WebKit-https.git
11 years ago2009-04-15 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Wed, 15 Apr 2009 17:26:11 +0000 (17:26 +0000)]
2009-04-15  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Alexey Proskuryakov.

        Use a media file inside of LayoutTests/http/ so a cgi isn't necessary to load
        it. Fixes a test failure introduced in r42533.

        * http/tests/resources/silence.mpg: Added.
        * http/tests/security/local-video-source-from-remote-expected.txt:
        * http/tests/security/local-video-source-from-remote.html: change the
        remote url, update a comment.
        * http/tests/security/resources/load-media.cgi: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-15 Eric Roman <eroman@chromium.org>
darin@chromium.org [Wed, 15 Apr 2009 17:20:21 +0000 (17:20 +0000)]
2009-04-15  Eric Roman  <eroman@chromium.org>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=25194

        Fix a caller of Settings::javaScriptCanOpenWindowsAutomatically() to
        use the new name.

        * bindings/v8/custom/V8DOMWindowCustom.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler.
ap@webkit.org [Wed, 15 Apr 2009 15:25:45 +0000 (15:25 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=25186
        There is no XSLT component in bugzilla

        * projects/xslt/index.html: Changed to search by bug title.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-15 Antonio Gomes <antonio.gomes@openbossa.org>
ariya@webkit.org [Wed, 15 Apr 2009 15:09:55 +0000 (15:09 +0000)]
2009-04-15  Antonio Gomes   <antonio.gomes@openbossa.org>

        Reviewed by Ariya Hidayat.

        Rename extend() method in QWebElementSelection to append().

        * Api/qwebelement.cpp:
        (QWebElementSelection::append):
        * Api/qwebelement.h:
        * tests/qwebelement/tst_qwebelement.cpp:
        (tst_QWebElement::appendSelection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-15 Antonio Gomes <antonio.gomes@openbossa.org>
hausmann@webkit.org [Wed, 15 Apr 2009 13:42:34 +0000 (13:42 +0000)]
2009-04-15  Antonio Gomes   <antonio.gomes@openbossa.org>

        Reviewed by Simon Hausmann.

        Fixed nit/typo in QWebElement documentation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMove CallFrameClosure from inside the Interpreter class to its own file.
oliver@apple.com [Wed, 15 Apr 2009 07:31:48 +0000 (07:31 +0000)]
Move CallFrameClosure from inside the Interpreter class to its own file.

Reviewed by NOBODY(Build fix)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBug 25202: Improve performance of repeated callbacks into the VM
oliver@apple.com [Wed, 15 Apr 2009 07:13:25 +0000 (07:13 +0000)]
Bug 25202: Improve performance of repeated callbacks into the VM

Reviewed by Cameron Zwarich

Add the concept of a CachedCall to native code for use in Array
prototype and similar functions where a single callback function
is called repeatedly with the same number of arguments.

Used Array.prototype.filter as the test function and got a 50% win
over a naive non-caching specialised version.  This makes the native
implementation of Array.prototype.filter faster than the JS one once
more.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Rubber-stamped by Darin Adler.
ap@webkit.org [Wed, 15 Apr 2009 06:43:35 +0000 (06:43 +0000)]
    Rubber-stamped by Darin Adler.

        Don't keep platform objects for authentication challenge in ResourceHandleInternal.
        We already have a copy in AuthenticationChallenge object.

        * platform/network/ResourceHandle.cpp:
        (WebCore::ResourceHandle::clearAuthentication):
        * platform/network/ResourceHandleInternal.h:
        (WebCore::ResourceHandleInternal::ResourceHandleInternal):
        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
        (WebCore::ResourceHandle::receivedCredential):
        (WebCore::ResourceHandle::receivedRequestToContinueWithoutCredential):
        (WebCore::ResourceHandle::receivedCancellation):
        * platform/network/mac/ResourceHandleMac.mm:
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
        (WebCore::ResourceHandle::didCancelAuthenticationChallenge):
        (WebCore::ResourceHandle::receivedCredential):
        (WebCore::ResourceHandle::receivedRequestToContinueWithoutCredential):
        (WebCore::ResourceHandle::receivedCancellation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Wed, 15 Apr 2009 05:06:37 +0000 (05:06 +0000)]
2009-04-14  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein

        https://bugs.webkit.org/show_bug.cgi?id=25157

        Move the run loop observer cleanup from -close
        to -_close.

