

FunctionStacks now contain FunctionBodyNodes not FuncDeclNodes.

  • parser/NodeConstructors.h:

(JSC::FuncExprNode::FuncExprNode):
(JSC::FuncDeclNode::FuncDeclNode):

  • parser/Nodes.cpp:

(JSC::ScopeNodeData::mark):
(JSC::FunctionBodyNode::finishParsing):

  • parser/Nodes.h:

(JSC::FunctionBodyNode::ident):

Move m_ident & make methods from FuncDeclNode & FuncExprNode to FunctionBodyNode.

  • runtime/JSFunction.h:

(JSC::FunctionBodyNode::make):

Make this method inline (was FuncDeclNode::makeFunction).

WebCore: Restrict use of FuncDeclNode & FuncExprNode to the parser.
https://bugs.webkit.org/show_bug.cgi?id=28209

Reviewed by Oliver Hunt.

  • inspector/JavaScriptDebugServer.cpp:

(WebCore::JavaScriptDebugServer::recompileAllJSFunctions):

Function signature change.

10:05 PM Changeset in webkit [47088] by abarth@webkit.org
  • 1 edit in trunk/WebKitSite/coding/major-objects.html

Hopefully the last tweak to major-objects.html

10:01 PM Changeset in webkit [47087] by abarth@webkit.org
  • 2 edits in trunk/WebKitSite

Unreviewed.

Patch by Adam Barth <abarth@webkit.org> on 2009-08-11
Remove a few characters from the ASCII art to make it fit better in the
template.

  • coding/major-objects.html:
9:56 PM Changeset in webkit [47086] by oliver@apple.com
  • 7 edits in trunk

Native JSON.stringify does not omit functions
https://bugs.webkit.org/show_bug.cgi?id=28117

Reviewed by Gavin Barraclough

Objects that are callable should be treated as undefined when
serialising to JSON.

9:55 PM Changeset in webkit [47085] by abarth@webkit.org
  • 1 edit
    1 add in trunk/WebKitSite

2009-08-11 Adam Barth <abarth@webkit.org>

Reviewed by Oliver Hunt.

Document relation between major objects in WebCore
https://bugs.webkit.org/show_bug.cgi?id=28210

Added a diagram showing the relation between major objects.

  • coding/major-objects.html: Added.
8:30 PM Changeset in webkit [47084] by mrowe@apple.com
  • 7 edits
    1 copy in trunk

2009-08-11 John Gregg <johnnyg@google.com>

Reviewed by Maciej Stachowiak.

WebCore:
Move the accessor for notifications presenter from WebUIDelegate
to WebUIDelegate2 which extends it. When making this call from
WebCoreSupport, check using QueryInterface before calling.

https://bugs.webkit.org/show_bug.cgi?id=28198

  • Interfaces/IWebUIDelegate.idl:
  • Interfaces/IWebUIDelegate2.idl: Added.
  • Interfaces/WebKit.idl:
  • WebCoreSupport/WebDesktopNotificationsDelegate.cpp:

(WebDesktopNotificationsDelegate::show):
(WebDesktopNotificationsDelegate::cancel):
(WebDesktopNotificationsDelegate::notificationObjectDestroyed):
(WebDesktopNotificationsDelegate::requestPermission):
(WebDesktopNotificationsDelegate::checkPermission):
(WebDesktopNotificationsDelegate::hasNotificationDelegate):
(WebDesktopNotificationsDelegate::notificationDelegate):

  • WebCoreSupport/WebDesktopNotificationsDelegate.h:

DumpRenderTree:
Switch DumpRenderTree to contain a WebUIDelegate2, which extends
WebUIDelegate, so that the notifications tests will still work.
https://bugs.webkit.org/show_bug.cgi?id=28198

  • DumpRenderTree/win/UIDelegate.h:
8:12 PM Changeset in webkit [47083] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

Rubberstamped by Mark Rowe.

<rdar://problem/7135864> fast/dom/frame-loading-via-document-write.html fails intermittently.

  • platform/mac-snowleopard/Skipped: Skip for now.
7:18 PM Changeset in webkit [47082] by staikos@webkit.org
  • 1 edit in trunk/WebCore/ChangeLog

Not reviewed: repair email address

6:30 PM Changeset in webkit [47081] by pkasting@chromium.org
  • 3 edits
    1 delete in trunk/WebCore

https://bugs.webkit.org/show_bug.cgi?id=27965
Merge skia/ImageSourceSkia.cpp into ImageSource.cpp.

Reviewed by Adam Treat.

  • WebCore.gypi: Remove deleted file, plus a file that was deleted a while back.
  • platform/graphics/ImageSource.cpp:

(WebCore::ImageSource::setData): Use slightly shorter syntax.
(WebCore::ImageSource::filenameExtension): Use shorter syntax.
(WebCore::ImageSource::createFrameAtIndex): Remove redundant call, clarify comments, reject all empty images.
(WebCore::ImageSource::frameHasAlphaAtIndex): Clarify comments, use shorter syntax.

  • platform/graphics/skia/ImageSourceSkia.cpp: Removed.
6:15 PM Changeset in webkit [47080] by Darin Adler
  • 2 edits in trunk/WebKitTools

Try to fix GTK build.

Patch by Darin Adler <Darin Adler> on 2009-08-11

  • DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:

(AccessibilityUIElement::stringForRange): Added.

6:09 PM Changeset in webkit [47079] by treat@webkit.org
  • 2 edits in trunk/LayoutTests

Remove tests from Skipped that now pass correctly after the addition
of keyDown named event handling to Qt's DumpRenderTree.

Patch by Mike Fenton <mike.fenton@torchmobile.com> on 2009-08-11
Reviewed by Adam Treat.

https://bugs.webkit.org/show_bug.cgi?id=28161

  • platform/qt/Skipped:
6:05 PM Changeset in webkit [47078] by atwilson@chromium.org
  • 11 edits in trunk

WebCore: SharedWorkers do not exit when the last parent document exits
https://bugs.webkit.org/show_bug.cgi?id=28170

Reviewed by David Levin.

Prevents Documents from being suspended/placed in the page cache if they are associated with shared workers.

Added vector cache instead of nested hash tables for SharedWorker repository.

Added SharedWorkerRepository::documentDetached API.

  • dom/Document.cpp:

(WebCore::Document::detach):
Notifies SharedWorkerRepository when the document is closing.

  • loader/FrameLoader.cpp:

Updated FrameLoader to not cache the Document if it is associated with a SharedWorker (since we can't suspend workers yet, we need to shut them down).
(WebCore::FrameLoader::canCachePageContainingThisFrame):
(WebCore::FrameLoader::logCanCacheFrameDecision):

  • workers/DefaultSharedWorkerRepository.cpp:

(WebCore::SharedWorkerProxy::create):
(WebCore::SharedWorkerProxy::isClosing):
Renamed from closing().
(WebCore::SharedWorkerProxy::matches):
Added manual equality function to replace old hash lookup.
(WebCore::SharedWorkerProxy::isDocumentInWorkerDocuments):
Checks to see if a document is in the worker's list of documents. Used to determine if page is suspendable.
(WebCore::SharedWorkerProxy::SharedWorkerProxy):
(WebCore::SharedWorkerProxy::addToWorkerDocuments):
Added tracking of the worker's list of documents for lifecycle purposes.
(WebCore::SharedWorkerProxy::documentDetached):
Shuts down the proxy when the last associated document is detached.
(WebCore::SharedWorkerProxy::close):
Marks the proxy as closed so it is no longer shared with new instances.
(WebCore::SharedWorkerProxy::workerContextDestroyed):
Removes the proxy from the repository/frees the proxy when the associated SharedWorkerContext is destroyed.
(WebCore::DefaultSharedWorkerRepository::workerScriptLoaded):
closing()->isClosing()
(WebCore::SharedWorkerRepository::documentDetached):
(WebCore::SharedWorkerRepository::hasSharedWorkers):
Used by FrameLoader to determine if a page has shared workers and so cannot be suspended/cached.
(WebCore::DefaultSharedWorkerRepository::hasSharedWorkers):
(WebCore::DefaultSharedWorkerRepository::removeProxy):
Invoked by workerContextDestroyed() to remove a SharedWorkerProxy from the repository.
(WebCore::DefaultSharedWorkerRepository::documentDetached):
(WebCore::DefaultSharedWorkerRepository::connectToWorker):
(WebCore::DefaultSharedWorkerRepository::getProxy):

  • workers/DefaultSharedWorkerRepository.h:
  • workers/SharedWorkerRepository.h:

LayoutTests: SharedWorkers do not exit when the last parent document exits.
https://bugs.webkit.org/show_bug.cgi?id=28170

Reviewed by David Levin.

Added more tests to check that previous incarnations of the SharedWorker "name" are shut down.

  • fast/workers/shared-worker-replace-global-constructor.html-disabled:

Fixed incorrect path to common script.

  • fast/workers/shared-worker-shared-expected.txt:
  • fast/workers/shared-worker-shared.html-disabled:

Added more tests for sharing, including initial test to make sure previous incarnations of shared worker were closed.

  • fast/workers/worker-replace-global-constructor.html:

Removed extraneous closing HTML tag.

6:04 PM Changeset in webkit [47077] by treat@webkit.org
  • 2 edits in trunk/WebKitTools

Add processing for string constants used by the tests to trigger
common actions like up, down, left, right, etc. for the Qt
implementation of DumpRenderTree.

Patch by Mike Fenton <mike.fenton@torchmobile.com> on 2009-08-10
Reviewed by Adam Treat.

Note this allows fast/forms/textarea-arrow-navigation.html to pass
correctly.

https://bugs.webkit.org/show_bug.cgi?id=28161

  • DumpRenderTree/qt/jsobjects.cpp:

(EventSender::keyDown):

5:58 PM Changeset in webkit [47076] by Chris Fleizach
  • 12 edits
    2 adds in trunk

WebCore: Bug 28200 - ListMarker should be included as part of the text value to parse
https://bugs.webkit.org/show_bug.cgi?id=28200

Reviewed by Darin Adler.

When returning text through the marker range attributes, the list marker text should
also be included. It isn't included by default because a TextIterator is used which
only looks at Nodes for text. A list marker does not have a corresponding node,
so its text is not returned through accessibility. That is, until now.

Test: platform/mac/accessibility/string-range-contains-listmarker.html

  • accessibility/AccessibilityObject.cpp:

(WebCore::AccessibilityObject::listMarkerTextForRange):
(WebCore::AccessibilityObject::stringForVisiblePositionRange):

  • accessibility/AccessibilityObject.h:
  • accessibility/mac/AccessibilityObjectWrapper.mm:

(-[AccessibilityObjectWrapper doAXAttributedStringForTextMarkerRange:]):
(-[AccessibilityObjectWrapper accessibilityParameterizedAttributeNames]):
(-[AccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]):

  • rendering/RenderListItem.h:

Make markerText() public so that accessibility code can access it.

WebKitTools: Bug 28200 - ListMarker should be included as part of the text value to parse
https://bugs.webkit.org/show_bug.cgi?id=28200

Reviewed by Darin Adler.

Add the ability to retrieve a string given a plain NSRange.

  • DumpRenderTree/AccessibilityUIElement.cpp:

(stringForRangeCallback):
(AccessibilityUIElement::getJSClass):

  • DumpRenderTree/AccessibilityUIElement.h:
  • DumpRenderTree/mac/AccessibilityUIElementMac.mm:

(AccessibilityUIElement::valueDescription):
(AccessibilityUIElement::stringForRange):

  • DumpRenderTree/win/AccessibilityUIElementWin.cpp:

(AccessibilityUIElement::stringForRange):

LayoutTests: Bug 28200 - ListMarker should be included as part of the text value to parse
https://bugs.webkit.org/show_bug.cgi?id=28200

Reviewed by Darin Adler.

Tests for checking that the list marker text is included when asking for text
through accessibility.

  • platform/mac/accessibility/bounds-for-range-expected.txt.
  • platform/mac/accessibility/string-range-contains-listmarker-expected.txt: Added.
  • platform/mac/accessibility/string-range-contains-listmarker.html: Added.
5:35 PM Changeset in webkit [47075] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

Rubberstamped by Mark Rowe.

<rdar://problem/7135553> plugins/root-object-premature-delete-crash.html crashes intermittently.

  • platform/mac-snowleopard/Skipped: Skip for now.
5:08 PM Changeset in webkit [47074] by dimich@chromium.org
  • 2 edits in trunk/WebKitTools

Fix fast/dom/prototype-inheritance.html
and fast/dom/prototype-inheritance-2.html
broken on Windows by http://trac.webkit.org/changeset/47018

Reviewed by NOBODY (Speculative fix for the layout test failure).

  • DumpRenderTree/win/DumpRenderTree.cpp:

(resetDefaultsToConsistentValues): enable app cache in Windows DRT.

4:20 PM Changeset in webkit [47073] by pkasting@chromium.org
  • 4 edits
    1 move in trunk/WebCore

https://bugs.webkit.org/show_bug.cgi?id=27965
Move cairo/ImageSourceCairo.cpp to ImageSource.cpp in preparation for
merging most of the other ImageSource*.cpp files with it.

Reviewed by Adam Treat.

  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.vcproj/WebCore.vcproj:
  • platform/graphics/ImageSource.cpp: Copied from WebCore/platform/graphics/cairo/ImageSourceCairo.cpp.
  • platform/graphics/cairo/ImageSourceCairo.cpp: Removed.
4:12 PM Changeset in webkit [47072] by ddkilzer@apple.com
  • 2 edits in trunk/BugsSite

Updated data/params parameter

  • data/params: Updated 'loginnetmask' parameter to allow users to choose whether their cookies are restricted to a single IP address or not.
4:09 PM Changeset in webkit [47071] by bfulgham@webkit.org
  • 4 edits in trunk/WebCore

Cairo-based Windows port does not handle the https protocol.
https://bugs.webkit.org/show_bug.cgi?id=28171.

Reviewed by Adam Roben.

No new tests needed. Existing https tests show this failure
for WinCairo builds.

  • WebCore.vcproj/WebCoreCURL.vsprops: Add a few preprocessor definitions for SSL support in cURL.
  • platform/network/curl/ResourceHandleManager.h: Add new certificate path member variable.
  • platform/network/curl/ResourceHandleManager.cpp:

(WebCore::certificateBundlePath): New method to retrieve the

proper certificate resource bundle path.

(WebCore::ResourceHandleManager::initializeHandle): Check for

the resource bundle and use if available to activate SSL
support.

3:44 PM Changeset in webkit [47070] by eric@webkit.org
  • 1 edit
    3 adds in trunk/LayoutTests

2009-08-11 Eric Seidel <eric@webkit.org>

No review, just adding results.

Cross-platform ICO decoder gets wrong bit depth for icons w/o color count
https://bugs.webkit.org/show_bug.cgi?id=28073

Land test results (as pkasting asked me to do hours ago, sorry).

  • fast/images/icon-0colors-expected.checksum: Added.
  • fast/images/icon-0colors-expected.png: Added.
  • fast/images/icon-0colors-expected.txt: Added.
3:33 PM Changeset in webkit [47069] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-11 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool: Re-factor shared landing logic into helper class to share more code
https://bugs.webkit.org/show_bug.cgi?id=28193

Added new WebKitLandingScripts class to hold this shared logic.
Also added a view_source_url function to move more webkit-specific urls out of bugzilla-tool core.

  • Scripts/bugzilla-tool:
3:25 PM Changeset in webkit [47068] by eric@webkit.org
  • 5 edits in trunk/WebCore

2009-08-11 Maxime Simon <Maxime Simon>

Reviewed by Eric Seidel.

[Haiku] Use Clipboard::create() instead of the constructor of Clipboard.
https://bugs.webkit.org/show_bug.cgi?id=28127

  • editing/haiku/EditorHaiku.cpp: (WebCore::Editor::newGeneralClipboard):
  • page/haiku/EventHandlerHaiku.cpp: (WebCore::EventHandler::passWheelEventToWidget): (WebCore::EventHandler::createDraggingClipboard):
  • platform/haiku/ClipboardHaiku.cpp: Style cleanup. There were many whitespace end of lines. (WebCore::ClipboardHaiku::ClipboardHaiku): (WebCore::ClipboardHaiku::clearAllData): (WebCore::ClipboardHaiku::getData): (WebCore::ClipboardHaiku::setData): (WebCore::ClipboardHaiku::types): (WebCore::ClipboardHaiku::dragLocation): (WebCore::ClipboardHaiku::dragImage): (WebCore::ClipboardHaiku::setDragImage): (WebCore::ClipboardHaiku::dragImageElement): (WebCore::ClipboardHaiku::createDragImage): (WebCore::ClipboardHaiku::declareAndWriteDragImage): (WebCore::ClipboardHaiku::writeURL): (WebCore::ClipboardHaiku::writeRange): (WebCore::ClipboardHaiku::hasData):
  • platform/haiku/DragDataHaiku.cpp: (WebCore::DragData::createClipboard):
3:15 PM Changeset in webkit [47067] by atwilson@chromium.org
  • 2 edits in trunk/WebCore

Speculative fix for GTK build break.

Reviewed by NOBODY (Build break).

  • WebKit.pro:
3:05 PM Changeset in webkit [47066] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-11 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Adam Treat.

Fix the current failures on the buildbot.

As Qt hooks up the maybeDump to loadFinished, we need to make
sure that calling dump() will not call maybeDump on loadFinished.

As dump is called my emitting done() which calls dump() and then
setting m_isLoading to false. So in the case m_isLoading is false,
do not dump again.

The current code is confusing, and should be made more clear
in another commit.

  • DumpRenderTree/qt/jsobjects.cpp: (LayoutTestController::maybeDump): (LayoutTestController::notifyDone):
2:35 PM Changeset in webkit [47065] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Fix the current failures on the buildbot.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-08-11
Reviewed by Adam Treat.

As Qt hooks up the maybeDump to loadFinished, we need to make
sure that calling dump() will not call maybeDump on loadFinished.

As dump is called my emitting done() which calls dump() and then
setting m_isLoading to false. So in the case m_isLoading is false,
do not dump again.

The current code is confusing, and should be made more clear
in another commit.

  • DumpRenderTree/qt/jsobjects.cpp:

(LayoutTestController::maybeDump):
(LayoutTestController::notifyDone):

2:31 PM Changeset in webkit [47064] by atwilson@chromium.org
  • 2 edits in trunk

WebCore: Speculative Windows build fix

Reviewed by NOBODY (build break).

  • WebCore.vcproj/WebCore.vcproj:

Added missing build steps for non-debug targets.

WebKit/win: Speculative Windows build fix - undo previous unnecessary change.

Reviewed by NOBODY (build break).

  • WebCoreSupport/WebDesktopNotificationsDelegate.h:
2:31 PM Changeset in webkit [47063] by atwilson@chromium.org
  • 2 edits in trunk

Speculative build fix for Qt build.

Reviewed by NOBODY (build break)

  • WebCore.pro: Added /notifications to include path
2:30 PM Changeset in webkit [47062] by oliver@apple.com
  • 7 edits in trunk

REGRESSION: Hang/crash in BytecodeGenerator::constRegisterFor loading simple page
https://bugs.webkit.org/show_bug.cgi?id=28169

Reviewed by Geoff Garen.

Handle the case where someone has attempted to shadow a property
on the global object with a constant.

2:13 PM Changeset in webkit [47061] by pkasting@chromium.org
  • 2 edits in trunk/WebKitTools

https://bugs.webkit.org/show_bug.cgi?id=27323
Handle arbitrary line endings when manufacturing patches for additions
with history.

Reviewed by Darin Adler.

  • Scripts/svn-create-patch:
2:04 PM Changeset in webkit [47060] by bweinstein@apple.com
  • 5 edits in trunk/WebKit/win

2009-08-11 Brian Weinstein <bweinstein@apple.com>

Reviewed by Steve Falkenburg.

Refactored MIMETypeforExtension to be a member of IWebViewPrivate
instead of IWebView.

  • Interfaces/IWebView.idl:
  • Interfaces/IWebViewPrivate.idl:
  • WebView.cpp: (WebView::MIMETypeForExtension):
  • WebView.h:
2:01 PM Changeset in webkit [47059] by atwilson@chromium.org
  • 2 edits in trunk/WebCore

Speculative build fix for Qt build.

Reviewed by NOBODY (build break)

  • WebCore.pro: Added /notifications to include path
1:45 PM Changeset in webkit [47058] by atwilson@chromium.org
  • 2 edits in trunk/WebKit/win

Speculative fix for windows build break.

