⚠ Archived content — this site is no longer maintained.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

Timeline



Apr 9, 2008:

11:52 PM Changeset in webkit [31769] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

2008-04-09 Brady Eidson <​beidson@apple.com>

Forgot to check in this updated result with my Storage-enumeration change
earlier today

  • fast/dom/Window/window-properties-expected.txt:
11:49 PM Changeset in webkit [31768] by beidson@apple.com
  • 4 edits
    4 adds in trunk

WebCore:

2008-04-09 Brady Eidson <​beidson@apple.com>

Reviewed by Mitz Pettel

Add the "onstorage" attribute for the body element to provide easy setup of StorageEvent handling.

Tests: storage/domstorage/sessionstorage/onstorage-attribute-markup.html

storage/domstorage/sessionstorage/onstorage-attribute-setattribute.html

  • html/HTMLAttributeNames.in: Added "onstorage"
  • html/HTMLBodyElement.cpp: (WebCore::HTMLBodyElement::parseMappedAttribute): Handle this attribute on the body element *only* because that's where the HTML5 spec says StorageEvents go. This might change in the future

LayoutTests:

2008-04-09 Brady Eidson <​beidson@apple.com>

Reviewed by Mitz Pettel

Add the "onstorage" attribute for the body element to provide easy setup of StorageEvent handling.

  • storage/domstorage/sessionstorage/onstorage-attribute-markup-expected.txt: Added.
  • storage/domstorage/sessionstorage/onstorage-attribute-markup.html: Added.
  • storage/domstorage/sessionstorage/onstorage-attribute-setattribute-expected.txt: Added.
  • storage/domstorage/sessionstorage/onstorage-attribute-setattribute.html: Added.
10:23 PM squirrelfish edited by oliver@apple.com
(diff)
10:22 PM squirrelfish edited by oliver@apple.com
(diff)
5:55 PM Changeset in webkit [31767] by Antti Koivisto
  • 4 edits in trunk/WebCore

2008-04-09 Antti Koivisto <Antti Koivisto>

Reviewed by Mitz.

Remove a less-than-critical bit from Node and replace it with a hash. Those bits are valuable!

Also moved setting of the guard to dispatchSimulatedClick for better consistency.
No one else calls dispatchSimulatedMouseEvent() currently and this is really a click() specific behavior.

  • dom/EventTargetNode.cpp: (WebCore::EventTargetNode::dispatchSimulatedMouseEvent): (WebCore::EventTargetNode::dispatchSimulatedClick):
  • dom/Node.cpp: (WebCore::Node::Node):
  • dom/Node.h:
5:23 PM Changeset in webkit [31766] by weinig@apple.com
  • 7 edits in trunk/WebCore

2008-04-09 Sam Weinig <​sam@webkit.org>

Reviewed by Geoffrey Garen.

Fix for ​https://bugs.webkit.org/show_bug.cgi?id=18389
Crash in JSDOMWindowWrapper::mark loading digg.com

  • Add a null check to JSDOMWindowWrapper::mark for the case when the collector runs during its allocation.
  • Cleans up the creation of the window to be a little more straight forward.
  • bindings/js/JSDOMWindowBase.cpp: (WebCore::JSDOMWindowBasePrivate::JSDOMWindowBasePrivate): (WebCore::JSDOMWindowBase::JSDOMWindowBase): (WebCore::JSDOMWindowBase::clear): (WebCore::JSDOMWindowBase::wrapper):
  • bindings/js/JSDOMWindowBase.h:
  • bindings/js/JSDOMWindowWrapper.cpp: (WebCore::JSDOMWindowWrapper::JSDOMWindowWrapper): (WebCore::JSDOMWindowWrapper::mark):
  • bindings/js/JSDOMWindowWrapper.h:
  • bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::initScript):
  • bindings/scripts/CodeGeneratorJS.pm:
5:21 PM Changeset in webkit [31765] by Adam Roben
  • 2 edits in trunk/JavaScriptCore

VC++ Express build fix

  • JavaScriptCore.vcproj/testkjs/testkjs.vcproj: Link against user32.lib.
4:29 PM Changeset in webkit [31764] by beidson@apple.com
  • 2 edits in trunk/WebCore

2008-04-09 Brady Eidson <​beidson@apple.com>

Rubberstamped by Mitzpettel

Added HTMLAttributeNames.in and HTMLTagNames.in to the xcodeproj for easy editing,
and sorted the html group

  • WebCore.xcodeproj/project.pbxproj:
4:20 PM Changeset in webkit [31763] by pewtermoose@webkit.org
  • 2 edits in trunk/WebCore

2008-04-09 Brent Fulgham <​bfulgham@gmail.com>

Reviewed by Mitz.

Correct build regression in Windows (Cairo) build due to changes
in FontDescription.h signature. (see ​http://bugs.webkit.org/show_bug.cgi?id=18394)

  • rendering/RenderThemeWin.cpp: Change for new font methods. (WebCore::RenderThemeWin::systemFont):
4:18 PM Changeset in webkit [31762] by beidson@apple.com
  • 6 edits in trunk

WebCore:

2008-04-09 Brady Eidson <​beidson@apple.com>

Reviewed by Sam Weinig

When enumerating a Storage object, the Storage built-ins should not be included, but the
prototype chain SHOULD be included.

  • bindings/js/JSStorageCustom.cpp: (WebCore::JSStorage::customGetPropertyNames): Return false so the prototype chain is walked
  • storage/Storage.idl: Mark all built-ins as "DontEnum"

LayoutTests:

2008-04-09 Brady Eidson <​beidson@apple.com>

Reviewed by Sam Weinig

When enumerating a Storage object, the Storage built-ins should not be included, but the
prototype chain SHOULD be included.

Updated the layout test to test this.

  • storage/domstorage/sessionstorage/enumerate-storage-expected.txt:
  • storage/domstorage/sessionstorage/enumerate-storage.html:
4:05 PM Changeset in webkit [31761] by ggaren@apple.com
  • 2 edits in branches/squirrelfish/JavaScriptCore

2008-04-09 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Oliver Hunt.

Next step toward re-entrant execution of function code (global code ->
built-in function -> JS function):


Made op_ret return from Machine::privateExecute if its calling codeBlock
is NULL.