        * WebView/WebView.mm:
        (-[WebView _close]):
        (-[WebView close]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Wed, 15 Apr 2009 04:19:08 +0000 (04:19 +0000)]
2009-04-14  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Nixed some now-defunct autogeneration code.

        * bindings/scripts/CodeGeneratorJS.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Wed, 15 Apr 2009 01:42:27 +0000 (01:42 +0000)]
2009-04-14  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by NOBODY (OOPS!).

        Fix <rdar://problem/6755724> <audio> and <video> elements can reference local
        file:/// URLs from remote in Safari

        Tests: http/tests/security/local-video-poster-from-remote.html
               http/tests/security/local-video-source-from-remote.html
               http/tests/security/local-video-src-from-remote.html

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::loadResource): Don't pass url to media engine if loader->canLoad()
        says it should not be loaded.

2009-04-14  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by NOBODY (OOPS!).

        Fix <rdar://problem/6755724> <audio> and <video> elements can reference local
        file:/// URLs from remote in Safari

        Test cases to ensure that local 'src', 'poster', and <source> are not loaded.

        * http/tests/security/local-video-poster-from-remote-expected.txt: Added.
        * http/tests/security/local-video-poster-from-remote.html: Added.
        * http/tests/security/local-video-source-from-remote-expected.txt: Added.
        * http/tests/security/local-video-source-from-remote.html: Added.
        * http/tests/security/local-video-src-from-remote-expected.txt: Added.
        * http/tests/security/local-video-src-from-remote.html: Added.
        * http/tests/security/resources/load-media.cgi: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
weinig@apple.com [Wed, 15 Apr 2009 01:15:07 +0000 (01:15 +0000)]
WebCore:

2009-04-14  Sam Weinig  <sam@webkit.org>

        Reviewed by Darin Adler.

        Part of <rdar://problem/6150868>
        Fix incorrect handling of content that needs to go into the head element
        once the head element has been removed.

        Test: fast/parser/head-content-after-head-removal.html

        * html/HTMLParser.cpp:
        (WebCore::HTMLParser::HTMLParser): Remove unneeded initializer of m_head.
        (WebCore::HTMLParser::handleError): Update since m_head is now a RefPtr.
        (WebCore::HTMLParser::createHead): Ditto.
        * html/HTMLParser.h: Make m_head a RefPtr.

LayoutTests:

2009-04-14  Sam Weinig  <sam@webkit.org>

        Reviewed by Darin Adler.

        Part of <rdar://problem/6150868>
        Test for incorrect handling of content that needs to go into the head element
        once the head element has been removed.

        * fast/parser/head-content-after-head-removal-expected.txt: Added.
        * fast/parser/head-content-after-head-removal.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 14 Apr 2009 23:11:24 +0000 (23:11 +0000)]
2009-04-14  Geoffrey Garen  <ggaren@apple.com>

        Used svn merge -r42529:42528 to roll out my last patch because it broke
        the build.

        * bindings/js/JSDOMGlobalObject.cpp:
        (WebCore::JSDOMGlobalObject::~JSDOMGlobalObject):
        (WebCore::JSDOMGlobalObject::findJSProtectedEventListener):
        (WebCore::JSDOMGlobalObject::findOrCreateJSProtectedEventListener):
        (WebCore::JSDOMGlobalObject::jsProtectedEventListeners):
        (WebCore::JSDOMGlobalObject::jsProtectedInlineEventListeners):
        * bindings/js/JSDOMGlobalObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 14 Apr 2009 22:54:25 +0000 (22:54 +0000)]
2009-04-14  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Nixed some now-dead code related to protected event listeners.