Reviewed by NOBODY (build break)

  • WebKitSupport/WebDesktopNotificationsDelegate.h:

WebCore/Notification*.h => WebCore/notifications/Notification*.h

1:37 PM Changeset in webkit [47057] by atwilson@chromium.org
  • 2 edits in trunk

Speculative fix for GTK build break.

Reviewed by NOBODY (Build break).

  • WebKit.pro:
1:21 PM Changeset in webkit [47056] by atwilson@chromium.org
  • 45 edits
    32 adds in trunk

JavaScriptCore: Desktop Notifications API
https://bugs.webkit.org/show_bug.cgi?id=25463

Patch by John Gregg <johnnyg@google.com> on 2009-08-11
Reviewed by Maciej Stachowiak.

Adds ENABLE_NOTIFICATION flag.

  • Configurations/FeatureDefines.xcconfig:
  • wtf/Platform.h:

WebCore: Desktop Notifications API
https://bugs.webkit.org/show_bug.cgi?id=25463

Patch by John Gregg <johnnyg@google.com> on 2009-08-11
Reviewed by Maciej Stachowiak.

Adds an implementation of desktop notifications, behind the compile-time
flag ENABLE(NOTIFICATIONS). This code simply relays calls on the JavaScript
API through a NotificationProvider interface, which must provide the services
to actually notify the user and manage the desktop. This provider is injected
into WebKit through the ChromeClient for normal page contexts and through the
WorkerThread for worker contexts.

A permissions API is defined on a per-origin basis, which allows a web page
to check if its origin is allowed to show desktop notifications, and to request
permission for its origin.

  • Configurations/FeatureDefines.xcconfig:
  • DerivedSources.cpp:
  • DerivedSources.make:
  • GNUmakefile.am:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.vcproj/WebCoreCommon.vsprops:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMWindowCustom.cpp:
  • bindings/js/JSDesktopNotificationsCustom.cpp: Added.

(WebCore::JSNotificationCenter::requestPermission):
(WebCore::JSNotification::addEventListener):
(WebCore::JSNotification::removeEventListener):

  • bindings/js/JSEventTarget.cpp:

(WebCore::toJS):
(WebCore::toEventTarget):

  • dom/EventTarget.cpp:

(WebCore::EventTarget::toNotification):

  • dom/EventTarget.h:
  • loader/EmptyClients.h:

(WebCore::EmptyChromeClient::notificationPresenter):

  • notifications: Added.
  • notifications/Notification.cpp: Added.

(WebCore::Notification::Notification):
(WebCore::Notification::~Notification):
(WebCore::Notification::show):
(WebCore::Notification::cancel):
(WebCore::Notification::ondisplay):
(WebCore::Notification::setOndisplay):
(WebCore::Notification::onerror):
(WebCore::Notification::setOnerror):
(WebCore::Notification::onclose):
(WebCore::Notification::setOnclose):
(WebCore::Notification::getAttributeEventListener):
(WebCore::Notification::setAttributeEventListener):
(WebCore::Notification::clearAttributeEventListener):
(WebCore::Notification::dispatchDisplayEvent):
(WebCore::Notification::dispatchErrorEvent):
(WebCore::Notification::dispatchCloseEvent):
(WebCore::Notification::addEventListener):
(WebCore::Notification::removeEventListener):
(WebCore::Notification::handleEvent):
(WebCore::Notification::dispatchEvent):

  • notifications/Notification.h: Added.

(WebCore::Notification::create):
(WebCore::Notification::isHTML):
(WebCore::Notification::url):
(WebCore::Notification::contents):
(WebCore::Notification::scriptExecutionContext):
(WebCore::Notification::toNotification):
(WebCore::Notification::refEventTarget):
(WebCore::Notification::derefEventTarget):

  • notifications/Notification.idl: Added.
  • notifications/NotificationCenter.cpp: Added.

(WebCore::NotificationCenter::NotificationCenter):
(WebCore::NotificationCenter::checkPermission):
(WebCore::NotificationCenter::requestPermission):

  • notifications/NotificationCenter.h: Added.

(WebCore::NotificationCenter::create):
(WebCore::NotificationCenter::createHTMLNotification):
(WebCore::NotificationCenter::createNotification):
(WebCore::NotificationCenter::context):
(WebCore::NotificationCenter::presenter):

  • notifications/NotificationCenter.idl: Added.
  • notifications/NotificationContents.h: Added.

(WebCore::NotificationContents::NotificationContents):
(WebCore::NotificationContents::icon):
(WebCore::NotificationContents::title):
(WebCore::NotificationContents::body):

  • notifications/NotificationPresenter.h: Added.

(WebCore::NotificationPresenter::):
(WebCore::NotificationPresenter::~NotificationPresenter):

  • page/Chrome.cpp:

(WebCore::Chrome::notificationPresenter):

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

(WebCore::DOMWindow::clear):
(WebCore::DOMWindow::webkitNotifications):

  • page/DOMWindow.h:
  • page/DOMWindow.idl:
  • page/Page.h:
  • workers/WorkerContext.cpp:

(WebCore::WorkerContext::~WorkerContext):
(WebCore::WorkerContext::webkitNotifications):

  • workers/WorkerContext.h:
  • workers/WorkerContext.idl:
  • workers/WorkerThread.h:

(WebCore::WorkerThread::getNotificationPresenter):
(WebCore::WorkerThread::setNotificationPresenter):

WebKit/mac: Desktop Notifications API
https://bugs.webkit.org/show_bug.cgi?id=25643

Patch by John Gregg <johnnyg@google.com> on 2009-08-11
Reviewed by Maciej Stachowiak.

Adds ENABLE_NOTIFICATION flag.

  • Configurations/FeatureDefines.xcconfig:

WebKit/win: Desktop Notifications API
https://bugs.webkit.org/show_bug.cgi?id=25643

Patch by John Gregg <johnnyg@google.com> on 2009-08-11
Reviewed by Maciej Stachowiak.

Adds test hooks for notification features to WebCoreSupport for
use in DumpRenderTree.

  • Interfaces/IWebDesktopNotificationsDelegate.idl: Added.
  • Interfaces/IWebUIDelegate.idl:
  • Interfaces/WebKit.idl:
  • WebCoreSupport/WebChromeClient.cpp:

(WebChromeClient::WebChromeClient):

  • WebCoreSupport/WebChromeClient.h:

(WebChromeClient::notificationPresenter):

  • WebCoreSupport/WebDesktopNotificationsDelegate.cpp: Added.

(NotificationCOMWrapper::create):
(NotificationCOMWrapper::NotificationCOMWrapper):
(NotificationCOMWrapper::QueryInterface):
(NotificationCOMWrapper::AddRef):
(NotificationCOMWrapper::Release):
(NotificationCOMWrapper::isHTML):
(NotificationCOMWrapper::contentsURL):
(NotificationCOMWrapper::iconURL):
(NotificationCOMWrapper::title):
(NotificationCOMWrapper::text):
(NotificationCOMWrapper::notifyDisplay):
(NotificationCOMWrapper::notifyError):
(NotificationCOMWrapper::notifyClose):
(WebDesktopNotificationsDelegate::WebDesktopNotificationsDelegate):
(WebDesktopNotificationsDelegate::show):
(WebDesktopNotificationsDelegate::cancel):
(WebDesktopNotificationsDelegate::notificationObjectDestroyed):
(WebDesktopNotificationsDelegate::requestPermission):
(WebDesktopNotificationsDelegate::checkPermission):
(WebDesktopNotificationsDelegate::notificationDelegate):

  • WebCoreSupport/WebDesktopNotificationsDelegate.h: Added.
  • WebCoreSupport/WebInspectorDelegate.h:

(WebInspectorDelegate::desktopNotificationsDelegate):

  • WebKit.vcproj/Interfaces.vcproj:
  • WebKit.vcproj/WebKit.vcproj:

WebKitTools: Add support for desktop notifications API to DumpRenderTree,
and support for ENABLE_NOTIFICATIONS flag to build-webkit.

Patch by John Gregg <johnnyg@google.com> on 2009-08-11
Reviewed by Maciej Stachowiak.

  • DumpRenderTree/LayoutTestController.cpp:

(grantDesktopNotificationPermissionCallback):
(LayoutTestController::staticFunctions):
(LayoutTestController::grantDesktopNotificationPermission):
(LayoutTestController::checkDesktopNotificationPermission):

  • DumpRenderTree/LayoutTestController.h:
  • DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp: Added.

(DRTDesktopNotificationPresenter::DRTDesktopNotificationPresenter):
(DRTDesktopNotificationPresenter::QueryInterface):
(DRTDesktopNotificationPresenter::AddRef):
(DRTDesktopNotificationPresenter::Release):
(DRTDesktopNotificationPresenter::showDesktopNotification):
(DRTDesktopNotificationPresenter::cancelDesktopNotification):
(DRTDesktopNotificationPresenter::notificationDestroyed):
(DRTDesktopNotificationPresenter::checkNotificationPermission):
(DRTDesktopNotificationPresenter::requestNotificationPermission):

  • DumpRenderTree/win/DRTDesktopNotificationPresenter.h: Added.
  • DumpRenderTree/win/DumpRenderTree.vcproj:
  • DumpRenderTree/win/UIDelegate.cpp:

(UIDelegate::UIDelegate):
(UIDelegate::desktopNotificationsDelegate):

  • DumpRenderTree/win/UIDelegate.h:
  • Scripts/build-webkit:

LayoutTests: Desktop Notifications API
https://bugs.webkit.org/show_bug.cgi?id=25463

Patch by John Gregg <johnnyg@google.com> on 2009-08-11
Reviewed by Maciej Stachowiak.

Adding layout tests for desktop notifications. Currently these
are all disabled because the flag for this feature is set to 0.

  • fast/notifications: Added.
  • fast/notifications/notifications-check-permission-expected.txt: Added.
  • fast/notifications/notifications-check-permission.html-disabled: Added.
    • Validates that checking permission for notifications works correctly.
  • fast/notifications/notifications-display-close-events-expected.txt: Added.
  • fast/notifications/notifications-display-close-events.html-disabled: Added.
    • Validates that the display and close events are fired correctly.
  • fast/notifications/notifications-double-show-expected.txt: Added.
  • fast/notifications/notifications-double-show.html-disabled: Added.
    • Validates that the same notification object cannot be shown twice.
  • fast/notifications/notifications-request-permission-expected.txt: Added.
  • fast/notifications/notifications-request-permission.html-disabled: Added.
    • Validates that requesting permission works correctly.
  • fast/notifications/notifications-with-permission-expected.txt: Added.
  • fast/notifications/notifications-with-permission.html-disabled: Added.
    • Validates that, with permission, showing a notification works correctly.
  • fast/notifications/notifications-without-permission-expected.txt: Added.
  • fast/notifications/notifications-without-permission.html-disabled: Added.
    • Validates that, without permission, showing a notification fails.
1:19 PM Changeset in webkit [47055] by mrowe@apple.com
  • 4 copies
    1 add in releases/Apple/Safari 4.0.3

Tag the Safari 4.0.3 release.

This release consists of JavaScriptCore-x531.5, JavaScriptGlue-x531.5, WebCore-x531.9 and WebKit-x531.9.

1:19 PM Changeset in webkit [47054] by andersca@apple.com
  • 2 edits in trunk/WebKit/mac

Remove dead declarations.

Reviewed by Sam Weinig.

  • WebView/WebViewInternal.h:
1:17 PM Changeset in webkit [47053] by mrowe@apple.com
  • 4 copies
    1 add in releases/Apple/Safari 4.0.2

Tag the Safari 4.0.2 release.

This release consists of JavaScriptCore-x530.19, WebCore-x530.19 and WebKit-x530.19.2. JavaScriptGlue remains unchanged from Safari 4.0.1.

1:14 PM Changeset in webkit [47052] by beidson@apple.com
  • 2 edits
    3 copies
    3 adds in trunk/LayoutTests

Rubberstamped by Mark Rowe.

-Stop skipping platform/mac/fast/text/myanmar-shaping.html on SnowLeopard.
-Move the previous results aside for Tiger and Leopard only.
-Generate new results for SnowLeopard.

  • platform/mac-snowleopard/Skipped: Remove platform/mac/fast/text/myanmar-shaping.html
  • platform/mac-leopard/fast/text/myanmar-shaping-expected.checksum: Copied from platform/mac/fast/text/myanmar-shaping-expected.checksum.
  • platform/mac-leopard/fast/text/myanmar-shaping-expected.png: Copied from platform/mac/fast/text/myanmar-shaping-expected.png.
  • platform/mac-leopard/fast/text/myanmar-shaping-expected.txt: Copied from platform/mac/fast/text/myanmar-shaping-expected.txt.
  • platform/mac/fast/text/myanmar-shaping-expected.checksum: Replaced.
  • platform/mac/fast/text/myanmar-shaping-expected.png: Replaced.
  • platform/mac/fast/text/myanmar-shaping-expected.txt: Replaced.
12:52 PM Changeset in webkit [47051] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Anders Carlsson.

Resolve <rdar://problem/6733642>.

  • platform/mac-snowleopard/Skipped: Re-enable plugins/netscape-plugin-setwindow-size.html and plugins/netscape-plugin-setwindow-size-2.html, as they now reliably pass.
12:51 PM Changeset in webkit [47050] by eric@webkit.org
  • 4 edits in trunk/JavaScriptCore

2009-08-11 Maxime Simon <Maxime Simon>

Reviewed by Eric Seidel.

Modifications on JavaScriptCore to allow Haiku port.
https://bugs.webkit.org/show_bug.cgi?id=28121

  • runtime/Collector.cpp: Haiku doesn't have sys/mman.h, using OS.h instead. (JSC::currentThreadStackBase): Haiku uses its own threading system.
  • wtf/Platform.h: Defining all Haiku platform values.
  • wtf/haiku/MainThreadHaiku.cpp: Adding a missing header (NotImplemented.h).
12:37 PM Changeset in webkit [47049] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-08-11 Joseph Pecoraro <joepeck02@gmail.com>

Reviewed by Timothy Hatcher.

Inspector: Console Drawer is Toggling instead of Staying Open
https://bugs.webkit.org/show_bug.cgi?id=28115

  • inspector/front-end/Drawer.js: (WebInspector.Drawer.prototype.showView): ensures the drawer will be open and showing the provided view
  • inspector/front-end/inspector.js: (WebInspector.showConsole): usees the new showView to guarantee the Drawer will be open (WebInspector.showChanges): uses the new showView guarantee the Drawer will be open
12:29 PM Changeset in webkit [47048] by Nikolas Zimmermann
  • 10 edits
    1 add in trunk/WebCore

2009-08-11 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] WMLPageState is not allowed to store the active card, it needs to be done per document
https://bugs.webkit.org/show_bug.cgi?id=28180

Don't store the active WMLCardElement in WMLPageState, but in WMLDocument.
Otherwhise this may lead to crashes related to intrinsic event exeuction.

Unfortunately select elements aren't testable by the layout tests, so adding
a new manual test reproducing the crash.

  • manual-tests/wml/select-onpick-event-crash.wml: Added.
  • wml/WMLCardElement.cpp: (WebCore::WMLCardElement::determineActiveCard):
  • wml/WMLDoElement.cpp: (WebCore::WMLDoElement::defaultEventHandler):
  • wml/WMLDocument.cpp: (WebCore::WMLDocument::finishedParsing):
  • wml/WMLDocument.h: (WebCore::WMLDocument::activeCard):
  • wml/WMLGoElement.cpp: (WebCore::WMLGoElement::executeTask):
  • wml/WMLPageState.cpp: (WebCore::WMLPageState::WMLPageState):
  • wml/WMLPageState.h:
  • wml/WMLPrevElement.cpp: (WebCore::WMLPrevElement::executeTask):
  • wml/WMLRefreshElement.cpp: (WebCore::WMLRefreshElement::executeTask):
12:28 PM Changeset in webkit [47047] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Mark Rowe.

Resolve <rdar://problem/6877291>.

  • platform/mac-snowleopard/Skipped: media/video-size-intrinsic-scale.html no longer fails - stop skipping it!
12:19 PM Changeset in webkit [47046] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

Rubberstamped by Mark Rowe.

  • platform/mac-snowleopard/Skipped: Remove http/tests/navigation/post-goback-repost-policy.html, as it has already been disabled globally, and re-enabling is tracked by <rdar://problem/6778137>
12:13 PM Changeset in webkit [47045] by pkasting@chromium.org
  • 3 edits in trunk/WebKitTools

https://bugs.webkit.org/show_bug.cgi?id=28183
Support VS2008 as well as VS2005 in a few scripts.

Reviewed by Darin Adler.

  • Scripts/pdevenv: Check both $VS80COMNTOOLS and $VS90COMNTOOLS.
  • Scripts/webkitdirs.pm: Use $VSINSTALLDIR if available instead of hardcoding the VS2005 dir.
12:10 PM Changeset in webkit [47044] by bweinstein@apple.com
  • 4 edits in trunk/WebKit/win

2009-08-11 Brian Weinstein <bweinstein@apple.com>

Reviewed by Adam Roben.

Fix of <rdar://5016504> SVG file cannot be dragged to webview,
but can be opened via Open File.

Added a new function to IWebView that gets the MIME type for a given
extension from the MIMETypeRegistry.

  • Interfaces/IWebView.idl:
  • WebView.cpp: (WebView::MIMETypeForExtension):
  • WebView.h:
12:09 PM Changeset in webkit [47043] by dimich@chromium.org
  • 2 edits in trunk/WebKitTools

Fix for layout tests failures. Need to initialize some preferences early
because WebView on Windows uses them during create time.

Reviewed by NOBODY (Windows layout tests fix).

  • DumpRenderTree/win/DumpRenderTree.cpp:

(main):

12:00 PM Changeset in webkit [47042] by pkasting@chromium.org
  • 3 edits
    2 adds in trunk

WebCore: https://bugs.webkit.org/show_bug.cgi?id=28073
Treat icons with no bit count and no color count as 256-color for
purposes of quality ranking. Also fix a couple cases of a style
violation.

Reviewed by Eric Seidel.

Test: fast/images/icon-0colors.html

  • platform/image-decoders/ico/ICOImageDecoder.cpp:

(WebCore::ICOImageDecoder::processDirectory):
(WebCore::ICOImageDecoder::readDirectoryEntry):

LayoutTests: https://bugs.webkit.org/show_bug.cgi?id=28073
Treat icons with no bit count and no color count as 256-color for
purposes of quality ranking.

Reviewed by Eric Seidel.

  • fast/images/icon-0colors.html: Added.
  • fast/images/resources/0colors.ico: Added.
11:29 AM Changeset in webkit [47041] by xan@webkit.org
  • 2 edits in trunk/WebKit/gtk

2009-08-11 Benjamin Otte <otte@gnome.org>

Reviewed by Xan Lopez.

When instructed to print an out-of-range page, print an empty page
instead of crashing.

  • webkit/webkitwebframe.cpp: (draw_page_callback):
11:27 AM Changeset in webkit [47040] by eric@webkit.org
  • 3 edits
    1 copy in trunk/WebKitTools

2009-08-11 Eric Seidel <eric@webkit.org>

Reviewed by Darin Adler.

Exception in land-patches
https://bugs.webkit.org/show_bug.cgi?id=27962

Use ("%s" % object) instead of ("" + object).
Added unit tests for logging.py.

  • Scripts/modules/logging.py:
  • Scripts/modules/logging_unittest.py: Added.
  • Scripts/run-webkit-unittests:
10:42 AM Changeset in webkit [47039] by dimich@chromium.org
  • 19 edits
    6 adds in trunk

Re-land the patch for https://bugs.webkit.org/show_bug.cgi?id=20534.
Previously reverted since incorrect merge on landing has caused Layout Tests failures on Windows.

WebKit/mac: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added new methods for overriding default WebPreference values
and for resetting preferences to their defaults.
See https://bugs.webkit.org/show_bug.cgi?id=20534

  • WebView/WebPreferences.mm:

(-[WebPreferences _setPreferenceForTestWithValue:withKey:]): added.

  • WebView/WebPreferencesPrivate.h: same.

WebKit/win: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added support for overriding default preferences per-test.
See https://bugs.webkit.org/show_bug.cgi?id=20534

  • Interfaces/IWebPreferencesPrivate.idl: added new method setPreferenceForTest
  • WebPreferences.cpp: same.
  • WebPreferences.h: same.