I'm checking this in by itself to demonstrate that a more clever
mechanism is not necessary for performance.


sunspider --squirrelfish reports no regression.

  • ChangeLog:
  • VM/Machine.cpp: (KJS::Machine::execute): (KJS::Machine::privateExecute):
3:36 PM Changeset in webkit [31760] by mitz@apple.com
  • 5 edits in trunk/LayoutTests

Reviewed by Sam Weinig.

  • fix more Windows layout test regressions from the font-weight patch

The patch silently introduced the ability to select some fonts
by their full name on Windows. Firefox and IE allow the same.
This made these tests use different, Windows-only fonts on
Windows. The patch removes references to these fonts from the
tests.

  • tables/mozilla/bugs/bug4093.html:
  • tables/mozilla/bugs/bug43854-1.html:
  • tables/mozilla/bugs/bug8381.html:
  • tables/mozilla/core/bloomberg.html:
2:21 PM Changeset in webkit [31759] by beidson@apple.com
  • 5 edits
    2 adds in trunk

WebCore:

2008-04-09 Brady Eidson <​beidson@apple.com>

Reviewed by Adam Roben

Per the HTML5 spec, enumerating a Storage object should walk the keys in the storage area,
and should not return the built-in properties.

Test: storage/domstorage/sessionstorage/enumerate-storage.html

  • bindings/js/JSLocationCustom.cpp: (WebCore::JSLocation::customGetPropertyNames): Removed the name from an unused parameter
  • bindings/js/JSStorageCustom.cpp: (WebCore::JSStorage::customGetPropertyNames): Copy all of the keys into the property names array
  • storage/Storage.idl: Add CustomGetPropertyNames

LayoutTests:

2008-04-09 Brady Eidson <​beidson@apple.com>

Reviewed by Adam Roben

Per the HTML5 spec, enumerating a Storage object should walk the keys in the storage area,
and should not return the built-in properties.

This test adds some keys to sessionStorage, enumerates the sessionStorage object, and logs
the sorted result. It should *not* contain the built in properties of the Storage object

  • storage/domstorage/sessionstorage/enumerate-storage-expected.txt: Added.
  • storage/domstorage/sessionstorage/enumerate-storage.html: Added.
1:36 PM Changeset in webkit [31758] by timothy@apple.com
  • 2 edits in trunk/WebCore

Fixes the regression where resource time/size pills do not
update during a load in the Inspector.

​http://bugs.webkit.org/show_bug.cgi?id=18374

Reviewed by John Sullivan.

  • page/inspector/ResourcesPanel.js:

(ResourcesPanel.prototype.show): Remove an unneeded call to _updateGraphBars,
since _updateSidebarWidth already calls _updateGraphBars.
(ResourcesPanel.prototype.addResource): Call updateGraphSideWidth.
(ResourcesPanel.prototype._updateGraphBars): Ditto.
(ResourceSidebarTreeElement.prototype.updateGraphSideWidth): Update the
style for graphSideElement with the passed in width.

11:36 AM Changeset in webkit [31757] by Antti Koivisto
  • 6 edits in trunk/WebCore

2008-04-09 Antti Koivisto <Antti Koivisto>

Reviewed by Mitz.

Avoid unnecessary calls to virtual updateStyleAttributeIfNeeded() method.

  • do the the checking in call sites instead of inside the function
  • rename to updateStyleAttribute()
  • use namedAttrMap directly (for getting id attribute) in insertedIntoDocument/removedFromDocument to avoid calling attributes() which may trigger style attribute update.


  • dom/Element.cpp: (WebCore::Element::attributes): (WebCore::Element::getAttribute): (WebCore::Element::hasAttributes): (WebCore::Element::insertedIntoDocument): (WebCore::Element::removedFromDocument):
  • dom/Element.h: (WebCore::Element::updateStyleAttribute):
  • dom/NamedAttrMap.cpp: (WebCore::NamedAttrMap::addAttribute):
  • dom/StyledElement.cpp: (WebCore::StyledElement::updateStyleAttribute):
  • dom/StyledElement.h:
11:09 AM Changeset in webkit [31756] by ggaren@apple.com
  • 5 edits in branches/squirrelfish/JavaScriptCore

2008-04-09 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Maciej Stachowiak.

Next step toward re-entrant execution of function code (global code ->
built-in function -> JS function):


Made Machine::execute return a value.


Sketched out some code for Machine::execute for functions -- still
doesn't work yet, though.

sunspider --squirrelfish reports no regression.

  • VM/Machine.cpp: (KJS::Machine::execute): (KJS::Machine::privateExecute):
  • VM/Machine.h:
  • kjs/interpreter.cpp: (KJS::Interpreter::evaluate):
  • kjs/testkjs.cpp: (runWithScripts):
10:01 AM Changeset in webkit [31755] by timothy@apple.com
  • 2 edits in trunk/WebCore

A more complete fix for: can't drag the Web Inspector when grabbing
in the toolbar area. Makes dragging under the search field in the
empty space work again.

​http://bugs.webkit.org/show_bug.cgi?id=18373

Reviewed by Adam Roben.

  • page/inspector/inspector.js:

(WebInspector.toolbarDragStart): Prevent dragging if the event target
is a toggleable toolbar item. Allow dragging when the target is a
normal toolbar item.

9:54 AM Changeset in webkit [31754] by ggaren@apple.com
  • 4 edits in branches/squirrelfish/JavaScriptCore

2008-04-09 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Sam Weinig.


First step toward re-entrant execution of function code (global code ->
built-in function -> JS function):


Tiny bit of refactoring in the Machine class.

sunspider --squirrelfish reports no regression.

  • VM/Machine.cpp: (KJS::Machine::dumpRegisters): (KJS::Machine::unwindCallFrame): (KJS::Machine::execute): (KJS::Machine::privateExecute):
  • VM/Machine.h: (KJS::Machine::isGlobalCallFrame):
  • kjs/interpreter.cpp: (KJS::Interpreter::evaluate):
8:28 AM Changeset in webkit [31753] by Adam Roben
  • 2 edits in trunk/JavaScriptCore

Build fix

8:18 AM Changeset in webkit [31752] by Adam Roben
  • 2 edits in trunk/JavaScriptCore

Build fix

  • wtf/AlwaysInline.h: Make sure to #include Platform.h before using the macros it defines.
12:06 AM squirrelfish edited by ggaren@apple.com
(diff)
12:04 AM Changeset in webkit [31751] by ggaren@apple.com
  • 11 edits
    2 adds in branches/squirrelfish/JavaScriptCore

2008-04-08 Geoffrey Garen <​ggaren@apple.com>

Reviewed by Oliver Hunt.


