WebKit-https.git
10 years ago2009-09-28 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Tue, 29 Sep 2009 01:48:55 +0000 (01:48 +0000)]
2009-09-28  Fumitoshi Ukai  <ukai@chromium.org>

        Unreviewed.

        Add myself to list of committers.

        * Scripts/modules/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48852 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Tue, 29 Sep 2009 01:10:11 +0000 (01:10 +0000)]
2009-09-28  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by David Kilzer.

        [Gtk][Layout tests] Generate results for css3 tests
        https://bugs.webkit.org/show_bug.cgi?id=29784

        Add results for css3 tests and enable these tests in the Skipped.

        * platform/gtk/Skipped:
        * platform/gtk/css3/css3-modsel-33-expected.txt: Added.
        * platform/gtk/css3/css3-modsel-35-expected.txt: Added.
        * platform/gtk/css3/css3-modsel-36-expected.txt: Added.
        * platform/gtk/css3/css3-modsel-37-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48851 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Tue, 29 Sep 2009 00:47:41 +0000 (00:47 +0000)]
2009-09-28  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by David Kilzer.

        [Gtk][Layout tests] Generate results for transitions tests
        https://bugs.webkit.org/show_bug.cgi?id=29786

        Add results for transitions tests and enable these tests in the Skipped.

        * platform/gtk/Skipped:
        * platform/gtk/transitions/transition-drt-api-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48850 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Tue, 29 Sep 2009 00:35:02 +0000 (00:35 +0000)]
2009-09-28  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by David Kilzer.

        [Gtk][Layout tests] Generate results for animations tests
        https://bugs.webkit.org/show_bug.cgi?id=29783

        Add results for animations tests and enable these tests in the Skipped.

        * platform/gtk/Skipped:
        * platform/gtk/animations/animation-drt-api-expected.txt: Added.
        * platform/gtk/animations/animation-drt-api-multiple-keyframes-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48849 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/7157288> Crash in RenderStyle::computedLineHeight()
mitz@apple.com [Tue, 29 Sep 2009 00:28:31 +0000 (00:28 +0000)]
<rdar://problem/7157288> Crash in RenderStyle::computedLineHeight()
when Times New Roman is not installed

Reviewed by Sam Weinig.

* platform/graphics/win/FontCacheWin.cpp:
(WebCore::FontCache::getLastResortFallbackFont): If Times New Roman is
not available, use the Windows default GUI font.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48848 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Tue, 29 Sep 2009 00:27:36 +0000 (00:27 +0000)]
2009-09-28  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by David Kilzer.

        [Gtk][Layout tests] Generate results for transforms tests
        https://bugs.webkit.org/show_bug.cgi?id=29785

        Add results for transforms tests and enable these tests in the Skipped.

        * platform/gtk/Skipped:
        * platform/gtk/transforms/2d/compound-transforms-vs-containers-expected.txt: Added.
        * platform/gtk/transforms/2d/transform-borderbox-expected.txt: Added.
        * platform/gtk/transforms/2d/transform-origin-borderbox-expected.txt: Added.
        * platform/gtk/transforms/no_transform_hit_testing-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48847 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoDo not add platform-specific methods to cross-platform header
jianli@chromium.org [Mon, 28 Sep 2009 23:56:42 +0000 (23:56 +0000)]
Do not add platform-specific methods to cross-platform header
FileSystem.h per Darin's feedback for 29109.
https://bugs.webkit.org/show_bug.cgi?id=29830

Reviewed by David Levin.

* platform/FileSystem.h:
* platform/chromium/DragDataChromium.cpp:
(WebCore::DragData::asURL):
* platform/chromium/FileSystemChromium.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Dumitru Daniliuc <dumi@chromium.org>
eric@webkit.org [Mon, 28 Sep 2009 23:33:25 +0000 (23:33 +0000)]
2009-09-28  Dumitru Daniliuc  <dumi@chromium.org>

        Reviewed by Dimitri Glazkov.

        Register Chromium's VFSs with a proper sqlite3_io_methods
        finder. This should only affect the POSIX implementation. The
        Windows change is included for consistency and in case sqlite
        starts using something similar in its Windows VFS in the future.

        https://bugs.webkit.org/show_bug.cgi?id=29743

        * platform/sql/chromium/SQLiteFileSystemChromiumPosix.cpp:
        (WebCore::SQLiteFileSystem::registerSQLiteVFS):
        * platform/sql/chromium/SQLiteFileSystemChromiumWin.cpp:
        (WebCore::SQLiteFileSystem::registerSQLiteVFS):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48845 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Jakub Wieczorek <faw217@gmail.com>
eric@webkit.org [Mon, 28 Sep 2009 23:25:43 +0000 (23:25 +0000)]
2009-09-28  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Simon Hausmann.

        [Qt] Update the Skipped list after adding XSLT support.
        https://bugs.webkit.org/show_bug.cgi?id=29817

        Remove the tests that are passing now and skip these that are not,
        with a reason of failure for each of them.

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Alex Milowski <alex@milowski.com>
ddkilzer@apple.com [Mon, 28 Sep 2009 22:58:47 +0000 (22:58 +0000)]
2009-09-28  Alex Milowski  <alex@milowski.com>

        <http://webkit.org/b/29098> Update to MathML project webpage

        Reviewed by David Kilzer.

        Updated MathML project page and added link to Wiki page.

        * projects/mathml/index.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48843 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Eric Seidel <eric@webkit.org>
eric@webkit.org [Mon, 28 Sep 2009 22:50:03 +0000 (22:50 +0000)]
2009-09-28  Eric Seidel  <eric@webkit.org>

        Reviewed by Simon Fraser.

        11 transforms/3d tests failing on the Leopard Debug bot after r48833
        https://bugs.webkit.org/show_bug.cgi?id=29828

        Leopard has HW compositing disabled until rdar://problem/7228836 can be fixed.
        So skip tests which depend on it being enabled per Simon's request.

        * platform/mac-leopard/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48842 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 28 Sep 2009 22:25:52 +0000 (22:25 +0000)]
2009-09-28  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam Barth.

        Change the V8 bindings' handling of window.toString(),
        so we return [object DOMWindow] (like JSC), not [object global].

        https://bugs.webkit.org/show_bug.cgi?id=29742

        This will fix a couple of Chromium port test failures and enable us to
        use the default expectations for a bunch more.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48841 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: added new exports.
ggaren@apple.com [Mon, 28 Sep 2009 22:23:10 +0000 (22:23 +0000)]
Windows build fix: added new exports.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-09-28
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48840 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRoll out r48825. It causes build failures by calling an undefined function from...
mrowe@apple.com [Mon, 28 Sep 2009 22:16:09 +0000 (22:16 +0000)]
Roll out r48825.  It causes build failures by calling an undefined function from within RefPtr.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48839 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 28 Sep 2009 22:06:52 +0000 (22:06 +0000)]
2009-09-28  Nate Chapin  <japhet@chromium.org>

        Rubber stamped by David Levin.

        Chromium build fix, add back in #include of V8Proxy.h in V8WebKitPointConstructor.cpp.

        * bindings/v8/custom/V8WebKitPointConstructor.cpp: Re-include V8Proxy.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48838 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: removed exports that no longer exist.
ggaren@apple.com [Mon, 28 Sep 2009 22:06:19 +0000 (22:06 +0000)]
Windows build fix: removed exports that no longer exist.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-09-28
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48837 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNotNullPassRefPtr: smart pointer optimized for passing references that are not null
ggaren@apple.com [Mon, 28 Sep 2009 21:45:58 +0000 (21:45 +0000)]
NotNullPassRefPtr: smart pointer optimized for passing references that are not null
https://bugs.webkit.org/show_bug.cgi?id=29822

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-09-28
Reviewed by Darin Adler.

JavaScriptCore:

Added NotNullPassRefPtr, and deployed it in all places that initialize
JavaScript objects.

2.2% speedup on bench-allocate-nonretained.js.

* API/JSCallbackConstructor.cpp:
(JSC::JSCallbackConstructor::JSCallbackConstructor):
* API/JSCallbackConstructor.h:
* API/JSCallbackObject.h:
* API/JSCallbackObjectFunctions.h:
(JSC::JSCallbackObject::JSCallbackObject):
* JavaScriptCore.exp:
* bytecode/CodeBlock.h:
(JSC::CodeBlock::addFunctionDecl):
(JSC::CodeBlock::addFunctionExpr):
* runtime/ArrayConstructor.cpp:
(JSC::ArrayConstructor::ArrayConstructor):
* runtime/ArrayConstructor.h:
* runtime/ArrayPrototype.cpp:
(JSC::ArrayPrototype::ArrayPrototype):
* runtime/ArrayPrototype.h:
* runtime/BooleanConstructor.cpp:
(JSC::BooleanConstructor::BooleanConstructor):
* runtime/BooleanConstructor.h:
* runtime/BooleanObject.cpp:
(JSC::BooleanObject::BooleanObject):
* runtime/BooleanObject.h:
* runtime/BooleanPrototype.cpp:
(JSC::BooleanPrototype::BooleanPrototype):
* runtime/BooleanPrototype.h:
* runtime/DateConstructor.cpp:
(JSC::DateConstructor::DateConstructor):
* runtime/DateConstructor.h:
* runtime/DateInstance.cpp:
(JSC::DateInstance::DateInstance):
* runtime/DateInstance.h:
* runtime/DatePrototype.cpp:
(JSC::DatePrototype::DatePrototype):
* runtime/DatePrototype.h:
* runtime/ErrorConstructor.cpp:
(JSC::ErrorConstructor::ErrorConstructor):
* runtime/ErrorConstructor.h:
* runtime/ErrorInstance.cpp:
(JSC::ErrorInstance::ErrorInstance):
* runtime/ErrorInstance.h:
* runtime/ErrorPrototype.cpp:
(JSC::ErrorPrototype::ErrorPrototype):
* runtime/ErrorPrototype.h:
* runtime/FunctionConstructor.cpp:
(JSC::FunctionConstructor::FunctionConstructor):
* runtime/FunctionConstructor.h:
* runtime/FunctionPrototype.cpp:
(JSC::FunctionPrototype::FunctionPrototype):
* runtime/FunctionPrototype.h:
* runtime/GlobalEvalFunction.cpp:
(JSC::GlobalEvalFunction::GlobalEvalFunction):
* runtime/GlobalEvalFunction.h:
* runtime/InternalFunction.cpp:
(JSC::InternalFunction::InternalFunction):
* runtime/InternalFunction.h:
(JSC::InternalFunction::InternalFunction):
* runtime/JSActivation.cpp:
(JSC::JSActivation::JSActivation):
* runtime/JSActivation.h:
(JSC::JSActivation::JSActivationData::JSActivationData):
* runtime/JSArray.cpp:
(JSC::JSArray::JSArray):
* runtime/JSArray.h:
* runtime/JSByteArray.cpp:
(JSC::JSByteArray::JSByteArray):
* runtime/JSByteArray.h:
* runtime/JSFunction.cpp:
(JSC::JSFunction::JSFunction):
* runtime/JSFunction.h:
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::JSGlobalObject):
* runtime/JSONObject.h:
(JSC::JSONObject::JSONObject):
* runtime/JSObject.h:
(JSC::JSObject::JSObject):
(JSC::JSObject::setStructure):
* runtime/JSVariableObject.h:
(JSC::JSVariableObject::JSVariableObject):
* runtime/JSWrapperObject.h:
(JSC::JSWrapperObject::JSWrapperObject):
* runtime/MathObject.cpp:
(JSC::MathObject::MathObject):
* runtime/MathObject.h:
* runtime/NativeErrorConstructor.cpp:
(JSC::NativeErrorConstructor::NativeErrorConstructor):
* runtime/NativeErrorConstructor.h:
* runtime/NativeErrorPrototype.cpp:
(JSC::NativeErrorPrototype::NativeErrorPrototype):
* runtime/NativeErrorPrototype.h:
* runtime/NumberConstructor.cpp:
(JSC::NumberConstructor::NumberConstructor):
* runtime/NumberConstructor.h:
* runtime/NumberObject.cpp:
(JSC::NumberObject::NumberObject):
* runtime/NumberObject.h:
* runtime/NumberPrototype.cpp:
(JSC::NumberPrototype::NumberPrototype):
* runtime/NumberPrototype.h:
* runtime/ObjectConstructor.cpp:
(JSC::ObjectConstructor::ObjectConstructor):
* runtime/ObjectConstructor.h:
* runtime/ObjectPrototype.cpp:
(JSC::ObjectPrototype::ObjectPrototype):
* runtime/ObjectPrototype.h:
* runtime/PropertyNameArray.h:
(JSC::PropertyNameArrayData::setCachedPrototypeChain):
* runtime/PrototypeFunction.cpp:
(JSC::PrototypeFunction::PrototypeFunction):
* runtime/PrototypeFunction.h:
* runtime/RegExpConstructor.cpp:
(JSC::RegExpConstructor::RegExpConstructor):
* runtime/RegExpConstructor.h:
* runtime/RegExpObject.cpp:
(JSC::RegExpObject::RegExpObject):
* runtime/RegExpObject.h:
(JSC::RegExpObject::RegExpObjectData::RegExpObjectData):
* runtime/RegExpPrototype.cpp:
(JSC::RegExpPrototype::RegExpPrototype):
* runtime/RegExpPrototype.h:
* runtime/StringConstructor.cpp:
(JSC::StringConstructor::StringConstructor):
* runtime/StringConstructor.h:
* runtime/StringObject.cpp:
(JSC::StringObject::StringObject):
* runtime/StringObject.h:
* runtime/StringObjectThatMasqueradesAsUndefined.h:
(JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined):
* runtime/StringPrototype.cpp:
(JSC::StringPrototype::StringPrototype):
* runtime/StringPrototype.h:
* wtf/PassRefPtr.h:
(WTF::NotNullPassRefPtr::NotNullPassRefPtr):
(WTF::NotNullPassRefPtr::~NotNullPassRefPtr):
(WTF::NotNullPassRefPtr::get):
(WTF::NotNullPassRefPtr::clear):
(WTF::NotNullPassRefPtr::releaseRef):
(WTF::NotNullPassRefPtr::operator*):
(WTF::NotNullPassRefPtr::operator->):
(WTF::NotNullPassRefPtr::operator!):
(WTF::NotNullPassRefPtr::operator UnspecifiedBoolType):
* wtf/RefPtr.h:
(WTF::RefPtr::RefPtr):
(WTF::operator==):