        * bindings/js/JSDOMGlobalObject.cpp:
        (WebCore::JSDOMGlobalObject::~JSDOMGlobalObject):
        * bindings/js/JSDOMGlobalObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 14 Apr 2009 22:48:42 +0000 (22:48 +0000)]
2009-04-14  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        More fix for https://bugs.webkit.org/show_bug.cgi?id=21260
        Unbounded memory growth when churning elements with anonymous event handler functions

        Stop using protected event listeners on SVGElementInstance.

        * bindings/js/JSSVGElementInstanceCustom.cpp:
        (WebCore::JSSVGElementInstance::mark): Mark our event listeners, since
        they're not protected anymore.

        (WebCore::JSSVGElementInstance::addEventListener):
        (WebCore::JSSVGElementInstance::removeEventListener): Make unprotected
        event listeners, since we mark them now.

        * bindings/scripts/CodeGeneratorJS.pm: Autogenerate event listener
        invalidation for SVGElementInstance.

        * svg/SVGElementInstance.h:
        (WebCore::SVGElementInstance::eventListeners): Added an accessor, for
        the sake of autogenerated code.

        * svg/SVGElementInstance.idl: Removed the ProtectedEventListener attribute
        from event listener properties. Added a CustomMarkFunction attribute,
        since we need to mark our event listeners.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Tue, 14 Apr 2009 22:47:38 +0000 (22:47 +0000)]
2009-04-14  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        - Speculative fix for <rdar://problem/6781422>

        Protect the plug-in instance proxy in case it's deleted while waiting for a reply.

        * Plugins/Hosted/NetscapePluginInstanceProxy.mm:
        (WebKit::NetscapePluginInstanceProxy::wheelEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows build fix
sfalken@apple.com [Tue, 14 Apr 2009 22:30:54 +0000 (22:30 +0000)]
Windows build fix

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoVersioning.
mrowe@apple.com [Tue, 14 Apr 2009 22:26:28 +0000 (22:26 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Skip another test that sometimes times out on Windows
aroben@apple.com [Tue, 14 Apr 2009 22:04:43 +0000 (22:04 +0000)]
    Skip another test that sometimes times out on Windows

        * platform/win/Skipped: Added dom/html/level2/html/table29.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42523 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 21:33:59 +0000 (21:33 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        Fix the build.

        Add the yarr headers (and only the headers) to the build, so that
        RegExp.cpp can compile. The headers are ifdefed out with yarr
        disabled, so we don't need anything else for now.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Add header guards to WebCore forwarding headers.
sfalken@apple.com [Tue, 14 Apr 2009 21:23:56 +0000 (21:23 +0000)]
    Add header guards to WebCore forwarding headers.

        Speeds up WebCore debug build on Windows by 12% on my system,
        since Visual Studio appears to not optimize for standard header
        guards in included files.

        Rubber stamped by Ada Chan.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Skip a test that sometimes times out on Windows
aroben@apple.com [Tue, 14 Apr 2009 20:56:12 +0000 (20:56 +0000)]
    Skip a test that sometimes times out on Windows

        * platform/win/Skipped: Added
        http/tests/incremental/slow-utf8-text.pl.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Sam Weinig <sam@webkit.org>
weinig@apple.com [Tue, 14 Apr 2009 20:53:49 +0000 (20:53 +0000)]
2009-04-14  Sam Weinig  <sam@webkit.org>

        Reviewed by Adele Peterson.

        Use a template function to generalize the way we create non-caching
        JS function getters.

        * bindings/js/JSDOMWindowCustom.cpp:
        * bindings/js/JSDOMWindowCustom.h:
        (WebCore::nonCachingStaticFunctionGetter):
        (WebCore::JSDOMWindow::customGetOwnPropertySlot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Skip another JavaScriptCore test that sometimes fails on Windows
aroben@apple.com [Tue, 14 Apr 2009 20:53:08 +0000 (20:53 +0000)]
    Skip another JavaScriptCore test that sometimes fails on Windows

        See Bug 25160: Various ecma/Date tests sometimes fail on Windows (but
        not Mac)
        <https://bugs.webkit.org/show_bug.cgi?id=25160>

        Rubber-stamped by Geoff Garen.