Support for re-entrant execution of global code (global code -> built-in
function -> global code).


Keep a stack of register files instead of just one. Globals propogate
between register files as the register files enter and exit the stack.


An activation still uses its own register file's base as its
registerBase, but the global object uses the register file *stack*'s
registerBase, which updates dynamically to match the register file at
the top of the stack.


sunspider --squirrelfish reports no regression.

Apr 8, 2008:

11:18 PM Changeset in webkit [31750] by mjs@apple.com
  • 3 edits in branches/squirrelfish/JavaScriptCore

2008-04-08 Maciej Stachowiak <​mjs@apple.com>

Reviewed by Geoff.


  • initial preparatory work for JS constructors


1) Allocate registers for the returnInfo block and "this" value when generating code for
op_construct. These are not used yet, but the JS branch of op_construct will use them.


2) Adjust argc and argv appropriately for native constructor calls.


3) Assign return value in a more straightforward way in op_ret since this is actually
a bit faster (and makes up for the allocation of extra registers above).

  • VM/CodeGenerator.cpp: (KJS::CodeGenerator::emitConstruct):
  • VM/Machine.cpp: (KJS::Machine::privateExecute):
11:18 PM Changeset in webkit [31749] by mrowe@apple.com
  • 2 edits in trunk/JavaScriptCore

2008-04-08 Mark Rowe <​mrowe@apple.com>

Export WTF::initializeThreading() from JavaScriptCore.

10:13 PM Changeset in webkit [31748] by weinig@apple.com
  • 2 edits in trunk/WebCore

2008-04-08 Sam Weinig <​sam@webkit.org>

Reviewed by Timothy Hatcher.

Fix for ​http://bugs.webkit.org/show_bug.cgi?id=18373
REGRESSION: Can't drag the Web Inspector when grabbing in the toolbar area

  • page/inspector/inspector.css:
8:21 PM squirrelfish edited by oliver@apple.com
(diff)
8:21 PM Changeset in webkit [31747] by alp@webkit.org
  • 2 edits in trunk/WebCore

2008-04-08 Christian Persch <​chpe@gnome.org>

Reviewed by Alp Toker.

​http://bugs.webkit.org/show_bug.cgi?id=18360
"missing sentinel" warnings in PluginDatabaseGtk.cpp

Use NULL, not 0 as sentinel in varargs calls.

  • plugins/gtk/PluginDatabaseGtk.cpp: (WebCore::PluginDatabase::getPluginPathsInDirectories): (WebCore::PluginDatabase::defaultPluginDirectories):
8:21 PM squirrelfish edited by oliver@apple.com
(diff)
7:17 PM Changeset in webkit [31746] by weinig@apple.com
  • 48 edits
    2 adds in trunk

JavaScriptCore:

2008-04-04 Sam Weinig <​sam@webkit.org>

Reviewed by Geoffrey Garen.

First step in implementing the "split window"

  • Add a GlobalThisValue to ExecState which should be used in places that used to implement the "use the global object as this if null" rule.
  • Factor out lookupGetter/lookupSetter into virtual methods on JSObject so that they can be forwarded.
  • Make defineGetter/defineSetter virtual methods for the same reason.
  • Have PrototypeReflexiveFunction store the globalObject used to create it so that it can be used to get the correct thisObject for eval.
  • API/JSObjectRef.cpp: (JSObjectCallAsFunction):
  • JavaScriptCore.exp:
  • kjs/Activation.h:
  • kjs/ExecState.cpp: (KJS::ExecState::ExecState): (KJS::GlobalExecState::GlobalExecState):
  • kjs/ExecState.h: (KJS::ExecState::globalThisValue):
  • kjs/ExecStateInlines.h: (KJS::ExecState::ExecState): (KJS::FunctionExecState::FunctionExecState):
  • kjs/JSGlobalObject.cpp: (KJS::JSGlobalObject::reset): (KJS::JSGlobalObject::toGlobalObject):
  • kjs/JSGlobalObject.h: (KJS::JSGlobalObject::JSGlobalObjectData::JSGlobalObjectData): (KJS::JSGlobalObject::JSGlobalObject):
  • kjs/array_instance.cpp: (KJS::CompareWithCompareFunctionArguments::CompareWithCompareFunctionArguments): (KJS::compareWithCompareFunctionForQSort):
  • kjs/array_object.cpp: (KJS::arrayProtoFuncSort): (KJS::arrayProtoFuncFilter): (KJS::arrayProtoFuncMap): (KJS::arrayProtoFuncEvery): (KJS::arrayProtoFuncForEach): (KJS::arrayProtoFuncSome):
  • kjs/function.cpp: (KJS::FunctionImp::callAsFunction): (KJS::ActivationImp::toThisObject): (KJS::globalFuncEval): (KJS::PrototypeReflexiveFunction::PrototypeReflexiveFunction): (KJS::PrototypeReflexiveFunction::mark):
  • kjs/function.h: (KJS::PrototypeReflexiveFunction::cachedGlobalObject):
  • kjs/function_object.cpp: (KJS::functionProtoFuncApply): (KJS::functionProtoFuncCall):
  • kjs/nodes.cpp: (KJS::ExpressionNode::resolveAndCall): (KJS::FunctionCallValueNode::evaluate): (KJS::LocalVarFunctionCallNode::inlineEvaluate): (KJS::ScopedVarFunctionCallNode::inlineEvaluate): (KJS::FunctionCallBracketNode::evaluate): (KJS::FunctionCallDotNode::inlineEvaluate):
  • kjs/object.cpp: (KJS::JSObject::call): (KJS::JSObject::put): (KJS::tryGetAndCallProperty): (KJS::JSObject::lookupGetter): (KJS::JSObject::lookupSetter): (KJS::JSObject::toThisObject): (KJS::JSObject::toGlobalObject): (KJS::JSObject::fillGetterPropertySlot):
  • kjs/object.h:
  • kjs/object_object.cpp: (KJS::objectProtoFuncLookupGetter): (KJS::objectProtoFuncLookupSetter):
  • kjs/string_object.cpp: (KJS::replace):

WebCore:

2008-04-04 Sam Weinig <​sam@webkit.org>

Reviewed by Geoffrey Garen.