WebKitTools: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added support for overriding default preferences per-test.
See https://bugs.webkit.org/show_bug.cgi?id=20534

  • DumpRenderTree/LayoutTestController.cpp:

(overridePreferenceCallback): add wiring for layoutTestController.overridePreference.
(LayoutTestController::staticFunctions): same.

  • DumpRenderTree/LayoutTestController.h: same.
  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:

(LayoutTestController::overridePreference): added empty overridePreference method.

  • DumpRenderTree/mac/DumpRenderTree.mm: implemented preference override.

(resetDefaultsToConsistentValues): new method, resets preferences to same set of value before every test.
(setDefaultsToConsistentValuesForTesting): new method, sets other details of testing environment, every time DRT starts.
(resetWebViewToConsistentStateBeforeTesting): move some preference setting from here to new resetDefaultsToConsistentValues().

  • DumpRenderTree/mac/LayoutTestControllerMac.mm:

(LayoutTestController::overridePreference):

  • DumpRenderTree/win/DumpRenderTree.cpp:

(resetDefaultsToConsistentValues): new method, resets preferences to same set of value before every test.
(resetWebViewToConsistentStateBeforeTesting): move some preference setting from here to new resetDefaultsToConsistentValues().
(createWebViewAndOffscreenWindow): same.
(main): same.

  • DumpRenderTree/win/LayoutTestControllerWin.cpp:

(LayoutTestController::overridePreference):

  • DumpRenderTree/wx/LayoutTestControllerWx.cpp:

(LayoutTestController::overridePreference):

LayoutTests: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added tests to verify that default preferences can be overridden per-test
and that overridden values are reset between each test.
See https://bugs.webkit.org/show_bug.cgi?id=20534

  • fast/harness/override-preferences-2-expected.txt: Added.
  • fast/harness/override-preferences-2.html: Added. Verifies overridePreverence("WebKitDefaultFontSize", "24").
  • fast/harness/override-preferences-expected.txt: Added.
  • fast/harness/override-preferences.html: Added. Verifies overridePreference("WebKitJavaScriptEnabled", false).
  • fast/harness/override-zzz-reset-expected.txt: Added.
  • fast/harness/override-zzz-reset.html: Added. Because of 'zzz' this test will run after the tests above and

verify that override of preferences does not 'spill' to the subsequent tests in a batch.

  • platform/gtk/Skipped: Tests above disabled for GTK until GTK DRT implements preference override.
10:26 AM QtWebKitJournal edited by ben@webkit.org
(diff)
9:58 AM Changeset in webkit [47038] by bweinstein@apple.com
  • 2 edits in trunk/WebKitTools

2009-08-11 Brian Weinstein <bweinstein@apple.com>

Reviewed by Sam Weinig.

Added support for DRT to support arguments for eventSender.mouseDown and eventSender.mouseUp for Windows.
https://bugs.webkit.org/show_bug.cgi?id=28166.

This is a step towards fixing fast/events/mouse-click-events.html on Windows.

  • DumpRenderTree/win/EventSender.cpp: (mouseDownCallback): (mouseUpCallback): (replaySavedEvents):
9:48 AM SuccessfulPortHowTo edited by eric@webkit.org
Added some recommended porting steps. (diff)
9:39 AM Changeset in webkit [47037] by jberlin@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-08-11 Jessie Berlin <jberlin@apple.com>

Fix windows build.


Reviewed by Adam Roben.

9:33 AM Changeset in webkit [47036] by atwilson@chromium.org
  • 18 edits
    1 copy in trunk/WebCore

Need to refactor WorkerObjectProxy.
https://bugs.webkit.org/show_bug.cgi?id=28136

Reviewed by David Levin.

Added WorkerReportingProxy for reporting worker state (exceptions, console messages, thread exited) to the parent.

Existing tests are sufficient, as this is only a refactoring.

  • GNUmakefile.am:

Added WorkerReportingProxy.h to build.

  • WebCore.gypi:

Added WorkerReportingProxy.h to build.

  • WebCore.vcproj/WebCore.vcproj:

Added WorkerReportingProxy.h to build.

  • WebCore.xcodeproj/project.pbxproj:

Added WorkerReportingProxy.h to build.

  • workers/DedicatedWorkerContext.cpp:
  • workers/DedicatedWorkerContext.h:

Removed addMessage/forwardException() APIs - now handled by WorkerReportingProxy.

  • workers/DedicatedWorkerThread.cpp:

(WebCore::DedicatedWorkerThread::DedicatedWorkerThread):
Changed constructor to pass WorkerReportingProxy to base class.

  • workers/DefaultSharedWorkerRepository.cpp:

Updated SharedWorkerProxy to implement WorkerReportingProxy interface.
(WebCore::SharedWorkerProxy::postExceptionToWorkerObject):
(WebCore::SharedWorkerProxy::postConsoleMessageToWorkerObject):
(WebCore::SharedWorkerProxy::workerContextClosed):
(WebCore::SharedWorkerProxy::workerContextDestroyed):
(WebCore::DefaultSharedWorkerRepository::workerScriptLoaded):
Now passes in WorkerReportingProxy when creating SharedWorkerThread.

  • workers/SharedWorkerContext.cpp:
  • workers/SharedWorkerContext.h:

Removed addMessage/forwardException() APIs - now handled by WorkerReportingProxy.

  • workers/SharedWorkerThread.cpp:

(WebCore::SharedWorkerThread::create):
(WebCore::SharedWorkerThread::SharedWorkerThread):
Passes WorkerReportingProxy to base class constructor.

  • workers/SharedWorkerThread.h:

Constructor now takes a WorkerReportingProxy.

  • workers/WorkerContext.cpp:

(WebCore::WorkerContext::~WorkerContext):
Moved code that calls workerContextDestroyed() into base class so shared workers share this functionality.
(WebCore::WorkerContext::close):
Now notifies WorkerReportingProxy when the thread is closing.
(WebCore::WorkerContext::reportException):
Reports exceptions via WorkerReportingProxy.
(WebCore::WorkerContext::addMessage):
Reports console messages via WorkerReportingProxy.

  • workers/WorkerContext.h:
  • workers/WorkerObjectProxy.h:

(WebCore::WorkerObjectProxy::workerContextClosed):
Default empty implementation for dedicated workers.

  • workers/WorkerReportingProxy.h: Added.

Base interface that contains APIs moved from WorkerObjectProxy.
(WebCore::WorkerReportingProxy::~WorkerReportingProxy):

  • workers/WorkerThread.cpp:

(WebCore::WorkerThread::WorkerThread):
Updated constructor to accept a WorkerReportingProxy.

  • workers/WorkerThread.h:

(WebCore::WorkerThread::workerReportingProxy):

8:39 AM Changeset in webkit [47035] by vestbo@webkit.org
  • 2 edits in trunk/JavaScriptCore

Buildfix for Qt-win platforms.

Patch by Csaba Osztrogonac <oszi@inf.u-szeged.hu> on 2009-08-11
Reviewed by Tor Arne Vestbø.

  • JavaScriptCore.pri: Choose MarkStackPosix.cpp or MarkStackWin.cpp depend on platform.
8:16 AM Changeset in webkit [47034] by kenneth@webkit.org
  • 2 edits
    1 add in trunk/LayoutTests

2009-08-11 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Add results for fast/borders/border-radius-split-inline and remove from platform/qt/Skipped
Note: Results will wrap to 4 lines even though it indicates 3 lines. This is due to font
size differences in Qt, and matches the output of FireFox.

https://bugs.webkit.org/show_bug.cgi?id=28147

  • platform/qt/Skipped:
  • platform/qt/fast/borders/border-radius-split-inline-expected.txt: Added.
7:56 AM Changeset in webkit [47033] by vestbo@webkit.org
  • 4 edits in trunk

Build the TestNetscapePlugin on Qt/Mac

Reviewed by Simon Hausmann.

  • WebKit.pro:
  • DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
7:56 AM Changeset in webkit [47032] by vestbo@webkit.org
  • 3 edits in trunk/WebKitTools

Fix a bunch of build warnings in TestNetscapePlugin

Reviewed by Simon Hausmann.

  • DumpRenderTree/TestNetscapePlugIn.subproj/TestObject.cpp:
  • DumpRenderTree/gtk/TestNetscapePlugin/TestNetscapePlugin.cpp:
3:37 AM Changeset in webkit [47031] by xan@webkit.org
  • 2 edits in trunk

2009-08-11 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

Prefix lengthy commands with AM_V_GEN so that they produce an
output like 'GEN <filename>' when in silent mode.

  • GNUmakefile.am:
2:40 AM Changeset in webkit [47030] by Simon Hausmann
  • 5 edits in trunk/WebCore

Fix the Qt build after r47022.

Patch by Simon Hausmann <simon.hausmann@nokia.com> on 2009-08-11
Reviewed by NOBODY (build fix).

Use the new markChildren() pattern for marking aggregate/child objects.

  • bridge/qt/qt_instance.cpp:

(JSC::Bindings::QtRuntimeObjectImp::markChildren):
(JSC::Bindings::QtInstance::markAggregate):

  • bridge/qt/qt_instance.h:
  • bridge/qt/qt_runtime.cpp:

(JSC::Bindings::QtRuntimeMetaMethod::markChildren):

  • bridge/qt/qt_runtime.h:
1:14 AM Changeset in webkit [47029] by dimich@chromium.org
  • 2 edits in trunk/WebKitTools

2009-08-11 Dmitry Titov <dimich@chromium.org>

Reviewed by NOBODY (build fix).

Attempt to fix the Windows test bot. Reverted r47015 caused one of the Windows
bots that runs Layout Tests to save "JS disabled" and "default font 24pt"
preferences permanently. Since these preferences are not initialized explicitly in DRT,
almost all Layout tests are failing. Attempt to fix by explicit
initialization of these into defaults (enabled, 16pt) at the beginning of the DRT.

  • DumpRenderTree/win/DumpRenderTree.cpp: (main): see above.
12:26 AM Changeset in webkit [47028] by Simon Hausmann
  • 2 edits in trunk/WebCore

Fix the build on the Qt build bot by making the generated JS bindings
files depend on the code generator itself.

Patch by Simon Hausmann <simon.hausmann@nokia.com> on 2009-08-11
Reviewed by Holger Freyther.

  • WebCore.pro:

Aug 10, 2009:

10:43 PM Changeset in webkit [47027] by oliver@apple.com
  • 3 edits in trunk/JavaScriptCore

Add new exports for MSVC

10:27 PM Changeset in webkit [47026] by oliver@apple.com
  • 3 edits in trunk/JavaScriptCore

Remove obsolete entries from MSVC exports file

10:18 PM Changeset in webkit [47025] by oliver@apple.com
  • 3 edits in trunk/JavaScriptCore

Add includes needed for non-allinonefile builds

10:15 PM Changeset in webkit [47024] by oliver@apple.com
  • 2 edits in trunk/JavaScriptCore

Fix export file for last build fix

10:07 PM Changeset in webkit [47023] by oliver@apple.com
  • 4 edits in trunk/JavaScriptCore

Build fix

Hoist page size initialization into platform specific code.

9:35 PM Changeset in webkit [47022] by oliver@apple.com
  • 103 edits
    3 copies
    4 adds in trunk

Stack overflow crash in JavaScript garbage collector mark pass
https://bugs.webkit.org/show_bug.cgi?id=12216

Reviewed by Gavin Barraclough and Sam Weinig

Make the GC mark phase iterative by using an explicit mark stack.
To do this marking any single object is performed in multiple stages

  • The object is appended to the MarkStack, this sets the marked bit for the object using the new markDirect() function, and then returns
  • When the MarkStack is drain()ed the object is popped off the stack and markChildren(MarkStack&) is called on the object to collect all of its children. drain() then repeats until the stack is empty.

Additionally I renamed a number of methods from 'mark' to 'markAggregate'
in order to make it more clear that marking of those object was not
going to result in an actual recursive mark.

9:17 PM Changeset in webkit [47021] by jorlow@chromium.org
  • 7 edits in trunk/WebCore

2009-08-10 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Eric Seidel.

Renamed all V8 NPN_ functions to _NPN_ to match JSC and make them "private".
https://bugs.webkit.org/show_bug.cgi?id=28089

Rename all the NPN_foo functions to _NPN_foo so that they can't directly be
called by mistake. This roughly matches the way JSC does it, which is an
additional advantage.

  • bindings/v8/NPV8Object.cpp: (npCreateV8ScriptObject): (_NPN_Invoke): (_NPN_InvokeDefault): (_NPN_Evaluate): (_NPN_EvaluateHelper): (_NPN_GetProperty): (_NPN_SetProperty): (_NPN_RemoveProperty): (_NPN_HasProperty): (_NPN_HasMethod): (_NPN_SetException): (_NPN_Enumerate): (_NPN_Construct):
  • bindings/v8/ScriptController.cpp: (WebCore::ScriptController::clearScriptObjects): (WebCore::ScriptController::createScriptInstanceForWidget): (WebCore::ScriptController::cleanupScriptObjectsForPlugin):
  • bindings/v8/V8NPObject.cpp: (npObjectInvokeImpl): (npObjectGetProperty): (npObjectIndexedPropertyGetter): (npObjectGetIndexedProperty): (npObjectSetProperty): (npObjectIndexedPropertySetter): (npObjectSetIndexedProperty): (weakNPObjectCallback): (createV8ObjectForNPObject): (forgetV8ObjectForNPObject):
  • bindings/v8/V8NPUtils.cpp: (getStringIdentifier):
  • bindings/v8/npruntime.cpp:
  • bindings/v8/npruntime_impl.h:
9:02 PM Changeset in webkit [47020] by jorlow@chromium.org
  • 16 edits in trunk

2009-08-10 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Brady Eidson.

Change _Storage.key() to match current spec behavior.
https://bugs.webkit.org/show_bug.cgi?id=28112

The current WebStorage spec says that _Storage.key() should return null when
the index is out of bounds, rather than an exception. A bunch of logic can be
simplified after this change.

  • bindings/js/JSStorageCustom.cpp: (WebCore::JSStorage::getPropertyNames):
  • bindings/v8/custom/V8StorageCustom.cpp: (WebCore::V8Custom::v8StorageNamedPropertyEnumerator):
  • storage/Storage.cpp: (WebCore::Storage::key):
  • storage/Storage.h:
  • storage/Storage.idl:
  • storage/StorageArea.h:
  • storage/StorageAreaImpl.cpp: (WebCore::StorageAreaImpl::key):
  • storage/StorageAreaImpl.h:
  • storage/StorageMap.cpp: (WebCore::StorageMap::key):
  • storage/StorageMap.h:

2009-08-10 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Brady Eidson.

_Storage.key() changed to match new spec behavior; update layout tests.
https://bugs.webkit.org/show_bug.cgi?id=28112

The current WebStorage spec says that _Storage.key() should return null when
the index is out of bounds, rather than an exception. Change the one layout
test that verified this behavior.

  • storage/domstorage/localstorage/simple-usage-expected.txt:
  • storage/domstorage/localstorage/simple-usage.html:
  • storage/domstorage/sessionstorage/simple-usage-expected.txt:
  • storage/domstorage/sessionstorage/simple-usage.html:
8:59 PM Changeset in webkit [47019] by jorlow@chromium.org
  • 1 edit in trunk/WebCore/ChangeLog

Fix silly change log bustage.

8:49 PM Changeset in webkit [47018] by jorlow@chromium.org
  • 4 edits in trunk/WebCore

2009-08-07 Michael Nordman <Michael Nordman>

Reviewed by Darin Fisher.

https://bugs.webkit.org/show_bug.cgi?id=28087

Mods for a chrome specific impl of ApplicationCacheHost.

  • The public section of the interface is the same, but we use PLATFORM(CHROMIUM) to identify which parts of the private section are shared. The chrome specific internals are encapsulated in a seperate ApplicationCacheHostInternal class. WebCore common code only contains a forward reference to this class.
  • Cleanup a couple of stale comments.
  • Have window.applicationCache return null to script when the feature is disabled in the preferences.


No new tests.

  • loader/appcache/ApplicationCacheHost.cpp:
  • loader/appcache/ApplicationCacheHost.h: (WebCore::ApplicationCacheHost::documentLoader):
  • page/DOMWindow.cpp: (WebCore::DOMWindow::applicationCache):

<<<<<<< .mine

7:55 PM Changeset in webkit [47017] by dimich@chromium.org
  • 19 edits
    6 deletes in trunk

Revert r47015 "Added support for overriding default preferences per-test."
https://bugs.webkit.org/show_bug.cgi?id=20534

Ir broke too many Windows tests and I don't have a simple fix right now.

7:53 PM Changeset in webkit [47016] by ben@webkit.org
  • 3 edits in trunk/WebCore

2009-08-10 Crystal Zhang <crystal.zhang@staikos.net>

Reviewed by George Staikos.

Implement GraphicsContext::fillRoundRect() for WINCE port
https://bugs.webkit.org/show_bug.cgi?id=27842

  • platform/graphics/GraphicsContext.h:
  • platform/graphics/wince/GraphicsContextWince.cpp: (WebCore::rectCenterPoint): (WebCore::GraphicsContext::fillRoundedRect): (WebCore::GraphicsContext::drawRoundCorner):
7:18 PM Changeset in webkit [47015] by dimich@chromium.org
  • 19 edits
    7 adds in trunk

WebKit/mac: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added new methods for overriding default WebPreference values
and for resetting preferences to their defaults.
https://bugs.webkit.org/show_bug.cgi?id=20534

  • WebView/WebPreferences.mm:

(-[WebPreferences _setPreferenceForTestWithValue:withKey:]): added.

  • WebView/WebPreferencesPrivate.h: same.

WebKit/win: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added support for overriding default preferences per-test.
https://bugs.webkit.org/show_bug.cgi?id=20534

  • Interfaces/IWebPreferencesPrivate.idl: added new method setPreferenceForTest
  • WebPreferences.cpp: same.
  • WebPreferences.h: same.

WebKitTools: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added support for overriding default preferences per-test.
https://bugs.webkit.org/show_bug.cgi?id=20534

  • DumpRenderTree/LayoutTestController.cpp:

(overridePreferenceCallback): add wiring for layoutTestController.overridePreference.
(LayoutTestController::staticFunctions): same.

  • DumpRenderTree/LayoutTestController.h: same.
  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:

(LayoutTestController::overridePreference): added empty overridePreference method.

  • DumpRenderTree/mac/DumpRenderTree.mm: implemented preference override.

(resetDefaultsToConsistentValues): new method, resets preferences to same set of value before every test.
(setDefaultsToConsistentValuesForTesting): new method, sets other details of testing environment, every time DRT starts.
(resetWebViewToConsistentStateBeforeTesting): move some preference setting from here to new resetDefaultsToConsistentValues().

  • DumpRenderTree/mac/LayoutTestControllerMac.mm:

(LayoutTestController::overridePreference):

  • DumpRenderTree/win/DumpRenderTree.cpp:

(resetDefaultsToConsistentValues): new method, resets preferences to same set of value before every test.
(resetWebViewToConsistentStateBeforeTesting): move some preference setting from here to new resetDefaultsToConsistentValues().
(createWebViewAndOffscreenWindow): same.
(main): same.

  • DumpRenderTree/win/LayoutTestControllerWin.cpp:

(LayoutTestController::overridePreference):

  • DumpRenderTree/wx/LayoutTestControllerWx.cpp:

(LayoutTestController::overridePreference):

LayoutTests: Added support for overriding default preferences per-test.
Originally implemented by Glenn Wilson <gwilson@chromium.org>.
Reviewed by Adam Roben.
Added tests to verify that default preferences can be overridden per-test
and that overridden values are reset between each test.
https://bugs.webkit.org/show_bug.cgi?id=20534

  • fast/harness/override-preferences-2-expected.txt: Added.
  • fast/harness/override-preferences-2.html: Added. Verifies overridePreverence("WebKitDefaultFontSize", "24").
  • fast/harness/override-preferences-expected.txt: Added.
  • fast/harness/override-preferences.html: Added. Verifies overridePreference("WebKitJavaScriptEnabled", false).
  • fast/harness/override-zzz-reset-expected.txt: Added.
  • fast/harness/override-zzz-reset.html: Added. Because of 'zzz' this test will run after the tests above and

verify that override of preferences does not 'spill' to the subsequent tests in a batch.

