WebKit-https.git
12 years ago2008-10-09 Chris Marrin <cmarrin@apple.com>
cmarrin@apple.com [Thu, 9 Oct 2008 23:04:07 +0000 (23:04 +0000)]
2008-10-09  Chris Marrin  <cmarrin@apple.com>

        Reviewed by Dan Bernstein.

        Fixed > 180 degree rotation bug
        Ensure that validateTransformFunctionList() is called after updating keyframe styles

        https://bugs.webkit.org/show_bug.cgi?id=21420

        Test: animations/big-rotation.html

        * page/animation/KeyframeAnimation.cpp:
        (WebCore::KeyframeAnimation::KeyframeAnimation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Cameron Zwarich <zwarich@apple.com>
cwzwarich@webkit.org [Thu, 9 Oct 2008 21:47:31 +0000 (21:47 +0000)]
2008-10-09  Cameron Zwarich  <zwarich@apple.com>

        Reviewed by Oliver Hunt.

        Bug 21459: REGRESSION (r37324): Safari crashes inside JavaScriptCore while browsing hulu.com
        <https://bugs.webkit.org/show_bug.cgi?id=21459>

        After r37324, an Arguments object does not mark an associated activation
        object. This change was made because Arguments no longer directly used
        the activation object in any way. However, if an activation is torn off,
        then the backing store of Arguments becomes the register array of the
        activation object. Arguments directly marks all of the arguments, but
        the activation object is being collected, which causes its register
        array to be freed and new memory to be allocated in its place.

        Unfortunately, it does not seem possible to reproduce this issue in a
        layout test.

        * kjs/Arguments.cpp:
        (JSC::Arguments::mark):
        * kjs/Arguments.h:
        (JSC::Arguments::setActivation):
        (JSC::Arguments::Arguments):
        (JSC::JSActivation::copyRegisters):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Brady Eidson <beidson@apple.com>
beidson@apple.com [Thu, 9 Oct 2008 20:43:24 +0000 (20:43 +0000)]
2008-10-09  Brady Eidson  <beidson@apple.com>

        Reviewed by Anders

        <rdar://problem/6250856> - Calling [WebView close] from within a redirection callback can cause bad things

        The API usage to reveal this crash was so particular that a layout test is not possible with our current infrastructure.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::changeLocation): Protect the Frame from deletion
        (WebCore::FrameLoader::continueLoadAfterNavigationPolicy): stopAllLoaders() might clear the Frame's page,
          so perform a second null check on the page.  Bail if it has been cleared.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Alp Toker <alp@nuanti.com>
alp@webkit.org [Thu, 9 Oct 2008 16:21:43 +0000 (16:21 +0000)]
2008-10-09  Alp Toker  <alp@nuanti.com>

        Build fix for recent DOM generation Makefile changes when custom
        CFLAGS/CXXFLAGS are passed in.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Jan Michael Alonzo <jmalonzo@webkit.org>
alp@webkit.org [Thu, 9 Oct 2008 16:13:53 +0000 (16:13 +0000)]
2008-10-09  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Alp Toker.

        https://bugs.webkit.org/show_bug.cgi?id=21390
        [Gtk] Linux/Gtk: GtkLauncher crashes on Acid3 (but after test 80 this time)

        * platform/gtk/ScrollViewGtk.cpp:
        (WebCore::ScrollView::platformRemoveChild):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Ariya Hidayat <ariya.hidayat@trolltech.com>
hausmann@webkit.org [Thu, 9 Oct 2008 14:26:27 +0000 (14:26 +0000)]
2008-10-09  Ariya Hidayat  <ariya.hidayat@trolltech.com>

        Reviewed by Simon.

        Build fix for MinGW.

        * wtf/AlwaysInline.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Darin Adler <darin@apple.com>
darin@apple.com [Thu, 9 Oct 2008 08:37:04 +0000 (08:37 +0000)]
2008-10-09  Darin Adler  <darin@apple.com>

       - try to fix Qt build

        * bridge/qt/qt_runtime.cpp:
        (JSC::Bindings::convertQVariantToValue): Pass JSGlobalData instead of ExecState to RegExp constructor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-09 Darin Adler <darin@apple.com>
darin@apple.com [Thu, 9 Oct 2008 07:25:46 +0000 (07:25 +0000)]
2008-10-09  Darin Adler  <darin@apple.com>

        - checked in a missing expected result

        * fast/dom/css-RGBValue-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRemove restriction on version number ending in a 4.
mrowe@apple.com [Thu, 9 Oct 2008 05:36:02 +0000 (05:36 +0000)]
Remove restriction on version number ending in a 4.

Rubber-stamped by Jon Honeycutt.

* win/tools/scripts/auto-version.sh:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Cameron Zwarich <zwarich@apple.com>
cwzwarich@webkit.org [Thu, 9 Oct 2008 05:30:22 +0000 (05:30 +0000)]
2008-10-08  Cameron Zwarich  <zwarich@apple.com>

        Reviewed by Maciej Stachowiak.

        Bug 21497: REGRESSION (r37433): Bytecode JSC tests are severely broken
        <https://bugs.webkit.org/show_bug.cgi?id=21497>

        Fix a typo in r37433 that causes the failure of a large number of JSC
        tests with the bytecode interpreter enabled.

        * VM/Machine.cpp:
        (JSC::Machine::privateExecute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix the Windows build after r37434.
mrowe@apple.com [Thu, 9 Oct 2008 05:19:36 +0000 (05:19 +0000)]
Fix the Windows build after r37434.

* DumpRenderTree/win/DumpRenderTree.cpp:
(dump): Update code to approximate valid C++ syntax.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Add SVGElementInstance to the list of exceptions that have
timothy@apple.com [Thu, 9 Oct 2008 04:34:27 +0000 (04:34 +0000)]
    Add SVGElementInstance to the list of exceptions that have
        global initializers in debug builds.

        Rubber-stamped by Mark Rowe.

        * Scripts/check-for-global-initializers: Add SVGElementInstance.o.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37439 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWindows build fix.
mrowe@apple.com [Thu, 9 Oct 2008 04:05:45 +0000 (04:05 +0000)]
Windows build fix.

* VM/CTI.cpp:
(JSC::): Update type of argument to ctiTrampoline.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37438 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoCorrect MIME types
mitz@apple.com [Thu, 9 Oct 2008 03:57:35 +0000 (03:57 +0000)]
Correct MIME types

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37437 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoNot reviewed. Fix Sams name :-)
zimmermann@webkit.org [Thu, 9 Oct 2008 03:54:25 +0000 (03:54 +0000)]
Not reviewed. Fix Sams name :-)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37436 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoReviewed by Oliver Hunt & Sam Weining.
zimmermann@webkit.org [Thu, 9 Oct 2008 03:34:43 +0000 (03:34 +0000)]
Reviewed by Oliver Hunt & Sam Weining.

 Fixes: https://bugs.webkit.org/show_bug.cgi?id=15413 (SVGElementInstance does not implement EventTarget)
 Fixes: https://bugs.webkit.org/show_bug.cgi?id=15430 (SVGElementInstances should rebuild themselves lazily)
 Fixes: https://bugs.webkit.org/show_bug.cgi?id=20550 (SVGUseElement.setAttribute crashes Safari)

 Rebuild SVG use element shadow tree lazily. Mark SVGElementInstance with the "needsUpdate" flag.
 Next time SVGUseElement::recalcStyle is invoked (ie. via Document::updateDocumentsRendering()) the
 use tree is rebuild.

 Make SVGElementInstance a real EventTarget, as the SVG specification demands.
 When dispatching events to a shadow tree element of a use element, the associated SVGElementInstance
 is used as event target. The SVGElementInstance, the shadow tree element and the corresponding element
 share an event listener list. Every event listener change on the SVGElementInstance object is forwared
 to the corresponding element. Each change marks the SVGElementInstance tree dirty, and causes a reclone.
 Each event listener defined via attributes (onclick/onkeydown/...) is copied from the correspondingElement
 to the shadow tree element - through the cloneNode calls - if the use element's shadow tree gets rebuild.
 Each listener, dynamically created using addEventListener, gets copied to the corersponding element as well.

 Now that the target/currentTarget properties of the Events are correct, event bubbling works as expected,
 see resources/use-instanceRoot-event-bubbling.js for details.

 Tests: svg/custom/use-elementInstance-event-target.svg (reenabled)
        svg/custom/use-elementInstance-methods.svg (reenabled)
        svg/custom/use-setAttribute-crash.svg (covers bug 20550)
        svg/custom/use-instanceRoot-as-event-target.xhtml (covers bug 15413)
        svg/custom/use-instanceRoot-event-bubbling.xhtml (covers bug 15413)
        svg/custom/use-instanceRoot-event-listeners.xhtml (covers bug 15413 & 15430)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37435 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Thu, 9 Oct 2008 00:45:48 +0000 (00:45 +0000)]
2008-10-08  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein

        Avoid disconnect between DRT and run-webkit-tests
        about whether to expect PNG dumps by having DRT always
        print two blocks terminated by #EOF, the second of which
        may be empty.

        https://bugs.webkit.org/show_bug.cgi?id=21483

        * DumpRenderTree/cg/PixelDumpSupportCG.cpp:
        (printPNG):
        (dumpWebViewAsPixelsAndCompareWithExpected):
        * DumpRenderTree/mac/DumpRenderTree.mm:
        (dump):
        * DumpRenderTree/win/DumpRenderTree.cpp:
        (dump):
        * Scripts/run-webkit-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37434 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScriptCore:
darin@apple.com [Thu, 9 Oct 2008 00:40:43 +0000 (00:40 +0000)]
JavaScriptCore:

2008-10-08  Darin Adler  <darin@apple.com>

        Reviewed by Cameron Zwarich.

        - https://bugs.webkit.org/show_bug.cgi?id=21403
          Bug 21403: use new CallFrame class rather than Register* for call frame manipulation

        Add CallFrame as a synonym for ExecState. Arguably, some day we should switch every
        client over to the new name.

        Use CallFrame* consistently rather than Register* or ExecState* in low-level code such
        as Machine.cpp and CTI.cpp. Similarly, use callFrame rather than r as its name and use
        accessor functions to get at things in the frame.

        Eliminate other uses of ExecState* that aren't needed, replacing in some cases with
        JSGlobalData* and in other cases eliminating them entirely.