First step in implementing the "split window"

  • This patch takes the first step in changing the window navigation model from clearing the window properties on navigation, to replacing an inner window. This is necessary to safely perform security checks using the lexical global object.

This first step adds a new class called JSDOMWindowWrapper, which wraps
the real window object. All JS calls that would go to the window object
now go to it, which it forwards to the current inner window. To accomplish
this, the wrapper window is used as the ThisValue wherever the window was used
before.

  • WebCore.base.exp:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMWindowBase.cpp: (WebCore::JSDOMWindowBase::JSDOMWindowBase): (WebCore::JSDOMWindowBase::clear): Reset the wrapper windows prototype too. (WebCore::JSDOMWindowBase::toThisObject): (WebCore::JSDOMWindowBase::wrapper): (WebCore::windowProtoFuncAToB): (WebCore::windowProtoFuncBToA): (WebCore::windowProtoFuncOpen): (WebCore::windowProtoFuncSetTimeout): (WebCore::windowProtoFuncClearTimeout): (WebCore::windowProtoFuncSetInterval): (WebCore::windowProtoFuncAddEventListener): (WebCore::windowProtoFuncRemoveEventListener): (WebCore::windowProtoFuncShowModalDialog): (WebCore::windowProtoFuncNotImplemented): (WebCore::toJS):
  • bindings/js/JSDOMWindowBase.h: Fix to expect the wrapper as the thisObj.
  • bindings/js/JSDOMWindowCustom.cpp: (WebCore::JSDOMWindow::postMessage): (WebCore::toDOMWindow):
  • bindings/js/JSDOMWindowWrapper.cpp: Added. (WebCore::): (WebCore::JSDOMWindowWrapper::JSDOMWindowWrapper): (WebCore::JSDOMWindowWrapper::~JSDOMWindowWrapper): (WebCore::JSDOMWindowWrapper::mark): (WebCore::JSDOMWindowWrapper::className): (WebCore::JSDOMWindowWrapper::getOwnPropertySlot): (WebCore::JSDOMWindowWrapper::put): (WebCore::JSDOMWindowWrapper::deleteProperty): (WebCore::JSDOMWindowWrapper::getPropertyNames): (WebCore::JSDOMWindowWrapper::getPropertyAttributes): (WebCore::JSDOMWindowWrapper::defineGetter): (WebCore::JSDOMWindowWrapper::defineSetter): (WebCore::JSDOMWindowWrapper::lookupGetter): (WebCore::JSDOMWindowWrapper::lookupSetter): (WebCore::JSDOMWindowWrapper::toGlobalObject): (WebCore::JSDOMWindowWrapper::impl): (WebCore::JSDOMWindowWrapper::disconnectFrame): (WebCore::JSDOMWindowWrapper::clear): (WebCore::toJS):
  • bindings/js/JSDOMWindowWrapper.h: Added. (WebCore::JSDOMWindowWrapper::innerWindow): (WebCore::JSDOMWindowWrapper::setInnerWindow): (WebCore::JSDOMWindowWrapper::classInfo): Forward methods to the innerWindow.
  • bindings/js/JSHTMLDocumentCustom.cpp: (WebCore::JSHTMLDocument::open):
  • bindings/js/ScheduledAction.cpp: (WebCore::ScheduledAction::execute):
  • bindings/js/kjs_events.cpp: (WebCore::JSAbstractEventListener::handleEvent):
  • bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::~KJSProxy): (WebCore::KJSProxy::evaluate): (WebCore::KJSProxy::clear): (WebCore::KJSProxy::initScript): (WebCore::KJSProxy::clearDocumentWrapper): (WebCore::KJSProxy::processingUserGesture): (WebCore::KJSProxy::attachDebugger):
  • bindings/js/kjs_proxy.h: (WebCore::KJSProxy::haveWindowWrapper): (WebCore::KJSProxy::windowWrapper): (WebCore::KJSProxy::globalObject): (WebCore::KJSProxy::initScriptIfNeeded): Hold onto the wrapper window instead of global object. As a convenience, keep the globalObject() as a forward to the inner window.
  • bindings/objc/DOMUtility.mm: (KJS::createDOMWrapper):
  • bindings/scripts/CodeGeneratorJS.pm:
  • dom/Document.cpp: (WebCore::Document::domWindow):
  • dom/Document.h: (WebCore::Document::defaultView):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::dispatchWindowObjectAvailable):
  • page/DOMWindow.idl:
  • page/Frame.cpp: (WebCore::Frame::~Frame): (WebCore::Frame::pageDestroyed):
6:02 PM Changeset in webkit [31745] by beidson@apple.com
  • 7 edits in trunk

JavaScriptCore:

2008-04-08 Brady Eidson <​beidson@apple.com>

Encourage Windows to rebuild - AGAIN...

  • kjs/DateMath.cpp:

WebCore:

2008-04-08 Brady Eidson <​beidson@apple.com>

Encourage Windows to rebuild - AGAIN...

  • WebCore.vcproj/build-generated-files.sh:
  • bindings/scripts/CodeGenerator.pm:
  • bindings/scripts/CodeGeneratorJS.pm:
  • config.h:
5:10 PM Changeset in webkit [31744] by beidson@apple.com
  • 2 edits in trunk/WebCore

2008-04-08 Brady Eidson <​beidson@apple.com>

OMG release builds didn't have this flag Windows build bots I'm so
sorry will you ever forgive me?

(Found by Stephanie, rubberstamped and landed by Brady)

  • WebCore.vcproj/WebCore.vcproj: Add ENABLE_DOM_STORAGE for release builds
5:06 PM Changeset in webkit [31743] by timothy@apple.com
  • 3 edits in trunk/WebCore

Fixes a crash in KJS::JSValue::toObject when closing
Safari with Inspector open.

​http://bugs.webkit.org/show_bug.cgi?id=18371

Reviewed by Adam Roben.

  • page/InspectorController.cpp:

(WebCore::inspectedWindow): NULL check the result of toJSDOMWindow.
Return JSNull if JSDOMWindow is 0.