  • platform/gtk/Skipped: Tests above disabled for GTK until GTK DRT implements preference override.
  • platform/qt/Skipped: same for QT.
6:11 PM Coding Style Guidelines edited by atwilson@chromium.org
(diff)
5:34 PM Changeset in webkit [47014] by mitz@apple.com
  • 2 edits in trunk/WebKit/win

Link drag images specify Lucida Grande on Windows and end up using Times
New Roman
<rdar://problem/7117835>

Reviewed by Darin Adler.

  • WebCoreSupport/WebDragClient.cpp:

(dragLabelFont): Added a RenderingMode parameter and changed to use the

same typeface as the Windows small caption font.

(WebDragClient::createDragImageForLink): Changed to respect the

page’s rendering mode.

5:15 PM Changeset in webkit [47013] by sfalken@apple.com
  • 2 edits in trunk/WebKitTools

Windows nightly fix.

Reviewed by Ada Chan.

  • FindSafari/FindSafari.cpp:

(_tmain): Copy Safari.dll if it exists.

4:47 PM Changeset in webkit [47012] by pkasting@chromium.org
  • 2 edits in trunk/WebKitLibraries

https://bugs.webkit.org/show_bug.cgi?id=27323
Even more line ending-stripping for auto-version.sh, based on output
provided by Jessie Berlin.

Reviewed by George Staikos.

  • win/tools/scripts/auto-version.sh:
4:29 PM Changeset in webkit [47011] by rik@webkit.org
  • 5 edits in trunk/WebCore

Inspector Resources / Graphs should support filtering
https://bugs.webkit.org/show_bug.cgi?id=20629

Patch by Anthony Ricaud <rik@webkit.org> on 2009-08-10
Reviewed by Timothy Hatcher.

Introduces a filter bar for resources.

Thanks to Matt Lilek for the CSS scope bar.

  • English.lproj/localizedStrings.js:
  • inspector/front-end/ResourcesPanel.js:

(WebInspector.ResourcesPanel.createFilterElement):
(WebInspector.ResourcesPanel):
(WebInspector.ResourcesPanel.prototype.toolbarItemClass.categoryOrder.filter):
(WebInspector.ResourcesPanel.prototype._updateFilter):
(WebInspector.ResourcesPanel.prototype._updateSummaryGraph):

  • inspector/front-end/inspector.css:
  • inspector/front-end/inspector.js:

(WebInspector.loaded):

4:17 PM Changeset in webkit [47010] by mrowe@apple.com
  • 2 edits in trunk/JavaScriptCore

Fix hundreds of "pointer being freed was not allocated" errors seen on the build bot.

Reviewed by Darin Adler.

  • wtf/FastMalloc.h: Implement nothrow variants of the delete and delete[] operators since

we implement the nothrow variants of new and new[]. The nothrow variant of delete is called
explicitly in the implementation of std::sort which was resulting in FastMalloc-allocated
memory being passed to the system allocator to free.

3:37 PM Changeset in webkit [47009] by bfulgham@webkit.org
  • 2 edits in trunk/WebCore

Unreviewed build correction after http://trac.webkit.org/changeset/46978

  • platform/network/curl/ResourceHandleManager.cpp:

(WebCore::ResourceHandleManager::initializeHandle): Correct

deprecated removeRef to new removeFragmentIdentifier.

3:27 PM Changeset in webkit [47008] by rniwa@webkit.org
  • 5 edits
    2 adds in trunk

WebCore: Background color formatting lost on enter
https://bugs.webkit.org/show_bug.cgi?id=20348

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-08-07
Reviewed by Justin Garcia.

This patch makes background color be preserved during editing.
In particular, editingStyleAtPosition would include the background color when obtaining styles.

  • editing/ApplyStyleCommand.cpp:

(WebCore::): Added CSSPropertyBackgroundColor to editingStyleProperties

LayoutTests: Background color formatting lost on enter
https://bugs.webkit.org/show_bug.cgi?id=20348

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-08-07
Reviewed by Justin Garcia.

This patch adds a test to make sure WebKit preserves background color when inserting a new paragraph.

  • editing/inserting/5994480-2-expected.txt: Because input element has explicit white background and the enclosing div has

transparent background, WebKit adds white background color in DeleteSelectionCommand::calculateTypingStyleAfterDelete upon deletion.

  • editing/style/highlight-insert-paragraph-expected.txt: Added.
  • editing/style/highlight-insert-paragraph.html: Added.
  • platform/mac/editing/pasteboard/select-element-1-expected.txt: Because select element has explicit white background color

and div has transparent background, WebKit adds white background color in calculateTypingStyleAfterDelete upon cut.

3:14 PM Changeset in webkit [47007] by Nikolas Zimmermann
  • 3 edits in trunk/LayoutTests

Rubber-stamped by George Staikos.
Forgot to update pixel test result in my last commit.

3:13 PM Changeset in webkit [47006] by levin@chromium.org
  • 3 edits in trunk/WebKitTools

2009-08-10 David Levin <levin@chromium.org>

Reviewed by David Kilzer.

check-webkit-style doesn't catch braces around single line statements when followed by else.
https://bugs.webkit.org/show_bug.cgi?id=28164

  • Scripts/modules/cpp_style.py: Added else to the regex that finds ending }'s when checking the no braces for "single lines" rule.
  • Scripts/modules/cpp_style_unittest.py: Added test case.
2:53 PM Changeset in webkit [47005] by ddkilzer@apple.com
  • 3 edits in trunk/WebKitTools

<http://webkit.org/b/28163> bugzilla-tool: scm module should not import bugzilla module

Reviewed by David Levin.

  • Scripts/bugzilla-tool: (parse_bug_id): Added. Moved from CommitMessage.parse_bug_id() in scm module.
  • Scripts/modules/scm.py: Removed import of bugzilla module. (CommitMessage.parse_bug_id): Deleted.
2:17 PM Changeset in webkit [47004] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

Stop sending QEvent::CursorChange to Qt for mouse move.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-08-10
Reviewed by Simon Hausmann.

Verify if the cursor didn't change in WidgetQt::setCursor,
just like the mac port does.

Thanks to Adriano Rezendo for spotting and debugging the
issue.

  • platform/qt/WidgetQt.cpp:

(WebCore::Widget::setCursor):

2:04 PM Changeset in webkit [47003] by jmalonzo@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-10 Jan Michael Alonzo <jmalonzo@webkit.org>

[Gtk] Unreviewed build fix. Move JSAPIValueWrapper.cpp/.h in the debug
section. This file is already part of AllInOneFile in Release builds.

  • GNUmakefile.am:
1:58 PM BuildingOnWindows edited by jberlin@apple.com
Add information about not being able to find debugging symbols (diff)
1:51 PM Changeset in webkit [47002] by ap@apple.com
  • 2 edits in trunk/WebCore

Reviewed by Brady Eidson.

https://bugs.webkit.org/show_bug.cgi?id=28165
Application cache maximum size cannot be set before the database is opened

  • loader/appcache/ApplicationCacheStorage.cpp: (WebCore::ApplicationCacheStorage::vacuumDatabaseFile): Open appcache database if it isn't open yet.
1:37 PM Changeset in webkit [47001] by Dimitri Glazkov
  • 11 edits in trunk/WebCore

2009-08-10 Vitaly Repeshko <vitalyr@quad.spb.corp.google.com>

Reviewed by Dimitri Glazkov.

[V8] Fix memory leak in node event listeners. See http://crbug.com/17400.
https://bugs.webkit.org/show_bug.cgi?id=28156

  • bindings/scripts/CodeGeneratorV8.pm: Custom event handler accessors now apply to all Node descendants.
  • bindings/v8/V8AbstractEventListener.h: (WebCore::V8AbstractEventListener::isObjectListener): Added virtual method to detect listeners that have weak references to JS objects.
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate):
  • bindings/v8/V8ObjectEventListener.cpp:
  • bindings/v8/V8ObjectEventListener.h: (WebCore::V8ObjectEventListener::isObjectListener):
  • bindings/v8/V8Utilities.cpp: (WebCore::removeHiddenDependency):
  • bindings/v8/custom/V8CustomBinding.h:
  • bindings/v8/custom/V8ElementCustom.cpp:
  • bindings/v8/custom/V8NodeCustom.cpp: Moved custom accessors from V8ElementCustom.cpp. (WebCore::toEventType): (WebCore::getEventListener): (WebCore::ACCESSOR_SETTER): (WebCore::ACCESSOR_GETTER): (WebCore::CALLBACK_FUNC_DECL):
  • bindings/v8/custom/V8XMLHttpRequestCustom.cpp: (WebCore::getEventListener): Made static.
12:46 PM Changeset in webkit [47000] by Darin Adler
  • 2 edits in trunk/JavaScriptCore
  • wtf/FastMalloc.h: Fix build.
12:42 PM Changeset in webkit [46999] by Darin Adler
  • 3 edits in trunk/JavaScriptCore

FastMalloc.h has cross-platform code but marked as WinCE-only
https://bugs.webkit.org/show_bug.cgi?id=28160

Patch by Darin Adler <Darin Adler> on 2009-08-10
Reviewed by Mark Rowe.

1) The support for nothrow was inside #if PLATFORM(WINCE) even though it is

not platform-specific.

2) The code tried to override operator delete nothrow, which does not exist.
3) The code in the header checks the value of USE_SYSTEM_MALLOC, but the code

in FastMalloc.cpp checks only if the macro is defined.

  • wtf/FastMalloc.h: See above.
  • wtf/FastMalloc.cpp: Ditto.
12:37 PM Changeset in webkit [46998] by jorlow@chromium.org
  • 1 edit in trunk/WebCore/ChangeLog

Fix date in ChangeLog

12:34 PM Changeset in webkit [46997] by jorlow@chromium.org
  • 5 edits in trunk/WebCore

2009-08-08 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Darin Fisher.

Cleanup DOM Storage namespace shutdown code + usage of security origin.
https://bugs.webkit.org/show_bug.cgi?id=28094

If a storage namespace is not explicitly closed, be sure to do so on
destruction of the object. In addition, the close call should wait
on the background thread finishing its syncing. (Not doing so is actually
a regression from the original LocalStorage code.)

There's no point to passing in the SecurityOrigin when copying a storage
area since what was passed in is exactly what is stored within each
storage area. In addition, the non-copy constructor should take in a
PassRefPtr rather than a pointer since that pointer was only passed into
the constuctor for RefPtr's anyway.

  • storage/StorageAreaImpl.cpp: (WebCore::StorageAreaImpl::StorageAreaImpl): (WebCore::StorageAreaImpl::copy):
  • storage/StorageAreaImpl.h:
  • storage/StorageNamespaceImpl.cpp: (WebCore::StorageNamespaceImpl::StorageNamespaceImpl): (WebCore::StorageNamespaceImpl::~StorageNamespaceImpl): (WebCore::StorageNamespaceImpl::copy): (WebCore::StorageNamespaceImpl::close):
  • storage/StorageNamespaceImpl.h:
12:30 PM Changeset in webkit [46996] by Nikolas Zimmermann
  • 4 edits in trunk/WebCore

Not reviewed, build fix.
Rename ref() to fragmentIdentifier() in the wml/ subdirectory.

12:15 PM Changeset in webkit [46995] by sfalken@apple.com
  • 2 edits in trunk/WebCore

Windows build fix.

Exclude JSMedia.cpp from build since it is included in DerivedSources.cpp.
Fixes linker warnings.

Corrected spelling of JSInspectorBackend.cpp.

Other changes related to actually using a Visual Studio to open/save the vcproj file.

  • WebCore.vcproj/WebCore.vcproj:
11:39 AM Changeset in webkit [46994] by weinig@apple.com
  • 2 edits in trunk/WebCore

Remove a dead #define.

Reviewed by Anders Carlsson.

  • bindings/js/JSDOMBinding.cpp:
11:36 AM Changeset in webkit [46993] by Dimitri Glazkov
  • 3 edits in trunk/WebCore

2009-08-10 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

[Chromium] Update KURLGoogle for ref* method renames in
http://trac.webkit.org/changeset/46978.
https://bugs.webkit.org/show_bug.cgi?id=28084

  • bindings/v8/custom/V8LocationCustom.cpp: (WebCore::ACCESSOR_SETTER): Renamed ref and setRef to fragmentIdentifier and

setFragmentIdentifier, respectively.

  • platform/KURLGoogle.cpp: Ditto, plus all other ref* method names.
11:30 AM Changeset in webkit [46992] by weinig@apple.com
  • 2 edits in trunk/JavaScriptCore

Fix an annoying indentation issue.

Reviewed by Anders Carlsson.

  • runtime/DateConstructor.cpp:

(JSC::constructDate):

11:26 AM Changeset in webkit [46991] by jmalonzo@webkit.org
  • 2 edits in trunk/WebCore

[Gtk] Unreviewed. Fix the nit as suggested by Darin in bug #28144 for r46989.

11:22 AM Changeset in webkit [46990] by bfulgham@webkit.org
  • 4 edits
    1 add in trunk

WebKit/win: https://bugs.webkit.org/show_bug.cgi?id=28048.
Move various WinCairo build settings into *.vsprops file.

Reviewed by Adam Roben.

  • WebKit.vcproj/WebKit.vcproj: Add new cURL.vsprops property sheet (which contains link settings for cURL and its dependencies), and remove the corresponding entries from the vcproj file. Remove png, jpeg, cairo library linking flags from vcproj, now that they are in the WinCairo.vsprops file.

WebKitLibraries: https://bugs.webkit.org/show_bug.cgi?id=28048.
Move various WinCairo build settings into *.vsprops file.

Reviewed by Adam Roben.

  • win/tools/vsprops/cURL.vsprops: Added.
11:20 AM Changeset in webkit [46989] by jmalonzo@webkit.org
  • 2 edits in trunk/WebCore

2009-08-10 John Kjellberg <john.kjellberg@power.alstom.com>

Reviewed by Darin Adler.

Memory leak in error handling code for SOUP interface
https://bugs.webkit.org/show_bug.cgi?id=28144

Freed memory leaked in error handling code for soup requests.

  • platform/network/soup/ResourceHandleSoup.cpp: (WebCore::finishedCallback): (WebCore::):
11:18 AM Changeset in webkit [46988] by pkasting@chromium.org
  • 2 edits in trunk/WebKitTools

https://bugs.webkit.org/show_bug.cgi?id=18599
Work around apparent bug in abs2rel() with symlinked directories.

Reviewed by David Kilzer.

  • Scripts/resolve-ChangeLogs:
11:10 AM Changeset in webkit [46987] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2009-08-10 Dimitri Glazkov <Dimitri Glazkov>

Reviewed by Darin Adler.

Provide graceful handling of the situation when an embedder invokes
visited link update methods before a Page constructor has been called.
https://bugs.webkit.org/show_bug.cgi?id=28149

  • page/Page.cpp: (WebCore::Page::allVisitedStateChanged): Added a null-check for allPages. (WebCore::Page::visitedStateChanged): Ditto.
11:03 AM Changeset in webkit [46986] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-08-10 Xan Lopez <xlopez@igalia.com>

Fix GTK+ build.

  • platform/network/soup/ResourceHandleSoup.cpp: (WebCore::):
10:56 AM Changeset in webkit [46985] by mitz@apple.com
  • 3 edits in trunk/WebCore

Remove the redundant, Windows CE-only GlyphBuffer::advnaceData()

Reviewed by Anders Carlsson.

  • platform/graphics/GlyphBuffer.h: Removed advanceData().
  • platform/graphics/wince/GraphicsContextWince.cpp:

(WebCore::GraphicsContext::drawText): Changed to use advances(from)

instead of advanceData() + from.

10:37 AM Changeset in webkit [46984] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-08-10 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

Remove workaround for older libsoup versions.

  • platform/network/soup/ResourceHandleSoup.cpp: (WebCore::fillResponseFromMessage):
10:37 AM Changeset in webkit [46983] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-08-10 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

Let startHttp validate the URI instead of doing it, with potential
crashes, in ::start.

  • platform/network/soup/ResourceHandleSoup.cpp: (WebCore::startHttp):
10:28 AM Changeset in webkit [46982] by Darin Adler
  • 6 edits
    3 adds in trunk

WebCore: When empty <input> is focused, getSelection() exposes shadow tree nodes
https://bugs.webkit.org/show_bug.cgi?id=15903

Patch by Darin Adler <Darin Adler> on 2009-08-10
Reviewed by Dan Bernstein.

Test: fast/forms/shadow-tree-exposure.html

  • page/DOMSelection.cpp:

(WebCore::selectionShadowAncestor): Added.
(WebCore::DOMSelection::anchorNode): Return the parent of the shadow
ancestor if the selection is in a shadow tree. Matches Firefox.
(WebCore::DOMSelection::anchorOffset): Return the index of the shadow
ancestor if the selection is in a shadow tree. Matches Firefox.
(WebCore::DOMSelection::focusNode): Ditto.
(WebCore::DOMSelection::focusOffset): Ditto.
(WebCore::DOMSelection::baseNode): More of the same, but since this
is a WebKit-invented property, it does not match Firefox.
(WebCore::DOMSelection::baseOffset): Ditto.
(WebCore::DOMSelection::extentNode): Ditto.
(WebCore::DOMSelection::extentOffset): Ditto.
(WebCore::DOMSelection::isCollapsed): Return true even when the
selection is a range, if the selection is in the shadow tree.
Matches Firefox. Also changed behavior when called and the frame
is gone to return true instead of false which makes more sense.
(WebCore::DOMSelection::empty): Use clear() just like the other
selection-clearing function in this file does.
(WebCore::DOMSelection::getRangeAt): Return a range that starts
and ends before the shadow ancestor. Matches Firefox.

LayoutTests: When empty <input> is focused, getSelection() exposes shadow tree nodes
https://bugs.webkit.org/show_bug.cgi?id=15903

Patch by Darin Adler <Darin Adler> on 2009-08-07
Reviewed by Dan Bernstein.

Also updated a recently introduced non-standard test to work in a more
standard way.

  • fast/forms/resources/shadow-tree-exposure.js: Added.
  • fast/forms/shadow-tree-exposure-expected.txt: Added.
  • fast/forms/shadow-tree-exposure.html: Added.
  • fast/js/method-check-expected.txt: Updated.
  • fast/js/method-check.html: Regenerated using the make-js-test-wrappers script.
  • fast/js/resources/method-check.js: Replaced a comment talking about the lack

of a gc() function with a call to GCController.collect(). Also removed the
shouldBeTrue call at the end of the fiel that wasn't needed.

10:25 AM Changeset in webkit [46981] by mitz@apple.com
  • 2 edits in trunk/WebKitTools

Build fix

  • DumpRenderTree/TestNetscapePlugIn.subproj/main.cpp:

(NPP_New):

10:07 AM Changeset in webkit [46980] by beidson@apple.com
  • 2 edits in trunk/WebKit/gtk

GTK build fix.

  • webkit/webkitdownload.cpp:

(webkit_download_get_suggested_filename):

9:53 AM Changeset in webkit [46979] by pfeldman@chromium.org
  • 10 edits
    15 adds
    12 deletes in trunk/WebCore

2009-08-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Make all status bar button images glyph-based.

https://bugs.webkit.org/show_bug.cgi?id=28124

  • inspector/front-end/ElementsPanel.js: (WebInspector.ElementsPanel):
  • inspector/front-end/Images/clearConsoleButtonGlyph.png: Added.
  • inspector/front-end/Images/clearConsoleButtons.png: Removed.
  • inspector/front-end/Images/consoleButtonGlyph.png: Added.
  • inspector/front-end/Images/consoleButtons.png: Removed.
  • inspector/front-end/Images/dockButtonGlyph.png: Added.
  • inspector/front-end/Images/dockButtons.png: Removed.
  • inspector/front-end/Images/enableButtons.png: Removed.
  • inspector/front-end/Images/enableOutlineButtonGlyph.png: Added.
  • inspector/front-end/Images/enableSolidButtonGlyph.png: Added.
  • inspector/front-end/Images/excludeButtonGlyph.png: Added.
  • inspector/front-end/Images/excludeButtons.png: Removed.
  • inspector/front-end/Images/focusButtonGlyph.png: Added.
  • inspector/front-end/Images/focusButtons.png: Removed.
  • inspector/front-end/Images/largerResourcesButtonGlyph.png: Added.
  • inspector/front-end/Images/largerResourcesButtons.png: Removed.
  • inspector/front-end/Images/nodeSearchButtonGlyph.png: Added.
  • inspector/front-end/Images/nodeSearchButtons.png: Removed.
  • inspector/front-end/Images/pauseOnExceptionButtonGlyph.png: Added.
  • inspector/front-end/Images/pauseOnExceptionButtons.png: Removed.
  • inspector/front-end/Images/percentButtonGlyph.png: Added.
  • inspector/front-end/Images/percentButtons.png: Removed.
  • inspector/front-end/Images/recordButtonGlyph.png: Added.
  • inspector/front-end/Images/recordButtons.png: Removed.
  • inspector/front-end/Images/recordToggledButtonGlyph.png: Added.
  • inspector/front-end/Images/reloadButtonGlyph.png: Added.
  • inspector/front-end/Images/reloadButtons.png: Removed.
  • inspector/front-end/Images/undockButtonGlyph.png: Added.
  • inspector/front-end/Panel.js: (WebInspector.Panel.prototype.createStatusBarButton):
  • inspector/front-end/ProfilesPanel.js: (WebInspector.ProfilesPanel):
  • inspector/front-end/ResourcesPanel.js: (WebInspector.ResourcesPanel):
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel):
  • inspector/front-end/WebKit.qrc:
  • inspector/front-end/inspector.css:
  • inspector/front-end/inspector.html:
9:49 AM QtWebKitFeatures46 edited by kenneth@webkit.org
(diff)
9:35 AM Changeset in webkit [46978] by beidson@apple.com
  • 19 edits in trunk/WebCore

KURL ref() methods should be fragmentIdentifier() methods
https://bugs.webkit.org/show_bug.cgi?id=28084

Reviewed by Darin Adler.