WebCore:

Added NotNullPassRefPtr, and deployed it in all places that initialize
JavaScript objects.

* bindings/js/DOMObjectWithSVGContext.h:
(WebCore::DOMObjectWithSVGContext::DOMObjectWithSVGContext):
* bindings/js/JSDOMBinding.cpp:
(WebCore::cacheDOMStructure):
* bindings/js/JSDOMBinding.h:
(WebCore::DOMObject::DOMObject):
(WebCore::DOMObjectWithGlobalPointer::DOMObjectWithGlobalPointer):
(WebCore::DOMConstructorObject::DOMConstructorObject):
(WebCore::DOMConstructorWithDocument::DOMConstructorWithDocument):
* bindings/js/JSDOMGlobalObject.cpp:
(WebCore::JSDOMGlobalObject::JSDOMGlobalObject):
* bindings/js/JSDOMGlobalObject.h:
* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::JSDOMWindowBase):
* bindings/js/JSDOMWindowBase.h:
* bindings/js/JSHTMLAllCollection.h:
(WebCore::JSHTMLAllCollection::JSHTMLAllCollection):
* bindings/js/JSInspectedObjectWrapper.cpp:
(WebCore::JSInspectedObjectWrapper::JSInspectedObjectWrapper):
* bindings/js/JSInspectedObjectWrapper.h:
* bindings/js/JSInspectorCallbackWrapper.cpp:
(WebCore::JSInspectorCallbackWrapper::JSInspectorCallbackWrapper):
* bindings/js/JSInspectorCallbackWrapper.h:
* bindings/js/JSQuarantinedObjectWrapper.cpp:
(WebCore::JSQuarantinedObjectWrapper::JSQuarantinedObjectWrapper):
* bindings/js/JSQuarantinedObjectWrapper.h:
* bindings/js/JSWorkerContextBase.cpp:
(WebCore::JSWorkerContextBase::JSWorkerContextBase):
* bindings/js/JSWorkerContextBase.h:
* bindings/scripts/CodeGeneratorJS.pm:
* bridge/runtime_object.cpp:
(JSC::RuntimeObjectImp::RuntimeObjectImp):
* bridge/runtime_object.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48836 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Jakub Wieczorek <faw217@gmail.com>
eric@webkit.org [Mon, 28 Sep 2009 21:38:08 +0000 (21:38 +0000)]
2009-09-28  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Simon Hausmann.

        [Qt] Add the expected result for the fast/xsl/document-function test.
        https://bugs.webkit.org/show_bug.cgi?id=29817

        * platform/qt/fast/xsl/document-function-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48835 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Nate Chapin <japhet@chromium.org>
eric@webkit.org [Mon, 28 Sep 2009 21:30:53 +0000 (21:30 +0000)]
2009-09-28  Nate Chapin  <japhet@chromium.org>

        Reviewed by Dimitri Glazkov.

        Allow V8 to handle x/y parameters in a WebKitPoint constructor.
        https://bugs.webkit.org/show_bug.cgi?id=29823

        Fixes V8's handling of LayoutTests/fast/dom/Window/webkitConvertPoint.html

        * bindings/v8/custom/V8WebKitPointConstructor.cpp: Allow for x/y parameters in constructor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48834 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Eric Seidel <eric@webkit.org>
eric@webkit.org [Mon, 28 Sep 2009 21:22:50 +0000 (21:22 +0000)]
2009-09-25  Eric Seidel  <eric@webkit.org>

        Reviewed by Simon Fraser.

        REGRESSION: media/video-pause-empty-events.html is occasionally timing out on bots
        https://bugs.webkit.org/show_bug.cgi?id=28624

        No longer skip media tests now that we're disabling hardware compositing instead.
        Add leopard-specific results for 2 tests expected to fail with HW compositing disabled.

        * platform/mac-leopard/Skipped:
        * platform/mac-leopard/fast/media/mq-transform-02-expected.checksum: Copied from LayoutTests/platform/mac-tiger/fast/media/mq-transform-02-expected.checksum.
        * platform/mac-leopard/fast/media/mq-transform-02-expected.png: Copied from LayoutTests/platform/mac-tiger/fast/media/mq-transform-02-expected.png.
        * platform/mac-leopard/fast/media/mq-transform-02-expected.txt: Copied from LayoutTests/platform/mac-tiger/fast/media/mq-transform-02-expected.txt.
        * platform/mac-leopard/fast/media/mq-transform-03-expected.checksum: Copied from LayoutTests/platform/mac-tiger/fast/media/mq-transform-03-expected.checksum.
        * platform/mac-leopard/fast/media/mq-transform-03-expected.png: Copied from LayoutTests/platform/mac-tiger/fast/media/mq-transform-03-expected.png.
        * platform/mac-leopard/fast/media/mq-transform-03-expected.txt: Copied from LayoutTests/platform/mac-tiger/fast/media/mq-transform-03-expected.txt.
2009-09-27  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Darin Adler.

        run-webkit-tests: Do not strip the metrics when there is no expected result for a test.
        https://bugs.webkit.org/show_bug.cgi?id=29771

        * Scripts/run-webkit-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48833 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix the build by doing something approximating reasonableness in the Xcode project.
mrowe@apple.com [Mon, 28 Sep 2009 20:57:52 +0000 (20:57 +0000)]
Fix the build by doing something approximating reasonableness in the Xcode project.

* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48832 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/7240911> REGRESSION (r48586): Crash occurs when loading
mitz@apple.com [Mon, 28 Sep 2009 20:55:38 +0000 (20:55 +0000)]
<rdar://problem/7240911> REGRESSION (r48586): Crash occurs when loading
a PDF

Reviewed by Anders Carlsson.

CGPDFObjectRef is not a CFTypeRef, and cannot be retained or released.
Its lifetime is managed by its container. Just use a Vector to store
CGPDFObjectRefs, relying on the CGPDFDocument to keep them alive.

* WebView/WebPDFDocumentExtras.mm:
(appendValuesInPDFNameSubtreeToVector):
(getAllValuesInPDFNameTree):
(web_PDFDocumentAllScripts):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48831 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoHard dependency on SSE2 instruction set with JIT
oliver@apple.com [Mon, 28 Sep 2009 20:42:15 +0000 (20:42 +0000)]
Hard dependency on SSE2 instruction set with JIT
https://bugs.webkit.org/show_bug.cgi?id=29779

Reviewed by Geoff Garen

Add floating point support checks to op_jfalse and op_jtrue, and
fix the logic for the slow case of op_add

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48830 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoProspective Tiger build fix.
hausmann@webkit.org [Mon, 28 Sep 2009 20:41:56 +0000 (20:41 +0000)]
Prospective Tiger build fix.

Patch by Simon Hausmann <hausmann@webkit.org> on 2009-09-28
* WebCore.xcodeproj/project.pbxproj: Try to use unique IDs for the new files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48829 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 28 Sep 2009 20:33:41 +0000 (20:33 +0000)]
2009-09-28  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Add explicit frontend event for commit load.

        https://bugs.webkit.org/show_bug.cgi?id=29814

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::didCommitLoad):
        * inspector/InspectorFrontend.cpp:
        (WebCore::InspectorFrontend::didCommitLoad):
        * inspector/InspectorFrontend.h:
        * inspector/front-end/DOMAgent.js:
        (WebInspector.DOMAgent.prototype._setDocument):
        * inspector/front-end/inspector.js:
        (WebInspector.addResource):
        (WebInspector.didCommitLoad):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48828 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Yaar Schnitman <yaar@chromium.org>
eric@webkit.org [Mon, 28 Sep 2009 20:30:52 +0000 (20:30 +0000)]
2009-09-28  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Dimitri Glazkov.

        Chromium port - recognize we are being built independently
        of chromium and look for dependencies under webkit/chromium rather
        than chromium/src.

        https://bugs.webkit.org/show_bug.cgi?id=29722

        * JavaScriptCore.gyp/JavaScriptCore.gyp:
2009-09-28  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Dimitri Glazkov.

        Chromium port - recognize we are being built independently
        of chromium and look for dependencies under webkit/chromium rather
        than chromium/src.

        https://bugs.webkit.org/show_bug.cgi?id=29722

        * WebCore.gyp/WebCore.gyp: See above. Also removed a few files from
          the sources list, since they are not supposed to be built here.
2009-09-28  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Dimitri Glazkov.

        https://bugs.webkit.org/show_bug.cgi?id=29722

        * chromium/DEPS: Describes the chromium port's dependencies and
          is used by gclient to fetch them.
        * chromium/webkit.gyp: Currently only builds webcore but in
          the future will also build the webkit api.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48827 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Implement XSLT support with QtXmlPatterns.
hausmann@webkit.org [Mon, 28 Sep 2009 20:11:36 +0000 (20:11 +0000)]
[Qt] Implement XSLT support with QtXmlPatterns.
https://bugs.webkit.org/show_bug.cgi?id=28303

Patch by Jakub Wieczorek <faw217@gmail.com> on 2009-09-28
Reviewed by Simon Hausmann.

JavaScriptCore:

* wtf/Platform.h: Add a WTF_USE_QXMLQUERY #define.