        * API/JSObjectRef.cpp:
        (JSObjectMakeFunctionWithCallback):
        (JSObjectMakeFunction):
        (JSObjectHasProperty):
        (JSObjectGetProperty):
        (JSObjectSetProperty):
        (JSObjectDeleteProperty):
        * API/OpaqueJSString.cpp:
        * API/OpaqueJSString.h:
        * VM/CTI.cpp:
        (JSC::CTI::getConstant):
        (JSC::CTI::emitGetArg):
        (JSC::CTI::emitGetPutArg):
        (JSC::CTI::getConstantImmediateNumericArg):
        (JSC::CTI::printOpcodeOperandTypes):
        (JSC::CTI::CTI):
        (JSC::CTI::compileOpCall):
        (JSC::CTI::compileBinaryArithOp):
        (JSC::CTI::privateCompileMainPass):
        (JSC::CTI::privateCompile):
        (JSC::CTI::privateCompileGetByIdProto):
        (JSC::CTI::privateCompileGetByIdChain):
        (JSC::CTI::compileRegExp):
        * VM/CTI.h:
        * VM/CodeBlock.h:
        * VM/CodeGenerator.cpp:
        (JSC::CodeGenerator::emitEqualityOp):
        (JSC::CodeGenerator::emitLoad):
        (JSC::CodeGenerator::emitUnexpectedLoad):
        (JSC::CodeGenerator::emitConstruct):
        * VM/CodeGenerator.h:
        * VM/Machine.cpp:
        (JSC::jsLess):
        (JSC::jsLessEq):
        (JSC::jsAddSlowCase):
        (JSC::jsAdd):
        (JSC::jsTypeStringForValue):
        (JSC::Machine::resolve):
        (JSC::Machine::resolveSkip):
        (JSC::Machine::resolveGlobal):
        (JSC::inlineResolveBase):
        (JSC::Machine::resolveBase):
        (JSC::Machine::resolveBaseAndProperty):
        (JSC::Machine::resolveBaseAndFunc):
        (JSC::Machine::slideRegisterWindowForCall):
        (JSC::isNotObject):
        (JSC::Machine::callEval):
        (JSC::Machine::dumpCallFrame):
        (JSC::Machine::dumpRegisters):
        (JSC::Machine::unwindCallFrame):
        (JSC::Machine::throwException):
        (JSC::DynamicGlobalObjectScope::DynamicGlobalObjectScope):
        (JSC::DynamicGlobalObjectScope::~DynamicGlobalObjectScope):
        (JSC::Machine::execute):
        (JSC::Machine::debug):
        (JSC::Machine::createExceptionScope):
        (JSC::cachePrototypeChain):
        (JSC::Machine::tryCachePutByID):
        (JSC::Machine::tryCacheGetByID):
        (JSC::Machine::privateExecute):
        (JSC::Machine::retrieveArguments):
        (JSC::Machine::retrieveCaller):
        (JSC::Machine::retrieveLastCaller):
        (JSC::Machine::findFunctionCallFrame):
        (JSC::Machine::getArgumentsData):
        (JSC::Machine::tryCTICachePutByID):
        (JSC::Machine::getCTIArrayLengthTrampoline):
        (JSC::Machine::getCTIStringLengthTrampoline):
        (JSC::Machine::tryCTICacheGetByID):
        (JSC::Machine::cti_op_convert_this):
        (JSC::Machine::cti_op_end):
        (JSC::Machine::cti_op_add):
        (JSC::Machine::cti_op_pre_inc):
        (JSC::Machine::cti_timeout_check):
        (JSC::Machine::cti_op_loop_if_less):
        (JSC::Machine::cti_op_loop_if_lesseq):
        (JSC::Machine::cti_op_new_object):
        (JSC::Machine::cti_op_put_by_id):
        (JSC::Machine::cti_op_put_by_id_second):
        (JSC::Machine::cti_op_put_by_id_generic):
        (JSC::Machine::cti_op_put_by_id_fail):
        (JSC::Machine::cti_op_get_by_id):
        (JSC::Machine::cti_op_get_by_id_second):
        (JSC::Machine::cti_op_get_by_id_generic):
        (JSC::Machine::cti_op_get_by_id_fail):
        (JSC::Machine::cti_op_instanceof):
        (JSC::Machine::cti_op_del_by_id):
        (JSC::Machine::cti_op_mul):
        (JSC::Machine::cti_op_new_func):
        (JSC::Machine::cti_op_call_JSFunction):
        (JSC::Machine::cti_vm_compile):
        (JSC::Machine::cti_op_push_activation):
        (JSC::Machine::cti_op_call_NotJSFunction):
        (JSC::Machine::cti_op_create_arguments):
        (JSC::Machine::cti_op_tear_off_activation):
        (JSC::Machine::cti_op_tear_off_arguments):
        (JSC::Machine::cti_op_ret_profiler):
        (JSC::Machine::cti_op_ret_scopeChain):
        (JSC::Machine::cti_op_new_array):
        (JSC::Machine::cti_op_resolve):
        (JSC::Machine::cti_op_construct_JSConstruct):
        (JSC::Machine::cti_op_construct_NotJSConstruct):
        (JSC::Machine::cti_op_get_by_val):
        (JSC::Machine::cti_op_resolve_func):
        (JSC::Machine::cti_op_sub):
        (JSC::Machine::cti_op_put_by_val):
        (JSC::Machine::cti_op_put_by_val_array):
        (JSC::Machine::cti_op_lesseq):
        (JSC::Machine::cti_op_loop_if_true):
        (JSC::Machine::cti_op_negate):
        (JSC::Machine::cti_op_resolve_base):
        (JSC::Machine::cti_op_resolve_skip):
        (JSC::Machine::cti_op_resolve_global):
        (JSC::Machine::cti_op_div):
        (JSC::Machine::cti_op_pre_dec):
        (JSC::Machine::cti_op_jless):
        (JSC::Machine::cti_op_not):
        (JSC::Machine::cti_op_jtrue):
        (JSC::Machine::cti_op_post_inc):
        (JSC::Machine::cti_op_eq):
        (JSC::Machine::cti_op_lshift):
        (JSC::Machine::cti_op_bitand):
        (JSC::Machine::cti_op_rshift):
        (JSC::Machine::cti_op_bitnot):
        (JSC::Machine::cti_op_resolve_with_base):
        (JSC::Machine::cti_op_new_func_exp):
        (JSC::Machine::cti_op_mod):
        (JSC::Machine::cti_op_less):
        (JSC::Machine::cti_op_neq):
        (JSC::Machine::cti_op_post_dec):
        (JSC::Machine::cti_op_urshift):
        (JSC::Machine::cti_op_bitxor):
        (JSC::Machine::cti_op_new_regexp):
        (JSC::Machine::cti_op_bitor):
        (JSC::Machine::cti_op_call_eval):
        (JSC::Machine::cti_op_throw):
        (JSC::Machine::cti_op_get_pnames):
        (JSC::Machine::cti_op_next_pname):
        (JSC::Machine::cti_op_push_scope):
        (JSC::Machine::cti_op_pop_scope):
        (JSC::Machine::cti_op_typeof):
        (JSC::Machine::cti_op_to_jsnumber):
        (JSC::Machine::cti_op_in):
        (JSC::Machine::cti_op_push_new_scope):
        (JSC::Machine::cti_op_jmp_scopes):
        (JSC::Machine::cti_op_put_by_index):
        (JSC::Machine::cti_op_switch_imm):
        (JSC::Machine::cti_op_switch_char):
        (JSC::Machine::cti_op_switch_string):
        (JSC::Machine::cti_op_del_by_val):
        (JSC::Machine::cti_op_put_getter):
        (JSC::Machine::cti_op_put_setter):
        (JSC::Machine::cti_op_new_error):
        (JSC::Machine::cti_op_debug):
        (JSC::Machine::cti_vm_throw):
        * VM/Machine.h:
        * VM/Register.h:
        * VM/RegisterFile.h:
        * kjs/Arguments.h:
        * kjs/DebuggerCallFrame.cpp:
        (JSC::DebuggerCallFrame::functionName):
        (JSC::DebuggerCallFrame::type):
        (JSC::DebuggerCallFrame::thisObject):
        (JSC::DebuggerCallFrame::evaluate):
        * kjs/DebuggerCallFrame.h:
        * kjs/ExecState.cpp:
        (JSC::CallFrame::thisValue):
        * kjs/ExecState.h:
        * kjs/FunctionConstructor.cpp:
        (JSC::constructFunction):
        * kjs/JSActivation.cpp:
        (JSC::JSActivation::JSActivation):
        (JSC::JSActivation::argumentsGetter):
        * kjs/JSActivation.h:
        * kjs/JSGlobalObject.cpp:
        (JSC::JSGlobalObject::init):
        * kjs/JSGlobalObjectFunctions.cpp:
        (JSC::globalFuncEval):
        * kjs/JSVariableObject.h:
        * kjs/Parser.cpp:
        (JSC::Parser::parse):
        * kjs/RegExpConstructor.cpp:
        (JSC::constructRegExp):
        * kjs/RegExpPrototype.cpp:
        (JSC::regExpProtoFuncCompile):
        * kjs/Shell.cpp:
        (prettyPrintScript):
        * kjs/StringPrototype.cpp:
        (JSC::stringProtoFuncMatch):
        (JSC::stringProtoFuncSearch):
        * kjs/identifier.cpp:
        (JSC::Identifier::checkSameIdentifierTable):
        * kjs/interpreter.cpp:
        (JSC::Interpreter::checkSyntax):
        (JSC::Interpreter::evaluate):
        * kjs/nodes.cpp:
        (JSC::ThrowableExpressionData::emitThrowError):
        (JSC::RegExpNode::emitCode):
        (JSC::ArrayNode::emitCode):
        (JSC::InstanceOfNode::emitCode):
        * kjs/nodes.h:
        * kjs/regexp.cpp:
        (JSC::RegExp::RegExp):
        (JSC::RegExp::create):
        * kjs/regexp.h:
        * profiler/HeavyProfile.h:
        * profiler/Profile.h:
        * wrec/WREC.cpp:
        * wrec/WREC.h:

WebKit/mac:

2008-10-08  Darin Adler  <darin@apple.com>

        Reviewed by Cameron Zwarich.

        - https://bugs.webkit.org/show_bug.cgi?id=21403
          Bug 21403: use new CallFrame class rather than Register* for call frame manipulation

        * WebView/WebScriptDebugger.mm:
        (WebScriptDebugger::WebScriptDebugger): Update since DebuggerCallFrame is simpler now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37433 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
weinig@apple.com [Thu, 9 Oct 2008 00:14:50 +0000 (00:14 +0000)]
WebCore:

2008-10-08  Sam Weinig  <sam@webkit.org>

        Reviewed by Cameron Zwarich.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=21241
        REGRESSION (r36977): getRGBColorValue().red returning incorrect value

        Update JSRGBColor to use the new static function per getter approach.