  • page/inspector/ElementsPanel.js:

(ElementsPanel.prototype.reset): NULL check the result of
InspectorController.inspectedWindow.

4:57 PM Changeset in webkit [31742] by Adam Roben
  • 3 edits in trunk/WebKit/win

Export callOnMainThread from WebKit.dll

Rubberstamped by Anders.

  • WebKit.vcproj/WebKit.def:
  • WebKit.vcproj/WebKit_debug.def:
4:07 PM Changeset in webkit [31741] by dino@apple.com
  • 1 edit
    11 adds in trunk/WebKitSite

Add some updated proposals for CSS specifications, such as
Transforms and Transitions which are already implemented in TOT.

2:57 PM Changeset in webkit [31740] by Adam Roben
  • 2 edits in trunk/WebCore

Wx build fix

  • webcore-wx.bkl: Removed entry for non-existent file.
2:52 PM Changeset in webkit [31739] by Adam Roben
  • 2 edits in trunk/WebCore

qmake build fixes

  • WebCore.pro: Removed entries for non-existent files.
2:34 PM Changeset in webkit [31738] by Adam Roben
  • 2 edits in trunk/JavaScriptCore

Mac build fix

2:28 PM Changeset in webkit [31737] by beidson@apple.com
  • 8 edits in trunk

Strongly encourage Windows bots to rebuild with DOM_STORAGE enabled

2:04 PM Changeset in webkit [31736] by timothy@apple.com
  • 31 edits
    37 adds in trunk/WebCore

Implements the majority of the Inspector UI refresh as shown at:
​http://trac.webkit.org/projects/webkit/wiki/ProposedWebInspectorUIRefresh

​http://bugs.webkit.org/show_bug.cgi?id=17773

A few areas that have not been re-implemented with the new UI are:

  • Search and search results.
  • Request and response headers in the Resources panel.
  • Changing the sorting, grouping or toggling small rows in Resources.
  • Image and font previews in the icon of resources.

Reviewed by Adam Roben.

  • English.lproj/InspectorLocalizedStrings.js: Updated to match actual strings.
  • WebCore.vcproj/WebCore.vcproj: Added new files.
  • page/InspectorController.cpp:

(WebCore::InspectorController::setWindowVisible): Call resetScriptObjects()
instead of individual clear functions.
(WebCore::InspectorController::populateScriptObjects): Renamed from
populateScriptResources.
(WebCore::InspectorController::addDatabaseScriptResource): Call
addDatabase instead of addResource.
(WebCore::InspectorController::removeDatabaseScriptResource): Call
removeDatabase instead of removeResource.
(WebCore::InspectorController::resetScriptObjects): Renamed from
clearScriptResources.
(WebCore::InspectorController::didCommitLoad): Call resetScriptObjects()
instead of individual clear functions.

  • page/InspectorController.h: Rename functions.
  • page/inspector/Console.js: Changed the object name to Console from

ConsolePanel. Made it inherit the prototype from View. Added code
to animate in and out.

  • page/inspector/Database.js: Removed title updating and Resource

pseudo-subclassing. Made more of a Model object that just encapsulates
data. Add a getter for table names.

  • page/inspector/DatabaseQueryView.js: Added. Implements the view seen

when selecting a Database in the DatabasesPanel. Implemented as an
interactive console-like area.

  • page/inspector/DatabaseTableView.js: Added. Implements the view seen

when selecting a Database Table in the DatabasesPanel. Matches the old
Browse view of Database panels.

  • page/inspector/DatabasesPanel.js: Changed the object name to DatabasesPanel

from DatabasePanel. Implements a panel that shows a sidebar of Databases
and Database Tables.

  • page/inspector/ElementsPanel.js: Changed the object name to ElementsPanel

from DocumentPanel. Implements the DOM tree that shows the DOM rooted at the
main resource.

  • page/inspector/FontView.js: Use this.contentElement instead of this.element.
  • page/inspector/ImageView.js: Ditto.
  • page/inspector/Images/clearConsoleButtons.png: Added.
  • page/inspector/Images/consoleButtons.png: Added.
  • page/inspector/Images/darkShadow.png: Flipped.
  • page/inspector/Images/database.png: Modified to be 32x32.
  • page/inspector/Images/databaseTable.png: Added.
  • page/inspector/Images/databasesIcon.png: Added.
  • page/inspector/Images/disclosureTriangleSmallDown.png: Added.
  • page/inspector/Images/disclosureTriangleSmallDownBlack.png: Added.
  • page/inspector/Images/disclosureTriangleSmallDownWhite.png: Added.
  • page/inspector/Images/disclosureTriangleSmallRight.png: Added.
  • page/inspector/Images/disclosureTriangleSmallRightBlack.png: Added.
  • page/inspector/Images/disclosureTriangleSmallRightDown.png: Added.
  • page/inspector/Images/disclosureTriangleSmallRightDownBlack.png: Added.
  • page/inspector/Images/disclosureTriangleSmallRightDownWhite.png: Added.
  • page/inspector/Images/disclosureTriangleSmallRightWhite.png: Added.
  • page/inspector/Images/dockButtons.png: Added.
  • page/inspector/Images/elementsIcon.png: Added.
  • page/inspector/Images/gradientHighlightBottom.png:
  • page/inspector/Images/resourceCSSIcon.png: Added.
  • page/inspector/Images/resourceDocumentIcon.png: Added.
  • page/inspector/Images/resourcePlainIcon.png: Added.
  • page/inspector/Images/resourcesIcon.png: Added.
  • page/inspector/Images/resourcesSizeGraphIcon.png: Added.
  • page/inspector/Images/resourcesTimeGraphIcon.png: Added.
  • page/inspector/Images/scriptsIcon.png: Added.
  • page/inspector/Images/segment.png: Modified to fit the taller status bar.
  • page/inspector/Images/segmentEnd.png: Ditto.
  • page/inspector/Images/segmentHover.png: Ditto.
  • page/inspector/Images/segmentHoverEnd.png: Ditto.
  • page/inspector/Images/segmentSelected.png: Ditto.
  • page/inspector/Images/segmentSelectedEnd.png: Ditto.
  • page/inspector/Images/sidebarSelectionBackground.png: Added.
  • page/inspector/Images/sidebarSelectionBackgroundFocused.png: Added.
  • page/inspector/Images/sidebarSelectionBackgroundInactive.png: Added.
  • page/inspector/Images/sidebarSmallSelectionBackground.png: Added.
  • page/inspector/Images/sidebarSmallSelectionBackgroundFocused.png: Added.
  • page/inspector/Images/sidebarSmallSelectionBackgroundInactive.png: Added.
  • page/inspector/Images/statusbarBackground.png: Added.
  • page/inspector/Images/statusbarBottomBackground.png: Added.
  • page/inspector/Images/statusbarButtons.png: Added.
  • page/inspector/Images/statusbarResizerVertical.png: Added.
  • page/inspector/Images/toolbarItemSelected.png: Added.
  • page/inspector/Panel.js: Added support for toolbar items and status bar items.