In addition to the renames, also took the chance to replace all use of:
setRef(String());
with:
removeFragmentIdentifier();

  • platform/KURL.cpp:

(WebCore::KURL::fragmentIdentifier):
(WebCore::KURL::hasFragmentIdentifier):
(WebCore::KURL::setFragmentIdentifier):
(WebCore::KURL::removeFragmentIdentifier):
(WebCore::KURL::prettyURL):
(WebCore::equalIgnoringFragmentIdentifier):

  • platform/KURL.h:
  • bindings/js/JSLocationCustom.cpp:

(WebCore::JSLocation::setHash): Rework with new names, and to use the new equalIgnoringNullity() method.

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::internalLinkElement):

  • css/CSSCursorImageValue.cpp:

(WebCore::isSVGCursorIdentifier):

  • history/HistoryItem.cpp:

(WebCore::HistoryItem::isCurrentDocument):

  • html/HTMLAnchorElement.cpp:

(WebCore::HTMLAnchorElement::hash):

  • html/HTMLFrameElementBase.cpp:

(WebCore::HTMLFrameElementBase::isURLAllowed):

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::begin):
(WebCore::FrameLoader::gotoAnchor):
(WebCore::FrameLoader::scheduleLocationChange):
(WebCore::FrameLoader::scrollToAnchor):
(WebCore::FrameLoader::shouldReload):
(WebCore::FrameLoader::urlsMatchItem):

  • loader/appcache/ApplicationCache.cpp:

(WebCore::ApplicationCache::resourceForURL):
(WebCore::ApplicationCache::resourceForRequest):

  • loader/appcache/ApplicationCacheGroup.cpp:

(WebCore::ApplicationCacheGroup::cacheForMainRequest):
(WebCore::ApplicationCacheGroup::fallbackCacheForMainRequest):
(WebCore::ApplicationCacheGroup::selectCache):
(WebCore::ApplicationCacheGroup::finishedLoadingMainResource):
(WebCore::ApplicationCacheGroup::didReceiveResponse):
(WebCore::ApplicationCacheGroup::didFail):
(WebCore::ApplicationCacheGroup::addEntry):

  • loader/appcache/ApplicationCacheResource.h:

(WebCore::ApplicationCacheResource::create):

  • loader/appcache/ApplicationCacheStorage.cpp:

(WebCore::ApplicationCacheStorage::findOrCreateCacheGroup):
(WebCore::ApplicationCacheStorage::cacheGroupForURL):
(WebCore::ApplicationCacheStorage::fallbackCacheGroupForURL):

  • loader/appcache/ManifestParser.cpp:

(WebCore::parseManifest):

  • page/Location.cpp:

(WebCore::Location::hash):

  • page/Page.cpp:

(WebCore::Page::goToItem):

  • rendering/RenderPartObject.cpp:

(WebCore::isURLAllowed):

  • workers/WorkerLocation.cpp:

(WebCore::WorkerLocation::hash):

9:19 AM Changeset in webkit [46977] by vestbo@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-10 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Adam Roben.

Allow TestNetscapePlugIn to fall back to the Carbon event model

Previously the plugin would fail, even though the Carbon event
model was available. The only way to get the Carbon event model
was to pass the "forcecarbon" argument in the test markup, but no
tests were using this.

Now the plugin uses carbon either if it's forced, or if the Cocoa
event model is not supported.

This helps us run most of the plugin tests on Qt/Mac, which still
does not support the Cocoa event model.

8:45 AM QtWebKitFeatures46 edited by ben@webkit.org
(diff)
8:27 AM QtWebKitJournal edited by Simon Hausmann
(diff)
8:26 AM QtWebKitFeatures46 created by Simon Hausmann
8:22 AM QtWebKitJournal edited by Simon Hausmann
(diff)
8:07 AM QtWebKitJournal edited by Simon Hausmann
(diff)
7:51 AM QtWebKitJournal edited by Simon Hausmann
(diff)
7:47 AM Changeset in webkit [46976] by xan@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-10 Xan Lopez <xlopez@igalia.com>

Unreviewed build fix.

Add new files to makefile.

  • GNUmakefile.am:
7:29 AM QtWebKitJournal edited by Simon Hausmann
(diff)
7:10 AM QtWebKitJournal edited by Simon Hausmann
(diff)
7:05 AM QtWebKitJournal edited by Simon Hausmann
(diff)
6:39 AM Changeset in webkit [46975] by Simon Hausmann
  • 3 edits in trunk/WebKit/qt

Various fixes and improvements to the QWebPluginInfo, QWebPluginDatabase and QWebSettings documentation.

Patch by Kavindra Palaraja <kavindra.palaraja@nokia.com> on 2009-08-10
Reviewed by Simon Hausmann.

  • Api/qwebplugindatabase.cpp:
  • Api/qwebsettings.cpp:
5:49 AM QtWebKit edited by Simon Hausmann
(diff)
5:46 AM Changeset in webkit [46974] by Simon Hausmann
  • 12 edits
    1 add in trunk/LayoutTests

[Qt] After ChangeSet https://bugs.webkit.org/show_bug.cgi?id=26896, the
DefaultFontSize was changed in Qt DumpRenderTree from 14 to 13. But
any expected file works with the original value. The expected values
should follow this change. This patch contains the new expected
files.

Patch by Renata Hodovan <hodovan.renata@stud.u-szeged.hu> on 2009-08-10
Reviewed by Simon Hausmann.

  • platform/qt/Skipped:
  • platform/qt/editing/pasteboard/5780697-2-expected.txt: Add.
  • platform/qt/fast/block/float/009-expected.txt:
  • platform/qt/fast/dynamic/genContentDestroyChildren-expected.txt:
  • platform/qt/fast/images/image-in-map-expected.txt:
  • platform/qt/fast/inline-block/004-expected.txt:
  • platform/qt/fast/inline-block/005-expected.txt:
  • platform/qt/fast/table/giantRowspan-expected.txt:
  • platform/qt/fast/table/height-percent-test-expected.txt:
  • platform/qt/tables/mozilla/bugs/adforce_imgis_com-expected.txt:
  • platform/qt/tables/mozilla/bugs/bug50695-2-expected.txt:
  • platform/qt/tables/mozilla_expected_failures/bugs/bug23847-expected.txt:
5:36 AM Changeset in webkit [46973] by Simon Hausmann
  • 2 edits in trunk/WebCore

Fix endless loop when the application is quit by a secondary thread during a sync xml http request.

Patch by Andy Shaw <andy.shaw@nokia.com> on 2009-08-10
Reviewed by Simon Hausmann.

Replace the m_finished state variable and the call to QCoreApplication::processEvents() with
a dedicated QEventLoop, that can be terminated from within WebCoreSynchronousLoader as well
as from Qt through its knowledge of globally registered eventloops.

There is no unit test for this bug as it would require exporting additional symbols and
introduce the synchronous loader as global variable just for the test.

  • platform/network/qt/ResourceHandleQt.cpp:

(WebCore::WebCoreSynchronousLoader::didFinishLoading):
(WebCore::WebCoreSynchronousLoader::didFail):
(WebCore::WebCoreSynchronousLoader::waitForCompletion):

3:02 AM Changeset in webkit [46972] by pfeldman@chromium.org
  • 10 edits
    1 add in trunk/WebCore

2009-08-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

WebInspector: Make properties inspection serialized.

https://bugs.webkit.org/show_bug.cgi?id=28078

  • inspector/front-end/DOMAgent.js: (InspectorController.getPrototypes): (InspectorController.getProperties):
  • inspector/front-end/InjectedScript.js: (InjectedScript.getPrototypes): (InjectedScript.getProperties): (InjectedScript.setPropertyValue):
  • inspector/front-end/ObjectPropertiesSection.js: (WebInspector.ObjectRef): (WebInspector.ObjectPropertyRef): (WebInspector.ObjectPropertiesSection.prototype.onpopulate): (WebInspector.ObjectPropertiesSection.prototype.update.callback): (WebInspector.ObjectPropertiesSection.prototype.update): (WebInspector.ObjectPropertiesSection.prototype._update): (WebInspector.ObjectPropertyTreeElement.prototype.onpopulate.callback): (WebInspector.ObjectPropertyTreeElement.prototype.onpopulate): (WebInspector.ObjectPropertyTreeElement.prototype.onattach): (WebInspector.ObjectPropertyTreeElement.prototype.update):
  • inspector/front-end/PropertiesSidebarPane.js: (WebInspector.PropertiesSidebarPane.prototype.update.callback): (WebInspector.PropertiesSidebarPane.prototype.update):
1:16 AM Changeset in webkit [46971] by Simon Hausmann
  • 2 edits in trunk/LayoutTests

Rebase Qt letter spacing DRT result after r46936

Reviewed by Ariya Hidayat.

  • platform/qt/fast/text/text-letter-spacing-expected.txt:
1:01 AM Changeset in webkit [46970] by Simon Hausmann
  • 2 edits in trunk/JavaScriptCore

Fix compilation with the interpreter instead of the JIT by including
PrototypeFunction.h as forward-declared through NativeFunctionWrapper.h.

Patch by Simon Hausmann <simon.hausmann@nokia.com> on 2009-08-10

  • runtime/ObjectConstructor.cpp:

Aug 9, 2009:

5:47 PM Changeset in webkit [46969] by Nikolas Zimmermann
  • 5 edits in trunk

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] wml.css lacks <pre> support
https://bugs.webkit.org/show_bug.cgi?id=28135

Add <pre> support for WML.
Extended fast/wml/text-emphasis.wml to cover <pre> support.

  • css/wml.css: <pre> section copied from html.css

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] wml.css lacks <pre> support
https://bugs.webkit.org/show_bug.cgi?id=28135

Extend fast/wml/text-emphasis.wml test, to cover <pre> formatting.

  • fast/wml/text-emphasis.wml:
  • platform/mac/fast/wml/text-emphasis-expected.txt:
5:01 PM Changeset in webkit [46968] by Nikolas Zimmermann
  • 2 edits in trunk/WebCore

Reviewed by Oliver Hunt.
Fix non-WML enabled builds. Remove ENABLE(WML) guards around the switch cases 'FrameLoadTypeBackWMLDeckNotAccessible'.

4:55 PM Changeset in webkit [46967] by oliver@apple.com
  • 4 edits
    3 adds in trunk

JSON.stringify replacer returning undefined does not omit object properties
https://bugs.webkit.org/show_bug.cgi?id=28118

Reviewed by George Staikos

Correct behaviour of stringify when using a replacer function that returns
undefined. This is a simple change to move the undefined value check to
after the replacer function is called. This means that the replacer function
is now called for properties with the value undefined, however i've confirmed
that this behaviour is correct.

In addition I've made the cyclic object exception have a more useful error
message.

4:33 PM Changeset in webkit [46966] by Nikolas Zimmermann
  • 21 edits
    24 adds in trunk

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] Deck access control is completly broken
https://bugs.webkit.org/show_bug.cgi?id=27721

Synchronize WebFrameLoadType with FrameLoadType enum. Append 'WebFrameLoadTypeBackWMLDeckNotAccessible'.

  • WebView/WebFramePrivate.h:

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] Deck access control is completly broken
https://bugs.webkit.org/show_bug.cgi?id=27721

Synchronize WebFrameLoadType with FrameLoadType enum. Append 'WebFrameLoadTypeBackWMLDeckNotAccessible'.

  • Interfaces/IWebFramePrivate.idl:

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] Deck access control is completly broken
https://bugs.webkit.org/show_bug.cgi?id=27721

Rewrite WML deck access control, actually works now standalone and within LayoutTests.
Added several new layout tests covering local & remote access support.

In order to detect WML task execution failures from within the WML layout tests a new
method Console::lastWMLErrorMessage() has been added, returning the last recorded WML
error message. Change WML log messages to dump to InspectorClient, just like it's done
for HTML - we used to behave like JSMessageSource, which is wrong.

Tests: http/tests/wml/access-target-domain-deny.html

http/tests/wml/access-target-path-deny.html
http/tests/wml/access-target.html
wml/access-target-deny.html
wml/access-target.html

  • inspector/ConsoleMessage.h: (WebCore::ConsoleMessage::source): Added public method to fetch the message source type. (WebCore::ConsoleMessage::message): Added public method to access the error message.
  • inspector/InspectorController.h: (WebCore::InspectorController::consoleMessages): Added for Console::lastWMLErrorMessage(), to be able to crawl through the list of dumped messages.
  • loader/FrameLoader.cpp: Recognize 'FrameLoadTypeBackWMLDeckNotAccessible', all guarded by ENABLE(WML). (WebCore::isBackForwardLoadType): (WebCore::FrameLoader::restoreDocumentState): (WebCore::FrameLoader::transitionToCommitted): (WebCore::FrameLoader::loadItem):
  • loader/FrameLoaderTypes.h: (WebCore::): Add new load type 'FrameLoadTypeBackWMLDeckNotAccessible', behaving like FrameLoadTypeBack plus WML specific information.
  • page/Console.cpp: (WebCore::Console::addMessage): Don't dump WML messages using ChromeClient (as it's done for JSMessageSource), but go through InspectorController. (WebCore::Console::lastWMLErrorMessage): Crawl InspectorController::consoleMssages() to look up the last WML error message, if present.
  • page/Console.h:
  • page/Console.idl: Add "[DontEnum] DOMString lastWMLErrorMessage()".
  • wml/WMLAccessElement.cpp: Store path/domain as member variables and pass them to WMLPageState once in insertedIntoDocument(). (WebCore::WMLAccessElement::parseMappedAttribute): (WebCore::WMLAccessElement::insertedIntoDocument):
  • wml/WMLAccessElement.h:
  • wml/WMLCardElement.cpp: Handle 'FrameLoadTypeBackWMLDeckNotAccessible' as deck-entry method. Report WML error in current deck. (WebCore::WMLCardElement::handleIntrinsicEventIfNeeded):
  • wml/WMLDocument.cpp: Enable deck-access control in finishedParsing. Take care of method name changes in WMLPageState. (WebCore::WMLDocument::finishedParsing): (WebCore::WMLDocument::initialize):
  • wml/WMLErrorHandling.cpp: Never report 'WMLErrorDeckNotAccessible' messages through the tokenizer, that would alter the page into error state. (WebCore::reportWMLError):
  • wml/WMLPageState.cpp: Completly rewrote access path/domain parsing & deck-access control, strictly following the WML spec. All covered by tests. (WebCore::WMLPageState::WMLPageState): (WebCore::normalizedHostName): (WebCore::hostFromURL): (WebCore::urlForHistoryItem): (WebCore::tryAccessHistoryURLs): (WebCore::WMLPageState::processAccessControlData): (WebCore::WMLPageState::resetAccessControlData): (WebCore::WMLPageState::canAccessDeck): (WebCore::WMLPageState::hostIsAllowedToAccess): (WebCore::WMLPageState::pathIsAllowedToAccess):
  • wml/WMLPageState.h:

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] Deck access control is completly broken
https://bugs.webkit.org/show_bug.cgi?id=27721

Add several new tests covering WML deck-access control.
Update variable-reference-invalid-character.html result, as console messages aren't dump to the results anymore.

  • http/tests/wml/access-target-domain-deny-expected.txt: Added.
  • http/tests/wml/access-target-domain-deny.html: Added.
  • http/tests/wml/access-target-expected.txt: Added.
  • http/tests/wml/access-target-path-deny-expected.txt: Added.
  • http/tests/wml/access-target-path-deny.html: Added.
  • http/tests/wml/access-target.html: Added.
  • http/tests/wml/resources/access-target-domain-deny.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • http/tests/wml/resources/access-target-domain-deny.wml: Added.
  • http/tests/wml/resources/access-target-path-deny.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • http/tests/wml/resources/access-target-path-deny.wml: Added.
  • http/tests/wml/resources/access-target.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • http/tests/wml/resources/access-target.wml: Added.
  • http/tests/wml/resources/locked-deck.wml: Added.
  • http/tests/wml/resources/unreachable-domain.wml: Added.
  • http/tests/wml/resources/unreachable-path.wml: Added.
  • wml/access-target-deny-expected.txt: Added.
  • wml/access-target-deny.html: Added.
  • wml/access-target-expected.txt: Added.
  • wml/access-target.html: Added.
  • wml/resources/access-target-deny.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • wml/resources/access-target-deny.wml: Added.
  • wml/resources/access-target.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • wml/resources/access-target.wml: Added.
  • wml/resources/locked-deck.wml: Added.
  • wml/variable-reference-invalid-character-expected.txt:
3:04 PM Changeset in webkit [46965] by Nikolas Zimmermann
  • 4 edits in trunk

Reviewed by Maciej Stachowiak.

FrameLoadType / WebFrameLoadType enums are out of sync
https://bugs.webkit.org/show_bug.cgi?id=28132

Modify FrameLoadType enum to match the order of WebFrameLoadType in WebFramePrivate.h / IWebFramePrivate.idl.
The mac/win public API exposes the old 'FrameLoadTypeReloadAllowingStaleData' value, which resides between
'FrameLoadTypeReload' and 'FrameLoadTypeSame'. Work around this problem, by assigning 'FrameLoadTypeSame = FrameLoadTypeReload + 2'.

Add a note to the FrameLoadType enum to keep it in sync with WebFramePrivate.h & IWebFramePrivate.idl.

8:41 AM Changeset in webkit [46964] by Nikolas Zimmermann
  • 25 edits in trunk

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] manual-tests/wml/onevent-prev.wml crashes
https://bugs.webkit.org/show_bug.cgi?id=27722

Fix intrinsic event execution, leading to crashes.
First finish parsing the document, then execute the events. Not the other way round.
-> exposes bugs all over the WML layout tests, which relied on that quirk.

Fixes manual-tests/wml/onevent-prev.wml.

  • wml/WMLDocument.cpp: (WebCore::WMLDocument::WMLDocument): (WebCore::WMLDocument::finishedParsing): (WebCore::WMLDocument::initialize):
  • wml/WMLDocument.h:

2009-08-09 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

[WML] manual-tests/wml/onevent-prev.wml crashes
https://bugs.webkit.org/show_bug.cgi?id=27722

Reorganize WML testcases to not rely on the fact that intrinsic events are executed before the document finished parsing.

  • http/tests/wml/resources/go-task-get-method-accept-charset.js:
  • http/tests/wml/resources/go-task-get-method.js:
  • http/tests/wml/resources/go-task-post-method-accept-charset.js:
  • http/tests/wml/resources/go-task-post-method.js:
  • http/tests/wml/resources/post-data-to-server.js:
  • wml/resources/WMLTestCase.js: (onloadHandler): (createStaticWMLTestCase): (createDynamicWMLTestCase): (createWMLTestCase.iframeElement.onload): (createWMLTestCase):
  • wml/resources/enter-card-with-events.js: (executeTest):
  • wml/resources/enter-first-card-with-events.js: (setupTestDocument): (prepareTest): (executeTest):
  • wml/resources/go-task-get-method-external-deck-with-href.js:
  • wml/resources/go-task-get-method-external-deck.js:
  • wml/resources/go-task-get-method-same-deck.js:
  • wml/resources/input-format.js:
  • wml/resources/newcontext-same-deck.js: (setupTestDocument): (executeTest):
  • wml/resources/onenterforward-event.js: (executeTest):
  • wml/resources/onenterforward-inline-event.js:
  • wml/resources/ontimer-event.js: (executeTest):
  • wml/resources/ontimer-inline-event.js: (executeTest):
  • wml/resources/option-element-onpick.js:
  • wml/resources/select-element-variables.js:
  • wml/resources/variable-reference-invalid-character.js:
  • wml/resources/variable-reference-valid.js:
3:33 AM Changeset in webkit [46963] by oliver@apple.com
  • 8 edits in trunk

[ES5] Implement Object.getPrototypeOf
https://bugs.webkit.org/show_bug.cgi?id=28114

Reviewed by Eric Seidel and Sam Weinig.