        Test: fast/dom/css-RGBValue.html

        * bindings/js/JSRGBColor.cpp:
        (WebCore::):
        (jsRGBColorRed):
        (jsRGBColorGreen):
        (jsRGBColorBlue):
        * bindings/js/JSRGBColor.h:

LayoutTests:

2008-10-08  Sam Weinig  <sam@webkit.org>

        Reviewed by Cameron Zwarich.

        Test for https://bugs.webkit.org/show_bug.cgi?id=21241
        REGRESSION (r36977): getRGBColorValue().red returning incorrect value

        * fast/dom/css-RGBValue.html: Added.
        * fast/dom/resources/css-RGBValue.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37432 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler, Nikolas Zimmermann and Dave Hyatt.
eric@webkit.org [Wed, 8 Oct 2008 22:48:55 +0000 (22:48 +0000)]
    Reviewed by Darin Adler, Nikolas Zimmermann and Dave Hyatt.

        svgElement.className.baseValue = "foo" does not work
        https://bugs.webkit.org/show_bug.cgi?id=20651

        * dom/StyledElement.cpp:
        (WebCore::StyledElement::classAttributeChanged):
        (WebCore::StyledElement::parseMappedAttribute):
        * dom/StyledElement.h:
        * svg/SVGStyledElement.cpp:
        (WebCore::SVGStyledElement::svgAttributeChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix crash in fast/js/constant-folding.html with CTI disabled.
mrowe@apple.com [Wed, 8 Oct 2008 22:38:44 +0000 (22:38 +0000)]
Fix crash in fast/js/constant-folding.html with CTI disabled.

Typed by Maciej Stachowiak, reviewed by Mark Rowe.

* VM/Machine.cpp:
(JSC::Machine::privateExecute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 20:30:04 +0000 (20:30 +0000)]
2008-10-08  Maciej Stachowiak  <mjs@apple.com>

        Not reviewed, trying to fix bot.

        Update -0 expectations better.

        * fast/xpath/4XPath/Core/test_numeric_expr-expected.txt:
        * fast/xpath/4XPath/Core/test_numeric_expr.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRoll out r37427 because it causes an infinite recursion loading about:blank.
timothy@apple.com [Wed, 8 Oct 2008 17:50:42 +0000 (17:50 +0000)]
Roll out r37427 because it causes an infinite recursion loading about:blank.

https://bugs.webkit.org/show_bug.cgi?id=21476

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScriptCore:
darin@apple.com [Wed, 8 Oct 2008 17:03:10 +0000 (17:03 +0000)]
JavaScriptCore:

2008-10-08  Darin Adler  <darin@apple.com>

        Reviewed by Cameron Zwarich.

        - https://bugs.webkit.org/show_bug.cgi?id=21403
          Bug 21403: use new CallFrame class rather than Register* for call frame manipulation

        Add CallFrame as a synonym for ExecState. Arguably, some day we should switch every
        client over to the new name.

        Use CallFrame* consistently rather than Register* or ExecState* in low-level code such
        as Machine.cpp and CTI.cpp. Similarly, use callFrame rather than r as its name and use
        accessor functions to get at things in the frame.

        Eliminate other uses of ExecState* that aren't needed, replacing in some cases with
        JSGlobalData* and in other cases eliminating them entirely.

        * API/JSObjectRef.cpp:
        (JSObjectMakeFunctionWithCallback):
        (JSObjectMakeFunction):
        (JSObjectHasProperty):
        (JSObjectGetProperty):
        (JSObjectSetProperty):
        (JSObjectDeleteProperty):
        * API/OpaqueJSString.cpp:
        * API/OpaqueJSString.h:
        * VM/CTI.cpp:
        (JSC::CTI::getConstant):
        (JSC::CTI::emitGetArg):
        (JSC::CTI::emitGetPutArg):
        (JSC::CTI::getConstantImmediateNumericArg):
        (JSC::CTI::printOpcodeOperandTypes):
        (JSC::CTI::CTI):
        (JSC::CTI::compileOpCall):
        (JSC::CTI::compileBinaryArithOp):
        (JSC::CTI::privateCompileMainPass):
        (JSC::CTI::privateCompile):
        (JSC::CTI::privateCompileGetByIdProto):
        (JSC::CTI::privateCompileGetByIdChain):
        (JSC::CTI::compileRegExp):
        * VM/CTI.h:
        * VM/CodeBlock.h:
        * VM/CodeGenerator.cpp:
        (JSC::CodeGenerator::emitEqualityOp):
        (JSC::CodeGenerator::emitLoad):
        (JSC::CodeGenerator::emitUnexpectedLoad):
        (JSC::CodeGenerator::emitConstruct):
        * VM/CodeGenerator.h:
        * VM/Machine.cpp:
        (JSC::jsLess):
        (JSC::jsLessEq):
        (JSC::jsAddSlowCase):
        (JSC::jsAdd):
        (JSC::jsTypeStringForValue):
        (JSC::Machine::resolve):
        (JSC::Machine::resolveSkip):
        (JSC::Machine::resolveGlobal):
        (JSC::inlineResolveBase):
        (JSC::Machine::resolveBase):
        (JSC::Machine::resolveBaseAndProperty):
        (JSC::Machine::resolveBaseAndFunc):
        (JSC::Machine::slideRegisterWindowForCall):
        (JSC::isNotObject):
        (JSC::Machine::callEval):
        (JSC::Machine::dumpCallFrame):
        (JSC::Machine::dumpRegisters):
        (JSC::Machine::unwindCallFrame):
        (JSC::Machine::throwException):
        (JSC::DynamicGlobalObjectScope::DynamicGlobalObjectScope):
        (JSC::DynamicGlobalObjectScope::~DynamicGlobalObjectScope):
        (JSC::Machine::execute):
        (JSC::Machine::debug):
        (JSC::Machine::createExceptionScope):
        (JSC::cachePrototypeChain):
        (JSC::Machine::tryCachePutByID):
        (JSC::Machine::tryCacheGetByID):
        (JSC::Machine::privateExecute):
        (JSC::Machine::retrieveArguments):
        (JSC::Machine::retrieveCaller):
        (JSC::Machine::retrieveLastCaller):
        (JSC::Machine::findFunctionCallFrame):
        (JSC::Machine::getArgumentsData):
        (JSC::Machine::tryCTICachePutByID):
        (JSC::Machine::getCTIArrayLengthTrampoline):
        (JSC::Machine::getCTIStringLengthTrampoline):
        (JSC::Machine::tryCTICacheGetByID):
        (JSC::Machine::cti_op_convert_this):
        (JSC::Machine::cti_op_end):
        (JSC::Machine::cti_op_add):
        (JSC::Machine::cti_op_pre_inc):
        (JSC::Machine::cti_timeout_check):
        (JSC::Machine::cti_op_loop_if_less):
        (JSC::Machine::cti_op_loop_if_lesseq):
        (JSC::Machine::cti_op_new_object):
        (JSC::Machine::cti_op_put_by_id):
        (JSC::Machine::cti_op_put_by_id_second):
        (JSC::Machine::cti_op_put_by_id_generic):
        (JSC::Machine::cti_op_put_by_id_fail):
        (JSC::Machine::cti_op_get_by_id):
        (JSC::Machine::cti_op_get_by_id_second):
        (JSC::Machine::cti_op_get_by_id_generic):
        (JSC::Machine::cti_op_get_by_id_fail):
        (JSC::Machine::cti_op_instanceof):
        (JSC::Machine::cti_op_del_by_id):
        (JSC::Machine::cti_op_mul):
        (JSC::Machine::cti_op_new_func):
        (JSC::Machine::cti_op_call_JSFunction):
        (JSC::Machine::cti_vm_compile):
        (JSC::Machine::cti_op_push_activation):
        (JSC::Machine::cti_op_call_NotJSFunction):
        (JSC::Machine::cti_op_create_arguments):
        (JSC::Machine::cti_op_tear_off_activation):
        (JSC::Machine::cti_op_tear_off_arguments):
        (JSC::Machine::cti_op_ret_profiler):
        (JSC::Machine::cti_op_ret_scopeChain):
        (JSC::Machine::cti_op_new_array):
        (JSC::Machine::cti_op_resolve):
        (JSC::Machine::cti_op_construct_JSConstruct):
        (JSC::Machine::cti_op_construct_NotJSConstruct):
        (JSC::Machine::cti_op_get_by_val):
        (JSC::Machine::cti_op_resolve_func):
        (JSC::Machine::cti_op_sub):
        (JSC::Machine::cti_op_put_by_val):
        (JSC::Machine::cti_op_put_by_val_array):
        (JSC::Machine::cti_op_lesseq):
        (JSC::Machine::cti_op_loop_if_true):
        (JSC::Machine::cti_op_negate):
        (JSC::Machine::cti_op_resolve_base):
        (JSC::Machine::cti_op_resolve_skip):
        (JSC::Machine::cti_op_resolve_global):
        (JSC::Machine::cti_op_div):
        (JSC::Machine::cti_op_pre_dec):
        (JSC::Machine::cti_op_jless):
        (JSC::Machine::cti_op_not):
        (JSC::Machine::cti_op_jtrue):
        (JSC::Machine::cti_op_post_inc):
        (JSC::Machine::cti_op_eq):
        (JSC::Machine::cti_op_lshift):
        (JSC::Machine::cti_op_bitand):
        (JSC::Machine::cti_op_rshift):
        (JSC::Machine::cti_op_bitnot):
        (JSC::Machine::cti_op_resolve_with_base):
        (JSC::Machine::cti_op_new_func_exp):
        (JSC::Machine::cti_op_mod):
        (JSC::Machine::cti_op_less):
        (JSC::Machine::cti_op_neq):
        (JSC::Machine::cti_op_post_dec):
        (JSC::Machine::cti_op_urshift):
        (JSC::Machine::cti_op_bitxor):
        (JSC::Machine::cti_op_new_regexp):
        (JSC::Machine::cti_op_bitor):
        (JSC::Machine::cti_op_call_eval):
        (JSC::Machine::cti_op_throw):
        (JSC::Machine::cti_op_get_pnames):
        (JSC::Machine::cti_op_next_pname):
        (JSC::Machine::cti_op_push_scope):
        (JSC::Machine::cti_op_pop_scope):
        (JSC::Machine::cti_op_typeof):
        (JSC::Machine::cti_op_to_jsnumber):
        (JSC::Machine::cti_op_in):
        (JSC::Machine::cti_op_push_new_scope):
        (JSC::Machine::cti_op_jmp_scopes):
        (JSC::Machine::cti_op_put_by_index):
        (JSC::Machine::cti_op_switch_imm):
        (JSC::Machine::cti_op_switch_char):
        (JSC::Machine::cti_op_switch_string):
        (JSC::Machine::cti_op_del_by_val):
        (JSC::Machine::cti_op_put_getter):
        (JSC::Machine::cti_op_put_setter):
        (JSC::Machine::cti_op_new_error):
        (JSC::Machine::cti_op_debug):
        (JSC::Machine::cti_vm_throw):
        * VM/Machine.h:
        * VM/Register.h:
        * VM/RegisterFile.h:
        * kjs/Arguments.h:
        * kjs/DebuggerCallFrame.cpp:
        (JSC::DebuggerCallFrame::functionName):
        (JSC::DebuggerCallFrame::type):
        (JSC::DebuggerCallFrame::thisObject):
        (JSC::DebuggerCallFrame::evaluate):
        * kjs/DebuggerCallFrame.h:
        * kjs/ExecState.cpp:
        (JSC::CallFrame::thisValue):
        * kjs/ExecState.h:
        * kjs/FunctionConstructor.cpp:
        (JSC::constructFunction):
        * kjs/JSActivation.cpp:
        (JSC::JSActivation::JSActivation):
        (JSC::JSActivation::argumentsGetter):
        * kjs/JSActivation.h:
        * kjs/JSGlobalObject.cpp:
        (JSC::JSGlobalObject::init):
        * kjs/JSGlobalObjectFunctions.cpp:
        (JSC::globalFuncEval):
        * kjs/JSVariableObject.h:
        * kjs/Parser.cpp:
        (JSC::Parser::parse):
        * kjs/RegExpConstructor.cpp:
        (JSC::constructRegExp):
        * kjs/RegExpPrototype.cpp:
        (JSC::regExpProtoFuncCompile):
        * kjs/Shell.cpp:
        (prettyPrintScript):
        * kjs/StringPrototype.cpp:
        (JSC::stringProtoFuncMatch):
        (JSC::stringProtoFuncSearch):
        * kjs/identifier.cpp:
        (JSC::Identifier::checkSameIdentifierTable):
        * kjs/interpreter.cpp:
        (JSC::Interpreter::checkSyntax):
        (JSC::Interpreter::evaluate):
        * kjs/nodes.cpp:
        (JSC::ThrowableExpressionData::emitThrowError):
        (JSC::RegExpNode::emitCode):
        (JSC::ArrayNode::emitCode):
        (JSC::InstanceOfNode::emitCode):
        * kjs/nodes.h:
        * kjs/regexp.cpp:
        (JSC::RegExp::RegExp):
        (JSC::RegExp::create):
        * kjs/regexp.h:
        * profiler/HeavyProfile.h:
        * profiler/Profile.h:
        * wrec/WREC.cpp:
        * wrec/WREC.h:

WebKit/mac:

2008-10-08  Darin Adler  <darin@apple.com>

        Reviewed by Cameron Zwarich.

        - https://bugs.webkit.org/show_bug.cgi?id=21403
          Bug 21403: use new CallFrame class rather than Register* for call frame manipulation

        * WebView/WebScriptDebugger.mm:
        (WebScriptDebugger::WebScriptDebugger): Update since DebuggerCallFrame is simpler now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37427 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Make the toolbar label text-shadow not disappear when clicking on the
timothy@apple.com [Wed, 8 Oct 2008 16:41:15 +0000 (16:41 +0000)]
    Make the toolbar label text-shadow not disappear when clicking on the
        search result count or around the search field.

        Reviewed by Timothy Hatcher.

        * inspector/front-end/inspector.css:
        (.toolbar-item:active .toolbar-label): Make this rule also require the
        toggleable class with the toolbar-item class.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37426 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Matthias Ettrich <ettrich@nokia.com>
hausmann@webkit.org [Wed, 8 Oct 2008 13:30:07 +0000 (13:30 +0000)]
2008-10-08  Matthias Ettrich <ettrich@nokia.com>

        Reviewed by Simon.

        Make QWebView ignore focus changes with PopupFocusReason to avoid
        massive updates when menus open and close.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37425 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Tor Arne Vestbø <tavestbo@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 13:13:24 +0000 (13:13 +0000)]
2008-10-08  Tor Arne Vestbø  <tavestbo@trolltech.com>

        Reviewed by Simon.

        Updated the qrc file for the Web Inspector to contain the current set
        of images.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37424 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Prasanth Ullattil <pullatti@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 12:43:13 +0000 (12:43 +0000)]
2008-10-08  Prasanth Ullattil  <pullatti@trolltech.com>

        Reviewed by Simon.

        Fix the linking of applications against WebKit on Qt/Windows.

        The prl files that qmake creates are buggy on Unix, but we they're
        fine on Windows and we have to have them there in order to get the
        dependencies correct.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37423 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Prasanth Ullattil <pullatti@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 12:20:41 +0000 (12:20 +0000)]
2008-10-08  Prasanth Ullattil  <pullatti@trolltech.com>

        Reviewed by Oliver Hunt.

        Avoid endless loops when compiling without the computed goto
        optimization.

        NEXT_OPCODE expands to "continue", which will not work inside
        loops.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37422 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Ariya Hidayat <ariya.hidayat@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 12:11:44 +0000 (12:11 +0000)]
2008-10-08  Ariya Hidayat  <ariya.hidayat@trolltech.com>

        Reviewed by Simon.

        Speed up rectangle filling by not re-creating a QBrush all the time.

        This triggers faster path in QPainter where the brush is reused.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37421 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Kavindra Palaraja <kdpalara@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 12:08:47 +0000 (12:08 +0000)]
2008-10-08  Kavindra Palaraja  <kdpalara@trolltech.com>

        Reviewed by Simon.

        Fixed a qdoc warning and adjusted some spacing

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37420 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 11:33:16 +0000 (11:33 +0000)]
2008-10-08  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        - update for new -0 handling in shouldBe

        (I confirmed the correctness of this change against other browsers, too.)

        * fast/xpath/4XPath/Core/test_numeric_expr.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37419 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Dean Jackson <dino@apple.com>
dino@apple.com [Wed, 8 Oct 2008 11:01:37 +0000 (11:01 +0000)]
2008-10-08  Dean Jackson  <dino@apple.com>

        Forgot to set the date on the spec documents with last commit.

        * CSSAnimations.html:
        * CSSTransforms.html:
        * CSSTransitions.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37418 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 10:53:13 +0000 (10:53 +0000)]
2008-10-08  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        Re-landing the following fix with the crashing bug in it fixed (r37405):

        - optimize away multiplication by constant 1.0

        2.3% speedup on v8 RayTrace benchmark

        Apparently it's not uncommon for JavaScript code to multiply by
        constant 1.0 in the mistaken belief that this converts integer to
        floating point and that there is any operational difference.

        * VM/CTI.cpp:
        (JSC::CTI::privateCompileMainPass): Optimize to_jsnumber for
        case where parameter is already number.
        (JSC::CTI::privateCompileSlowCases): ditto
        * VM/Machine.cpp:
        (JSC::Machine::privateExecute): ditto
        * kjs/grammar.y:
        (makeMultNode): Transform as follows:
        +FOO * BAR ==> FOO * BAR
        FOO * +BAR ==> FOO * BAR
        FOO * 1 ==> +FOO
        1 * FOO ==> +FOO
        (makeDivNode): Transform as follows:
        +FOO / BAR ==> FOO / BAR
        FOO / +BAR ==> FOO / BAR
        (makeSubNode): Transform as follows:
        +FOO - BAR ==> FOO - BAR
        FOO - +BAR ==> FOO - BAR
        * kjs/nodes.h:
        (JSC::ExpressionNode::stripUnaryPlus): Helper for above
        grammar.y changes
        (JSC::UnaryPlusNode::stripUnaryPlus): ditto

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37417 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 10:27:08 +0000 (10:27 +0000)]
2008-10-08  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        - update regression testing system to be able to tell 0 from -0

        * fast/js/resources/js-test-pre.js:

        Update tests and results appropriately.

        * fast/js/resources/constant-folding.js:
        * fast/js/constant-folding-expected.txt:
        * fast/js/math-transforms-expected.txt:
        * fast/js/resources/math-transforms.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37416 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScriptCore:
mjs@apple.com [Wed, 8 Oct 2008 09:18:08 +0000 (09:18 +0000)]
JavaScriptCore:

2008-10-08  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        - correctly handle appending -0 to a string, it should stringify as just 0

        * kjs/ustring.cpp:
        (JSC::concatenate):

LayoutTests:

2008-10-08  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        - added test case for: correctly handle appending -0 to a string, it should stringify as just 0
        (added to existing test)

        * fast/js/number-toString-expected.txt:
        * fast/js/resources/number-toString.js:

        - updated results in light of above fix

        * fast/js/math-transforms-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37415 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Thiago Macieira <thiago.macieira@nokia.com>
hausmann@webkit.org [Wed, 8 Oct 2008 09:10:44 +0000 (09:10 +0000)]
2008-10-08  Thiago Macieira  <thiago.macieira@nokia.com>

        Reviewed by Simon.

        Fixes: Encoding of Qt URLs

        This encoding was added by Simon and I because QUrl's
        tolerant parser wasn't tolerant enough. Now it is, so we don't need
        this anymore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37414 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Marius Storm-Olsen <marius@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 09:10:32 +0000 (09:10 +0000)]
2008-10-08  Marius Storm-Olsen  <marius@trolltech.com>

        Reviewed by Simon.

        Fixes dependencies in qmake generated visual studio projects.

        Only add debug lib name if we're in the debug build_pass, else the release version.
        The MSVC solution generator would pick up the debug javascriptcore lib as a dependency.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37413 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Prasanth Ullattil <pullatti@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 09:10:23 +0000 (09:10 +0000)]
2008-10-08  Prasanth Ullattil  <pullatti@trolltech.com>

        Reviewed by Simon.

        Fix WebKit compilation with VC2008SP1

        Apply the TR1 workaround for JavaScriptCore, too.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Rhys Weatherley <rhys.weatherley@nokia.com>
hausmann@webkit.org [Wed, 8 Oct 2008 09:10:14 +0000 (09:10 +0000)]
2008-10-08  Rhys Weatherley  <rhys.weatherley@nokia.com>

        Reviewed by Simon.

        Compile for platforms without context menus.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37411 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 Prasanth Ullattil <pullatti@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 09:10:04 +0000 (09:10 +0000)]
2008-10-08  Prasanth Ullattil  <pullatti@trolltech.com>

        Reviewed by Simon.

        Fix compilation errors on VS2008 64Bit

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37410 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-08 André Pönitz <apoenitz@trolltech.com>
hausmann@webkit.org [Wed, 8 Oct 2008 09:09:42 +0000 (09:09 +0000)]
2008-10-08  André Pönitz  <apoenitz@trolltech.com>

        Reviewed by Simon.