WebCore:

* GNUmakefile.am:
* WebCore.gypi:
* WebCore.pro:
* WebCore.vcproj/WebCore.vcproj:
* WebCoreSources.bkl:
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::~Document):
(WebCore::Document::setTransformSource):
* dom/Document.h:
(WebCore::Document::transformSource):
* dom/TransformSource.h: Added.
(WebCore::TransformSource::platformSource):
* dom/TransformSourceLibxslt.cpp: Added.
(WebCore::TransformSource::TransformSource): Wraps a libxml2 document.
(WebCore::TransformSource::~TransformSource):
* dom/TransformSourceQt.cpp: Added.
(WebCore::TransformSource::TransformSource): Wraps a plain string.
(WebCore::TransformSource::~TransformSource):
* dom/XMLTokenizerLibxml2.cpp:
(WebCore::XMLTokenizer::doEnd):
* dom/XMLTokenizerQt.cpp:
(WebCore::XMLTokenizer::doEnd):
(WebCore::XMLTokenizer::parseProcessingInstruction):
* xml/XSLStyleSheet.h:
(WebCore::XSLStyleSheet::sheetString):
* xml/XSLStyleSheetLibxslt.cpp:
(WebCore::XSLStyleSheet::document):
* xml/XSLStyleSheetQt.cpp: Added.
* xml/XSLTProcessor.h:
* xml/XSLTProcessorLibxslt.cpp:
(WebCore::xmlDocPtrFromNode):
* xml/XSLTProcessorQt.cpp: Added.
(WebCore::XSLTMessageHandler::XSLTMessageHandler): A subclass of QAbstractMessageHandler.
(WebCore::XSLTMessageHandler::handleMessage): Forwards all processor messages to the Console.
(WebCore::XSLTProcessor::transformToString): Uses QXmlQuery.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48826 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Yongjun Zhang <yongjun.zhang@nokia.com>
eric@webkit.org [Mon, 28 Sep 2009 20:09:40 +0000 (20:09 +0000)]
2009-09-28  Yongjun Zhang  <yongjun.zhang@nokia.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=28054

        Use derefInNotNull() to work around winscw compiler forward declaration bug
        regarding templated classes.

        The compiler bug is reported at
        https://xdabug001.ext.nokia.com/bugzilla/show_bug.cgi?id=9812.

        The change should be reverted when the above bug is fixed in winscw compiler.

        Add parenthesis around (RefPtr::*UnspecifiedBoolType) to make winscw compiler
        work with the default UnSpecifiedBoolType() operator, which removes the winscw hack.

        * wtf/RefPtr.h:
        (WTF::RefPtr::~RefPtr):
        (WTF::RefPtr::clear):
        (WTF::RefPtr::operator UnspecifiedBoolType):
2009-09-28  Yongjun Zhang  <yongjun.zhang@nokia.com>

        Reviewed by Eric Seidel.

        Make JObjectWrapper::ref() and deref() public accessible to derefIfNull().

        Will be reverted when the following winscw compiler bug is fixed.
        https://xdabug001.ext.nokia.com/bugzilla/show_bug.cgi?id=9812

        * bridge/jni/jni_instance.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48825 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRemove __clear_cache which is an internal function of GCC
hausmann@webkit.org [Mon, 28 Sep 2009 18:57:31 +0000 (18:57 +0000)]
Remove __clear_cache which is an internal function of GCC
https://bugs.webkit.org/show_bug.cgi?id=28886

Patch by Gabor Loki <loki@inf.u-szeged.hu> on 2009-09-28
Reviewed by Simon Hausmann.

Although __clear_cache is exported from GCC, this is an internal
function. GCC makes no promises about it.

* jit/ExecutableAllocator.h:
(JSC::ExecutableAllocator::cacheFlush):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48824 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Shinichiro Hamaji <hamaji@chromium.org>
eric@webkit.org [Mon, 28 Sep 2009 18:51:10 +0000 (18:51 +0000)]
2009-09-28  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        Move message-port-multi.js from resources to script-tests
        https://bugs.webkit.org/show_bug.cgi?id=29798

        * fast/events/message-port-multi-expected.txt:
        * fast/events/message-port-multi.html:
        * fast/events/resources/message-port-multi.js: Removed.
        * fast/events/script-tests/message-port-multi.js: Added.
        (channel.port2.onmessage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48823 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Andrew Scherkus <scherkus@chromium.org>
eric@webkit.org [Mon, 28 Sep 2009 18:43:54 +0000 (18:43 +0000)]
2009-09-28  Andrew Scherkus  <scherkus@chromium.org>

        Reviewed by Eric Carlson.

        Delegate implementation of rendererIsNeeded() for media control elements to RenderTheme.

        https://bugs.webkit.org/show_bug.cgi?id=28689

        Covered by LayoutTests/media/video-no-audio.html

        * rendering/MediaControlElements.cpp: Removed subclass implementations of rendererIsNeeded().
        (WebCore::MediaControlElement::rendererIsNeeded): Include calling shouldRenderMediaControlPart().
        (WebCore::MediaControlInputElement::rendererIsNeeded): Include calling shouldRenderMediaControlPart().
        * rendering/MediaControlElements.h:
        * rendering/RenderTheme.cpp:
        (WebCore::RenderTheme::shouldRenderMediaControlPart): New method.  Contains logic moved from subclass implementations of rendererIsNeeded().
        * rendering/RenderTheme.h:
        * rendering/RenderThemeChromiumSkia.cpp:
        (WebCore::RenderThemeChromiumSkia::shouldRenderMediaControlPart): Override to always render mute button.
        * rendering/RenderThemeChromiumSkia.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48822 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] DragData::asURL should return file URL.
jianli@chromium.org [Mon, 28 Sep 2009 18:13:00 +0000 (18:13 +0000)]
[chromium] DragData::asURL should return file URL.
https://bugs.webkit.org/show_bug.cgi?id=29109

Reviewed by David Levin.

Tested by LayoutTests/fast/events/drag-to-navigate.html.

* platform/FileSystem.h:
* platform/chromium/ChromiumBridge.h:
* platform/chromium/DragDataChromium.cpp:
(WebCore::DragData::asURL):
* platform/chromium/FileSystemChromium.cpp:
(WebCore::getAbsolutePath):
(WebCore::isDirectory):
(WebCore::filePathToURL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48821 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 28 Sep 2009 17:57:17 +0000 (17:57 +0000)]
2009-09-28  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by David Levin.

        [V8] Add a special case to handle index arguments, because we need to be
        able to see if they're < 0.
        https://bugs.webkit.org/show_bug.cgi?id=29810

        Test: LayoutTests/dom/html/level1/core/hc_characterdataindexsizeerrdeletedatacountnegative.html

        * bindings/scripts/CodeGeneratorV8.pm: Added special case (matching CodeGeneratorJS.pm)
          for index arguments.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48820 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix an absolute path to somewhere in Oliver's machine to a relative path
weinig@apple.com [Mon, 28 Sep 2009 17:55:40 +0000 (17:55 +0000)]
Fix an absolute path to somewhere in Oliver's machine to a relative path
for derived JSONObject.lut.h.

Reviewed by Oliver Hunt.

* JavaScriptCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48819 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoExtend the MediaWiki/KHTMLFixes.css workaround to cover older MediaWiki versions
mitz@apple.com [Mon, 28 Sep 2009 16:34:53 +0000 (16:34 +0000)]
Extend the MediaWiki/KHTMLFixes.css workaround to cover older MediaWiki versions
https://bugs.webkit.org/show_bug.cgi?id=29792

Reviewed by Darin Adler.

* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::setCSSStyleSheet): If site specific hacks are
enabled, check if the linked style sheet is one of two versions of the
MediaWiki KHTMLFixes.css. If so, remove the offending rule.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48818 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 28 Sep 2009 16:29:14 +0000 (16:29 +0000)]
2009-09-28  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Adam Barth.

        [V8] Fix an error in type logic in CodeGeneratorV8.pm, where
        unsigned ints are accidentally used as signed.
        https://bugs.webkit.org/show_bug.cgi?id=29810

        Test: LayoutTests/fast/forms/textarea-maxlength.html

        * bindings/scripts/CodeGeneratorV8.pm: Made sure "unsigned long" in IDL is
          properly generates on "unsigned" return value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48817 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Mon, 28 Sep 2009 16:14:18 +0000 (16:14 +0000)]
2009-09-25  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Darin Adler.

        ASSERTION FAILED: !repaintContainer || repaintContainer == this
        https://bugs.webkit.org/show_bug.cgi?id=29755

        It's possible for RenderObject::container() to return an object that is higher
        in the hierarchy than the repaintContainer that is being used to repaint an
        object. For example, this can happen when running an accelerated opacity
        transition on an element with a position:absolute parent, and a position:absolute child.

        In this case we need to detect when RenderObject::container() will skip over
        repaintContainer, and in that case adjust the coords to shift into repaintContainer's space.

        Test: compositing/repaint/opacity-between-absolute.html

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::mapLocalToContainer):
        (WebCore::RenderBox::computeRectForRepaint):
        * rendering/RenderInline.cpp:
        (WebCore::RenderInline::computeRectForRepaint):
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::container):
        * rendering/RenderObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48816 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRename XSLStyleSheet.cpp to XSLStyleSheetLibxslt.cpp.
hausmann@webkit.org [Mon, 28 Sep 2009 15:06:24 +0000 (15:06 +0000)]
Rename XSLStyleSheet.cpp to XSLStyleSheetLibxslt.cpp.
https://bugs.webkit.org/show_bug.cgi?id=28303

Patch by Jakub Wieczorek <faw217@gmail.com> on 2009-09-28
Reviewed by Simon Hausmann.

In preparation for adding XSLT support with QtXmlPatterns to the Qt
port, rename XSLStyleSheet.cpp to XSLStyleSheetLibxslt.cpp as we will
have our own implementation which most likely will not share any code
with the current one.

Additionally, fix some coding style issues.

* GNUmakefile.am:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* WebCoreSources.bkl:
* xml/XSLStyleSheet.cpp: Renamed to XSLStyleSheetLibxslt.cpp.
* xml/XSLStyleSheetLibxslt.cpp: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48815 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoProspective build fix for r48812.
hausmann@webkit.org [Mon, 28 Sep 2009 14:45:12 +0000 (14:45 +0000)]
Prospective build fix for r48812.

Patch by Simon Hausmann <simon.hausmann@nokia.com> on 2009-09-28
Forgot to remove the inline reset() implementation when reverting the indentation
changes in the file.

* xml/XSLTProcessor.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48814 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 28 Sep 2009 14:40:17 +0000 (14:40 +0000)]
2009-09-28  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by nobody (trivial follow up fix), Joseph Pecoraro LGTM-ed.

        Web Inspector: Follow up to r48809.
        InspectorController.wrapObject should only be called on the inspectable page side.

        https://bugs.webkit.org/show_bug.cgi?id=17429

        * inspector/front-end/EventListenersSidebarPane.js:
        (WebInspector.EventListenersSidebarPane.prototype.update.callback):
        (WebInspector.EventListenersSidebarPane.prototype.update):
        (WebInspector.EventListenersSection):
        (WebInspector.EventListenersSection.prototype.update):
        (WebInspector.EventListenerBar):
        (WebInspector.EventListenerBar.prototype.update):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48813 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMove the libxslt specific part of XSLTProcessor to a separate file.
hausmann@webkit.org [Mon, 28 Sep 2009 14:36:24 +0000 (14:36 +0000)]
Move the libxslt specific part of XSLTProcessor to a separate file.
https://bugs.webkit.org/show_bug.cgi?id=28303

Patch by Jakub Wieczorek <faw217@gmail.com> on 2009-09-28
Reviewed by Simon Hausmann.

In preparation for adding XSLT support with QtXmlPatterns to the Qt
port, move the libxslt part of the XSLTProcessor implementation into
another file and leave the part that can be reused and shared.

Additionally, fix some coding style issues.