Implement getPrototypeOf

Aug 8, 2009:

6:36 PM Changeset in webkit [46962] by Dimitri Glazkov
  • 1 edit in trunk/WebCore/ChangeLog

Updated ChangeLog with bug URLs.

6:31 PM Changeset in webkit [46961] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2009-08-08 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

[Chromium] Update WebCore.gyp to add ImageBuffer.cpp,
introduced by http://trac.webkit.org/changeset/46956.

  • WebCore.gypi: Added ImageBuffer.cpp.
4:28 PM Changeset in webkit [46960] by jmalonzo@webkit.org
  • 2 edits in trunk/LayoutTests

2009-08-08 Jan Michael Alonzo <jmalonzo@webkit.org>

[Gtk] Skip media/controls-right-click-on-timebar.html as it needs
eventSender which we don't have at the moment.

  • platform/gtk/Skipped:
3:39 PM Changeset in webkit [46959] by Nikolas Zimmermann
  • 2 edits in trunk/WebCore

Rubber-stamped by Jan Michael Alonzo.

Add [DontEnum] to WML specific initializeWMLPageState/resetWMLPageState methods, to assure
fast/dom/Window/window-properties.html doesn't show any differences when WML is enabled.

1:34 PM Changeset in webkit [46958] by krit@webkit.org
  • 2 edits in trunk/WebCore

2009-08-08 Dirk Schulze <krit@webkit.org>

Reviewed by Eric Seidel.

SVG Filter need feColorMatrix implementation
https://bugs.webkit.org/show_bug.cgi?id=27711

Adds the filter effect feColorMatrix to SVG filters.

There is already a test in the w3c directory.
Test: svg/W3C-SVG-1.1/filters-color-01-b.svg

  • platform/graphics/filters/FEColorMatrix.cpp: (WebCore::matrix): (WebCore::saturate): (WebCore::huerotate): (WebCore::luminance): (WebCore::FEColorMatrix::apply):
12:32 PM Changeset in webkit [46957] by krit@webkit.org
  • 2 edits in trunk/WebCore

2009-08-08 Dirk Schulze <krit@webkit.org>

Reviewed by Oliver Hunt.

Build fix for Tiger.

  • platform/graphics/cg/ImageBufferCG.cpp: (WebCore::ImageBuffer::ImageBuffer):
12:11 PM Changeset in webkit [46956] by krit@webkit.org
  • 27 edits
    1 add in trunk/WebCore

2009-08-08 Dirk Schulze <krit@webkit.org>

Reviewed by Oliver Hunt.

This is an initial implementation for linearRGB and sRGB support for all
platforms. Some platforms like Cg support different color spaces
natively, others like Qt, Cairo and Skia do not. This patch uses the
native implementation of platforms if possible and offers a common version
if not.

Different color spaces are used in SVG Filters. Nearly every Filter test
case is a test for this patch and will be enabled, once Filters are enabled.

  • GNUmakefile.am:
  • WebCore.pro:
  • html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::drawTextInternal):
  • html/HTMLCanvasElement.cpp: (WebCore::HTMLCanvasElement::createImageBuffer):
  • platform/graphics/GeneratedImage.cpp: (WebCore::GeneratedImage::drawPattern):
  • platform/graphics/ImageBuffer.cpp: Added. (WebCore::ImageBuffer::transformColorSpace):
  • platform/graphics/ImageBuffer.h: (WebCore::): (WebCore::ImageBuffer::create):
  • platform/graphics/cairo/ImageBufferCairo.cpp: (WebCore::ImageBuffer::ImageBuffer): (WebCore::ImageBuffer::platformTransformColorSpace):
  • platform/graphics/cairo/ImageCairo.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/cg/ImageBufferCG.cpp: (WebCore::ImageBuffer::ImageBuffer):
  • platform/graphics/chromium/TransparencyWin.cpp: (WebCore::TransparencyWin::OwnedBuffers::OwnedBuffers):
  • platform/graphics/filters/FilterEffect.cpp: (WebCore::FilterEffect::getEffectContext):
  • platform/graphics/qt/ImageBufferQt.cpp: (WebCore::ImageBuffer::ImageBuffer): (WebCore::ImageBuffer::platformTransformColorSpace):
  • platform/graphics/qt/PathQt.cpp: (WebCore::Path::strokeContains): (WebCore::Path::strokeBoundingRect):
  • platform/graphics/skia/ImageBufferSkia.cpp: (WebCore::ImageBuffer::ImageBuffer): (WebCore::ImageBuffer::platformTransformColorSpace):
  • platform/graphics/skia/SkiaUtils.cpp: (WebCore::scratchContext):
  • platform/graphics/wx/ImageBufferWx.cpp: (WebCore::ImageBuffer::ImageBuffer):
  • platform/mac/ScrollbarThemeMac.mm: (WebCore::ScrollbarThemeMac::paint):
  • rendering/RenderBoxModelObject.cpp: (WebCore::RenderBoxModelObject::paintFillLayerExtended):
  • svg/SVGMaskElement.cpp: (WebCore::SVGMaskElement::drawMaskerContent):
  • svg/SVGPatternElement.cpp: (WebCore::SVGPatternElement::buildPattern):
  • svg/graphics/SVGImage.cpp: (WebCore::SVGImage::nativeImageForCurrentFrame):
  • svg/graphics/SVGPaintServerGradient.cpp: (WebCore::createMaskAndSwapContextForTextGradient):
  • svg/graphics/SVGPaintServerPattern.cpp: (WebCore::SVGPaintServerPattern::setup):
  • svg/graphics/SVGResourceFilter.cpp: (WebCore::SVGResourceFilter::prepareFilter): (WebCore::SVGResourceFilter::applyFilter):
  • svg/graphics/SVGResourceMasker.cpp: (WebCore::SVGResourceMasker::applyMask):
  • svg/graphics/filters/SVGFETile.cpp: (WebCore::FETile::apply):
10:13 AM Changeset in webkit [46955] by xan@webkit.org
  • 2 edits
    3 deletes in trunk

WebCore:

2009-08-08 Xan Lopez <xlopez@igalia.com>

Roll out r46928, since it wasn't done in accordance with the
agreed naming.

  • bindings/gdom/GdomDOMObject.cpp: Removed.

WebKit/gtk:

2009-08-08 Xan Lopez <xlopez@igalia.com>

Roll out r46928, since it wasn't done in accordance with the
agreed naming.

  • ChangeLog:

Remove the ChangeLog entry, since it was added at the bottom of
the file.

  • gdom/GdomDOMObject.h: Removed.
  • gdom/GdomDOMObjectPrivate.h: Removed.
9:01 AM Changeset in webkit [46954] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-08 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

bugzilla-tool needs a --quiet option
https://bugs.webkit.org/show_bug.cgi?id=28060

A first pass at a --quiet option. This doesn't do all the excting
things we might want, but it addresses 80% of the use case.

  • Scripts/bugzilla-tool:
2:29 AM Changeset in webkit [46953] by Simon Hausmann
  • 2 edits in trunk/WebKit/qt

Doc: there is no group of explicitly shared classes, only one class uses this.

Patch by Volker Hilsheimer <volker.hilsheimer@nokia.com> on 2009-08-08
Reviewed by Simon Hausmann.

Explain the implications in the QWebHistoryItem documentation, and get rid
of the "group".

  • Api/qwebhistory.cpp:
12:39 AM Changeset in webkit [46952] by jmalonzo@webkit.org
  • 12 edits
    2 adds in trunk

2009-08-08 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed by Xan Lopez.

[Gtk] Enable accessibility in Gtk DRT
https://bugs.webkit.org/show_bug.cgi?id=25989

Add SPI to get the focused accessible element to be used by DRT.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (webkit_accessible_get_focused_element):
  • accessibility/gtk/AccessibilityObjectWrapperAtk.h:
  • webkit/webkitprivate.h:
  • webkit/webkitwebframe.cpp: (webkit_web_frame_get_focused_accessible_element):

Remove passing accessibility tests from the Skipped list

  • platform/gtk/Skipped:

Add Accessibility support to the GTK DRT.

  • DumpRenderTree/AccessibilityUIElement.h:
  • DumpRenderTree/gtk/AccessibilityControllerGtk.cpp: Added. (AccessibilityController::AccessibilityController): (AccessibilityController::~AccessibilityController): (AccessibilityController::focusedElement): (AccessibilityController::rootElement):
  • DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp: Added. (AccessibilityUIElement::AccessibilityUIElement): (AccessibilityUIElement::~AccessibilityUIElement): (AccessibilityUIElement::getLinkedUIElements): (AccessibilityUIElement::getDocumentLinks): (AccessibilityUIElement::getChildren): (AccessibilityUIElement::getChildrenWithRange): (AccessibilityUIElement::childrenCount): (AccessibilityUIElement::elementAtPoint): (AccessibilityUIElement::getChildAtIndex): (AccessibilityUIElement::allAttributes): (AccessibilityUIElement::attributesOfLinkedUIElements): (AccessibilityUIElement::attributesOfDocumentLinks): (AccessibilityUIElement::titleUIElement): (AccessibilityUIElement::parentElement): (AccessibilityUIElement::attributesOfChildren): (AccessibilityUIElement::parameterizedAttributeNames): (AccessibilityUIElement::role): (AccessibilityUIElement::title): (AccessibilityUIElement::description): (AccessibilityUIElement::language): (AccessibilityUIElement::x): (AccessibilityUIElement::y): (AccessibilityUIElement::width): (AccessibilityUIElement::height): (AccessibilityUIElement::clickPointX): (AccessibilityUIElement::clickPointY): (AccessibilityUIElement::intValue): (AccessibilityUIElement::minValue): (AccessibilityUIElement::maxValue): (AccessibilityUIElement::valueDescription): (AccessibilityUIElement::isEnabled): (AccessibilityUIElement::insertionPointLineNumber): (AccessibilityUIElement::isActionSupported): (AccessibilityUIElement::isRequired): (AccessibilityUIElement::attributesOfColumnHeaders): (AccessibilityUIElement::attributesOfRowHeaders): (AccessibilityUIElement::attributesOfColumns): (AccessibilityUIElement::attributesOfRows): (AccessibilityUIElement::attributesOfVisibleCells): (AccessibilityUIElement::attributesOfHeader): (AccessibilityUIElement::indexInTable): (AccessibilityUIElement::rowIndexRange): (AccessibilityUIElement::columnIndexRange): (AccessibilityUIElement::lineForIndex): (AccessibilityUIElement::boundsForRange): (AccessibilityUIElement::cellForColumnAndRow): (AccessibilityUIElement::selectedTextRange): (AccessibilityUIElement::setSelectedTextRange): (AccessibilityUIElement::attributeValue): (AccessibilityUIElement::isAttributeSettable): (AccessibilityUIElement::increment): (AccessibilityUIElement::decrement):
  • DumpRenderTree/gtk/DumpRenderTree.cpp: (webViewWindowObjectCleared): (main):
  • GNUmakefile.am:

Aug 7, 2009:

11:08 PM Changeset in webkit [46951] by jmalonzo@webkit.org
  • 2 edits in trunk/WebCore

Gtk build fix - add html/canvas to the binding generator's include path

10:17 PM Changeset in webkit [46950] by oliver@apple.com
  • 1 edit
    3 adds in trunk/LayoutTests

Test function declaration parsing

Reviewed by Cameron Zwarich

9:42 PM Changeset in webkit [46949] by Simon Fraser
  • 2 edits in trunk/WebKit/mac

2009-08-07 Simon Fraser <Simon Fraser>

Fix the build on 10.6.

  • WebView/WebHTMLView.mm: (-[WebHTMLView attachRootLayer:]):
9:14 PM Changeset in webkit [46948] by Simon Fraser
  • 2 edits in trunk/WebKit/mac

2009-08-07 Simon Fraser <Simon Fraser>

Fix a stylistic nit related to the location of the *, which, for some
ridiculous reason, WebKit style dictates to be different between Objective-C and C++.

  • WebView/WebDynamicScrollBarsView.mm: (-[WebDynamicScrollBarsView reflectScrolledClipView:]):
9:08 PM Changeset in webkit [46947] by Simon Fraser
  • 4 edits in trunk/WebKit/mac

2009-08-07 Simon Fraser <Simon Fraser>

Reviewed by Dan Bernstein.

Work around an issue on Leopard where composited layers are misplaced and squished when the page
gets over 4096px tall.
https://bugs.webkit.org/show_bug.cgi?id=27272
<rdar://problem/7067892> [Leopard] Composisted layers are misplaced and squished on on long pages
<rdar://problem/7068252> [Leopard] When switching to a tab with HW layers, they fade in

  • WebView/WebDynamicScrollBarsView.mm: (-[WebDynamicScrollBarsView reflectScrolledClipView:]): Call -_updateLayerHostingViewPosition after scrolling.
  • WebView/WebHTMLView.mm: (-[WebHTMLView _frameOrBoundsChanged]): Call -_updateLayerHostingViewPosition after the view changes size.

(-[WebHTMLView attachRootLayer:]):
Turn off default animations, to avoid animations of sublayer transform, and fading-in
when tab switching.

  • WebView/WebHTMLViewInternal.h: (-[WebHTMLView _updateLayerHostingViewPosition]): New method that constrains the height of the layer-hosting view to a max height of 4096px, and compensates for the height restriction by placing the layer-hosting view at the top of the visible part of the WebHTMLView, and adjusting the position of the hosted layers via sublayer transform.
8:31 PM Changeset in webkit [46946] by jmalonzo@webkit.org
  • 2 edits in trunk/LayoutTests

2009-08-07 Jan Michael Alonzo <jmalonzo@webkit.org>

[Gtk] Skip two new tests (from r46896) needing gtk-specific results.

  • platform/gtk/Skipped:
8:06 PM Changeset in webkit [46945] by oliver@apple.com
  • 2 edits in trunk/WebKitSite

2009-08-07 Joseph Pecoraro <joepeck02@gmail.com>

Reviewed by Oliver Hunt.

Fix webkit-jobs link

  • contact.html:
7:33 PM Changeset in webkit [46944] by jmalonzo@webkit.org
  • 2 edits in trunk/WebKitTools

Unreviewed Gtk build fix

7:22 PM Changeset in webkit [46943] by oliver@apple.com
  • 3 edits in trunk/WebCore

Another windows build fix attempt

7:11 PM Changeset in webkit [46942] by oliver@apple.com
  • 2 edits in trunk/WebCore

Windows build fix

6:57 PM Changeset in webkit [46941] by jmalonzo@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Jan Michael Alonzo <jmalonzo@webkit.org>

Gtk build fix - add html/canvas to the include path.

  • GNUmakefile.am:
6:43 PM Changeset in webkit [46940] by treat@webkit.org
  • 2 edits in trunk/WebCore

Prospective build fix for Qt following r46937.

Patch by Adam Treat <adam.treat@torchmobile.com> on 2009-08-07

  • WebCore.pro:
6:25 PM Changeset in webkit [46939] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Yong Li <yong.li@torchmobile.com>

WINCE PORT: store only width for GlyphBufferAdvance, to save space on low-memory devices
https://bugs.webkit.org/show_bug.cgi?id=27734

  • platform/graphics/GlyphBuffer.h: (WebCore::GlyphBuffer::advanceAt): (WebCore::GlyphBuffer::advanceData): (WebCore::GlyphBuffer::add):
6:24 PM Changeset in webkit [46938] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Yong Li <yong.li@torchmobile.com>

Reviewed by Eric Seidel.

WINCE PORT: pass unrecognized glyphs to GDI to handle
https://bugs.webkit.org/show_bug.cgi?id=27734

  • platform/graphics/FontFastPath.cpp: (WebCore::Font::glyphDataForCharacter):
6:01 PM Changeset in webkit [46937] by cmarrin@apple.com
  • 8 edits
    14 moves
    1 add in trunk/WebCore

Move Canvas related file to a subdirectories
https://bugs.webkit.org/show_bug.cgi?id=28026

5:54 PM Changeset in webkit [46936] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Szabo Carol <carol.szabo@nokia.com>

Reviewed by David Hyatt.

QtWebkit ignores letter-spacing and word-spacing styles.
QtWebkit handles custom letter-spacing and word-spacing in the native QFont
class, but the values for these options are not not passed on from
the platform independent part of the Font class.
Fixed QtWebkit behavior by applying the options to QFont when it is
returned via Font::font()

http://bugs.webkit.org/show_bug.cgi?id=27988

LayoutTests:
fast/css/word-space-extra.html
css2.1/t1604-c542-letter-sp-00-b-a.html
css2.1/t1604-c541-word-sp-01-b-a.html
css1/text_properties/word_spacing.html

  • platform/graphics/qt/FontQt.cpp: (WebCore::Font::font):
5:52 PM Changeset in webkit [46935] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2009-08-07 Nicolas Weber <thakis@chromium.org>

Reviewed by Darin Adler.

Use correct highlight color for listboxes and completion popups.

https://bugs.webkit.org/show_bug.cgi?id=28090

  • rendering/RenderThemeChromiumMac.h:
  • rendering/RenderThemeChromiumMac.mm: Rename activeListBoxSelectionBackgroundColor to platformActiveListBoxSelectionBackgroundColor, so that it actually overrides a virtual method.
5:48 PM Changeset in webkit [46934] by abarth@webkit.org
  • 1 edit
    4 adds in trunk/WebCore

2009-08-07 Maxime Simon <Maxime Simon>

Reviewed by Eric Seidel.

Added a fourth bunch of Haiku-specific files for WebCore.
https://bugs.webkit.org/show_bug.cgi?id=26988

  • platform/haiku/PasteboardHaiku.cpp: Added. (WebCore::Pasteboard::Pasteboard): (WebCore::Pasteboard::generalPasteboard): (WebCore::Pasteboard::writeSelection): (WebCore::Pasteboard::canSmartReplace): (WebCore::Pasteboard::plainText): (WebCore::Pasteboard::documentFragment): (WebCore::Pasteboard::writeURL): (WebCore::Pasteboard::writeImage): (WebCore::Pasteboard::clear):
  • platform/haiku/PlatformKeyboardEventHaiku.cpp: Added. (WebCore::keyIdentifierForHaikuKeyCode): (WebCore::windowsKeyCodeForKeyEvent): (WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent): (WebCore::PlatformKeyboardEvent::disambiguateKeyDownEvent): (WebCore::PlatformKeyboardEvent::currentCapsLockState):
  • platform/haiku/PlatformMouseEventHaiku.cpp: Added. (WebCore::PlatformMouseEvent::PlatformMouseEvent):
  • platform/haiku/PlatformWheelEventHaiku.cpp: Added. (WebCore::PlatformWheelEvent::PlatformWheelEvent):
5:41 PM Changeset in webkit [46933] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-07 Zoltan Horvath <hzoltan@inf.u-szeged.hu>

Reviewed by Eric Seidel.

Allow custom memory allocation control for Noncopyable class
https://bugs.webkit.org/show_bug.cgi?id=27879

Several classes which are inherited from Noncopyable are instantiated by
operator new, so Noncopyable class has been inherited from FastAllocBase.