        * Scripts/run-javascriptcore-tests: Skip ecma/Date/15.9.2.2-3.js.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Skip a sometimes-failing test on Windows
aroben@apple.com [Tue, 14 Apr 2009 20:46:23 +0000 (20:46 +0000)]
    Skip a sometimes-failing test on Windows

        Rubber-stamped by Simon Fraser.

        * platform/win/Skipped: Added
        transitions/zero-duration-with-non-zero-delay-end.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Benjamin C Meyer <benjamin.meyer@torchmobile.com>
hausmann@webkit.org [Tue, 14 Apr 2009 20:46:15 +0000 (20:46 +0000)]
2009-04-14  Benjamin C Meyer  <benjamin.meyer@torchmobile.com>

        Reviewed by George Staikos.

        https://bugs.webkit.org/show_bug.cgi?id=25099

        When creating a QNetworkRequest make sure to populate the
        CacheLoadControlAttribute with the value set by the ResourceRequest::cachePolicy() so that the cache will be used as WebKit expects.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Laszlo Gombos <laszlo.1.gombos@nokia.com>
hausmann@webkit.org [Tue, 14 Apr 2009 20:14:47 +0000 (20:14 +0000)]
2009-04-14  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Simon Hausmann.

        https://bugs.webkit.org/show_bug.cgi?id=24841

        Fix linking against QtWebKit for Symbian and other platforms
        where the OS ABI distinguishes between an import or an export situation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Adele Peterson <adele@apple.com>
adele@apple.com [Tue, 14 Apr 2009 19:53:16 +0000 (19:53 +0000)]
2009-04-14  Adele Peterson  <adele@apple.com>

        Reviewed by Darin.

        Initialize WebKitSystemInterface in class methods that could get called before a WebView/WebFrame is set up.  This was causing Mail to crash on launch.

        * Misc/WebCache.mm: (+[WebCache initialize]):
        * WebView/WebView.mm: (+[WebView initialize]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix. Move the DerivedSources.make calls into build-wxwebkit so that the FEATURE...
kevino@webkit.org [Tue, 14 Apr 2009 19:25:36 +0000 (19:25 +0000)]
Build fix. Move the DerivedSources.make calls into build-wxwebkit so that the FEATURE_DEFINES are parsed properly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Fix a world leak caused by opening the Web Inspector. This was
timothy@apple.com [Tue, 14 Apr 2009 19:06:53 +0000 (19:06 +0000)]
    Fix a world leak caused by opening the Web Inspector. This was
        a regression caused by the InspectorController becoming refcounted.

        <rdar://problem/6782944>

        Reviewed by Darin Adler.

        * bindings/js/ScriptObject.cpp:
        (WebCore::ScriptGlobalObject::get): Renamed to better match the
        other functions.
        (WebCore::ScriptGlobalObject::remove): Added. Deletes the property.
        * bindings/js/ScriptObject.h:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::~InspectorController): Do less work
        at destruction time since the object lifetime is tied to JS GC.
        Assert that inspectedPageDestroyed cleared everything needed.
        (WebCore::InspectorController::inspectedPageDestroyed): Do most
        of the work that ~InspectorController was doing.
        (WebCore::InspectorController::scriptObjectReady): Renamed getObject.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Dmitry Titov <dimich@chromium.org>
dimich@chromium.org [Tue, 14 Apr 2009 18:59:43 +0000 (18:59 +0000)]
2009-04-14  Dmitry Titov  <dimich@chromium.org>

    Fixing ChangeLog.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Antony Sargent <asargent@chromium.org>
dimich@chromium.org [Tue, 14 Apr 2009 18:55:13 +0000 (18:55 +0000)]
2009-04-14  Antony Sargent  <asargent@chromium.org>

        Reviewed by Dimitri Glazkov.