* GNUmakefile.am:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* WebCoreSources.bkl:
* xml/XSLTProcessor.cpp:
(WebCore::XSLTProcessor::createDocumentFromSource): Remove trailing whitespaces.
(WebCore::createFragmentFromSource): Remove trailing whitespaces.
(WebCore::XSLTProcessor::transformToFragment): Remove trailing whitespaces.
(WebCore::XSLTProcessor::reset): Moved out of the class definition.
* xml/XSLTProcessorLibxslt.cpp: Added.
(WebCore::XSLTProcessor::genericErrorFunc): Moved.
(WebCore::XSLTProcessor::parseErrorFunc): Moved.
(WebCore::docLoaderFunc): Moved.
(WebCore::setXSLTLoadCallBack): Moved.
(WebCore::writeToVector): Moved.
(WebCore::saveResultToString): Moved.
(WebCore::xsltParamArrayFromParameterMap): Moved.
(WebCore::freeXsltParamArray): Moved.
(WebCore::xsltStylesheetPointer): Moved.
(WebCore::xmlDocPtrFromNode): Moved.
(WebCore::resultMIMEType): Moved.
(WebCore::XSLTProcessor::transformToString): Moved.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48812 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMinor correction to Copyright notice. Not reviewed.
staikos@webkit.org [Mon, 28 Sep 2009 14:28:23 +0000 (14:28 +0000)]
Minor correction to Copyright notice.  Not reviewed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48811 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Charles Wei <charles.wei@torchmobile.com.cn>
eric@webkit.org [Mon, 28 Sep 2009 13:35:39 +0000 (13:35 +0000)]
2009-09-28  Charles Wei  <charles.wei@torchmobile.com.cn>

        Reviewed by Nikolas Zimmermann.

        Fix the crash of SVG that crashes when use a non-exist symbol
        https://bugs.webkit.org/show_bug.cgi?id=27693

        * svg/custom/use-non-existing-symbol-crash-expected.txt: Added.
        * svg/custom/use-non-existing-symbol-crash.svg: Added.
2009-09-28  Charles Wei  <charles.wei@torchmobile.com.cn>

        Reviewed by Nikolas Zimmermann.

        Fix the crash of SVG that crashes when use a non-exist symbol
        https://bugs.webkit.org/show_bug.cgi?id=27693

        Test: svg/custom/use-non-existing-symbol-crash.svg

        * svg/SVGUseElement.cpp:
        (WebCore::shadowTreeContainsChangedNodes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-28 Joseph Pecoraro <joepeck@webkit.org>
pfeldman@chromium.org [Mon, 28 Sep 2009 12:36:44 +0000 (12:36 +0000)]
2009-09-28  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by Timothy Hatcher.

        Inspector Should Show Event Listeners/Handlers Registered on each Node
        https://bugs.webkit.org/show_bug.cgi?id=17429

          Extracted a method from dispatchEvent to get the event ancestor chain

        * dom/Node.cpp:
        (WebCore::Node::eventAncestors): the extracted method
        (WebCore::Node::dispatchGenericEvent): use eventAncestors
        * dom/Node.h:

          Asynchronous Flow For the Inspector, Backend -> DOM Agent -> Frontend
          The DOMAgent's getEventListenersForNode handles the logic of finding
          all the relevant listeners in the event flow.

        * inspector/InspectorBackend.cpp:
        (WebCore::InspectorBackend::getEventListenersForNode):
        * inspector/InspectorBackend.h:
        * inspector/InspectorBackend.idl:
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::getEventListenersForNode):
        (WebCore::InspectorDOMAgent::buildObjectForEventListener):
        (WebCore::InspectorDOMAgent::buildObjectForNode): added localName
        * inspector/InspectorDOMAgent.h:
        (WebCore::EventListenerInfo::EventListenerInfo):
        * inspector/InspectorFrontend.cpp:
        (WebCore::InspectorFrontend::didGetEventListenersForNode):
        * inspector/InspectorFrontend.h:
        (WebCore::InspectorFrontend::scriptState):
        * inspector/front-end/DOMAgent.js: added localName to WebInspector.DOMNode from payload
        (WebInspector.EventListeners.getEventListenersForNodeAsync.mycallback):
        (WebInspector.EventListeners.getEventListenersForNodeAsync):

          New Sidebar Pane in the Element's Panel
          Includes Gear Menu for filtering the Event Listeners on the
          "Selected Node Only" or "All Nodes"

        * inspector/front-end/ElementsPanel.js: Handles refreshing the Pane when necessary
        (WebInspector.ElementsPanel.this.treeOutline.focusedNodeChanged):
        (WebInspector.ElementsPanel):
        (WebInspector.ElementsPanel.prototype.updateEventListeners):
        * inspector/front-end/EventListenersSidebarPane.js: Added.
        (WebInspector.EventListenersSidebarPane): The 1st level in the Pane
        (WebInspector.EventListenersSidebarPane.prototype.update.callback):
        (WebInspector.EventListenersSidebarPane.prototype.update):
        (WebInspector.EventListenersSection): The 2nd level in the Pane
        (WebInspector.EventListenersSection.prototype.update): filters on Preference
        (WebInspector.EventListenersSection.prototype.addListener):
        (WebInspector.EventListenerBar): The 3rd level in the Pane
        (WebInspector.EventListenerBar.prototype._getNodeDisplayName):
        (WebInspector.EventListenerBar.prototype._getFunctionDisplayName):
        (WebInspector.EventListenersSidebarPane.prototype._changeSetting): For the Gear Menu

          Consolidated "appropriateSelectorForNode"

        * inspector/front-end/StylesSidebarPane.js:
        * inspector/front-end/utilities.js:

          Miscellaneous Updates

        * English.lproj/localizedStrings.js: "Event Listeners", "No Event Listeners", "Selected Node Only", "All Nodes"
        * WebCore.gypi: included the new inspector files
        * WebCore.vcproj/WebCore.vcproj: included source files that were missing
        * inspector/front-end/Images/grayConnectorPoint.png: Added. Thanks to Timothy Hatcher.
        * inspector/front-end/Images/whiteConnectorPoint.png: Added. Thanks to Timothy Hatcher.
        * inspector/front-end/inspector.js: Preferences for the Gear Menu Event Listeners filter
        * inspector/front-end/inspector.css: reused as much as possible
        * inspector/front-end/inspector.html: include the new script
        * inspector/front-end/WebKit.qrc: included the new inspector files

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoCompile fix with namespaced Qt.
hausmann@webkit.org [Mon, 28 Sep 2009 11:16:17 +0000 (11:16 +0000)]
Compile fix with namespaced Qt.

Patch by Andre Poenitz <andre.poenitz@trolltech.com> on 2009-09-28
Reviewed by Simon Hausmann.

* Api/qwebinspector_p.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48808 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd ARM version detection for Windows CE.
hausmann@webkit.org [Mon, 28 Sep 2009 09:17:10 +0000 (09:17 +0000)]
Add ARM version detection for Windows CE.

Patch by Joerg Bornemann <joerg.bornemann@nokia.com> on 2009-09-28
Reviewed by Simon Hausmann.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=29760
weinig@apple.com [Sun, 27 Sep 2009 23:59:05 +0000 (23:59 +0000)]
WebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=29760
Implement CSSOM Range.getClientRects/getBoundingClientRect

Reviewed by Dan Bernstein.

Tests: fast/dom/Range/getBoundingClientRect-getClientRects-relative-to-viewport.html
       fast/dom/Range/getBoundingClientRect.html
       fast/dom/Range/getClientRects.html

* dom/Range.cpp:
(WebCore::Range::getClientRects):
(WebCore::Range::getBoundingClientRect):
(WebCore::adjustFloatQuadsForScrollAndAbsoluteZoom):
(WebCore::Range::getBorderAndTextQuads):
* dom/Range.h:
* dom/Range.idl:
Implement Range.getClientRects/getBoundingClientRect.

* dom/Element.cpp:
* rendering/RenderObject.h:
(WebCore::adjustForAbsoluteZoom):
(WebCore::adjustIntRectForAbsoluteZoom):
(WebCore::adjustFloatPointForAbsoluteZoom):
(WebCore::adjustFloatQuadForAbsoluteZoom):
Move point/quad adjustment methods from Element.cpp to RenderObject.h
so that Range.cpp can use them as well.

LayoutTests: Tests for https://bugs.webkit.org/show_bug.cgi?id=29760
Implement CSSOM Range.getClientRects/getBoundingClientRect

Reviewed by Dan Bernstein.

* fast/dom/Element/getBoundingClientRect-expected.txt: Copied from LayoutTests/fast/dom/getBoundingClientRect-expected.txt.
* fast/dom/Element/getBoundingClientRect-getClientRects-relative-to-viewport-expected.txt: Copied from LayoutTests/fast/dom/getBoundingClientRect-getClientRects-relative-to-viewport-expected.txt.
* fast/dom/Element/getBoundingClientRect-getClientRects-relative-to-viewport.html: Copied from LayoutTests/fast/dom/getBoundingClientRect-getClientRects-relative-to-viewport.html.
* fast/dom/Element/getBoundingClientRect.html: Copied from LayoutTests/fast/dom/getBoundingClientRect.html.
* fast/dom/Element/getClientRects-expected.txt: Copied from LayoutTests/fast/dom/getClientRects-expected.txt.
* fast/dom/Element/getClientRects.html: Copied from LayoutTests/fast/dom/getClientRects.html.
* fast/dom/Range/getBoundingClientRect-expected.txt: Added.
* fast/dom/Range/getBoundingClientRect-getClientRects-relative-to-viewport-expected.txt: Added.
* fast/dom/Range/getBoundingClientRect-getClientRects-relative-to-viewport.html: Added.
* fast/dom/Range/getBoundingClientRect.html: Added.
* fast/dom/Range/getClientRects-expected.txt: Added.
* fast/dom/Range/getClientRects.html: Added.
* fast/dom/Window/window-properties-expected.txt:
* fast/dom/getBoundingClientRect-expected.txt: Removed.
* fast/dom/getBoundingClientRect-getClientRects-relative-to-viewport-expected.txt: Removed.
* fast/dom/getBoundingClientRect-getClientRects-relative-to-viewport.html: Removed.
* fast/dom/getBoundingClientRect.html: Removed.
* fast/dom/getClientRects-expected.txt: Removed.
* fast/dom/getClientRects.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-27 Joe Ligman <joseph.ligman@mindspring.com>
eric@webkit.org [Sun, 27 Sep 2009 21:33:42 +0000 (21:33 +0000)]
2009-09-27  Joe Ligman  <joseph.ligman@mindspring.com>

        Reviewed by Simon Hausmann.

        [Qt] Adding API setFocus and hasFocus to QWebElement. This API is needed for
        clients that want to check/set the focus node of the document.
        https://bugs.webkit.org/show_bug.cgi?id=29682

        * Api/qwebelement.cpp:
        (QWebElement::hasFocus):
        (QWebElement::setFocus):
        * Api/qwebelement.h:
        * tests/qwebelement/tst_qwebelement.cpp:
        (tst_QWebElement::hasSetFocus):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed fix for WebInspector with Qt build.
hausmann@webkit.org [Sun, 27 Sep 2009 21:32:52 +0000 (21:32 +0000)]
Unreviewed fix for WebInspector with Qt build.

Patch by Simon Hausmann <hausmann@webkit.org> on 2009-09-27
Simply re-generate the Qt resource file by running
WebKitTools/Scripts/generate-qt-inspector-resource

* inspector/front-end/WebKit.qrc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-27 Jakub Wieczorek <faw217@gmail.com>
eric@webkit.org [Sun, 27 Sep 2009 19:45:25 +0000 (19:45 +0000)]
2009-09-27  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Darin Adler.

        run-webkit-tests: Do not strip the metrics when there is no expected result for a test.
        https://bugs.webkit.org/show_bug.cgi?id=29771

        * Scripts/run-webkit-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48803 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-27 Jakub Wieczorek <faw217@gmail.com>
eric@webkit.org [Sun, 27 Sep 2009 19:12:49 +0000 (19:12 +0000)]
2009-09-27  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Simon Hausmann.

        [Qt] Remove tests that are passing from the Skipped list after implementing
        the waitForPolicyDelegate() function in LayoutTestController.
        https://bugs.webkit.org/show_bug.cgi?id=25037

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-27 Jakub Wieczorek <faw217@gmail.com>
eric@webkit.org [Sun, 27 Sep 2009 19:05:44 +0000 (19:05 +0000)]
2009-09-27  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Simon Hausmann.