Inherits from the View prototype.

  • page/inspector/Resource.js: Removed title updating and the ResourceTreeElement.

Made more of a Model object that just encapsulates data.

  • page/inspector/ResourceCategory.js: Removed the ResourceCategoryTreeElement.

Made more of a Model object that just encapsulates data.

  • page/inspector/ResourceView.js: A simple base object for Resource views

shown in ResourcesPanel. Inherits from the View prototype.

  • page/inspector/ResourcesPanel.js: Changed the object name to ResourcesPanel

from NetworkPanel. Implements the timeline graph, size graph and resource
viewing by using ResourceViews.

  • page/inspector/SidebarTreeElement.js: Added. Inherits from TreeElement and implements

a section element and a regular element that has an icon, title and optional subtitle.

  • page/inspector/SourceView.js: Use this.contentElement instead of this.element.
  • page/inspector/StylesSidebarPane.js: Null check WebInspector.mainResource.
  • page/inspector/View.js: Added. A simple object to manage showing and hiding an element.
  • page/inspector/WebKit.qrc: Added new files.
  • page/inspector/inspector.css: New and changed style rules to support the new UI.
  • page/inspector/inspector.html: New and changed HTML for the UI. The search field

is disabled until search is reimplemented.

  • page/inspector/inspector.js: Removed code related to the sidebar and back-forward lists.

Also removed code related to navigation to panels. Added code to instantiate the new
panels and setup the toolbar.

1:59 PM Changeset in webkit [31735] by mitz@apple.com
  • 5 edits in trunk/WebCore

Reviewed by Antti Koivisto.

  • fix a regression from r31313 which made the inspector highlight and yellow find highlight for table cells with extra top appear in the wrong position
  • rendering/LayoutState.cpp: (WebCore::LayoutState::LayoutState): Changed absolutePosition() to absolutePositionForContent(), to get the right position for children of table cells. This is the correct fix for this part of r31313.
  • rendering/RenderBox.cpp: (WebCore::RenderBox::absolutePosition): Reverted the change from r31313 by changing absolutePosition() to absolutePositionForContent() here too.
  • rendering/RenderObject.h: (WebCore::RenderObject::absolutePositionForContent): Changed to return the result from absolutePosition().
  • rendering/RenderTableCell.cpp: (WebCore::RenderTableCell::absoluteClippedOverflowRect): Turned the "else" case into an early return. (WebCore::RenderTableCell::absolutePosition): Reverted the change from r31313.
1:58 PM Changeset in webkit [31734] by mitz@apple.com
  • 3 edits
    4 adds in trunk

WebCore:

Reviewed by Antti Koivisto.

Test: fast/block/float/br-with-clear-2.html

  • rendering/bidi.cpp: (WebCore::RenderBlock::layoutInlineChildren): The existing code for handling a float-clearing <br> in incremental layout had several bugs in it: it only applied when the <br> was on the last line of the block; it assumed that the <br> was always the rightmost box on the line (which is false in right-to-left runs); and in some places it looked at the last child of the block instead of the last box on the line. Fixed these bugs.

LayoutTests:

Reviewed by Antti Koivisto.

  • fast/block/float/br-with-clear-2.html: Added.
  • platform/mac/fast/block/float/br-with-clear-2-expected.checksum: Added.
  • platform/mac/fast/block/float/br-with-clear-2-expected.png: Added.
  • platform/mac/fast/block/float/br-with-clear-2-expected.txt: Added.
1:50 PM Changeset in webkit [31733] by Adam Roben
  • 2 edits in trunk/WebCore

Mac build fix

  • WebCore.xcodeproj/project.pbxproj: Removed MainThread* files which were moved to WTF.
1:36 PM Changeset in webkit [31732] by timothy@apple.com
  • 3 edits in trunk/WebCore

Change WebInspector.elementDragStart and WebInspector.elementDragEnd to have more
state, so calling elementDragEnd does not need the listener functions. Also makes
calling elementDragStart again without elementDragEnd work correctly.

Reviewed by Adam Roben.

  • page/inspector/ElementsPanel.js:

(DocumentPanel.prototype.rightSidebarResizerDragStart): Just call WebInspector.elementDragStart.
Removed code that stored away the listeners now that elementDragStart does it.
(DocumentPanel.prototype.rightSidebarResizerDragEnd): Just call WebInspector.elementDragEnd.
Removed code that deleted listeners properties now that elementDragEnd does it.

  • page/inspector/inspector.js:

(WebInspector.sidebarResizerDragEnd): Pass less arguments to elementDragEnd.
(WebInspector.searchResultsResizerDragEnd): Ditto.
(WebInspector.elementDragStart): Store the listeners as private properties on WebInspector.
If the properties are already defined, call elementDragEnd first.
(WebInspector.elementDragEnd): Remove the event listeners using the stored properties.
Delete the private listener properties.

1:30 PM Changeset in webkit [31731] by Adam Roben
  • 1 edit
    1 add in trunk/WebCore

Build fix

  • ForwardingHeaders/wtf/MainThread.h: Added. Forgot to commit this.
1:12 PM Changeset in webkit [31730] by Adam Roben
  • 21 edits
    7 moves
    5 adds in trunk

Move callOnMainThread to WTF

JavaScriptCore:

Move callOnMainThread to WTF

Reviewed by Alexey Proskuryakov.