  • wtf/Noncopyable.h:

2009-08-07 George Staikos <george.staikos@torchmobile.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=27305
Implement WinCE-specific unicode layer.
Written by George Staikos <george.staikos@torchmobile.com>
with bug fixes by Yong Li <yong.li@torchmobile.com>
refactored by Joe Mason <joe.mason@torchmobile.com>

  • wtf/Platform.h:
  • wtf/unicode/Unicode.h:
  • wtf/unicode/wince/UnicodeWince.cpp: Added. (WTF::Unicode::toLower): (WTF::Unicode::toUpper): (WTF::Unicode::foldCase): (WTF::Unicode::isPrintableChar): (WTF::Unicode::isSpace): (WTF::Unicode::isLetter): (WTF::Unicode::isUpper): (WTF::Unicode::isLower): (WTF::Unicode::isDigit): (WTF::Unicode::isPunct): (WTF::Unicode::toTitleCase): (WTF::Unicode::direction): (WTF::Unicode::category): (WTF::Unicode::decompositionType): (WTF::Unicode::combiningClass): (WTF::Unicode::mirroredChar): (WTF::Unicode::digitValue):
  • wtf/unicode/wince/UnicodeWince.h: Added. (WTF::Unicode::): (WTF::Unicode::isSeparatorSpace): (WTF::Unicode::isHighSurrogate): (WTF::Unicode::isLowSurrogate): (WTF::Unicode::isArabicChar): (WTF::Unicode::hasLineBreakingPropertyComplexContext): (WTF::Unicode::umemcasecmp): (WTF::Unicode::surrogateToUcs4):
5:06 PM Changeset in webkit [46932] by Darin Adler
  • 2 edits in trunk/LayoutTests
  • platform/gtk/Skipped: Skip a newly-failing test. I filed a bug about it too.
5:03 PM Changeset in webkit [46931] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Nicolas Weber <thakis@chromium.org>

Teach WebKit how to decode jpegs in cmyk and ycck color spaces.
Heavily inspired by Firefox's take on this issue.

https://bugs.webkit.org/show_bug.cgi?id=27909

Reviewed by Eric Seidel.

  • platform/image-decoders/jpeg/JPEGImageDecoder.cpp: (WebCore::JPEGImageReader::decode): Tell jpeglib to convert ycck to cmyk and keep cmyk. (WebCore::convertCMYKToRGBA): Added function to convert cmyk to rgb. (WebCore::convertRGBToRGBA): Extracted existing conversion logic into its own function. (WebCore::JPEGImageDecoder::outputScanlines): Call convertCMYKToRGBA for cmyk images, convertRGBToRGBA for rgb images.
5:03 PM Changeset in webkit [46930] by Darin Adler
  • 2 edits in trunk/LayoutTests
  • platform/gtk/Skipped: Updated since textarea-arrow-navigation.html moved.
5:02 PM Changeset in webkit [46929] by bweinstein@apple.com
  • 2 edits in trunk/LayoutTests

2009-08-07 Brian Weinstein <bweinstein@apple.com>

Rubber-stamped by Mark Rowe.

Add http/tests/misc/favicon-loads-with-images-disabled.html to get the
Windows Debug bot back green.

  • platform/win/Skipped:
4:51 PM Changeset in webkit [46928] by abarth@webkit.org
  • 1 edit
    4 adds in trunk

2009-07-30 Martin Robinson <mrobinson@appcelerator.com>

Reviewed by Xan Lopez, Gustavo Noronha and Jan Alonzo.

[GTK] new-window-policy-decision-requested provides no information about the target frame
https://bugs.webkit.org/show_bug.cgi?id=27792

Add frameName property to the WebKitWebNavigation object.

  • WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::getNavigationAction): (WebKit::FrameLoaderClient::dispatchDecidePolicyForNewWindowAction): (WebKit::FrameLoaderClient::dispatchDecidePolicyForNavigationAction):
  • webkit/webkitwebnavigationaction.cpp: (webkit_web_navigation_action_get_property): (webkit_web_navigation_action_set_property): (webkit_web_navigation_action_class_init): (webkit_web_navigation_action_get_target_frame): (webkit_web_navigation_action_set_target_frame):
  • webkit/webkitwebnavigationaction.h:
4:34 PM Changeset in webkit [46927] by Darin Adler
  • 2 edits in trunk/WebKitTools
  • Scripts/commit-log-editor: Fix unchecked access to environment

variable that may not be there.

4:25 PM Changeset in webkit [46926] by Darin Adler
  • 16 edits
    2 moves
    1 add
    5 deletes in trunk/LayoutTests

"Shadow tree" DOM nodes are being accidentally exposed to JavaScript
https://bugs.webkit.org/show_bug.cgi?id=28086

Patch by Darin Adler <Darin Adler> on 2009-08-07
Reviewed by Adele Peterson.

Rewrite tests that depend on selection inside <input> and <textarea> elements
from showing up with getSelection(), since does not match Firefox and needs to
change soon.

  • editing/deleting/delete-ligature-002-expected.txt: Updated.
  • editing/deleting/delete-ligature-002.html: Changed to use a contenteditable

<div> element instead of a <textarea> element.

  • editing/deleting/delete-ligature-003-expected.txt: Updated.
  • editing/deleting/delete-ligature-003.html: Changed to use a contenteditable

<div> element instead of a <textarea> element.

  • editing/deleting/skip-virama-001-expected.txt: Updated.
  • editing/deleting/skip-virama-001.html: Changed to use a contenteditable

<div> element instead of a <textarea> element.

  • editing/input/textarea-arrow-navigation-expected.txt: Moved to fast/forms.
  • editing/input/textarea-arrow-navigation.html: Ditto.
  • editing/inserting/insert-thai-characters-001-expected.txt: Updated.
  • editing/inserting/insert-thai-characters-001.html: Changed to use a contenteditable

<div> element instead of a <textarea> element.

  • editing/selection/5213963-expected.txt: Added. This is now a text-only test.
  • editing/selection/5213963.html: Added a call to dumpAsText(). Changed selection

testing to use selectionEnd instead of getSelection().type.

  • editing/selection/find-in-text-control-expected.txt: Updated.
  • editing/selection/find-in-text-control.html: Changed test to distinguish the

different text controls. Made it use activeElement to find out if the selection
is inside a control.

  • editing/selection/move-begin-end-expected.txt: Updated.
  • editing/selection/move-begin-end.html: Changed to use a contenteditable

<div> element instead of a <textarea> element.

  • editing/text-iterator/thai-cursor-movement-expected.txt: Updated.
  • editing/text-iterator/thai-cursor-movement.html: Changed to use a contenteditable

<div> element instead of a <textarea> element.

  • fast/forms/textarea-arrow-navigation-expected.txt: Moved from editing/input since

this is a <textarea> test, not a general editing test.

  • fast/forms/textarea-arrow-navigation.html: Ditto.

Changed to use eventSender instead of the editing machinery to test the down arrow.

  • platform/mac/editing/selection/5213963-expected.checksum: Removed.

This is now a text-only test.

  • platform/mac/editing/selection/5213963-expected.png: Ditto.
  • platform/mac/editing/selection/5213963-expected.txt: Ditto.
  • platform/qt/editing/selection/5213963-expected.txt: Ditto.
  • platform/win/editing/selection/5213963-expected.txt: Ditto.
4:24 PM Changeset in webkit [46925] by atwilson@chromium.org
  • 6 edits
    2 adds in trunk

WebCore:

2009-08-07 Drew Wilson <atwilson@google.com>

Reviewed by David Levin.

SharedWorkers should be shared
https://bugs.webkit.org/show_bug.cgi?id=28058

Added tracking of shared workers to the SharedWorkerRepository.

  • workers/DefaultSharedWorkerRepository.cpp: (WebCore::SharedWorkerProxy::create): Changed SharedWorkerProxy to derive from ThreadSafeShared, as it needs to outlive the repository (if closed while some documents are still loading). (WebCore::SharedWorkerProxy::thread): (WebCore::SharedWorkerProxy::closing): Added flag used to determine whether the worker has closed itself while loading. (WebCore::SharedWorkerProxy::url): (WebCore::SharedWorkerProxy::name): Proxy now tracks worker URL and name to allow validity checks per section 4.8.3 of the WebWorkers spec. (WebCore::SharedWorkerProxy::SharedWorkerProxy): (WebCore::SharedWorkerProxy::addToDocumentSet): Added API to track the document set for a worker to allow worker shutdown. Currently does nothing. (WebCore::SharedWorkerScriptLoader::SharedWorkerScriptLoader): Renamed SharedWorkerLoader to be SharedWorkerScriptLoader for clarity. (WebCore::SharedWorkerScriptLoader::load): (WebCore::SharedWorkerScriptLoader::notifyFinished): (WebCore::DefaultSharedWorkerRepository::instance): (WebCore::DefaultSharedWorkerRepository::workerScriptLoaded): Now handles "worker already loaded" case. (WebCore::SharedWorkerRepository::connect): Delegates to DefaultSharedWorkerRepository::connectToWorker (WebCore::DefaultSharedWorkerRepository::connectToWorker): Added code to track the SharedWorkerProxy in a map, and to handle the "worker already loaded" case. (WebCore::DefaultSharedWorkerRepository::getProxy): Atomic get-or-create operation on the internal tracking HashMap.
  • workers/DefaultSharedWorkerRepository.h:
  • workers/SharedWorker.cpp: (WebCore::SharedWorker::SharedWorker):
  • workers/SharedWorkerRepository.h: Removed instance() API and made constructor private to prevent instantiation. (WebCore::SharedWorkerRepository::SharedWorkerRepository):

LayoutTests:

2009-08-07 Drew Wilson <atwilson@google.com>

Reviewed by David Levin.

SharedWorkers should be shared
https://bugs.webkit.org/show_bug.cgi?id=28058

Added tracking to the SharedWorkerRepository so instances of SharedWorkers are shared across pages.

  • fast/workers/shared-worker-shared-expected.txt: Added.
  • fast/workers/shared-worker-shared.html-disabled: Added sharing test.
4:14 PM Changeset in webkit [46924] by abarth@webkit.org
  • 1 edit
    4 adds in trunk/WebCore

2009-08-07 Maxime Simon <Maxime Simon>

Reviewed by Eric Seidel.

Added four Haiku-specific files for WebCore:
PopupMenuHaiku.cpp, ScreenHaiku.cpp,
SearchPopupMenuHaiku.cpp and SoundHaiku.cpp
https://bugs.webkit.org/show_bug.cgi?id=28080

  • platform/haiku/PopupMenuHaiku.cpp: Added. (WebCore::PopupMenu::PopupMenu): (WebCore::PopupMenu::~PopupMenu): (WebCore::PopupMenu::show): (WebCore::PopupMenu::hide): (WebCore::PopupMenu::updateFromElement): (WebCore::PopupMenu::itemWritingDirectionIsNatural):
  • platform/haiku/ScreenHaiku.cpp: Added. (WebCore::screenRect): (WebCore::screenAvailableRect): (WebCore::screenDepth): (WebCore::screenDepthPerComponent): (WebCore::screenIsMonochrome):
  • platform/haiku/SearchPopupMenuHaiku.cpp: Added. (WebCore::SearchPopupMenu::SearchPopupMenu): (WebCore::SearchPopupMenu::saveRecentSearches): (WebCore::SearchPopupMenu::loadRecentSearches): (WebCore::SearchPopupMenu::enabled):
  • platform/haiku/SoundHaiku.cpp: Added. (WebCore::systemBeep):
4:14 PM Changeset in webkit [46923] by abarth@webkit.org
  • 1 edit
    2 adds in trunk/WebCore

2009-08-07 Maxime Simon <Maxime Simon>

Reviewed by Eric Seidel.

Added two Haiku-specific files to WebCore:
TemporaryLinkStubs.cpp and WidgetHaiku.cpp
https://bugs.webkit.org/show_bug.cgi?id=28080

  • platform/haiku/TemporaryLinkStubs.cpp: Added. (loadResourceIntoArray): (WebCore::historyContains): (WebCore::supportedKeySizes): (WebCore::signedPublicKeyAndChallengeString): (WebCore::userIdleTime): (WebCore::callOnMainThread): (WebCore::SharedBuffer::createWithContentsOfFile): (WebCore::KURL::fileSystemPath): (WebCore::getSupportedKeySizes):
  • platform/haiku/WidgetHaiku.cpp: Added. (WebCore::Widget::Widget): (WebCore::Widget::~Widget): (WebCore::Widget::frameRect): (WebCore::Widget::setFrameRect): (WebCore::Widget::setFocus): (WebCore::Widget::setCursor): (WebCore::Widget::show): (WebCore::Widget::hide): (WebCore::Widget::paint): (WebCore::Widget::setIsSelected):
4:12 PM Changeset in webkit [46922] by abarth@webkit.org
  • 4 edits in trunk/WebCore

2009-08-07 Jian Li <jianli@chromium.org>

Reviewed by Dimitri Glazkov.

[V8] Fix the problem that isAttribute is set to false for onerror and
onmessage in worker custom code.
https://bugs.webkit.org/show_bug.cgi?id=28083

  • bindings/v8/custom/V8AbstractWorkerCustom.cpp: (WebCore::getEventListener): (WebCore::ACCESSOR_SETTER): (WebCore::CALLBACK_FUNC_DECL):
  • bindings/v8/custom/V8WorkerContextCustom.cpp: (WebCore::ACCESSOR_SETTER):
  • bindings/v8/custom/V8WorkerCustom.cpp: (WebCore::getEventListener): (WebCore::ACCESSOR_SETTER):
4:11 PM Changeset in webkit [46921] by abarth@webkit.org
  • 3 edits in trunk/WebCore

2009-08-07 Vitaly Repeshko <vitalyr@quad.spb.corp.google.com>

Reviewed by Dimitri Glazkov.

V8 bindings: speed up lookupDOMWrapper by using new V8 API function.

https://bugs.webkit.org/show_bug.cgi?id=28071

  • bindings/v8/V8DOMWrapper.cpp:
  • bindings/v8/V8DOMWrapper.h: (WebCore::V8DOMWrapper::lookupDOMWrapper):
4:05 PM Changeset in webkit [46920] by rniwa@webkit.org
  • 5 edits
    8 adds in trunk

WebCore: selectionHasStyle doesn't handle text-specific properties properly
https://bugs.webkit.org/show_bug.cgi?id=27858

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-08-07
Reviewed by Justin Garcia.

This patch modifies selectionHasStyle so that it returns TrueTriState when text styles
(text-decoration, font-style, font-weight, & color) are present in all text nodes,
regardless of whether the style is also present in their ancestor nodes or not.
e.g. New behavior concludes that <b><i>hello</i><i>world</i></b> has italic style.

WebKit now applies the text styles (bold, italic, etc) if the specified style was not present on
at least one text node, rather than at the beginning of selection except on Mac.
On Mac, WebKit applies the text style if the style was not present at the beginning of style.

Test: editing/execCommand/toggle-compound-styles.html

  • editing/ApplyStyleCommand.cpp:

(WebCore::getPropertiesNotInComputedStyle): bug fix, must use copy instead of makeMutable for a mutable style.

  • editing/Editor.cpp:

(WebCore::):
(WebCore::triStateOfStyleInComputedStyle): Added a boolean type to ignore text-specific styles
(WebCore::Editor::selectionHasStyle): Ignores text-specific styles for all but text nodes

  • editing/EditorCommand.cpp:

(WebCore::executeToggleStyle): Uses selectionHasStyle instead of selectionStartHasStyle

LayoutTests: selectionHasStyle doesn't handle text-specific properties properly
https://bugs.webkit.org/show_bug.cgi?id=27858

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-08-07
Reviewed by Justin Garcia.

This patch adds toggle-compound-styles to make sure WebKit can handle nested text styles
such as <b><i>hello</i>world</b> (italicized) and <b><i>hello</i>world</b> (not italicized)
on all platforms except on Mac (to match TextEdit's behavior).

  • editing/execCommand/resources/toggle-compound-styles.js: Added.

(testSingleToggle):

  • editing/execCommand/toggle-compound-styles.html: Added.
  • platform/gtk/editing/execCommand: Added.
  • platform/gtk/editing/execCommand/toggle-compound-styles-expected.txt: Added.
  • platform/mac/editing/execCommand/toggle-compound-styles-expected.txt: Added.
  • platform/qt/editing/execCommand/toggle-compound-styles-expected.txt: Added.
  • platform/win/editing/execCommand: Added.
  • platform/win/editing/execCommand/toggle-compound-styles-expected.txt: Added.
3:36 PM Changeset in webkit [46919] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Jian Li <jianli@chromium.org>

Reviewed by Dimitri Glazkov.

[V8] Do not register V8 error message listener in WorkerContextExecutionProxy.
https://bugs.webkit.org/show_bug.cgi?id=28082

  • bindings/v8/WorkerContextExecutionProxy.cpp: (WebCore::WorkerContextExecutionProxy::initV8IfNeeded):
3:25 PM Changeset in webkit [46918] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Kwang Yul Seo <skyul@company100.net>

Reviewed by Eric Seidel.

Cairo-based Windows port does not handle cookies properly
https://bugs.webkit.org/show_bug.cgi?id=27414

curl handles cookies by itself, so using WinINet functions
to get and set cookies is wrong.

Replace CookieJarWin.cpp with CookieJarCurl.cpp so that
cookies can be implemented later once curl provides an API to
get and set cookies.

  • WebCore.vcproj/WebCore.vcproj:
3:09 PM Changeset in webkit [46917] by Dimitri Glazkov
  • 9 edits
    1 add in trunk/WebCore

2009-08-07 Michael Nordman <Michael Nordman>

Reviewed by Dimitri Glazkov.

https://bugs.webkit.org/show_bug.cgi?id=28074

V8 bindings for the window.applicationCache attribute.

  • WebCore.gypi:
  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/DOMObjectsInclude.h:
  • bindings/v8/DerivedSourcesAllInOne.cpp:
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate): (WebCore::V8DOMWrapper::convertEventTargetToV8Object):
  • bindings/v8/V8Index.cpp:
  • bindings/v8/V8Index.h:
  • bindings/v8/custom/V8CustomBinding.h:
  • bindings/v8/custom/V8DOMApplicationCacheCustom.cpp: Added.
2:41 PM Changeset in webkit [46916] by Darin Adler
  • 4 edits in trunk/LayoutTests

2009-08-07 Darin Adler <Darin Adler>

Reviewed by Mark Rowe.

Fix three tests so they are not dependent on the getSelection() result
reflecting selections inside <input> and <textarea> elements.

  • fast/events/right-click-focus.html: Removed unneeded check of anchorNode. The test works fine by just testing the actual result of typing.
  • fast/forms/focus-style-pending.html: Detect focus with an onfocus attribute rather than by looking at getSelection().focusNode.
  • fast/forms/search-click-in-placeholder.html: Detect focus with an onfocus attribute rather than by looking at getSelection().baseNode.
2:39 PM Changeset in webkit [46915] by ajwong@chromium.org
  • 3 edits
    2 adds in trunk

WebCore: Right click on timeline of media controls panel cause seek
https://bugs.webkit.org/show_bug.cgi?id=27920

Patch by Alpha Lam <hclam@chromium.org> on 2009-08-07
Reviewed by Eric Seidel.

Add a test simulate the behavior of seeking by right clicking
on the time bar of media controls while playing.

  • media/controls-right-click-on-timebar-expected.txt: Added.
  • media/controls-right-click-on-timebar.html: Added.

LayoutTests: Right click on timeline of media controls panel cause seek
https://bugs.webkit.org/show_bug.cgi?id=27920

Patch by Alpha Lam <hclam@chromium.org> on 2009-08-07
Reviewed by Eric Seidel.

Add a test simulate the behavior of seeking by right clicking
on the time bar of media controls while playing.

  • media/controls-right-click-on-timebar-expected.txt: Added.
  • media/controls-right-click-on-timebar.html: Added.
2:36 PM Changeset in webkit [46914] by rniwa@webkit.org
  • 13 edits in trunk/WebCore

deprecatedCopyInheritableProperties must be replaced by two different functions
https://bugs.webkit.org/show_bug.cgi?id=28057

Patch by Ryosuke Niwa <rniwa@webkit.org> on 2009-08-07
Reviewed by Justin Garcia.

This patch deletes deprecatedInheritableProperties and modifies call callees to call either editingStyleAtPosition
or prepareEditingStyleToApplyAt. The concept of editing style is introduced in this patch,
which consists of all CSS properties need to be preserved under editing operations.

No test is added since this patch does not change any behavior.