        This is some cleanup motivated by the crash in http://crbug.com/9775 , which
        happens because of calling window.open inside a window.onload handler.

        These changes are just part of the fix, along with some asserts to help prevent
        breakage on future changes.

        https://bugs.webkit.org/show_bug.cgi?id=25132

        * bindings/v8/V8EventListenerList.cpp:
        (WebCore::V8EventListenerList::add):
        (WebCore::V8EventListenerList::remove):
        (WebCore::V8EventListenerList::clear):
        * bindings/v8/V8EventListenerList.h:
        (WebCore::V8EventListenerList::size):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Evan Martin <evan@chromium.org>
dimich@chromium.org [Tue, 14 Apr 2009 18:55:07 +0000 (18:55 +0000)]
2009-04-14  Evan Martin  <evan@chromium.org>

        Reviewed by Darin Adler.

        Fix a race that can occur between flex and the perl script that parses
        its output.

        * css/maketokenizer: Consume all input.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Rohit Rao <rohitrao@google.com>
dimich@chromium.org [Tue, 14 Apr 2009 18:55:00 +0000 (18:55 +0000)]
2009-04-14  Rohit Rao  <rohitrao@google.com>

        Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=25184
        Do not change the text color for selected text in Mac Chromium.

        * rendering/RenderThemeChromiumMac.h:
        (WebCore::RenderThemeChromiumMac::supportsSelectionForegroundColors):
        Now returns false, matching the behavior of Safari on Mac.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Justin Garcia <justin.garcia@apple.com>
justin.garcia@apple.com [Tue, 14 Apr 2009 18:48:16 +0000 (18:48 +0000)]
2009-04-13  Justin Garcia  <justin.garcia@apple.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=25172
        Extra partial layout during the first keypress in an empty block

        Determining if a placeholder needs to be removed requires the creation of a VisiblePosition
        which was triggering layout before we were finished modifying the DOM for the insert operation.

        Find out if we need to remove a placeholder near the start of the operation, and if we do remember
        where it's located.  Then, later, remove it along with all of the other DOM operations we perform.

        Renamed lineBreakExistsAtPosition to lineBreakExistsAtVisiblePosition and added a lineBreakExistsAtPosition
        that takes in a Position.  In InsertTextCommand, we can use the latter, since we normalize manually.

        removePlaceholderAt now takes in a Position that it assumes has already been checked for a placeholder.

        No test added as this only effects performance.

        * editing/BreakBlockquoteCommand.cpp:
        (WebCore::BreakBlockquoteCommand::doApply):
        * editing/CompositeEditCommand.cpp:
        (WebCore::CompositeEditCommand::removePlaceholderAt):
        (WebCore::CompositeEditCommand::moveParagraphs):
        (WebCore::CompositeEditCommand::breakOutOfEmptyMailBlockquotedParagraph):
        (WebCore::CompositeEditCommand::positionAvoidingSpecialElementBoundary):
        * editing/CompositeEditCommand.h:
        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::calculateTypingStyleAfterDelete):
        (WebCore::DeleteSelectionCommand::doApply):
        * editing/FormatBlockCommand.cpp:
        (WebCore::FormatBlockCommand::doApply):
        * editing/InsertLineBreakCommand.cpp:
        (WebCore::InsertLineBreakCommand::doApply):
        * editing/InsertParagraphSeparatorCommand.cpp:
        (WebCore::InsertParagraphSeparatorCommand::doApply):
        * editing/InsertTextCommand.cpp:
        (WebCore::InsertTextCommand::input):
        * editing/htmlediting.cpp:
        (WebCore::lineBreakExistsAtVisiblePosition):
        (WebCore::lineBreakExistsAtPosition):
        * editing/htmlediting.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemove support for profile-guided optimization on Windows
aroben@apple.com [Tue, 14 Apr 2009 18:35:52 +0000 (18:35 +0000)]
Remove support for profile-guided optimization on Windows

JavaScriptCore:

        Remove support for profile-guided optimization on Windows

        Rubber-stamped by Steve Falkenburg.