  • wtf/MainThread.cpp:
  • wtf/MainThread.h:
  • wtf/gtk/MainThreadGtk.cpp:
  • wtf/mac/MainThreadMac.mm:
  • wtf/qt/MainThreadQt.cpp:
  • wtf/win/MainThreadWin.cpp:
  • wtf/wx/MainThreadWx.cpp: Moved here from WebCore/platform. Replaced all instances of "WebCore" with "WTF".
  • kjs/bool_object.cpp: Touched to force JavaScriptCore.vcproj to build. to the WTF namespace.
  • wtf/ThreadingWin.cpp: (WTF::initializeThreading): Call initializeMainThread.

WebCore:

Move callOnMainThread to WTF

Reviewed by Alexey Proskuryakov.

  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCoreSources.bkl: Removed MainThread files.
  • bindings/js/JSCustomSQLTransactionCallback.cpp:
  • loader/icon/IconDatabase.cpp: (WebCore::iconDatabase):
  • storage/Database.cpp: (WebCore::Database::Database):
  • storage/DatabaseTracker.cpp: Updated #includes and replaced calls to WebCore::initializeThreadingAndMainThread with calls to KJS::initializeThreading.
  • platform/MainThread.cpp: Removed.
  • platform/MainThread.h: Removed.
  • platform/gtk/MainThreadGtk.cpp: Removed.
  • platform/mac/MainThreadMac.mm: Removed.
  • platform/qt/MainThreadQt.cpp: Removed.
  • platform/win/MainThreadWin.cpp: Removed.
  • platform/wx/MainThreadWx.cpp: Removed.

WebKit/gtk:

Move callOnMainThread to WTF

Reviewed by Alexey Proskuryakov.

  • webkit/webkitprivate.cpp: Updated #include. (webkit_init): Changed to call KJS::initializeThreading.

WebKit/win:

Move callOnMainThread to WTF

Reviewed by Alexey Proskuryakov.

  • WebIconDatabase.cpp: Updated #include
12:48 PM Changeset in webkit [31729] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

2008-04-08 Mark Rowe <​mrowe@apple.com>

Clean up after Brady.

  • DumpRenderTree/mac/DumpRenderTree.mm: Add an include so that NSInteger can be found.
11:56 AM Changeset in webkit [31728] by Adam Roben
  • 2 edits in trunk/BugsSite

Combine :equal operations with the following operation if they are fewer than 3 characters long

This keeps us from showing lots of small changes on long lines, just
because some letters happened to be the same.

Inspired by
​http://code.google.com/p/reviewboard/source/browse/trunk/reviewboard/diffviewer/diffutils.py?r=1264#147

  • PrettyPatch/PrettyPatch.rb: (PrettyPatch::DiffSection.initialize): Remove :equal operations fewer than 3 characters long. The characters from the removed operations become part of the subsequent operation. (PrettyPatch::CodeLine.text_as_html): Don't wrap 0-length strings in <ins>/<del> tags. Also removed the @fromLineNumber.nil? and @toLineNumber.nil? checks as they are no longer reliable now that we're removing operations.
11:28 AM Changeset in webkit [31727] by sfalken@apple.com
  • 1 edit in trunk/WebKit/win/ChangeLog

Fix ChangeLog

11:28 AM Changeset in webkit [31726] by sfalken@apple.com
  • 11 edits in trunk/WebKit/win

Add missing increment/decrement COM class counts to detect leaks.


Reviewed by Adam Roben.

  • CFDictionaryPropertyBag.cpp: (CFDictionaryPropertyBag::CFDictionaryPropertyBag): (CFDictionaryPropertyBag::~CFDictionaryPropertyBag):
  • CFDictionaryPropertyBag.h:
  • HTTPHeaderPropertyBag.cpp: (HTTPHeaderPropertyBag::HTTPHeaderPropertyBag): (HTTPHeaderPropertyBag::~HTTPHeaderPropertyBag):
  • HTTPHeaderPropertyBag.h:
  • WebActionPropertyBag.cpp: (WebActionPropertyBag::WebActionPropertyBag): (WebActionPropertyBag::~WebActionPropertyBag):
  • WebDocumentLoader.cpp: (WebDocumentLoader::WebDocumentLoader): (WebDocumentLoader::~WebDocumentLoader):
  • WebDropSource.cpp: (WebDropSource::WebDropSource): (WebDropSource::~WebDropSource):
  • WebDropSource.h:
  • WebElementPropertyBag.cpp: (WebElementPropertyBag::WebElementPropertyBag): (WebElementPropertyBag::~WebElementPropertyBag):
  • WebTextRenderer.cpp: (WebTextRenderer::WebTextRenderer): (WebTextRenderer::~WebTextRenderer):
  • WebView.cpp: (WebView::updateActiveState):
11:08 AM Changeset in webkit [31725] by Adam Roben
  • 3 edits in trunk/BugsSite

Change PrettyPatch to use DiffBuilder for intra-line diffs

This gives us much prettier intra-line diffs, largely because it can
distinguish multiple changes on the same line. e.g., if a line changes
from:

const int myConstant = 0;

to

static const unsigned myConstant;

You will see that "static " was inserted, "int" changed to "unsigned",
and " = 0" was deleted.

This seems to have also gotten rid of some spurious instances of
"<ins></ins>" and "<del></del>" at the end of a line.

  • PrettyPatch/PrettyPatch.rb: (PrettyPatch::DiffSection.initialize): Use DiffBuilder instead of setChangeExtentFromLine (which had been copied from Trac's diffing code). (PrettyPatch::CodeLine):
    • Removed the changeExtent attribute
    • Added the operations attribute
    • Removed the setChangeExtentFromLine method

(PrettyPatch::CodeLine.text_as_html): Uses @operations to determine
what text to wrap in <ins> and <del> tags.

  • PrettyPatch/diff.rb: Deleted a bunch of code we don't use. (HTMLDiff::DiffBuilder.initialize): Moved code here from the old build method, but left out the calls to perform_operation, since we build the diff HTML outside of this class. (HTMLDiff::DiffBuilder.split_inputs_to_words): Removed calls to convert_html_to_list_of_words, since we're not diffing HTML.
11:06 AM Changeset in webkit [31724] by Adam Roben
  • 1 edit
    1 add in trunk/BugsSite

Check in diff.rb

This came from
​http://instiki.rubyforge.org/svn/instiki/trunk/lib/diff.rb
revision 28521

  • PrettyPatch/diff.rb: Added.
11:06 AM Changeset in webkit [31723] by timothy@apple.com
  • 3 edits in trunk/WebCore

Makes the Network timeline graph resource sizes on the bar graph
when Transfer Size is selected.