        [Qt] Implement layoutTestController.waitForPolicyDelegate.
        https://bugs.webkit.org/show_bug.cgi?id=25037

        * DumpRenderTree/qt/DumpRenderTree.cpp:
        (WebCore::WebPage::acceptNavigationRequest):
        * DumpRenderTree/qt/DumpRenderTree.pro:
        * DumpRenderTree/qt/jsobjects.cpp:
        (LayoutTestController::reset):
        (LayoutTestController::notifyDone):
        (LayoutTestController::waitForPolicyDelegate):
        * DumpRenderTree/qt/jsobjects.h:
        (LayoutTestController::waitForPolicy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-27 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sun, 27 Sep 2009 09:44:19 +0000 (09:44 +0000)]
2009-09-27  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by nobody (trivial ChangeLog fix).

        Restore WebCore/ChangeLog truncated in r48778.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-27 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sun, 27 Sep 2009 09:19:13 +0000 (09:19 +0000)]
2009-09-27  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: DOM store is being unbound twice, leading to assertion failure.

        https://bugs.webkit.org/show_bug.cgi?id=29770

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::didOpenDatabase):
        (WebCore::InspectorController::didUseDOMStorage):
        * inspector/InspectorDOMStorageResource.cpp:
        (WebCore::InspectorDOMStorageResource::unbind):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-26 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sun, 27 Sep 2009 09:14:19 +0000 (09:14 +0000)]
2009-09-26  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Do not track DOM changes while inspector window is closed.

        https://bugs.webkit.org/show_bug.cgi?id=29769

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::inspectedWindowScriptObjectCleared):
        (WebCore::InspectorController::populateScriptObjects):
        (WebCore::InspectorController::resetScriptObjects):
        (WebCore::InspectorController::didCommitLoad):
        * inspector/InspectorController.h:
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::setDocument):
        * inspector/InspectorDOMAgent.h:
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::dispatchWindowObjectAvailable):
        * page/android/InspectorControllerAndroid.cpp:
        (WebCore::InspectorController::inspectedWindowScriptObjectCleared):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-26 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sun, 27 Sep 2009 07:08:10 +0000 (07:08 +0000)]
2009-09-26  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: [REGRESSION] Double Clicking Resources Fails to Open in New Window

        https://bugs.webkit.org/show_bug.cgi?id=29762

        * inspector/front-end/InjectedScript.js:
        (InjectedScript.setStyleText):
        (InjectedScript.openInInspectedWindow):
        * inspector/front-end/InjectedScriptAccess.js:
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourceSidebarTreeElement.prototype.ondblclick):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago GTK BUILD FIX: add ENABLE_ORIENTATION_EVENTS support to configure.ac
ddkilzer@apple.com [Sun, 27 Sep 2009 03:15:37 +0000 (03:15 +0000)]
    GTK BUILD FIX: add ENABLE_ORIENTATION_EVENTS support to configure.ac

        * configure.ac: Added support for ENABLE_ORIENTATION_EVENTS.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Part 2 of 2: <http://webkit.org/b/29753> DerivedSources.make broken for non...
ddkilzer@apple.com [Sun, 27 Sep 2009 03:02:24 +0000 (03:02 +0000)]
    Part 2 of 2: <webkit.org/b/29753> DerivedSources.make broken for non-Mac targets

        Reviewed by Darin Adler.

        Fix ENABLE_ORIENTATION_EVENTS for non-Mac platforms.

WebCore:

        * DerivedSources.make: Moved Platform.h check for
        ENABLE_ORIENTATION_EVENTS into Mac-only section and added
        default of ENABLE_ORIENTATION_EVENTS = 0 to non-Mac section.
        Added ifndef test to make it possible to override both
        ENABLE_DASHBOARD_SUPPORT and ENABLE_ORIENTATION_EVENTS external
        to the makefile.  Moved addition of ENABLE_ORIENTATION_EVENTS to
        ADDITIONAL_IDL_DEFINES to common section.
        * GNUmakefile.am: Added support for ENABLE_ORIENTATION_EVENTS if
        it is ever used.
        * WebCore.pro: Ditto.

WebKit:

        * chromium/features.gypi: Added 'ENABLE_ORIENTATION_EVENTS=0' to
        disable this feature by default.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago <http://webkit.org/b/29764> mark-bug-fixed: add -o|--open switch
ddkilzer@apple.com [Sat, 26 Sep 2009 23:14:21 +0000 (23:14 +0000)]
    <webkit.org/b/29764> mark-bug-fixed: add -o|--open switch

        Reviewed by Eric Seidel.

        The -o|--open switch uses the open(1) command on Mac OS X to
        open the bug URL in the default web browser.  If there are
        similar mechanisms on other platforms, they may be added later.

        * Scripts/mark-bug-fixed:
        (MarkBugFixed.__init__): Added -o|--open switch to list of parse
        options.
        (MarkBugFixed._determine_bug_id_and_svn_revision): Moved logging
        code into main() and extracted prompting code into
        _prompt_user_for_correctness().
        (MarkBugFixed._open_bug_in_web_browser): Added.
        (MarkBugFixed._prompt_user_for_correctness): Added.
        (MarkBugFixed.main): Added logging code from
        _determine_bug_id_and_svn_revision().  Added code to call
        _open_bug_in_web_browser() if the switch is set.  Added code to
        call _prompt_user_for_correctness() when needed.
        * Scripts/modules/bugzilla.py:
        (Bugzilla.short_bug_url_for_bug_id): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48794 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago svn-unapply and svn-apply don't work when used outside multiple svn working...
ddkilzer@apple.com [Sat, 26 Sep 2009 22:57:25 +0000 (22:57 +0000)]
    svn-unapply and svn-apply don't work when used outside multiple svn working directories

        <http://webkit.org/b/29744>
        <rdar://problem/7252905>

        Reviewed by Eric Seidel.

        Some users have a workflow where svn-create-patch, svn-apply and
        svn-unapply are used outside of multiple svn working
        directories.  Instead of aborting the scripts in these cases,
        print a warning and assume that Subversion is being used.

        * Scripts/VCSUtils.pm:
        (determineVCSRoot): Call warn() instead of die() if both isGit()
        and isSVN() initially return false.  Set $VCSUtils::isSVN to 1
        to enforce the assumption about Subversion, then return
        determineSVNRoot().
        * Scripts/svn-apply: Switch to using isGit() and isSVN() from
        VCSUtils.pm.  They both already cache their values and checking
        here is redundant since determineVCSRoot() is called later.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48793 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-26 Kent Tamura <tkent@chromium.org>
eric@webkit.org [Sat, 26 Sep 2009 18:56:04 +0000 (18:56 +0000)]
2009-09-26  Kent Tamura  <tkent@chromium.org>

        Reviewed by David Kilzer.

        Move placeholder-related code to HTMLTextFormControlElement from
        HTMLInputElement, WMLInputElement, InputElement, and
        HTMLTextAreaElement.
        https://bugs.webkit.org/show_bug.cgi?id=28703

        * dom/InputElement.cpp:
        (WebCore::InputElement::dispatchFocusEvent):
        (WebCore::InputElement::dispatchBlurEvent):
        (WebCore::InputElement::setValueFromRenderer):
        * dom/InputElement.h:
        * html/HTMLFormControlElement.cpp:
        (WebCore::HTMLTextFormControlElement::HTMLTextFormControlElement):
        (WebCore::HTMLTextFormControlElement::~HTMLTextFormControlElement):
        (WebCore::HTMLTextFormControlElement::dispatchFocusEvent):
        (WebCore::HTMLTextFormControlElement::dispatchBlurEvent):
        (WebCore::HTMLTextFormControlElement::placeholderShouldBeVisible):
        (WebCore::HTMLTextFormControlElement::updatePlaceholderVisibility):
        * html/HTMLFormControlElement.h:
        (WebCore::HTMLTextFormControlElement::handleFocusEvent):
        (WebCore::HTMLTextFormControlElement::handleBlurEvent):
        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::HTMLInputElement):
        (WebCore::HTMLInputElement::handleFocusEvent):
        (WebCore::HTMLInputElement::handleBlurEvent):
        (WebCore::HTMLInputElement::parseMappedAttribute):
        (WebCore::HTMLInputElement::createRenderer):
        (WebCore::HTMLInputElement::setValue):
        (WebCore::HTMLInputElement::setValueFromRenderer):
        * html/HTMLInputElement.h:
        (WebCore::HTMLInputElement::supportsPlaceholder):
        (WebCore::HTMLInputElement::isEmptyValue):
        * html/HTMLIsIndexElement.cpp:
        (WebCore::HTMLIsIndexElement::parseMappedAttribute):
        * html/HTMLTextAreaElement.cpp:
        (WebCore::HTMLTextAreaElement::HTMLTextAreaElement):
        (WebCore::HTMLTextAreaElement::createRenderer):
        * html/HTMLTextAreaElement.h:
        (WebCore::HTMLTextAreaElement::supportsPlaceholder):
        (WebCore::HTMLTextAreaElement::isEmptyValue):
        * rendering/RenderTextControl.cpp:
        (WebCore::RenderTextControl::RenderTextControl):
        * rendering/RenderTextControl.h:
        * rendering/RenderTextControlMultiLine.cpp:
        (WebCore::RenderTextControlMultiLine::RenderTextControlMultiLine):
        * rendering/RenderTextControlMultiLine.h:
        * rendering/RenderTextControlSingleLine.cpp:
        (WebCore::RenderTextControlSingleLine::RenderTextControlSingleLine):
        (WebCore::RenderTextControlSingleLine::updateFromElement):
        * rendering/RenderTextControlSingleLine.h:
        * wml/WMLInputElement.cpp:
        (WebCore::WMLInputElement::setValue):
        (WebCore::WMLInputElement::createRenderer):
        * wml/WMLInputElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48792 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-26 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Sat, 26 Sep 2009 17:51:12 +0000 (17:51 +0000)]
2009-09-26  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        [Layout tests] [Gtk] Gtk DumpRenderTree should use WebKit test fonts
        https://bugs.webkit.org/show_bug.cgi?id=29689

        Load test fonts through FontConfig before each test.
        This ensures a more proper rendering of the tests.

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (initializeFonts):
        (runTest):
        (main):
        * DumpRenderTree/gtk/fonts.conf: Copied from WebKitTools/DumpRenderTree/qt/fonts.conf.
        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-26 Shu Chang <Chang.Shu@nokia.com>
eric@webkit.org [Sat, 26 Sep 2009 15:00:27 +0000 (15:00 +0000)]
2009-09-26  Shu Chang  <Chang.Shu@nokia.com>

        Reviewed by Alexey Proskuryakov.

        Optimize the code so only the text from start to end is scanned.
        https://bugs.webkit.org/show_bug.cgi?id=29092

        On a platform with webkit+Qt+Symbian, the parsing time for a 600K text
        file improved from 400ms to 40ms (10x faster).

        * dom/Text.cpp:
        (WebCore::Text::createWithLengthLimit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48790 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-26 Xiaomei Ji <xji@chromium.org>
eric@webkit.org [Sat, 26 Sep 2009 13:12:20 +0000 (13:12 +0000)]
2009-09-26  Xiaomei Ji  <xji@chromium.org>

        Reviewed by Eric Seidel.

        This Patch fixes [chromium] the drop-down is always left-aligned even
        for RTL element.
        https://bugs.webkit.org/show_bug.cgi?id=29612

        For auto-complete, the items in drop-down should be right-aligned if
        the directionality of <input> field is RTL.
        For <select><option>, the items in drop-down should be right-aligned
        if the directionality of <select> is RTL.

        No automatic test is possible. Manual tests are added.

        * manual-tests/autofill_alignment.html: Added.
        * manual-tests/select_alignment.html: Added.
        * platform/chromium/PopupMenuChromium.cpp:
        (WebCore::PopupListBox::paintRow): Adjust the starting x-axis of text to
        be paint if it should be right-aligned.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd MarkStackSymbian.cpp to build JavascriptCore for Symbian.
hausmann@webkit.org [Sat, 26 Sep 2009 12:47:01 +0000 (12:47 +0000)]
Add MarkStackSymbian.cpp to build JavascriptCore for Symbian.

Patch by Yongjun Zhang <yongjun.zhang@nokia.com> on 2009-09-26
Reviewed by Simon Hausmann.