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Removed
        the Release_PGO configuration. Also let VS re-order the source files
        list.

WebCore:

        Remove support for profile-guided optimization on Windows

        Rubber-stamped by Steve Falkenburg.

        * WebCore.vcproj/WebCore.vcproj: Removed the Release_PGO
        configuration.

WebKit/win:

        Remove support for profile-guided optimization on Windows

        Rubber-stamped by Steve Falkenburg.

        * WebKit.vcproj/WebKit.vcproj: Removed the Release_PGOInstrument and
        Release_PGOOptimize configurations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 18:32:40 +0000 (18:32 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        Unreviewed build fix.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Dmitry Titov <dimich@chromium.org>
dimich@chromium.org [Tue, 14 Apr 2009 18:15:31 +0000 (18:15 +0000)]
2009-04-14  Dmitry Titov  <dimich@chromium.org>

        Reviewed by Dimitri Glazkov.

        https://bugs.webkit.org/show_bug.cgi?id=25163
        Upstream fix for releasing v8 objects on worker thread termination in Chromium.

        * bindings/v8/V8DOMMap.cpp:
        (WebCore::domDataList): Now uses Vector instead of HashMap.
        (WebCore::domDataListMutex):
        (WebCore::ThreadSpecificDOMData::ThreadSpecificDOMData):  remove usage of currentThread();
        (WebCore::ThreadSpecificDOMData::~ThreadSpecificDOMData): ditto.
        (WebCore::NonMainThreadSpecificDOMData::~NonMainThreadSpecificDOMData): moved call to removeAllDOMObjectsInCurrentThread() to ~WorkerScriptController.
        (WebCore::handleWeakObjectInOwningThread):
        (WebCore::derefDelayedObjects):
        (WebCore::removeAllDOMObjectsInCurrentThread): not static anymore.
        * bindings/v8/V8DOMMap.h:
        * bindings/v8/WorkerContextExecutionProxy.cpp:
        (WebCore::WorkerContextExecutionProxy::dispose): removed code that avoided dual-dereference of WorkerContext.
        (WebCore::WorkerContextExecutionProxy::initContextIfNeeded): this ref() is countered in removeAllDOMObjectsInCurrentThread(), when the WorkerContext is removed from the v8 map.
        * bindings/v8/WorkerScriptController.cpp:
        (WebCore::WorkerScriptController::~WorkerScriptController):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Fix Bug 25183: Split up WebCore.vcproj's settings into vsprops files
aroben@apple.com [Tue, 14 Apr 2009 18:08:38 +0000 (18:08 +0000)]
    Fix Bug 25183: Split up WebCore.vcproj's settings into vsprops files

        <https://bugs.webkit.org/show_bug.cgi?id=25183>

        WebCore.vcproj now uses a set of .vsprops files to store most of its
        settings.

        Reviewed by Darin Adler.

        * WebCore.vcproj/WebCore.vcproj: Moved settings from here into the new
        .vsprops files.

        * WebCore.vcproj/WebCoreCFNetwork.vsprops: Added. Contains settings
        for ports that build against CFNetwork.
        * WebCore.vcproj/WebCoreCG.vsprops: Added. Contains settings for ports
        that build against CoreGraphics.
        * WebCore.vcproj/WebCoreCURL.vsprops: Added. Contains settings for
        ports that build against libcurl.
        * WebCore.vcproj/WebCoreCairo.vsprops: Added. Contains settings for
        ports that build against Cairo.
        * WebCore.vcproj/WebCoreCommon.vsprops: Added. Contains settings
        shared by all ports.
        * WebCore.vcproj/WebCoreMediaQT.vsprops: Added. Contains settings for
        ports that use QTMovieWin to implement the MediaPlayerPrivate
        interface.
        * WebCore.vcproj/WebCorePthreads.vsprops: Added. Contains settings for
        ports that build against pthreads.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler.
ap@webkit.org [Tue, 14 Apr 2009 17:56:07 +0000 (17:56 +0000)]
    Reviewed by Darin Adler.

        http/tests/xmlhttprequest/basic-auth.html fails on Tiger.