        Fix compilation with Qt namespaces.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Sam Weinig <sam@webkit.org>
weinig@apple.com [Wed, 8 Oct 2008 06:36:41 +0000 (06:36 +0000)]
2008-10-07  Sam Weinig  <sam@webkit.org>

        Roll out r37405.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoSwitch CTI runtime calls to the fastcall calling convention
oliver@apple.com [Wed, 8 Oct 2008 04:56:38 +0000 (04:56 +0000)]
Switch CTI runtime calls to the fastcall calling convention

Reviewed by Cameron Zwarich

Basically this means that we get to store the argument for CTI
calls in the ECX register, which saves a register->memory write
and subsequent memory->register read.

This is a 1.7% progression in SunSpider and 2.4% on commandline
v8 tests on Windows

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 03:46:34 +0000 (03:46 +0000)]
2008-10-07  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Mark Rowe.

        - optimize away multiplication by constant 1.0

        2.3% speedup on v8 RayTrace benchmark

        Apparently it's not uncommon for JavaScript code to multiply by
        constant 1.0 in the mistaken belief that this converts integer to
        floating point and that there is any operational difference.

        * VM/CTI.cpp:
        (JSC::CTI::privateCompileMainPass): Optimize to_jsnumber for
        case where parameter is already number.
        (JSC::CTI::privateCompileSlowCases): ditto
        * VM/Machine.cpp:
        (JSC::Machine::privateExecute): ditto
        * kjs/grammar.y:
        (makeMultNode): Transform as follows:
        +FOO * BAR ==> FOO * BAR
        FOO * +BAR ==> FOO * BAR
        FOO * 1 ==> +FOO
        1 * FOO ==> +FOO
        (makeDivNode): Transform as follows:
        +FOO / BAR ==> FOO / BAR
        FOO / +BAR ==> FOO / BAR
        (makeSubNode): Transform as follows:
        +FOO - BAR ==> FOO - BAR
        FOO - +BAR ==> FOO - BAR
        * kjs/nodes.h:
        (JSC::ExpressionNode::stripUnaryPlus): Helper for above
        grammar.y changes
        (JSC::UnaryPlusNode::stripUnaryPlus): ditto

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 03:46:04 +0000 (03:46 +0000)]
2008-10-07  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Mark Rowe.

        - test cases that will check some possible edge cases for math optimizations

        * fast/js/math-transforms-expected.txt: Added.
        * fast/js/math-transforms.html: Added.
        * fast/js/resources/math-transforms.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Alp Toker <alp@nuanti.com>
alp@webkit.org [Wed, 8 Oct 2008 00:42:37 +0000 (00:42 +0000)]
2008-10-07  Alp Toker  <alp@nuanti.com>

        GTK+ build fix for older automake versions (1.7). Discussed in bug
        #21392.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 8 Oct 2008 00:15:20 +0000 (00:15 +0000)]
2008-10-07  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        - make constant folding code more consistent

        Added a makeSubNode to match add, mult and div; use the makeFooNode functions always,
        instead of allocating nodes directly in other places in the grammar.

        * kjs/grammar.y:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRoll out r37391 as it appears to have broken 40+ regression tests.
mrowe@apple.com [Wed, 8 Oct 2008 00:12:08 +0000 (00:12 +0000)]
Roll out r37391 as it appears to have broken 40+ regression tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Sam Weinig <sam@webkit.org>
weinig@apple.com [Tue, 7 Oct 2008 23:49:59 +0000 (23:49 +0000)]
2008-10-07  Sam Weinig  <sam@webkit.org>

        Reviewed by Cameron Zwarich.

        Move hasGetterSetterProperties flag from PropertyMap to StructureID.

        * kjs/JSObject.cpp:
        (JSC::JSObject::put):
        (JSC::JSObject::defineGetter):
        (JSC::JSObject::defineSetter):
        * kjs/JSObject.h:
        (JSC::JSObject::hasGetterSetterProperties):
        (JSC::JSObject::getOwnPropertySlotForWrite):
        (JSC::JSObject::getOwnPropertySlot):
        * kjs/PropertyMap.h:
        * kjs/StructureID.cpp:
        (JSC::StructureID::StructureID):
        (JSC::StructureID::addPropertyTransition):
        (JSC::StructureID::toDictionaryTransition):
        (JSC::StructureID::changePrototypeTransition):
        (JSC::StructureID::getterSetterTransition):
        * kjs/StructureID.h:
        (JSC::StructureID::hasGetterSetterProperties):
        (JSC::StructureID::setHasGetterSetterProperties):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Tue, 7 Oct 2008 23:31:34 +0000 (23:31 +0000)]
2008-10-07  Anders Carlsson  <andersca@apple.com>

        Reviewed by Antti Koivisto.

        <rdar://problem/6273887> Crash in ApplicationCacheGroup

        Make sure to stop loading even before a cache update is in progress so that the
        manifest load will be stopped.

        * loader/appcache/ApplicationCacheGroup.cpp:
        (WebCore::ApplicationCacheGroup::~ApplicationCacheGroup):
        (WebCore::ApplicationCacheGroup::stopLoading):
        (WebCore::ApplicationCacheGroup::cacheUpdateFailed):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix the build due to a bad patch apply.
timothy@apple.com [Tue, 7 Oct 2008 22:09:42 +0000 (22:09 +0000)]
Fix the build due to a bad patch apply.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix the build due to a bad patch apply.
timothy@apple.com [Tue, 7 Oct 2008 22:08:02 +0000 (22:08 +0000)]
Fix the build due to a bad patch apply.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Auto-generate DOMDocument's createNodeIterator: and createTreeWalker: methods.
timothy@apple.com [Tue, 7 Oct 2008 21:49:55 +0000 (21:49 +0000)]
    Auto-generate DOMDocument's createNodeIterator: and createTreeWalker: methods.

        https://bugs.webkit.org/show_bug.cgi?id=21433

        Reviewed by Sam Weinig.

        * WebCore.xcodeproj/project.pbxproj: Add ObjCNodeFilterCondition.{mm,h}.
        * bindings/objc/DOM.mm: Remove previous category implementations
        on DOMDocument for createNodeIterator: and createTreeWalker:. Also
        moves ObjCNodeFilterCondition to its own file.
        * bindings/objc/DOMTraversal.h: Remove the category methods that added
        createNodeIterator: and createTreeWalker: to DOMDocument. This is fine
        to do since DOmDocument.h is included still, and has these methods.
        * bindings/objc/DOMUtility.mm:
        (JSC::createDOMWrapper): Remove special case for JSNodeIterator and
        JSTreeWalker now that the ObjC binding conforms to the standard wrap.
        * bindings/objc/ObjCNodeFilterCondition.h: Split out from DOM.mm.
        (WebCore::ObjCNodeFilterCondition::create): Moved from DOM.mm.
        (WebCore::ObjCNodeFilterCondition::ObjCNodeFilterCondition): Ditto.
        * bindings/objc/ObjCNodeFilterCondition.mm: Split out from DOM.mm.
        (WebCore::ObjCNodeFilterCondition::acceptNode): Moved from DOM.mm.
        * bindings/objc/PublicDOMInterfaces.h: Add the createNodeIterator: and
        createTreeWalker: methods to DOMDocument.
        * bindings/scripts/CodeGeneratorObjC.pm:
        (GetObjCTypeGetter): Add a case for NodeFilter.
        (AddIncludesForType): Include ObjCNodeFilterCondition.h for NodeFilter.
        (GenerateImplementation): Remove existing NodeFilter special case that
        used the m_filter member variable. Add a new special getter for protocol
        types that aren't EventTarget, so the right class is used for NodeFilter.
        Add a special case for NodeFilter where it creates an ObjCNodeFilterCondition.
        * dom/Document.idl: Remove the #ifdef LANGUAGE_OBJECTIVE_C. Add the
        OldStyleObjC extended attribute to createNodeIterator and createTreeWalker.
        Rename the entityReferenceExpansion parameter to expandEntityReferences to
        match the previous ObjC API.
        * dom/NodeIterator.idl: Remove ObjCIvar from the filter attribute. The
        m_filter member variable was never used in practice, it was always nil!
        We can remove it and not pad the object because this can't be subclassed.
        * dom/TreeWalker.idl: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Auto-generate the DOMEventTarget protocol implementation for
timothy@apple.com [Tue, 7 Oct 2008 21:49:43 +0000 (21:49 +0000)]
    Auto-generate the DOMEventTarget protocol implementation for
        DOMNode and DOMSVGElementInstance.

        https://bugs.webkit.org/show_bug.cgi?id=21432

        Reviewed by Darin Adler.

        * WebCore.xcodeproj/project.pbxproj: Add ObjCEventListener.{mm,h}.
        * bindings/objc/DOM.mm: Remove many unneeded header includes. Move
        ObjCEventListener to it's own file. Remove the manual impelmentations
        of the DOMEventTarget protocol for DOMNode and DOMSVGElementInstance.
        * bindings/objc/DOMEvents.h: Remove the categories that defined
        DOMEventTarget for DOMNode and DOMSVGElementInstance.
        * bindings/objc/ObjCEventListener.h: Split out from DOM.mm.
        * bindings/objc/ObjCEventListener.mm: Split out from DOM.mm.
        (WebCore::ObjCEventListener::find): Moved from DOM.mm.
        (WebCore::ObjCEventListener::wrap): Use PassRefPtr to prevent the
        callers from doing a manual deref.
        (WebCore::ObjCEventListener::ObjCEventListener): Moved from DOM.mm.
        (WebCore::ObjCEventListener::~ObjCEventListener): Ditto.
        * bindings/scripts/CodeGeneratorObjC.pm:
        (GetObjCTypeGetter): Add a case for EventListener and use WTF::getPtr.
        (AddIncludesForType): Include ObjCEventListener.h for EventListener.
        And include EventTargetSVGElementInstance.h for SVGElementInstance.
        (GenerateHeader): Remove the check for multiple parents.
        (GenerateImplementation): Remove the check for multiple parents. Remove
        the @deprecatedFunctions array since deprecated methods get generated
        into the main @interface now to work with protocols. Add support
        for the EventTargetNodeCast extended attribute. Add support for
        EventListener parameters.
        * dom/Node.idl: Define superclasses for ObjC so the implementation
        and interface implement the DOMEventTarget protocol. Explicitly
        specify Object as a superclass to use DOMEventTarget. Object will turn
        into DOMObject. This is needed to take the code generator down the right
        path of multiple super-classes as protocols. It is ObjC only for legacy
        reasons. The event target methods are normally on NodeEventTarget, a
        subclass of Node. But the ObjC API has never has this subclass and
        they are on DOMNode.
        * svg/SVGElementInstance.idl: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 21:36:39 +0000 (21:36 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Add new pseudo-elements and pseudo-classes that will enable scrollbars to be styled by CSS.