​http://bugs.webkit.org/show_bug.cgi?id=18351

Reviewed by Adam Roben.

  • page/inspector/Resource.js:

(WebInspector.Resource.CompareByTime): Array sorting function for time.
Copied from WebInspector.NetworkPanel.timelineEntryCompare.

  • page/inspector/ResourcesPanel.js:

(WebInspector.NetworkPanel): Set the selectedIndex of graphModeSelectElement to
make Transfer Time the default graph.
(WebInspector.NetworkPanel.prototype.changeGraphMode): Call calculator.reset,
_refreshAllResources and _updateGraphDividersIfNeeded.
(WebInspector.NetworkPanel.prototype.get totalDuration): Removed.
(WebInspector.NetworkPanel.prototype._updateGraphBoundriesIfNeeded): Removed boundary
checking code and just call calculator.updateBoundries.
(WebInspector.NetworkPanel.prototype._updateGraphDividersIfNeeded): Add a force argument.
Use calculator.boundarySpan when computing the slice. If the slice is NaN, don't make a label.
Ask the calculator to format the label values for each slice.
(WebInspector.NetworkPanel.prototype.clearTimeline): Call calculator.reset.
(WebInspector.NetworkPanel.timelineEntryCompare): Call WebInspector.Resource.CompareByTime.

(WebInspector.NetworkTimelineEntry.prototype.refresh): Ask the panel's calculator to compute
bar graph percentages for a resource. Use those percentages to change the bar style.

(WebInspector.ResourceCalculator): Renamed from WebInspector.TimelineValueCalculator.
(WebInspector.ResourceCalculator.prototype.computeSummaryValues): Changed the Array.forEach
to a normal for loop. Work with resources instead of TimelineEntries.
(WebInspector.ResourceCalculator.prototype.computeBarGraphPercentages): Return basic bar graph
percents with start always being zero.
(WebInspector.ResourceCalculator.prototype.get boundarySpan): Return the difference between
maximumBoundary and minimumBoundary.
(WebInspector.ResourceCalculator.prototype.updateBoundries): Update maximumBoundary based on _value.
The minimumBoundary is always zero.
(WebInspector.ResourceCalculator.prototype.reset): Delete maximumBoundary and minimumBoundary.
(WebInspector.ResourceCalculator.prototype._value): Takes a Resource instead of TimelineEntry.

(WebInspector.ResourceTransferTimeCalculator): Renamed from TransferTimeCalculator.
(WebInspector.ResourceTransferTimeCalculator.prototype.computeSummaryValues): Work with resources
instead of entries. Changed the Array.forEach to a normal for loop.
(WebInspector.ResourceTransferTimeCalculator.prototype.computeBarGraphPercentages): Return percents
just like WebInspector.NetworkTimelineEntry.prototype.refresh was doing before.
(WebInspector.ResourceTransferTimeCalculator.prototype.updateBoundries): Update minimumBoundary and
maximumBoundary just like the checks WebInspector.NetworkPanel.prototype._updateGraphBoundriesIfNeeded was doing.

(WebInspector.ResourceTransferSizeCalculator): Renamed from TransferSizeCalculator.
(WebInspector.ResourceTransferSizeCalculator.prototype._value): Takes a Resource instead of TimelineEntry.

10:29 AM Changeset in webkit [31722] by beidson@apple.com
  • 6 edits in trunk

WebKitTools:

2008-04-08 Brady Eidson <​beidson@apple.com>

Reviewed by Mitzpettel

Fixed ​http://bugs.webkit.org/show_bug.cgi?id=18302
-WebArchive subresources dump in random order, intermittent failures

  • DumpRenderTree/mac/DumpRenderTree.mm: (compareResourceURLs): Sorting function based on the resource URLs (serializeWebArchiveToXML): Sort the subresource array

LayoutTests:

2008-04-08 Brady Eidson <​beidson@apple.com>

Reviewed by Mitzpettel

Fixed ​http://bugs.webkit.org/show_bug.cgi?id=18302
-WebArchive subresources dump in random order, intermittent failures

Now that a DRT fix has set the order, these 3 tests needed new results

  • webarchive/test-css-import-expected.txt:
  • webarchive/test-duplicate-resources-expected.txt:
  • webarchive/test-frameset-expected.txt:
10:27 AM Changeset in webkit [31721] by rwlbuis@webkit.org
  • 2 edits in trunk/WebCore

Fix the SVG build.

  • svg/SVGFEFloodElement.cpp: (WebCore::SVGFEFloodElement::filterEffect):
9:50 AM Changeset in webkit [31720] by alice.liu@apple.com
  • 2 edits in trunk/WebCore

2008-04-08 Alice Liu <​alice.liu@apple.com>

Reviewed by Dan Bernstein.


fixed ​http://bugs.webkit.org/show_bug.cgi?id=18329
REGRESSION: Assertion failure in -[WebViewFactory endOfTextMarkerRange:]
when invoking the Dictionary pop-up

  • page/mac/AccessibilityObjectWrapper.mm: (-[AccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]): In the section of code where WebCore-equivalents of mac-specific types are prepared, some functions weren't meant to operate on the default initialized values. Checking for nil in certain places will better maintain the original pre-refactoring code path.
1:15 AM Changeset in webkit [31719] by mitz@apple.com
  • 2 edits in trunk/WebCore

2008-04-08 Dan Bernstein <​mitz@apple.com>

  • fix most of the Windows layout test regressions from the font-weight patch
  • platform/graphics/win/FontCacheWin.cpp: (WebCore::matchImprovingEnumProc): Changed to be symmetric with respect to italics: always prefer (reject) a candidate that changes the italics trait to (from) the desired value. (WebCore::createGDIFont): Added code to set more members of the LOGFONT structure before creating the font.
12:20 AM Changeset in webkit [31718] by rwlbuis@webkit.org
  • 4 edits in trunk/WebCore

Reviewed by Ollie.

​http://bugs.webkit.org/show_bug.cgi?id=18354
feFlood in attribute should be removed according to SVG 1.1 Errata

Remove the in attribute from SVGFEFlood DOM.

Note: See TracTimeline for information about the timeline view.