        Turns out that NSURLCredentialPersistenceNone doesn't work on Tiger, so we have to use
        session persistence.

        * platform/network/mac/ResourceHandleMac.mm:
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge): Route the credential through
        receivedCredential(), ensuring that the Tiger-specific need to use session credentials
        is honored.
        (WebCore::ResourceHandle::receivedCredential): On Tiger, change PersistenceNone to
        PersistenceForSession, because the former doesn't work.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Adele Peterson <adele@apple.com>
adele@apple.com [Tue, 14 Apr 2009 17:40:28 +0000 (17:40 +0000)]
2009-04-13  Adele Peterson  <adele@apple.com>

        Reviewed by Darin Adler.

        Fix for <rdar://problem/6703873> Triple-click quoted line and type Return creates an extra quoted blank line

        Test: editing/inserting/6703873.html

        * editing/BreakBlockquoteCommand.cpp: (WebCore::BreakBlockquoteCommand::doApply):
        If the startNode's original parent is now empty, we can remove it.  Its already been cloned and copied with the startNode,
        so we only need to keep it around if it actually holds some original content, otherwise it will look like an extra empty
        container in the document.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Tue, 14 Apr 2009 16:52:01 +0000 (16:52 +0000)]
2009-04-14  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed. Update gtk-doc control and base sgml files for 1.1.5.

        * docs/webkitgtk-docs.sgml:
        * docs/webkitgtk-sections.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 16:44:47 +0000 (16:44 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        * NEWS: updated for 1.1.5 release.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 16:14:37 +0000 (16:14 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        Unreviewed version bump in preparation for 1.1.5 release.

        * configure.ac:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 15:56:43 +0000 (15:56 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Holger Freyther.

        https://bugs.webkit.org/show_bug.cgi?id=21546
        [GTK] ATK accessibility enhancements

        Add a proper implementation of AtkText::get_text and
        AtkText::get_character_count, which takes into account non
        TextControl text. With this the 'Text' field in Accerciser's
        Interface Viewer is filled correctly.

        * page/gtk/AccessibilityObjectWrapperAtk.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 15:56:28 +0000 (15:56 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Holger Freyther.

        https://bugs.webkit.org/show_bug.cgi?id=21546
        [GTK] ATK accessibility enhancements

        Remove some dead code and g_return_if_fail from static functions.

        * page/gtk/AccessibilityObjectWrapperAtk.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 14 Apr 2009 15:56:10 +0000 (15:56 +0000)]
2009-04-14  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Holger Freyther.

        https://bugs.webkit.org/show_bug.cgi?id=21546
        [GTK] ATK accessibility enhancements

        Also implement text interface for StaticTextRole, which covers
        most of the text in web pages.

        * page/gtk/AccessibilityObjectWrapperAtk.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Kenneth Rohde Christiansen <kenneth.christiansen@openbossa.org>
hausmann@webkit.org [Tue, 14 Apr 2009 15:07:03 +0000 (15:07 +0000)]
2009-04-14  Kenneth Rohde Christiansen  <kenneth.christiansen@openbossa.org>

        Reviewed by Tor Arne Vestbø.

        Rename QWebElement::setHtml and html to setXml and toXml respectivily.

        Also add a mean to define the scope (inner or other).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Land updated Windows results for two tests changed by r42471
aroben@apple.com [Tue, 14 Apr 2009 14:53:32 +0000 (14:53 +0000)]
    Land updated Windows results for two tests changed by r42471

        Rubber-stamped by Anders Carlsson.