        The new pseudo-elements are:
            scrollbar
            scrollbar-button
            scrollbar-corner
            scrollbar-thumb
            scrollbar-track

        These elements will work with all the usual pseudo-classes (:hover, :active, :enabled, :disabled, etc.) and with
        the following new pseudo-classes:
            scrollbar-active
            scrollbar-back
            scrollbar-forward
            scrollbar-horizontal
            scrollbar-vertical

        Reviewed by Adele

        * css/CSSSelector.cpp:
        (WebCore::CSSSelector::extractPseudoType):
        * css/CSSSelector.h:
        (WebCore::CSSSelector::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Make sure short functions get included in ChangeLog output for git repositories
aroben@apple.com [Tue, 7 Oct 2008 21:35:28 +0000 (21:35 +0000)]
    Make sure short functions get included in ChangeLog output for git repositories

        Reviewed by Dave Kilzer.

        * Scripts/prepare-ChangeLog:
        (sub diffCommand): Pass -U0 to git diff so that each contiguous change
        will get its own chunk without any surrounding context.
        (sub extractLineRange): Use the line numbers from the chunk header
        without modifying them now that they're accurate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Tue, 7 Oct 2008 21:29:48 +0000 (21:29 +0000)]
2008-10-07  Anders Carlsson  <andersca@apple.com>

        Reviewed by Mitz Pettel.

        Update WebKitSystemInterface.

        * libWebKitSystemInterfaceLeopard.a:
        * libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler, Nikolas Zimmermann and Dave Hyatt.
eric@webkit.org [Tue, 7 Oct 2008 21:27:54 +0000 (21:27 +0000)]
    Reviewed by Darin Adler, Nikolas Zimmermann and Dave Hyatt.

        svgElement.className.baseValue = "foo" does not work
        https://bugs.webkit.org/show_bug.cgi?id=20651

        * dom/StyledElement.cpp:
        (WebCore::StyledElement::classAttributeChanged):
        (WebCore::StyledElement::parseMappedAttribute):
        * dom/StyledElement.h:
        * svg/SVGStyledElement.cpp:
        (WebCore::SVGStyledElement::svgAttributeChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37391 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 21:22:05 +0000 (21:22 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        https://bugs.webkit.org/show_bug.cgi?id=18989

        Remove global substitution from SunSpider regexp test, since only Firefox was honoring it and
        we want all browsers to be doing the same thing on the test.

        Reviewed by Maciej

        * tests/regexp-dna.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Add support to the Resources panel for queries like "#123", "foo #123",
timothy@apple.com [Tue, 7 Oct 2008 21:08:29 +0000 (21:08 +0000)]
    Add support to the Resources panel for queries like "#123", "foo #123",
        "line: 123" and "foo line: 123". These will match the query limiting
        the search only to the line specified. If only a line is specified,
        the whole line is matched.

        https://bugs.webkit.org/show_bug.cgi?id=21422

        Reviewed by Darin Adler.

        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype.sourceRow): Don't return the last
        row if the index is greater than the rows collection. Let it return
        undefined by indexing out-of-bounds.
        * inspector/front-end/SourceView.js:
        (WebInspector.SourceView.prototype.performSearch): Add support for
        queries like "#123", "foo #123", "line: 123" and "foo line: 123".
        Also match the whole query to the whole document in case there are
        colors like "#333".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37389 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Sam Weinig <sam@webkit.org>
weinig@apple.com [Tue, 7 Oct 2008 20:49:36 +0000 (20:49 +0000)]
2008-10-07  Sam Weinig  <sam@webkit.org>

        Reviewed by Cameron Zwarich.

        Roll r37370 back in with bug fixes.

        - PropertyMap::storageSize() should reflect the number of keys + deletedOffsets
          and has nothing to do with the internal deletedSentinel count anymore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37388 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 20:32:12 +0000 (20:32 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Move viewless WebKit methods that Safari needs from WebViewInternal to WebViewPrivate.

        Reviewed by Tim Hatcher

        * WebView/WebView.mm:
        (WebKitInitializeApplicationCachePathIfNecessary):
        (-[WebView _registerDraggedTypes]):
        (-[WebView _usesDocumentViews]):
        (-[WebView _commonInitializationWithFrameName:groupName:usesDocumentViews:]):
        (-[WebView _initWithFrame:frameName:groupName:usesDocumentViews:]):
        (-[WebView isFlipped]):
        (-[WebView viewWillDraw]):
        * WebView/WebViewInternal.h:
        * WebView/WebViewPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37387 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Gavin Barraclough <barraclough@apple.com>
barraclough@apple.com [Tue, 7 Oct 2008 20:27:50 +0000 (20:27 +0000)]
2008-10-07  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver Hunt.

        Move callframe initialization into JIT code, again.

        As a part of the restructuring the second result from functions is now
        returned in edx, allowing the new value of 'r' to be returned via a
        register, and stored to the stack from JIT code, too.

        4.5% progression on v8-tests. (3% in their harness)

        * VM/CTI.cpp:
        (JSC::):
        (JSC::CTI::emitCall):
        (JSC::CTI::compileOpCall):
        (JSC::CTI::privateCompileMainPass):
        (JSC::CTI::privateCompileSlowCases):
        (JSC::CTI::privateCompile):
        * VM/CTI.h:
        (JSC::CallRecord::CallRecord):
        * VM/Machine.cpp:
        (JSC::Machine::cti_op_call_JSFunction):
        (JSC::Machine::cti_op_construct_JSConstruct):
        (JSC::Machine::cti_op_resolve_func):
        (JSC::Machine::cti_op_post_inc):
        (JSC::Machine::cti_op_resolve_with_base):
        (JSC::Machine::cti_op_post_dec):
        * VM/Machine.h:
        * kjs/JSFunction.h:
        * kjs/ScopeChain.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37386 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 20:17:58 +0000 (20:17 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Fix crash in isFlipped.  Null check _private since isFlipped can get called from within AppKit
        machinery during teardown of the WebView.

        Reviewed by Adam Roben

        * WebView/WebView.mm:
        (-[WebView isFlipped]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37385 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 20:15:02 +0000 (20:15 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Make viewless WebKit update focused and active state when the window becomes and loses key.
        The focus controller has been patched to understand that in viewless mode it can recur down and
        update all of the frames, which is why this code works now when placed just on the WebView.

        Reviewed by Adam Roben

        * WebView/WebView.mm:
        (-[WebView addWindowObservers]):
        (-[WebView removeWindowObservers]):
        (-[WebView viewWillMoveToWindow:]):
        (-[WebView viewDidMoveToWindow]):
        (-[WebView _updateFocusedAndActiveState]):
        (-[WebView _windowDidBecomeKey:]):
        (-[WebView _windowDidResignKey:]):
        (-[WebView _windowWillOrderOnScreen:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37384 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix typo in method name.
mrowe@apple.com [Tue, 7 Oct 2008 19:35:11 +0000 (19:35 +0000)]
Fix typo in method name.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37383 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Alp Toker <alp@nuanti.com>
alp@webkit.org [Tue, 7 Oct 2008 18:28:43 +0000 (18:28 +0000)]
2008-10-07  Alp Toker  <alp@nuanti.com>

        Reviewed by Mark Rowe.

        https://bugs.webkit.org/show_bug.cgi?id=21392
        [GTK] Auto-generate JS DOM binding sources list

        Remove the huge lists of generated DOM binding sources and headers in
        the build system. These are difficult to maintain and can be derived
        automatically.

        The new strategy is to re-use the existing lists of IDL sources (which
        are needed for dist support anyway). This will also ease the addition
        of new language bindings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37382 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Cameron Zwarich <zwarich@apple.com>
cwzwarich@webkit.org [Tue, 7 Oct 2008 18:17:37 +0000 (18:17 +0000)]
2008-10-07  Cameron Zwarich  <zwarich@apple.com>

        Rubber-stamped by Mark Rowe.

        Roll out r37370.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37381 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Darin Adler <darin@apple.com>
darin@apple.com [Tue, 7 Oct 2008 17:27:12 +0000 (17:27 +0000)]
2008-10-07  Darin Adler  <darin@apple.com>

        * coding/RefPtr.html: Second draft.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37378 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Tor Arne Vestbø <tavestbo@trolltech.com>
vestbo@webkit.org [Tue, 7 Oct 2008 16:30:40 +0000 (16:30 +0000)]
2008-10-07  Tor Arne Vestbø  <tavestbo@trolltech.com>

        Reviewed by Simon.

        Don't attempt to paint when updating control tints

        We don't have a valid PlatformGraphicsContext so schedule
        the dirty scrollbar/scrollview area for repaint instead.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37377 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 Tor Arne Vestbø <tavestbo@trolltech.com>
vestbo@webkit.org [Tue, 7 Oct 2008 11:20:56 +0000 (11:20 +0000)]
2008-10-07  Tor Arne Vestbø  <tavestbo@trolltech.com>

        QtWebKit build fix after changes to FrameLoaderClient.h in r37371

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37376 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 08:09:15 +0000 (08:09 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Make sure the parent visibility state is set up properly on the outermost ScrollView widget.
        This fixes the repainting issues with viewless WebKit on Mac.

        Reviewed by Sam Weinig

        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::transitionToCommittedForNewPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37375 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago[qt] Build fix after Scrollbar.h and Widget.h changes.
zecke@webkit.org [Tue, 7 Oct 2008 07:54:06 +0000 (07:54 +0000)]
[qt] Build fix after Scrollbar.h and Widget.h changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37374 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 07:50:37 +0000 (07:50 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Make sure viewless Mac WebKit does a layout if needed before drawing.

        Reviewed by Sam Weinig

        * WebView/WebView.mm:
        (-[WebView viewWillDraw]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37373 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-07 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 7 Oct 2008 07:47:23 +0000 (07:47 +0000)]
2008-10-07  David Hyatt  <hyatt@apple.com>

        Make sure observers get hooked up to watch for size changes in viewless WebKit mode.

        Reviewed by Sam Weinig

        * ChangeLog:
        * WebView/WebFrame.mm:
        (-[WebFrame _drawRect:contentsOnly:]):
        * WebView/WebView.mm:
        (-[WebView viewWillDraw]):
        (-[WebView _boundsChanged]):
        (-[WebView addSizeObservers]):
        (-[WebView viewDidMoveToWindow]):
        (-[WebView viewDidMoveToSuperview]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37372 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
hyatt@apple.com [Tue, 7 Oct 2008 06:58:08 +0000 (06:58 +0000)]
WebCore:

2008-10-06  David Hyatt  <hyatt@apple.com>

        Enable viewless Mac WebKit to (kinda sorta) paint basic pages (with no frames on them).