  • css/CSSComputedStyleDeclaration.cpp: Removed deprecatedCopyInheritableProperties
  • css/CSSComputedStyleDeclaration.h: Removed deprecatedCopyInheritableProperties
  • editing/ApplyStyleCommand.cpp:

(WebCore::):
(WebCore::editingStyleAtPosition): Obtains the editing-specific computed style at the position
(WebCore::prepareEditingStyleToApplyAt): Removes redundant editing styles at the specified position
(WebCore::removeStylesAddedByNode): Removes the style derived from the specified node

  • editing/ApplyStyleCommand.h:

(WebCore::):

  • editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::moveParagraphs): Uses editingStyleAtPosition
(WebCore::CompositeEditCommand::breakOutOfEmptyListItem): Uses editingStyleAtPosition

  • editing/DeleteSelectionCommand.cpp:

(WebCore::removeEnclosingAnchorStyle): Uses removeStyleOfNode
(WebCore::DeleteSelectionCommand::saveTypingStyleState): Uses editingStyleAtPosition
(WebCore::DeleteSelectionCommand::calculateTypingStyleAfterDelete): Uses prepareEditingStyleToApplyAt

  • editing/EditCommand.cpp:
  • editing/EditCommand.h:
  • editing/InsertParagraphSeparatorCommand.cpp:

(WebCore::InsertParagraphSeparatorCommand::calculateStyleBeforeInsertion): Uses editingStyleAtPosition
(WebCore::InsertParagraphSeparatorCommand::applyStyleAfterInsertion): Uses prepareEditingStyleToApplyAt

  • editing/RemoveFormatCommand.cpp:

(WebCore::RemoveFormatCommand::doApply): Uses editingStyleAtPosition

  • editing/ReplaceSelectionCommand.cpp:

(WebCore::handleStyleSpansBeforeInsertion): Uses editingStyleAtPosition
(WebCore::ReplaceSelectionCommand::handleStyleSpans): Uses editingStyleAtPosition and prepareEditingStyleToApplyAt
(WebCore::ReplaceSelectionCommand::doApply): Uses prepareEditingStyleToApplyAt

  • editing/markup.cpp:

(WebCore::removeEnclosingMailBlockquoteStyle): Uses removeStyleOfNode
(WebCore::removeDefaultStyles): Uses prepareEditingStyleToApplyAt
(WebCore::createMarkup): Uses editingStyleAtPosition

2:36 PM Changeset in webkit [46913] by Darin Adler
  • 2 edits in trunk/LayoutTests

2009-08-07 Darin Adler <Darin Adler>

Reviewed by Mark Rowe.

Another try at making the gc-9.html test less prone to intermittent failure.

  • fast/dom/gc-9.html: Changed this to do the "before garbage collection" test right after setting the properties. This practically eliminates the possibility that we will get unlucky and get a garbage collection between when theproperty is set and tested, which seems to have happened at least once on the buildbot.
2:10 PM Changeset in webkit [46912] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Handle the case where only a single test is missing results.

  • BuildSlaveSupport/build.webkit.org-config/master.cfg:
2:07 PM Changeset in webkit [46911] by abarth@webkit.org
  • 3 edits
    3 adds in trunk/JavaScriptCore

2009-08-07 George Staikos <george.staikos@torchmobile.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=27305
Implement WinCE-specific unicode layer.
Written by George Staikos <george.staikos@torchmobile.com>
with bug fixes by Yong Li <yong.li@torchmobile.com>
refactored by Joe Mason <joe.mason@torchmobile.com>

  • wtf/Platform.h:
  • wtf/unicode/Unicode.h:
  • wtf/unicode/wince/UnicodeWince.cpp: Added. (WTF::Unicode::toLower): (WTF::Unicode::toUpper): (WTF::Unicode::foldCase): (WTF::Unicode::isPrintableChar): (WTF::Unicode::isSpace): (WTF::Unicode::isLetter): (WTF::Unicode::isUpper): (WTF::Unicode::isLower): (WTF::Unicode::isDigit): (WTF::Unicode::isPunct): (WTF::Unicode::toTitleCase): (WTF::Unicode::direction): (WTF::Unicode::category): (WTF::Unicode::decompositionType): (WTF::Unicode::combiningClass): (WTF::Unicode::mirroredChar): (WTF::Unicode::digitValue):
  • wtf/unicode/wince/UnicodeWince.h: Added. (WTF::Unicode::): (WTF::Unicode::isSeparatorSpace): (WTF::Unicode::isHighSurrogate): (WTF::Unicode::isLowSurrogate): (WTF::Unicode::isArabicChar): (WTF::Unicode::hasLineBreakingPropertyComplexContext): (WTF::Unicode::umemcasecmp): (WTF::Unicode::surrogateToUcs4):
1:19 PM CreatingLayoutTests created by mike.fenton@torchmobile.com
Initial Version
1:15 PM Changeset in webkit [46910] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-07 Yongjun Zhang <yongjun.zhang@nokia.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=28069

Add inline to help winscw compiler resolve specialized argument in
templated functions.

  • runtime/LiteralParser.cpp: (JSC::LiteralParser::Lexer::lexString):
1:09 PM Changeset in webkit [46909] by kenneth@webkit.org
  • 3 edits in trunk/LayoutTests

2009-08-07 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Move fast/forms/menulist-style-color.html to Skipped list for mac/win.
Bug filed - https://bugs.webkit.org/show_bug.cgi?id=28079

  • platform/mac/Skipped:
  • platform/win/Skipped:
12:52 PM Changeset in webkit [46908] by andersca@apple.com
  • 2 edits in trunk/WebKit/mac

2009-08-07 Anders Carlsson <andersca@apple.com>

Fix Tiger build.


  • WebView/WebRenderNode.mm: (-[WebRenderNode _initWithCoreFrame:]):
12:51 PM Changeset in webkit [46907] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-07 Zoltan Horvath <hzoltan@inf.u-szeged.hu>

Reviewed by Eric Seidel.

Allow custom memory allocation control for RegExpObjectData struct
http://bugs.webkit.org/show_bug.cgi?id=26750

Inherits RegExpObjectData struct from FastAllocBase because
it has been instantiated by 'new' in JavaScriptCore/runtime/RegExpObject.cpp:62

  • runtime/RegExpObject.h:
12:28 PM Creating and Submitting Layout Tests and Patches edited by mike.fenton@torchmobile.com
(diff)
12:27 PM Changeset in webkit [46906] by abarth@webkit.org
  • 4 edits
    2 deletes in trunk/WebCore

2009-08-07 Steve Block <steveblock@google.com>

Reviewed by Darin Adler.

Bug 27250: Geolocation callback function IDL files are superfluous
https://bugs.webkit.org/show_bug.cgi?id=27250

The objects for the Geolocation success and error callbacks are created 'manually' in
WebCore/bindings/js/JSGeolocationCustom.cpp. Furthermore, the callback interfaces are
marked 'NoInterfaceObject' in the W3C spec, so the prototype should not
appear on the window object. Hence IDL files for these callbacks are not required.

See http://www.w3.org/TR/geolocation-API/#geolocation_interface and
http://www.w3.org/TR/WebIDL/#NoInterfaceObject.

No new tests required.

  • DerivedSources.make: Modified. Removed reference to IDL files.
  • WebCore.gypi: Modified. Removed reference to IDL files.
  • WebCore.xcodeproj/project.pbxproj: Modified. Removed reference to IDL files.
  • page/PositionCallback.idl: Removed.
  • page/PositionErrorCallback.idl: Removed.
12:17 PM Changeset in webkit [46905] by ben@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-07 George Wright <george.wright@torchmobile.com>

Reviewed by Adam Treat.

Fix DumpRenderTree for the Qt port to always dump the
PNG data unless the expected and actual hashes match.

https://bugs.webkit.org/show_bug.cgi?id=28077

  • DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::DumpRenderTree::dump):
11:51 AM Changeset in webkit [46904] by kenneth@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

2009-08-07 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Add Test results for mac and windows for new tests.

https://bugs.webkit.org/show_bug.cgi?id=28050

  • platform/mac/fast/forms/text-style-color-expected.txt: Added.
  • platform/win/fast/forms/text-style-color-expected.txt: Added.
11:51 AM Changeset in webkit [46903] by abarth@webkit.org
  • 1 edit
    2 adds in trunk

2009-08-07 Adam Treat <adam.treat@torchmobile.com>

Reviewed by David Levin.

Build fix for Qt when ENABLE_VIDEO = 0. This was necessitated by r46890 which
put the idl files back into play regardless of compile time defines.

  • html/TimeRanges.idl:
11:38 AM Changeset in webkit [46902] by pkasting@chromium.org
  • 2 edits in trunk/LayoutTests

https://bugs.webkit.org/show_bug.cgi?id=28076
Add fast/images/icon-decoding.html to Qt skip list until it has
results.

Reviewed by Adam Treat.

  • platform/qt/Skipped:
11:36 AM Changeset in webkit [46901] by andersca@apple.com
  • 3 edits in trunk/WebKit/mac

2009-08-07 Anders Carlsson <andersca@apple.com>

Reviewed by Timothy Hatcher and Sam Weinig.

Change WebRenderNode to take a WebFrame instead of a WebFrameView.


  • WebView/WebRenderNode.h:
  • WebView/WebRenderNode.mm: (-[WebRenderNode _initWithName:position:rect:coreFrame:children:]): (copyRenderNode): (-[WebRenderNode _initWithCoreFrame:]): (-[WebRenderNode initWithWebFrame:]):
11:17 AM Changeset in webkit [46900] by treat@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Adam Treat <adam.treat@torchmobile.com>

Reviewed by David Levin.

Build fix for Qt when ENABLE_VIDEO = 0. This was necessitated by r46890 which
put the idl files back into play regardless of compile time defines.

  • html/TimeRanges.idl:
11:14 AM Changeset in webkit [46899] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-07 Pierre d'Herbemont <pdherbemont@apple.com>

Reviewed by Eric Seidel.

commit-log-editor does not produce a git commit log that is git friendly.
https://bugs.webkit.org/show_bug.cgi?id=27754

We make sure we end up with:

  • A first paragraph describing the bug. It is eventually prefixed by "WebKit: <line>" or "WebCore: <line>". This used to be "WebCore:\n\n<line>".
  • The Reviewed By line.
  • An eventual Patch By line if author and committer doesn't match.
  • The rest of the commit.
  • Scripts/commit-log-editor:
10:56 AM Changeset in webkit [46898] by abarth@webkit.org
  • 1 edit in trunk/WebKitTools/Scripts/run-webkit-unittests

Make run-webkit-unittests executable.

10:37 AM Changeset in webkit [46897] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-07 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by George Staikos.

Coding style patches for RenderThemeQt.cpp based on results from cpp_style.py.

https://bugs.webkit.org/show_bug.cgi?id=28050

  • platform/qt/RenderThemeQt.cpp: (WebCore::RenderThemeQt::baselinePosition): (WebCore::inflateButtonRect): (WebCore::RenderThemeQt::supportsFocus): (WebCore::RenderThemeQt::applyTheme):
10:37 AM Changeset in webkit [46896] by kenneth@webkit.org
  • 4 edits
    4 adds in trunk

2009-08-07 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Eric Seidel.

Prevent RenderThemeQt AdjustStyle based calls for TextField, MenuList and MenuListButton
from discarding style colour.

Add new tests and remove tests from platform/qt/Skipped that now pass.

https://bugs.webkit.org/show_bug.cgi?id=28050

  • fast/forms/menulist-style-color.html: Added.
  • fast/forms/text-style-color.html: Added.
  • platform/qt/Skipped:
  • platform/qt/fast/forms/menulist-style-color-expected.txt: Added.
  • platform/qt/fast/forms/text-style-color-expected.txt: Added.

2009-08-07 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Eric Seidel.

Prevent RenderThemeQt AdjustStyle based calls for TextField, MenuList and MenuListButton
from discarding style colour.

Add tests and remove tests from platform/qt/Skipped that now pass.

https://bugs.webkit.org/show_bug.cgi?id=28050

Tests: fast/forms/menulist-style-color.html

fast/forms/text-style-color.html

  • platform/qt/RenderThemeQt.cpp: (WebCore::RenderThemeQt::adjustTextFieldStyle): (WebCore::RenderThemeQt::adjustMenuListStyle): (WebCore::RenderThemeQt::adjustMenuListButtonStyle):
10:22 AM Changeset in webkit [46895] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-08-07 Adam Barth <abarth@webkit.org>

Unreviewed. (Darin Adler indicated he'd prefer if we landed these
kinds of changes unreviewed, like editing the WebKit Team wiki page.)

Added Pierre d'Herbemont to list of committers.

  • Scripts/modules/committers.py:
9:15 AM QtWebKitTodo edited by andre.pedralho@openbossa.org
(diff)
8:57 AM Changeset in webkit [46894] by Dimitri Glazkov
  • 5 edits in trunk/WebCore

2009-08-07 Dimitri Glazkov <Dimitri Glazkov>

Unreviewed, build fix.

https://bugs.webkit.org/show_bug.cgi?id=22495
[V8] Re-land http://trac.webkit.org/changeset/46821, now that
http://trac.webkit.org/changeset/46874 re-landed

  • bindings/v8/DOMObjectsInclude.h:
  • bindings/v8/DerivedSourcesAllInOne.cpp:
  • bindings/v8/V8Index.cpp:
  • bindings/v8/V8Index.h:
8:52 AM Changeset in webkit [46893] by treat@webkit.org
  • 3 edits in trunk/WebCore

2009-08-07 Adam Treat <adam.treat@torchmobile.com>

Reviewed by Darin Adler and George Staikos.

Add rectToRect convenience function which returns the TransformationMatrix
which maps the 'from' rectangle to the 'to' rectangle.

https://bugs.webkit.org/show_bug.cgi?id=28067

No tests as this would require binding the method to js and since it is static
that doesn't seem possible. Please check the math.

  • platform/graphics/transforms/TransformationMatrix.cpp: (WebCore::TransformationMatrix::rectToRect):
  • platform/graphics/transforms/TransformationMatrix.h:
8:35 AM Changeset in webkit [46892] by beidson@apple.com
  • 10 edits
    2 adds in trunk

WebCore:

2009-08-07 Brady Eidson <beidson@apple.com>

Reviewed by Darin Adler.

Implement the HTML5 hashchange event.
https://bugs.webkit.org/show_bug.cgi?id=21605

Test: fast/loader/hashchange-event.html

  • dom/EventNames.h: Add "hashchange"
  • html/HTMLAttributeNames.in: Add "onhashchange"
  • loader/FrameLoader.cpp: (WebCore::HashChangeEventTask::create): (WebCore::HashChangeEventTask::performTask): (WebCore::HashChangeEventTask::HashChangeEventTask): (WebCore::FrameLoader::scrollToAnchor): When an anchor navigation is completed and the new fragment identifier is different from the old one, queue a hash change event on the Document.

Add a window event listener for the hashchange event if onhashchange is encountered:

  • html/HTMLBodyElement.cpp: (WebCore::HTMLBodyElement::parseMappedAttribute):
  • html/HTMLFrameSetElement.cpp: (WebCore::HTMLFrameSetElement::parseMappedAttribute):

Add a new string utility method that gives "null and empty are equivalent" behavior:

  • platform/text/PlatformString.h: (WebCore::equalIgnoringNullity):
  • platform/text/StringImpl.cpp: (WebCore::equalIgnoringNullity):
  • platform/text/StringImpl.h:

LayoutTests:

2009-08-07 Brady Eidson <beidson@apple.com>

Reviewed by Darin Adler.

Implement the HTML5 hashchange event.
https://bugs.webkit.org/show_bug.cgi?id=21605

  • fast/loader/hashchange-event-expected.txt: Added.
  • fast/loader/hashchange-event.html: Added.
7:57 AM Changeset in webkit [46891] by jberlin@apple.com
  • 7 edits in trunk/WebCore

2009-08-07 Jessie Berlin <jberlin@apple.com>

Add the initial implementation of the ability to resize the columns in
a DataGrid in the Web Inspector. Enables that functionality in both the
Profile View and in the Local Storage View.


Reviewed by Timothy Hatcher.


https://bugs.webkit.org/show_bug.cgi?id=26182


In this version, resizing a column can change the width of at most two
columns (the columns to the left and right of the resizer being dragged).


Possible changes in future patches:


1) Be able to specify the minimum size of a column and the maximum size

of a column. If the minimum size is equal to the maximum size, the
column can't be resized.

2) Make the behavior more in line with the following description:


Everything before the column that is resized stays fixed size. The
resized column changes sizes. The other columns except the last stay
the same size. And the last column adjusts to fit with a minimum size.
If the last column is already the minimum, the next to last column
shrinks, etc.


3) Make it possible to scale the column sizes appropriately when the

Web Inspector window gets resized.


  • inspector/front-end/DOMStorageItemsView.js: (WebInspector.DOMStorageItemsView.prototype.update): Once the DataGrid is added to the DOM, update the widths of the DataGrid in order to put the resizers in place. (WebInspector.DOMStorageItemsView.prototype.resize): Update the widths of the DataGrid in order to correctly adjust the positions of the resizers.


  • inspector/front-end/DataGrid.js: (WebInspector.DataGrid): Store column groups whose widths can later be changed when the columns are resized by the user. Also, keep track of whether the widths of the columns have been initialized. (WebInspector.DataGrid.prototype.updateWidths): Create the resizers for the columns. (WebInspector.DataGrid.prototype._clickInDataTable): Didn't actually change anything in this method, just added a comma to the end of the function. (WebInspector.DataGrid.prototype._startResizerDragging): Store the resizer that is currently dragging. (WebInspector.DataGrid.prototype._resizerDragging): Constrain the area that the resizer can be dragged to the column to it's left and right, with some padding added to make sure that the neighboring columns don't disappear. (WebInspector.DataGrid.prototype._endResizerDragging):


  • inspector/front-end/DatabasesPanel.js: (WebInspector.DatabasesPanel.prototype.resize): Resize the view. (WebInspector.DatabasesPanel.prototype._updateSidebarWidth): ditto.


  • inspector/front-end/ProfileView.js: (WebInspector.ProfileView.prototype.show): Now that the DataGrid is actually attached to the DOM, update the widths of the DataGrid in order to put the resizers in place. (WebInspector.ProfileView.prototype.resize): Update the widths of the DataGrid in order to correctly adjust the positions of the resizers.


  • inspector/front-end/ProfilesPanel.js: (WebInspector.ProfilesPanel.prototype.resize): Resize the view. (WebInspector.ProfilesPanel.prototype._updateSidebarWidth): ditto.


  • inspector/front-end/inspector.css: Add in a style rule for a resizer in the Data Grid.
3:56 AM UsingGitWithWebKit edited by pdherbemont@free.fr
(diff)
3:22 AM WebKit Team edited by jmalonzo@webkit.org
Fix my entry (diff)
2:55 AM Changeset in webkit [46890] by Simon Hausmann
  • 5 edits in trunk/WebCore

2009-08-07 Simon Hausmann <simon.hausmann@nokia.com>

Reviewed by Tor Arne Vestbø.

Back out r46847 and fix the Qt build system to always generate
all the JS binding files from IDL files. The generated files
have #if feature guards, so the #ifs are not needed in the
files using them.

  • WebCore.pro:
  • bindings/js/JSEventTarget.cpp:
  • bindings/js/JSWorkerContextBase.cpp:
  • bindings/js/WorkerScriptController.cpp:
2:44 AM WebKit Team edited by zoltan@webkit.org
Add my entry. (diff)
1:21 AM WebKit Team edited by pdherbemont@apple.com
Add my entry. (diff)
12:32 AM Changeset in webkit [46889] by abarth@webkit.org
  • 7 edits
    3 adds in trunk

2009-08-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Adele Peterson.

Cannot scroll for box-reflect:right
https://bugs.webkit.org/show_bug.cgi?id=27979

Update m_overflowLeft an m_overflowWidth for reflection just like
m_overflowTop and m_overflowHeight.

  • fast/reflections/reflection-overflow-scroll-expected.txt: Added.
  • fast/reflections/reflection-overflow-scroll.html: Added.
  • fast/reflections/resources/reflection-overflow-scroll.js: Added.
  • platform/mac/fast/reflections/reflection-nesting-expected.txt:
  • platform/mac/fast/repaint/reflection-redraw-expected.txt:

2009-08-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Adele Peterson.

Cannot scroll for box-reflect:right
https://bugs.webkit.org/show_bug.cgi?id=27979

Update m_overflowLeft an m_overflowWidth for reflection just like
m_overflowTop and m_overflowHeight.

Test: fast/reflections/reflection-overflow-scroll.html

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::layoutBlock):

2009-08-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Adele Peterson.

Cannot scroll for box-reflect:right
https://bugs.webkit.org/show_bug.cgi?id=27979

Update m_overflowLeft an m_overflowWidth for reflection just like
m_overflowTop and m_overflowHeight.

  • Scripts/make-js-test-wrappers: Added regexp to skip box-shadow-overflo

w-scroll.js

Note: See TracTimeline for information about the timeline view.