Re-use Windows shrinkAllocation implementation because Symbian doesn't
support releasing part of memory region.

Use fastMalloc and fastFree to implement allocateStack and releaseStack
for Symbian port.

* JavaScriptCore.pri:
* runtime/MarkStack.h:
(JSC::MarkStack::MarkStackArray::shrinkAllocation):
* runtime/MarkStackSymbian.cpp: Added.
(JSC::MarkStack::initializePagesize):
(JSC::MarkStack::allocateStack):
(JSC::MarkStack::releaseStack):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48788 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoREGRESSION (r48775) FontList.plist written by TOT WebKit causes Safari 4
mitz@apple.com [Sat, 26 Sep 2009 06:43:25 +0000 (06:43 +0000)]
REGRESSION (r48775) FontList.plist written by TOT WebKit causes Safari 4
to crash on launch
https://bugs.webkit.org/show_bug.cgi?id=29759

Reviewed by Sam Weinig.

* platform/graphics/win/FontDatabase.cpp:
(WebCore::writeFontDatabaseToPlist): Reverted to saving the CG font DB
property list at the root of FontList.plist, but with an additional
key for the last value of the Fonts registry key.
(WebCore::populateFontDatabase): Pass the FontList.plist in its entirety
to populatFontDatabaseFromPlist.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agowx build fix. Add EmptyPluginHalterClient when creating Page.
kevino@webkit.org [Sat, 26 Sep 2009 06:41:34 +0000 (06:41 +0000)]
wx build fix. Add EmptyPluginHalterClient when creating Page.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoBuild fix. Adding missing header files.
kevino@webkit.org [Sat, 26 Sep 2009 06:31:43 +0000 (06:31 +0000)]
Build fix. Adding missing header files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Part 1 of 2: <http://webkit.org/b/29753> DerivedSources.make broken for non...
ddkilzer@apple.com [Sat, 26 Sep 2009 04:32:58 +0000 (04:32 +0000)]
    Part 1 of 2: <webkit.org/b/29753> DerivedSources.make broken for non-Mac targets

        Reviewed by Darin Adler.

        * DerivedSources.make: Move tests for ENABLE_CONTEXT_MENUS,
        ENABLE_DRAG_SUPPORT and ENABLE_INSPECTOR into Mac-only section.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Adam Barth <abarth@webkit.org>
eric@webkit.org [Sat, 26 Sep 2009 02:35:15 +0000 (02:35 +0000)]
2009-09-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Load blocks during unload should not affect targeted loads
        https://bugs.webkit.org/show_bug.cgi?id=29747

        Test that we can follow targeted hyperlinks in an unload handler.

        * fast/loader/resources/unload-hyperlink-targeted-subframe.html: Added.
        * fast/loader/unload-hyperlink-targeted-expected.txt: Added.
        * fast/loader/unload-hyperlink-targeted.html: Added.
2009-09-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Load blocks during unload should not affect targeted loads
        https://bugs.webkit.org/show_bug.cgi?id=29747

        Move the check of the unload state after checking for targeted links.

        Test: fast/loader/unload-hyperlink-targeted.html

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::loadURL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Gabor Loki <loki@inf.u-szeged.hu>
eric@webkit.org [Sat, 26 Sep 2009 02:27:02 +0000 (02:27 +0000)]
2009-09-25  Gabor Loki  <loki@inf.u-szeged.hu>

        Reviewed by Gavin Barraclough.

        Fix unaligned data access in YARR_JIT on ARMv5 and below.
        https://bugs.webkit.org/show_bug.cgi?id=29695

        On ARMv5 and below all data access should be naturally aligned.
        In the YARR_JIT there is a case when character pairs are
        loaded from the input string, but this data access is not
        naturally aligned. This fix introduces load32WithUnalignedHalfWords
        and branch32WithUnalignedHalfWords functions which contain
        naturally aligned memory loads - half word loads - on ARMv5 and below.

        * assembler/MacroAssemblerARM.cpp:
        (JSC::MacroAssemblerARM::load32WithUnalignedHalfWords):
        * assembler/MacroAssemblerARM.h:
        (JSC::MacroAssemblerARM::load32WithUnalignedHalfWords):
        (JSC::MacroAssemblerARM::branch32WithUnalignedHalfWords):
        * assembler/MacroAssemblerARMv7.h:
        (JSC::MacroAssemblerARMv7::load32WithUnalignedHalfWords):
        (JSC::MacroAssemblerARMv7::branch32):
        (JSC::MacroAssemblerARMv7::branch32WithUnalignedHalfWords):
        * assembler/MacroAssemblerX86Common.h:
        (JSC::MacroAssemblerX86Common::load32WithUnalignedHalfWords):
        (JSC::MacroAssemblerX86Common::branch32WithUnalignedHalfWords):
        * wtf/Platform.h:
        * yarr/RegexJIT.cpp:
        (JSC::Yarr::RegexGenerator::generatePatternCharacterPair):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Kenneth Russell <kbr@google.com>
eric@webkit.org [Sat, 26 Sep 2009 02:13:16 +0000 (02:13 +0000)]
2009-09-25  Kenneth Russell  <kbr@google.com>

        Reviewed by Dimitri Glazkov.

        [Chromium] Add initial V8 bindings for WebGL
        https://bugs.webkit.org/show_bug.cgi?id=29664

        * WebCore.gypi:
        * bindings/scripts/CodeGeneratorV8.pm:
        * bindings/v8/DOMObjectsInclude.h:
        * bindings/v8/DerivedSourcesAllInOne.cpp:
        * bindings/v8/V8DOMWrapper.cpp:
        (WebCore::V8DOMWrapper::getTemplate):
        * bindings/v8/V8Index.cpp:
        * bindings/v8/V8Index.h:
        * bindings/v8/custom/V8CanvasArrayBufferCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        * bindings/v8/custom/V8CanvasArrayCustom.h: Added.
        (WebCore::constructCanvasArray):
        * bindings/v8/custom/V8CanvasByteArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CanvasFloatArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CanvasIntArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CanvasRenderingContext3DCustom.cpp: Added.
        (WebCore::jsArrayToFloatArray):
        (WebCore::jsArrayToIntArray):
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::):
        (WebCore::vertexAttribAndUniformHelperf):
        (WebCore::uniformHelperi):
        (WebCore::uniformMatrixHelper):
        * bindings/v8/custom/V8CanvasShortArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CanvasUnsignedByteArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CanvasUnsignedIntArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CanvasUnsignedShortArrayCustom.cpp: Added.
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_PROPERTY_GETTER):
        (WebCore::INDEXED_PROPERTY_SETTER):
        * bindings/v8/custom/V8CustomBinding.h:
        * bindings/v8/custom/V8DocumentCustom.cpp:
        (WebCore::CALLBACK_FUNC_DECL):
        * bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
        (WebCore::CALLBACK_FUNC_DECL):
        * platform/graphics/GraphicsContext3D.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48781 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: This is breaking Chromium try bots, so I'm counting this as a build...
jorlow@chromium.org [Sat, 26 Sep 2009 01:45:44 +0000 (01:45 +0000)]
JavaScriptCore: This is breaking Chromium try bots, so I'm counting this as a build fix.

Patch by Jeremy Orlow <jorlow@chromium.org> on 2009-09-25
Add more svn:ignore exceptions.  On different platforms, these files are
generated with different case for JavaScriptCore.  Also there are some
wtf project files that get built apparently.

* JavaScriptCore.gyp: Changed property svn:ignore.

WebCore: This is breaking Chromium try bots, so I'm counting this as a build fix.

Patch by Jeremy Orlow <jorlow@chromium.org> on 2009-09-25
Add more svn:ignore exceptions.  On different platforms, these files are
generated with different case for WebCore.

* WebCore.gyp: Changed property svn:ignore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Build fix.
adachan@apple.com [Sat, 26 Sep 2009 00:57:12 +0000 (00:57 +0000)]
    Build fix.

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reverting r48767, as it broke Windows build in a non-trivial way.
ap@apple.com [Sat, 26 Sep 2009 00:21:50 +0000 (00:21 +0000)]
    Reverting r48767, as it broke Windows build in a non-trivial way.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-24 Tony Chang <tony@chromium.org>
agl@chromium.org [Fri, 25 Sep 2009 23:18:55 +0000 (23:18 +0000)]
2009-09-24  Tony Chang  <tony@chromium.org>

        Reviewed by David Levin.

        Add a gyp variable to allow building a debug webcore without debug
        symbols.  This allows for faster compile, link, and gdb times.

        https://bugs.webkit.org/show_bug.cgi?id=29721

        No new tests, build config change.

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Darin Fisher <darin@chromium.org>
darin@chromium.org [Fri, 25 Sep 2009 23:11:46 +0000 (23:11 +0000)]
2009-09-25  Darin Fisher  <darin@chromium.org>

        Reviewed by Dimitri Glazkov.

        Declare RegisteredEventListener as a class instead of a struct.
        This fixes a warning in the Chromium build.

        * dom/RegisteredEventListener.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: WebCore part of
mitz@apple.com [Fri, 25 Sep 2009 23:07:21 +0000 (23:07 +0000)]
WebCore: WebCore part of
<rdar://problem/7211635> 2 byte characters are displayed as garbaged
<rdar://problem/7212626> garbled/gibberish text (off-by-one)

Reviewed by Jon Honeycutt.

When the Windows Fonts directory contains more than one font file for a
given font name, which of the fonts gets assigned to the name in the
Core Graphics font database was determined arbitrarily and did not
always match the font GDI used for the same font name. The mismatch
caused character-to-glyph mapping to use one font and glyph rendering to
use another.

The fix is to update the Core Graphics font database from the registry
entries (that reflect the name-to-font mapping that GDI uses) after
populating it with the result of scanning the Fonts directory. As a
consequence, the directory needs to be scanned at startup every time the
registry key changes, so the last value of the registry key is kept
in the property list on disk so that it could be compared to the current
value on startup.

* platform/graphics/win/FontDatabase.cpp:
(WebCore::populateFontDatabaseFromPlist): Now takes a property list as
a parameter and avoids round-tripping through XML by calling
wkAddFontsFromPlist() instead of wkAddFontsFromPlistRepresentation().
(WebCore::fontFilenamesFromRegistryKey):
(WebCore::cgFontDBKey):
(WebCore::writeFontDatabaseToPlist): Now takes the CG font DB property
list and a property list with the font filenames from the registry and
writes a dictionary with those property lists as values.
(WebCore::fontFilenamesFromRegistry): Added. Returns an array with the
values in the HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Fonts
registry key.
(WebCore::populateFontDatabase): Changed to read the contents of the
Fonts registry key and compare it with the last-saved value from the
property list, and to call wkAddFontsFromRegistry() after populating the
CG font DB from the file system. Uses wkCreateFontsPlist() instead of
wkCreateFontsPlistRepresentation() to avoid round-tripping through XML.

WebKitLibraries: WebKitSystemInterface changes for
<rdar://problem/7211635> 2 byte characters are displayed as garbaged
<rdar://problem/7212626> garbled/gibberish text (off-by-one)

Reviewed by Adam Roben.

* win/include/WebKitSystemInterface/WebKitSystemInterface.h: Added
wkAddFontsFromPlistRepresentation() and replaced
wkCreateFontsPlistRepresentation() with wkCreateFontsPlist() and
wkAddFontsFromPlistRepresentation() with wkAddFontsFromPlist().
* win/lib/WebKitSystemInterface.lib:
* win/lib/WebKitSystemInterface_debug.lib:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: Inlined some object creation code, including lexicalGlobalObject...
ggaren@apple.com [Fri, 25 Sep 2009 22:26:44 +0000 (22:26 +0000)]
JavaScriptCore: Inlined some object creation code, including lexicalGlobalObject access
https://bugs.webkit.org/show_bug.cgi?id=29750

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-09-25
Reviewed by Darin Adler.

SunSpider says 0.5% faster.

0.8% speedup on bench-alloc-nonretained.js.
2.5% speedup on v8-splay.js.