        Reviewed by Sam Weinig

        * WebCore.base.exp:
        * WebCore.xcodeproj/project.pbxproj:
        * loader/EmptyClients.h:
        (WebCore::EmptyFrameLoaderClient::hasWebView):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::loadWithDocumentLoader):
        (WebCore::FrameLoader::transitionToCommitted):
        * loader/FrameLoaderClient.h:
        * page/FocusController.cpp:
        (WebCore::FocusController::setActive):
        * page/FrameView.cpp:
        (WebCore::FrameView::FrameView):
        (WebCore::FrameView::init):
        (WebCore::FrameView::layoutIfNeededRecursive):
        * page/FrameView.h:

WebKit/gtk:

2008-10-06  David Hyatt  <hyatt@apple.com>

        Enable viewless Mac WebKit to paint some basic pages.

        Reviewed by Sam Weinig

        * WebCoreSupport/FrameLoaderClientGtk.cpp:
        (WebKit::FrameLoaderClient::transitionToCommittedForNewPage):
        * WebCoreSupport/FrameLoaderClientGtk.h:

WebKit/mac:

2008-10-06  David Hyatt  <hyatt@apple.com>

        Enable viewless Mac WebKit to paint some basic pages.

        Reviewed by Sam Weinig

        * WebCoreSupport/WebChromeClient.mm:
        (WebChromeClient::windowResizerRect):
        (WebChromeClient::repaint):
        (WebChromeClient::screenToWindow):
        (WebChromeClient::windowToScreen):
        (WebChromeClient::platformWindow):
        * WebCoreSupport/WebFrameLoaderClient.h:
        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::forceLayoutForNonHTML):
        (WebFrameLoaderClient::dispatchDidCommitLoad):
        (WebFrameLoaderClient::transitionToCommittedForNewPage):
        (WebFrameLoaderClient::createFrame):
        * WebView/WebFrame.mm:
        (-[WebFrame _drawRect:contentsOnly:]):
        * WebView/WebFrameInternal.h:
        * WebView/WebFrameView.mm:
        * WebView/WebFrameViewInternal.h:
        * WebView/WebHTMLView.mm:
        (-[WebHTMLView drawSingleRect:]):
        (-[WebHTMLView drawRect:]):
        * WebView/WebView.mm:
        (-[WebView isFlipped]):
        (-[WebView _boundsChanged]):
        (-[WebView _mustDrawUnionedRect:singleRects:count:]):
        (-[WebView drawSingleRect:]):
        (-[WebView drawRect:]):
        (-[WebView _commonInitializationWithFrameName:groupName:usesDocumentViews:]):
        (-[WebView initWithFrame:frameName:groupName:]):
        (-[WebView _initWithFrame:frameName:groupName:usesDocumentViews:]):
        (-[WebView initWithCoder:]):
        (-[WebView removeSizeObservers]):
        (-[WebView viewWillMoveToWindow:]):
        (-[WebView viewWillMoveToSuperview:]):
        (-[WebView _usesDocumentViews]):
        * WebView/WebViewInternal.h:

WebKit/qt:

2008-10-06  David Hyatt  <hyatt@apple.com>

        Enable viewless Mac WebKit to paint some basic pages.

        Reviewed by Sam Weinig

        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::transitionToCommittedForNewPage):

WebKit/win:

2008-10-06  David Hyatt  <hyatt@apple.com>

        Enable viewless Mac WebKit to paint some basic pages.

        Reviewed by Sam Weinig

        * WebCoreSupport/WebFrameLoaderClient.cpp:
        (WebFrameLoaderClient::createFrame):
        * WebCoreSupport/WebFrameLoaderClient.h:

WebKit/wx:

2008-10-06  David Hyatt  <hyatt@apple.com>

        Enable viewless Mac WebKit to paint some basic pages.

        Reviewed by Sam Weinig

        * WebKitSupport/FrameLoaderClientWx.cpp:
        * WebKitSupport/FrameLoaderClientWx.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37371 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Sam Weinig <sam@webkit.org>
weinig@apple.com [Tue, 7 Oct 2008 05:43:06 +0000 (05:43 +0000)]
2008-10-06  Sam Weinig  <sam@webkit.org>

        Reviewed by Cameron Zwarich.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=21415
        Improve the division between PropertyStorageArray and PropertyMap

        - Rework ProperyMap to store offsets in the value so that they don't
          change when rehashing.  This allows us not to have to keep the
          PropertyStorageArray in sync and thus not have to pass it in.
        - Rename PropertyMap::getOffset -> PropertyMap::get since put/remove
          now also return offsets.
        - A Vector of deleted offsets is now needed since the storage is out of
          band.

        1% win on SunSpider.  Wash on V8 suite.

        * JavaScriptCore.exp:
        * VM/CTI.cpp:
        (JSC::transitionWillNeedStorageRealloc):
        * VM/Machine.cpp:
        (JSC::Machine::privateExecute):
        Transition logic can be greatly simplified by the fact that
        the storage capacity is always known, and is correct for the
        inline case.
        * kjs/JSObject.cpp:
        (JSC::JSObject::put): Rename getOffset -> get.
        (JSC::JSObject::deleteProperty): Ditto.
        (JSC::JSObject::getPropertyAttributes): Ditto.
        (JSC::JSObject::removeDirect): Use returned offset to
        clear the value in the PropertyNameArray.
        (JSC::JSObject::allocatePropertyStorage): Add assert.
        * kjs/JSObject.h:
        (JSC::JSObject::getDirect): Rename getOffset -> get
        (JSC::JSObject::getDirectLocation): Rename getOffset -> get
        (JSC::JSObject::putDirect): Use propertyStorageCapacity to determine whether
        or not to resize.  Also, since put now returns an offset (and thus
        addPropertyTransition does also) setting of the PropertyStorageArray is
        now done here.
        (JSC::JSObject::transitionTo):
        * kjs/PropertyMap.cpp:
        (JSC::PropertyMap::checkConsistency): PropertyStorageArray is no longer
        passed in.
        (JSC::PropertyMap::operator=): Copy the delete offsets vector.
        (JSC::PropertyMap::put): Instead of setting the PropertyNameArray
        explicitly, return the offset where the value should go.
        (JSC::PropertyMap::remove): Instead of removing from the PropertyNameArray
        explicitly, return the offset where the value should be removed.
        (JSC::PropertyMap::get): Switch to using the stored offset, instead
        of the implicit one.
        (JSC::PropertyMap::insert):
        (JSC::PropertyMap::expand): This is never called when m_table is null,
        so remove that branch and add it as an assertion.
        (JSC::PropertyMap::createTable): Consistency checks no longer take
        a PropertyNameArray.
        (JSC::PropertyMap::rehash): No need to rehash the PropertyNameArray
        now that it is completely out of band.
        * kjs/PropertyMap.h:
        (JSC::PropertyMapEntry::PropertyMapEntry): Store offset into PropertyNameArray.
        (JSC::PropertyMap::get): Switch to using the stored offset, instead
        of the implicit one.
        * kjs/StructureID.cpp:
        (JSC::StructureID::StructureID): Initialize the propertyStorageCapacity to
        JSObject::inlineStorageCapacity.
        (JSC::StructureID::growPropertyStorageCapacity): Grow the storage capacity as
        described below.
        (JSC::StructureID::addPropertyTransition): Copy the storage capacity.
        (JSC::StructureID::toDictionaryTransition): Ditto.
        (JSC::StructureID::changePrototypeTransition): Ditto.
        (JSC::StructureID::getterSetterTransition): Ditto.
        * kjs/StructureID.h:
        (JSC::StructureID::propertyStorageCapacity): Add propertyStorageCapacity
        which is the current capacity for the JSObjects PropertyStorageArray.
        It starts at the JSObject::inlineStorageCapacity (currently 2), then
        when it first needs to be resized moves to the JSObject::nonInlineBaseStorageCapacity
        (currently 16), and after that doubles each time.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37370 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago - build fix
mitz@apple.com [Tue, 7 Oct 2008 04:42:51 +0000 (04:42 +0000)]
    - build fix

        * bindings/objc/DOMEvents.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37369 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Cameron Zwarich <zwarich@apple.com>
cwzwarich@webkit.org [Tue, 7 Oct 2008 03:53:47 +0000 (03:53 +0000)]
2008-10-06  Cameron Zwarich  <zwarich@apple.com>

        Reviewed by Oliver Hunt.

        Bug 21396: Remove the OptionalCalleeActivation call frame slot
        <https://bugs.webkit.org/show_bug.cgi?id=21396>

        Remove the OptionalCalleeActivation call frame slot. We have to be
        careful to store the activation object in a register, because objects
        in the scope chain do not get marked.

        This is a 0.3% speedup on both SunSpider and the V8 benchmark.

        * VM/CTI.cpp:
        (JSC::CTI::privateCompileMainPass):
        * VM/CodeBlock.cpp:
        (JSC::CodeBlock::dump):
        * VM/CodeGenerator.cpp:
        (JSC::CodeGenerator::CodeGenerator):
        (JSC::CodeGenerator::emitReturn):
        * VM/CodeGenerator.h:
        * VM/Machine.cpp:
        (JSC::Machine::dumpRegisters):
        (JSC::Machine::unwindCallFrame):
        (JSC::Machine::privateExecute):
        (JSC::Machine::cti_op_call_JSFunction):
        (JSC::Machine::cti_op_push_activation):
        (JSC::Machine::cti_op_tear_off_activation):
        (JSC::Machine::cti_op_construct_JSConstruct):
        * VM/Machine.h:
        (JSC::Machine::initializeCallFrame):
        * VM/RegisterFile.h:
        (JSC::RegisterFile::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37366 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Tony Chang <tony@chromium.org>
eric@webkit.org [Tue, 7 Oct 2008 01:21:50 +0000 (01:21 +0000)]
2008-10-06  Tony Chang  <tony@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Chromium doesn't use pthreads on windows, so make its use conditional.

        Also convert a WORD to a DWORD to avoid a compiler warning.  This
        matches the other methods around it.

        * wtf/ThreadingWin.cpp:
        (WTF::wtfThreadEntryPoint):
        (WTF::ThreadCondition::broadcast):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37364 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Mark Mentovai <mark@moxienet.com>
eric@webkit.org [Tue, 7 Oct 2008 01:15:43 +0000 (01:15 +0000)]
2008-10-06  Mark Mentovai  <mark@moxienet.com>

        Reviewed by Tim Hatcher.

        Allow ENABLE_DASHBOARD_SUPPORT and ENABLE_MAC_JAVA_BRIDGE to be
        disabled on the Mac.

        https://bugs.webkit.org/show_bug.cgi?id=21333

        * wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37363 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Mark Mentovai <mark@moxienet.com>
eric@webkit.org [Tue, 7 Oct 2008 01:14:05 +0000 (01:14 +0000)]
2008-10-06  Mark Mentovai  <mark@moxienet.com>

        Reviewed by Tim Hatcher.

        Use #if ENABLE(feature) where possible, and #if ENABLE_feature where
        Platform.h is not available, in preference to #ifdef ENABLE_feature.
        #ifdef is wrong now that features are disabled by #defining
        ENABLE_feature to 0.

        https://bugs.webkit.org/show_bug.cgi?id=21338

        * bindings/objc/DOMEvents.h:
        * bindings/objc/PublicDOMInterfaces.h:
        * dom/Document.idl:
        * page/DOMWindow.idl:
        * svg/svgtags.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37362 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Mark Mentovai <mark@moxienet.com>
eric@webkit.org [Tue, 7 Oct 2008 01:11:38 +0000 (01:11 +0000)]
2008-10-06  Mark Mentovai  <mark@moxienet.com>

        Reviewed by Sam Weinig.

        * platform/network/mac/FormDataStreamMac.mm: #import <wtf/Threading.h>
        to get the declaration for isMainThread().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37361 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Jeremy Moskovich <jeremy@chromium.org>
eric@webkit.org [Tue, 7 Oct 2008 01:09:23 +0000 (01:09 +0000)]
2008-10-06  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by Tim Hatcher.

        WebCoreObjCExtras.c is actually an obj-c++ file, so change its name
        to reflect that.

        * WebCore.xcodeproj/project.pbxproj:
        * platform/mac/WebCoreObjCExtras.c: Removed.
        * platform/mac/WebCoreObjCExtras.mm: Copied from WebCore/platform/mac/WebCoreObjCExtras.c.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37358 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Jeremy Moskovich <jeremy@chromium.org>
eric@webkit.org [Tue, 7 Oct 2008 01:05:34 +0000 (01:05 +0000)]
2008-10-06  Jeremy Moskovich  <jeremy@chromium.org>

        Reviewed by Dan Bernstein.

        Added C++ forward declaration for the NSURLAuthenticationChallenge class
        so that the m_currentMacChallenge variable doesn't cause a
        compilation error when ResourceHandleInternal.h is included from a C++ file.
        Fixes: https://bugs.webkit.org/show_bug.cgi?id=21411

        * platform/network/ResourceHandleInternal.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37357 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWindows build fix.
sfalken@apple.com [Tue, 7 Oct 2008 00:17:26 +0000 (00:17 +0000)]
Windows build fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37354 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScriptCore:
sfalken@apple.com [Mon, 6 Oct 2008 23:47:08 +0000 (23:47 +0000)]
JavaScriptCore:

        https://bugs.webkit.org/show_bug.cgi?id=21416
        Pass 0 for size to VirtualAlloc, as documented by MSDN.
        Identified by Application Verifier.

        Reviewed by Darin Adler.

        * kjs/collector.cpp:
        (KJS::freeBlock):

WebCore:

        https://bugs.webkit.org/show_bug.cgi?id=21416
        Add missing null checks identified by Application Verifier.

        Reviewed by Darin Adler.

        * platform/win/SharedTimerWin.cpp:
        (WebCore::clearTimer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37353 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Kevin McCullough <kmccullough@apple.com>
kmccullough@apple.com [Mon, 6 Oct 2008 23:15:11 +0000 (23:15 +0000)]
2008-10-06  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcher.

        Removed accidentally left in debugging statement.

        * inspector/front-end/inspector.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37352 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Kevin McCullough <kmccullough@apple.com>
kmccullough@apple.com [Mon, 6 Oct 2008 23:10:09 +0000 (23:10 +0000)]
2008-10-06  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcher and Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=21412
        Bug 21412: Refactor user initiated profile count to be more stable

        * inspector/InspectorController.cpp: Keep track of the user-initiated
        profiles here now.
        (WebCore::InspectorController::InspectorController):
        (WebCore::InspectorController::startUserInitiatedProfiling):
        (WebCore::InspectorController::stopUserInitiatedProfiling):
        * inspector/InspectorController.h:
        The front end will now need to check for the existence of the user-
        initiated profile title and use its count instead of keeping its own.
        * inspector/front-end/ProfilesPanel.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37351 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Kevin McCullough <kmccullough@apple.com>
kmccullough@apple.com [Mon, 6 Oct 2008 23:09:53 +0000 (23:09 +0000)]
2008-10-06  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcheri and Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=21412
        Bug 21412: Refactor user initiated profile count to be more stable
        - Export UString::from for use with creating the profile title.

        * JavaScriptCore.exp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37350 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScriptCore:
kmccullough@apple.com [Mon, 6 Oct 2008 23:08:16 +0000 (23:08 +0000)]
JavaScriptCore:

2008-10-06  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcher.

        https://bugs.webkit.org/show_bug.cgi?id=21412
        Bug 21412: Refactor user initiated profile count to be more stable
        - Export UString::from for use with creating the profile title.

        * JavaScriptCore.exp:

WebCore:

2008-10-06  Kevin McCullough  <kmccullough@apple.com>

        Reviewed by Tim Hatcher.

        https://bugs.webkit.org/show_bug.cgi?id=21412
        Bug 21412: Refactor user initiated profile count to be more stable

        * inspector/InspectorController.cpp: Keep track of the user-initiated
        profiles here now.
        (WebCore::InspectorController::InspectorController):
        (WebCore::InspectorController::startUserInitiatedProfiling):
        (WebCore::InspectorController::stopUserInitiatedProfiling):
        * inspector/InspectorController.h:
        The front end will now need to check for the existence of the user-
        initiated profile title and use its count instead of keeping its own.
        * inspector/front-end/ProfilesPanel.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37349 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Rubber-stamped by Anders Carlsson.
kdecker@apple.com [Mon, 6 Oct 2008 22:41:15 +0000 (22:41 +0000)]
    Rubber-stamped by Anders Carlsson.

        Rename _webkit_applicationCacheDirectoryWithBundleIdentifier to _webkit_localCacheDirectoryWithBundleIdentifier.

        * Misc/WebKitNSStringExtras.h:
        * Misc/WebKitNSStringExtras.m:
        (+[NSString _webkit_localCacheDirectoryWithBundleIdentifier:]):
        * WebView/WebDataSource.mm:
        (-[WebDataSource _transferApplicationCache:]):
        * WebView/WebView.mm:
        (WebKitInitializeApplicationCachePathIfNecessary):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37348 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Oct 2008 22:22:35 +0000 (22:22 +0000)]
2008-10-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Make WebNetscapePluginStream a WebCore::NetscapePlugInStreamLoaderClient.

        * Plugins/WebBaseNetscapePluginStream.h:
        (WebNetscapePluginStream::~WebNetscapePluginStream):
        * Plugins/WebBaseNetscapePluginStream.mm:
        (WebNetscapePluginStream::didReceiveResponse):
        (-[WebBaseNetscapePluginStream startStreamWithResponse:]):
        (WebNetscapePluginStream::wantsAllStreams):
        (-[WebBaseNetscapePluginStream wantsAllStreams]):
        (WebNetscapePluginStream::didFail):
        (-[WebBaseNetscapePluginStream destroyStreamWithError:]):
        (WebNetscapePluginStream::didFinishLoading):
        (-[WebBaseNetscapePluginStream finishedLoading]):
        (WebNetscapePluginStream::didReceiveData):
        (-[WebBaseNetscapePluginStream receivedData:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37347 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Oct 2008 21:56:30 +0000 (21:56 +0000)]
2008-10-06  Anders Carlsson  <andersca@apple.com>

        Bring back the stop method. It was called through performSelector, which
        is why I couldn't find any references to it.

        * Plugins/WebBaseNetscapePluginStream.h:
        * Plugins/WebBaseNetscapePluginStream.mm:
        (-[WebBaseNetscapePluginStream stop]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37346 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Oct 2008 21:28:52 +0000 (21:28 +0000)]
2008-10-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        More conversion and cleanup.

        * Plugins/WebBaseNetscapePluginStream.h:
        * Plugins/WebBaseNetscapePluginStream.mm:
        (WebNetscapePluginStream::startStream):
        (-[WebBaseNetscapePluginStream startStreamWithResponse:]):
        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView evaluateJavaScriptPluginRequest:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37345 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Oct 2008 21:05:39 +0000 (21:05 +0000)]
2008-10-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Fold initWithRequestURL into initWithRequest.

        * Plugins/WebBaseNetscapePluginStream.h:
        * Plugins/WebBaseNetscapePluginStream.mm:
        (-[WebBaseNetscapePluginStream initWithRequest:plugin:notifyData:sendNotification:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37344 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Oct 2008 20:57:02 +0000 (20:57 +0000)]
2008-10-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Change the init methods not to return nil on failures. (These failures never occur anyway)

        Use initWithRequest as the initializer, initWithRequestURL is going to be merged with it.

        * Plugins/WebBaseNetscapePluginStream.mm:
        (-[WebBaseNetscapePluginStream initWithRequest:plugin:notifyData:sendNotification:]):
        (-[WebBaseNetscapePluginStream initWithRequestURL:plugin:notifyData:sendNotification:]):
        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView evaluateJavaScriptPluginRequest:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37343 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2008-10-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Oct 2008 20:34:16 +0000 (20:34 +0000)]
2008-10-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by David Hyatt.

        Convert more methods over to C++.

        * Plugins/WebBaseNetscapePluginStream.h:
        * Plugins/WebBaseNetscapePluginStream.mm:
        (WebNetscapePluginStream::ownerForStream):
        (WebNetscapePluginStream::pluginCancelledConnectionError):
        (WebNetscapePluginStream::errorForReason):
        (-[WebBaseNetscapePluginStream errorForReason:]):
        (-[WebBaseNetscapePluginStream startStreamResponseURL:expectedContentLength:lastModifiedDate:MIMEType:headers:]):
        (WebNetscapePluginStream::cancelLoadAndDestroyStreamWithError):
        (-[WebBaseNetscapePluginStream cancelLoadAndDestroyStreamWithError:]):
        (WebNetscapePluginStream::deliverData):
        (WebNetscapePluginStream::deliverDataTimerFired):
        (WebNetscapePluginStream::deliverDataToFile):
        (-[WebBaseNetscapePluginStream finishedLoading]):
        (-[WebBaseNetscapePluginStream receivedData:]):
        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView destroyStream:reason:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37342 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Sam Weinig.
mitz@apple.com [Mon, 6 Oct 2008 20:25:46 +0000 (20:25 +0000)]
    Reviewed by Sam Weinig.

        - separate GDI text drawing into its own function

        * platform/graphics/win/FontCGWin.cpp:
        (WebCore::drawGDIGlyphs):
        (WebCore::Font::drawGlyphs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@37341 268f45cc-cd09-0410-ab3c-d52691b4dbfc