* interpreter/CachedCall.h:
(JSC::CachedCall::CachedCall):
* interpreter/CallFrame.h:
(JSC::ExecState::lexicalGlobalObject):
(JSC::ExecState::globalThisValue):
* interpreter/Interpreter.cpp:
(JSC::Interpreter::dumpRegisters):
(JSC::Interpreter::execute):
(JSC::Interpreter::privateExecute):
* jit/JITStubs.cpp:
(JSC::DEFINE_STUB_FUNCTION):
* runtime/FunctionConstructor.cpp:
(JSC::constructFunction):
* runtime/ScopeChain.cpp:
(JSC::ScopeChainNode::print):
* runtime/ScopeChain.h:
(JSC::ScopeChainNode::ScopeChainNode):
(JSC::ScopeChainNode::~ScopeChainNode):
(JSC::ScopeChainNode::push):
(JSC::ScopeChain::ScopeChain):
(JSC::ScopeChain::globalObject): Added a globalObject data member to ScopeChainNode.
Replaced accessor function for globalObject() with data member. Replaced
globalThisObject() accessor with direct access to globalThis, to match.

* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::init):
* runtime/JSGlobalObject.h: Inlined array and object construction.

WebCore: Inlined some object creation code, including lexicalGlobalObject access
https://bugs.webkit.org/show_bug.cgi?id=29750

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-09-25
Reviewed by Darin Adler.

* bindings/js/JSInspectorBackendCustom.cpp:
(WebCore::JSInspectorBackend::currentCallFrame):
* inspector/JavaScriptDebugServer.cpp:
(WebCore::JavaScriptDebugServer::hasBreakpoint): Updated for JavaScriptCore
API changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=24399
hyatt@apple.com [Fri, 25 Sep 2009 22:24:45 +0000 (22:24 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=24399
Make @import work in user stylesheets.  The first bug was that the URL wasn't being set on the
user sheets themselves, so relative @import URLs couldn't resolve properly.  The second bug
was that the loads would be denied.  This is fixed by using the requestUserCSSStyleSheet method
instead of the normal request method.  In order to know when to do this, CSSStyleSheets now have
a propagated boolean, m_isUserStyleSheet, that lets them know if they are user stylesheets or not.

Reviewed by Anders Carlsson.

* css/CSSImportRule.cpp:
(WebCore::CSSImportRule::insertedIntoParent):
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::CSSStyleSheet):
* css/CSSStyleSheet.h:
(WebCore::CSSStyleSheet::setIsUserStyleSheet):
(WebCore::CSSStyleSheet::isUserStyleSheet):
* dom/Document.cpp:
(WebCore::Document::pageUserSheet):
(WebCore::Document::pageGroupUserSheets):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago <http://webkit.org/b/29718> mark-bug-fixed: add -u|--update-only switch
ddkilzer@apple.com [Fri, 25 Sep 2009 22:02:38 +0000 (22:02 +0000)]
    <webkit.org/b/29718> mark-bug-fixed: add -u|--update-only switch

        Reviewed by Eric Seidel.

        * Scripts/mark-bug-fixed:
        (MarkBugFixed.__init__): Added -u|--update-only switch to list
        of parse options.
        (MarkBugFixed.main): When -u|--update-only is specified, add a
        comment to the bug without marking it Resolved/Fixed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Fri, 25 Sep 2009 22:01:43 +0000 (22:01 +0000)]
2009-09-25  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Darin Adler.

        Crash with hardware accelerated rotation of a PDF image in a data URL
        <rdar://problem/7250378>

        PDF images don't return a color space from CGImageGetColorSpace(),
        so we need to null-check the return value before use.

        Test: compositing/color-matching/pdf-image-match.html

        * platform/graphics/mac/GraphicsLayerCA.mm:
        (WebCore::GraphicsLayerCA::setContentsToImage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSpeed up access to history items by caching date computation.
darin@apple.com [Fri, 25 Sep 2009 21:40:50 +0000 (21:40 +0000)]
Speed up access to history items by caching date computation.

Patch by Darin Adler <darin@apple.com> on 2009-09-25
Reviewed by Steve Falkenburg.

* History/WebHistory.mm:
(getDayBoundaries): Refactored from timeIntervalForBeginningOfDay.
Returns the beginning of the day that the passed time is within and
the beginning of the next day.
(beginningOfDay): Added. Uses getDayBoundaries so it can be fast for
multiple dates within the same day, which is the common case.
(dateKey): Added. Calls beginningOfDay and converts to an integer.
(WebHistory::findKey): Changed to call dateKey instead of
timeIntervalForBeginningOfDay.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNull-deref when first access to an Attr node is after its Element is destroyed
darin@apple.com [Fri, 25 Sep 2009 21:08:17 +0000 (21:08 +0000)]
Null-deref when first access to an Attr node is after its Element is destroyed
https://bugs.webkit.org/show_bug.cgi?id=29748

Patch by Darin Adler <darin@apple.com> on 2009-09-25
Reviewed by Geoffrey Garen.

WebCore:

Test: fast/dom/Attr/access-after-element-destruction.html

* bindings/js/JSAttrCustom.cpp:
(WebCore::JSAttr::markChildren): Added. Keeps the ownerElement alive as
long as the Attr is alive.

* bindings/js/JSNamedNodeMapCustom.cpp:
(WebCore::JSNamedNodeMap::markChildren): Added. Keeps the Element alive as
long as the NamedNodeMap is alive.

* dom/Attr.idl: Added CustomMarkFunction attribute.

* dom/NamedAttrMap.cpp:
(WebCore::NamedNodeMap::getAttributeItem): Tweaked formatting.
(WebCore::NamedNodeMap::detachFromElement): Call clearAttributes so we don't
have attributes hanging around that might need an Attr node created; that way
we won't crash with a null-dereference trying to deal with one of them. This
can't happen when working with JavaScript since the Element will be kept
alive due to the change above.
(WebCore::NamedNodeMap::addAttribute): Fix function name in comment.
(WebCore::NamedNodeMap::removeAttribute): Removed unneeded "+ 1" and added
missing braces.

* dom/NamedAttrMap.h: Made the element function public so it can be used by
the JavaScript binding to keep the Element alive.

* dom/NamedNodeMap.idl: Added CustomMarkFunction attribute.

LayoutTests:

* fast/dom/Attr/access-after-element-destruction-expected.txt: Added.
* fast/dom/Attr/access-after-element-destruction.html: Added.
* fast/dom/Attr/script-tests/TEMPLATE.html: Copied from LayoutTests/fast/dom/Node/script-tests/TEMPLATE.html.
* fast/dom/Attr/script-tests/access-after-element-destruction.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago* Scripts/prepare-ChangeLog: Leave files from the script-tests directory
darin@apple.com [Fri, 25 Sep 2009 21:07:59 +0000 (21:07 +0000)]
* Scripts/prepare-ChangeLog: Leave files from the script-tests directory
out, just as we do for the resources directory.

Reviewed by Geoffrey Garen.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Darin Adler and Sam Weinig.
ap@apple.com [Fri, 25 Sep 2009 20:44:11 +0000 (20:44 +0000)]
    Reviewed by Darin Adler and Sam Weinig.

        Onclick not fired for an element copied with cloneContents() or cloneNode()
        https://bugs.webkit.org/show_bug.cgi?id=25130

        The change here is that JS event listeners don't keep a reference to a global object from
        where they were created, and instead take it as a parameter when parsing source code. Also,
        the listener creation won't fail just because it happens for an element in a frameless
        document.
        Thus, moving nodes between documents no longer results in having incorrect registered
        lazy event listeners on them.

        Tests: fast/events/attribute-listener-cloned-from-frameless-doc-context-2.html
               fast/events/attribute-listener-cloned-from-frameless-doc-context.html
               fast/events/attribute-listener-cloned-from-frameless-doc.xhtml
               fast/events/attribute-listener-extracted-from-frameless-doc-context-2.html
               fast/events/attribute-listener-extracted-from-frameless-doc-context.html

        * bindings/js/JSEventListener.cpp:
        (WebCore::JSEventListener::JSEventListener): Don't take a reference to JSDOMGlobalObject.
        (WebCore::JSEventListener::jsFunction): Take ScriptExecutionContext as a parameter for
        getting to JSDOMGlobalObject. It's not used in base class, but is in JSLazyEventListner.
        (WebCore::JSEventListener::markJSFunction): Don't mark the global object.
        (WebCore::JSEventListener::handleEvent): Get global object from ScriptExecutionContext.
        (WebCore::JSEventListener::reportError): Ditto.

        * bindings/js/JSEventListener.h: (WebCore::JSEventListener::create): Don't keep a reference
        to JSDOMGlobalObject.

        * bindings/js/JSLazyEventListener.cpp: (WebCore::JSLazyEventListener::parseCode): Listener
        creation was split between this function and ScriptEventListener; moved it here, as JS
        global object can be different now.

        * bindings/js/JSLazyEventListener.h: (WebCore::JSLazyEventListener::create): Keep source URL,
        which can not be determined at parsing time.

        * bindings/js/ScriptEventListener.cpp: (WebCore::createAttributeEventListener): Moved code
        for listener creation to JSLazyEventListener. XSSAuditor code remains here, because tests
        expect that errors are logged at document parsing time, and because I don't know what other
        side effects moving it vould have.

        * dom/EventListener.h: handleEvent() and reportError() now take ScriptExecutionContext,
        because JSC needs a global context here.

        * bindings/js/JSAbstractWorkerCustom.cpp:
        (WebCore::JSAbstractWorker::addEventListener):
        (WebCore::JSAbstractWorker::removeEventListener):
        * bindings/js/JSDOMApplicationCacheCustom.cpp:
        (WebCore::JSDOMApplicationCache::addEventListener):
        (WebCore::JSDOMApplicationCache::removeEventListener):
        * bindings/js/JSDOMGlobalObject.cpp:
        (WebCore::JSDOMGlobalObject::createJSAttributeEventListener):
        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::addEventListener):
        (WebCore::JSDOMWindow::removeEventListener):
        * bindings/js/JSEventSourceCustom.cpp:
        (WebCore::JSEventSource::addEventListener):
        (WebCore::JSEventSource::removeEventListener):
        * bindings/js/JSMessagePortCustom.cpp:
        (WebCore::JSMessagePort::addEventListener):
        (WebCore::JSMessagePort::removeEventListener):
        * bindings/js/JSNodeCustom.cpp:
        (WebCore::JSNode::addEventListener):
        (WebCore::JSNode::removeEventListener):
        * bindings/js/JSSVGElementInstanceCustom.cpp:
        (WebCore::JSSVGElementInstance::addEventListener):
        (WebCore::JSSVGElementInstance::removeEventListener):
        * bindings/js/JSWorkerContextCustom.cpp:
        (WebCore::JSWorkerContext::addEventListener):
        (WebCore::JSWorkerContext::removeEventListener):
        * bindings/js/JSXMLHttpRequestCustom.cpp:
        (WebCore::JSXMLHttpRequest::addEventListener):
        (WebCore::JSXMLHttpRequest::removeEventListener):
        * bindings/js/JSXMLHttpRequestUploadCustom.cpp:
        (WebCore::JSXMLHttpRequestUpload::addEventListener):
        (WebCore::JSXMLHttpRequestUpload::removeEventListener):
        * bindings/objc/ObjCEventListener.h:
        * bindings/objc/ObjCEventListener.mm:
        (WebCore::ObjCEventListener::handleEvent):
        * bindings/scripts/CodeGeneratorJS.pm:
        * dom/EventTarget.cpp:
        (WebCore::EventTarget::fireEventListeners):
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::handleEvent):
        * inspector/InspectorDOMAgent.h:
        * inspector/InspectorDOMStorageResource.cpp:
        (WebCore::InspectorDOMStorageResource::handleEvent):
        * inspector/InspectorDOMStorageResource.h:
        * loader/ImageDocument.cpp:
        (WebCore::ImageEventListener::handleEvent):
        * svg/animation/SVGSMILElement.cpp:
        (WebCore::ConditionEventListener::handleEvent):
        * workers/WorkerContext.cpp:
        (WebCore::WorkerContext::reportException):
        Don't pass global object to JSEventListener::create(), which no longer needs it.
        Note that some of these functions still have an early return for null global object, which
        can probably be removed in a later patch.
        Pass ScriptExecutionContext to EventListener methods that now need it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoChecked in a file that was created by make-script-test-wrappers.
darin@apple.com [Fri, 25 Sep 2009 20:24:55 +0000 (20:24 +0000)]
Checked in a file that was created by make-script-test-wrappers.

Patch by Darin Adler <darin@apple.com> on 2009-09-25
* platform/gtk/editing/pasteboard/middle-click-onpaste.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 25 Sep 2009 20:00:56 +0000 (20:00 +0000)]
2009-09-25  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Added Tony to committers.py because he's a
        committer now.

        * Scripts/modules/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=29740
adele@apple.com [Fri, 25 Sep 2009 18:55:42 +0000 (18:55 +0000)]
WebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=29740
<rdar://problem/7168738> Gmail: After changing a foreground text color, pressing return doesn't apply background to new line

Patch by Enrica Casucci <enrica@apple.com> on 2009-09-25
Reviewed by Darin Adler, Dan Bernstein, Adele Peterson, and others.

Change the way style is preserved when inserting a new paragraph.
The original code handled insertion at the beginning and at the end of a paragraph as special
cases. The newly created paragraph contained a set of nodes generated starting from the
computed style of the insertion node. This approach has two problems:
1. if the insertion node has a non opaque background color and one of the parent element did have
a solid background color the new paragraph did not have the element with the solid color in the tree.
2. in some circumstances it generated more markup than the original paragraph had (a span with bold, italic,
background color and some font attribute was being reproduced as span + bold + italic + font as separate tags.
The new approach is to recreate in the new paragraph the same hierarchy of nodes found in the
paragraph where the insertion point is.

Test: editing/inserting/insert-bg-font.html

* editing/InsertParagraphSeparatorCommand.cpp:
(WebCore::InsertParagraphSeparatorCommand::getAncestorsInsideBlock): retrieves the list of all the ancestors
between the insert node and the outer block.
(WebCore::InsertParagraphSeparatorCommand::cloneHierarchyUnderNewBlock): uses the list of ancestors to recreate
in the new paragraph the same element hierarchy present in the starting paragraph.
(WebCore::InsertParagraphSeparatorCommand::doApply): changed the code to handle the general case of insertion
in the middle of the paragraph to use the new methods. Changed the handling of the insertion at the beginning and
at the end of the paragraph to use the new methods instead of applying the calculated style.
* editing/InsertParagraphSeparatorCommand.h: added methods getAncestorsInsideBlock and cloneHierarchyUnderNewBlock.

LayoutTests: Updated the expected results to reflect the changes in the way the new paragraph
is created and added test case for https://bugs.webkit.org/show_bug.cgi?id=29740
<rdar://problem/7168738> Gmail: After changing a foreground text color, pressing return doesn't apply background to new line

Patch by Enrica Casucci <enrica@apple.com> on 2009-09-25
Reviewed by Darin Adler, Dan Bernstein, Adele Peterson, and others.

* editing/inserting/insert-bg-font.html: Added.
* platform/mac/editing/inserting/insert-bg-font-expected.txt: Added.
* platform/mac/editing/pasteboard/5478250-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Patrick Mueller <Patrick_Mueller@us.ibm.com>
eric@webkit.org [Fri, 25 Sep 2009 18:36:56 +0000 (18:36 +0000)]
2009-09-25  Patrick Mueller  <Patrick_Mueller@us.ibm.com>

        Reviewed by Timothy Hatcher.

        Content-type parameters not taken into account when building form-data
        https://bugs.webkit.org/show_bug.cgi?id=28970

        existing manual test case extended with new tests

        * English.lproj/localizedStrings.js:
        * inspector/front-end/ResourceView.js:
        (WebInspector.ResourceView.prototype._refreshFormData):
        (WebInspector.ResourceView.prototype._refreshParms):
        * manual-tests/inspector/display-form-data.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Eric Seidel <eric@webkit.org>
eric@webkit.org [Fri, 25 Sep 2009 18:27:53 +0000 (18:27 +0000)]
2009-09-25  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        commit-queue should auto-retry patches which fail to commit due to out of date files
        https://bugs.webkit.org/show_bug.cgi?id=28316

        * Scripts/bugzilla-tool:
         - Handle new CheckoutNeedsUpdate exception.
        * Scripts/modules/logging_unittest.py:
         - Call the ScriptError constructor correctly (this test had regressed).
        * Scripts/modules/scm.py:
         - Added the ability to define custom error handlers for run_command
           and added a commit_error_handler which throws CheckoutNeedsUpdate
           instead of ScriptError.
         - Re-ordered ScriptError constructor arguments to make ScriptError("message text") usage possible.
        * Scripts/modules/scm_unittest.py:
         - Added tests of new error handlers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Yuan Song <song.yuan@ericsson.com>
eric@webkit.org [Fri, 25 Sep 2009 17:53:43 +0000 (17:53 +0000)]
2009-09-25  Yuan Song  <song.yuan@ericsson.com>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=14566

        Add test "invalid-domain-change-throws-exception.html" to ensure a SECURITY_ERR exception is raised if an attempt is made to change document.domain to an invalid value. In the existing test case "basic-textareas.html", fix the faulty behavior of setting document.domain to invalid value, and replace the data URL to a file URL that points to "basic-textareas-standards.html" in order to make "basic-textareas.html" runnable in a browser.

        * fast/forms/basic-textareas.html:
        * fast/forms/resources/basic-textareas-standards.html: Added.
        * fast/js/invalid-domain-change-throws-exception-expected.txt: Added.
        * fast/js/invalid-domain-change-throws-exception.html: Added.
        * fast/js/resources/invalid-domain-change-throws-exception.js: Added.
2009-09-25  Yuan Song  <song.yuan@ericsson.com>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=14566

        Raise SECURITY_ERR exception if an attempt is made to change document.domain to an invalid value.

        Test: fast/js/invalid-domain-change-throws-exception.html

        * dom/Document.cpp:
        (WebCore::Document::setDomain):
        * dom/Document.h:
        * dom/Document.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Eric Seidel <eric@webkit.org>
eric@webkit.org [Fri, 25 Sep 2009 17:40:47 +0000 (17:40 +0000)]
2009-09-25  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        commit-queue should give better feedback when failing a patch
        https://bugs.webkit.org/show_bug.cgi?id=29316

        * Scripts/bugzilla-tool:
         - Update ScriptError uses to the new constructor format.
         - Move CommitQueue._run_command to WebKitLandingScripts.run_command_with_teed_output
           so that we can print to both stdout as well as an output buffer for error reporting.
         - Update run_and_throw_if_fail to use teed output so that it can report the "output" as part of ScriptError.
         - Use e.message_with_output() when failing a patch (this is the real fix here).
           I also removed use of "This patch will require manual commit." as that's not always true.
         - Add missing word "bug" from log message.
        * Scripts/modules/scm.py:
         - Make ScriptError save a bunch more data so that error messages can be nicer.
         - Update ScriptError callers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Adam Barth <abarth@webkit.org>
eric@webkit.org [Fri, 25 Sep 2009 16:36:57 +0000 (16:36 +0000)]
2009-09-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Dimitri Glazkov.

        [V8] Teach ScheduledAction::execute about isolated worlds
        https://bugs.webkit.org/show_bug.cgi?id=27703

        Test a strange error condition after clearing a setInterval
        timer.  The key point is not to crash on this test.

        * fast/dom/timer-clear-interval-in-handler-and-generate-error-expected.txt: Added.
        * fast/dom/timer-clear-interval-in-handler-and-generate-error.html: Added.
2009-09-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Dimitri Glazkov.

        [V8] Teach ScheduledAction::execute about isolated worlds
        https://bugs.webkit.org/show_bug.cgi?id=27703

        When setTimeout is called with a string argument in an isolated
        world, we now compile the string in the isolated world.

        Last time we tried this change, we got a lot of crashes.  This
        time we're using a fresh local handle as our context to avoid
        trouble if the peristent handle gets disposed before we leave
        the context.

        Test: fast/dom/timer-clear-interval-in-handler-and-generate-error.html

        * bindings/v8/ScheduledAction.cpp:
        (WebCore::ScheduledAction::execute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Csaba Osztrogonac <oszi@inf.u-szeged.hu>
eric@webkit.org [Fri, 25 Sep 2009 16:30:00 +0000 (16:30 +0000)]
2009-09-25  Csaba Osztrogonac  <oszi@inf.u-szeged.hu>

        Reviewed by Simon Hausmann.

        [Qt] Make tst_qwebframe work if Qt built without SSL support
        https://bugs.webkit.org/show_bug.cgi?id=29735

        * tests/qwebframe/tst_qwebframe.cpp: Missing #ifndef blocks added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Paul Godavari <paul@chromium.org>
eric@webkit.org [Fri, 25 Sep 2009 16:23:06 +0000 (16:23 +0000)]
2009-09-25  Paul Godavari  <paul@chromium.org>

        Reviewed by Darin Fisher.

        Fix a regression in Mac Chromium popup menus, where the user's
        selection was ignored and the popup became unresponsive.
        https://bugs.webkit.org/show_bug.cgi?id=29726

        The fix is to notify the popup's client that the popup was hidden,
        even if the popup has no parent.

        * platform/chromium/PopupMenuChromium.cpp:
        (WebCore::PopupListBox::hidePopup):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Laszlo Gombos <laszlo.1.gombos@nokia.com>
eric@webkit.org [Fri, 25 Sep 2009 16:10:49 +0000 (16:10 +0000)]
2009-09-25  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Gavin Barraclough.

        Add ARM version detection rules for Symbian
        https://bugs.webkit.org/show_bug.cgi?id=29715

        * wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Alexander Pavlov <apavlov@chromium.org>
pfeldman@chromium.org [Fri, 25 Sep 2009 15:47:56 +0000 (15:47 +0000)]
2009-09-25  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Dan Bernstein.

        Enable Pasteboard::writePlainText for Chromium and fix code style nits.
        https://bugs.webkit.org/show_bug.cgi?id=29734

        * platform/chromium/PasteboardChromium.cpp:
        (WebCore::Pasteboard::writePlainText):
        * platform/gtk/PasteboardGtk.cpp:
        (WebCore::Pasteboard::writePlainText):
        (WebCore::Pasteboard::writeURL):
        * platform/mac/PasteboardMac.mm:
        (WebCore::Pasteboard::writeSelection):
        (WebCore::Pasteboard::writePlainText):
        (WebCore::Pasteboard::writeURL):
        * platform/qt/PasteboardQt.cpp:
        (WebCore::Pasteboard::writePlainText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd SPI to invalidate a WebView's backing store
aroben@apple.com [Fri, 25 Sep 2009 15:32:41 +0000 (15:32 +0000)]
Add SPI to invalidate a WebView's backing store

This is useful in conjunction with
IWebUIDelegatePrivate2::drawBackground. This provides a way for the UI
delegate to force WebView to get its background redrawn.

Fixes <http://webkit.org/b/29737> Need SPI to invalidate a WebView's
backing store

Reviewed by Anders Carlsson.

* Interfaces/IWebViewPrivate.idl: Added invalidateBackingStore.
* WebView.cpp:
(WebView::invalidateBackingStore): If rect is 0, invalidate our whole
client rect. Otherwise invalidate the intersection of rect with our
client rect.
* WebView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-09-25 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Fri, 25 Sep 2009 14:01:52 +0000 (14:01 +0000)]
2009-09-25  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Xan Lopez.

        Pop down the context menu when the page transitions to committed,
        to avoid keeping a context menu out of context.

        * WebCoreSupport/FrameLoaderClientGtk.cpp:
        (WebKit::FrameLoaderClient::transitionToCommittedForNewPage):
        * webkit/webkitprivate.h:
        * webkit/webkitwebview.cpp:
        (destroy_menu_cb):
        (webkit_web_view_forward_context_menu_event):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@48753 268f45cc-cd09-0410-ab3c-d52691b4dbfc