        * platform/win/css1/text_properties/text_transform-expected.txt: Copied from LayoutTests/platform/mac-leopard/css1/text_properties/text_transform-expected.txt.
        * platform/win/css2.1/t1605-c545-txttrans-00-b-ag-expected.txt: Copied from LayoutTests/platform/mac-leopard/css2.1/t1605-c545-txttrans-00-b-ag-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Small run-javascriptcore-tests cleanup
aroben@apple.com [Tue, 14 Apr 2009 14:53:10 +0000 (14:53 +0000)]
    Small run-javascriptcore-tests cleanup

        Reviewed by Cameron Zwarich.

        * Scripts/run-javascriptcore-tests: Move the list of tests to skip
        into its own variable, annotated with the bugs that are filed about
        the various failures.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Skip another sometimes-failing ecma/Date test on Windows
aroben@apple.com [Tue, 14 Apr 2009 14:52:55 +0000 (14:52 +0000)]
    Skip another sometimes-failing ecma/Date test on Windows

        These failing tests are covered by Bug 25160: Various ecma/Date tests
        sometimes fail on Windows (but not Mac)
        <https://bugs.webkit.org/show_bug.cgi?id=25160>

        Reviewed by Cameron Zwarich.

        * Scripts/run-javascriptcore-tests: Added ecma/Date/15.9.2.1.js to the
        list of tests to skip.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Kenneth Rohde Christiansen <kenneth.christiansen@openbossa.org>
hausmann@webkit.org [Tue, 14 Apr 2009 14:08:48 +0000 (14:08 +0000)]
2009-04-14  Kenneth Rohde Christiansen  <kenneth.christiansen@openbossa.org>

        Reviewed by Tor Arne Vestbø.

        Rename QWebElement::setText() and text() to ::setPlainText and
        ::toPlainText, respectively.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Tue, 14 Apr 2009 11:30:16 +0000 (11:30 +0000)]
2009-04-14  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Holger Freyther.

        [Gtk] disable soup auth dialog when running DRT
        https://bugs.webkit.org/show_bug.cgi?id=24598

        Disable soup's auth dialog when running the tests.

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (setDefaultsToConsistentStateValuesForTesting):
        * GNUmakefile.am:

2009-04-14  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Holger Freyther.

        [Gtk] disable soup auth dialog when running DRT
        https://bugs.webkit.org/show_bug.cgi?id=24598

        Enable http/tests/xmlhttprequest/failed-auth.html test. We should
        be able to run this test in non-interactive mode

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-14 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Tue, 14 Apr 2009 11:16:01 +0000 (11:16 +0000)]
2009-04-14  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Gtk build fix when building minidom. Not reviewed.

        Use C-style comment instead of C++ style since autotools builds
        minidom using gcc and not g++.

        * wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdate WebKitSystemInterface so that the Tiger portion supports PowerPC.
mrowe@apple.com [Tue, 14 Apr 2009 10:19:13 +0000 (10:19 +0000)]
Update WebKitSystemInterface so that the Tiger portion supports PowerPC.

* libWebKitSystemInterfaceLeopard.a:
* libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Another Windows build fix.
ap@webkit.org [Tue, 14 Apr 2009 08:41:09 +0000 (08:41 +0000)]
    Another Windows build fix.

        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::WebCoreSynchronousLoader::load):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Windows build fix.
ap@webkit.org [Tue, 14 Apr 2009 08:24:37 +0000 (08:24 +0000)]
    Windows build fix.

        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::WebCoreSynchronousLoader::willSendRequest):
        (WebCore::WebCoreSynchronousLoader::didReceiveChallenge):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemoved mentions of tests from ChangeLog - I'll land those once OS-level support...
ap@webkit.org [Tue, 14 Apr 2009 08:16:01 +0000 (08:16 +0000)]
Removed mentions of tests from ChangeLog - I'll land those once OS-level support for this feature is in.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Sam Weinig.
ap@webkit.org [Tue, 14 Apr 2009 08:11:42 +0000 (08:11 +0000)]
    Reviewed by Sam Weinig.

        <rdar://problem/6698851> Implement XMLHttpRequest withCredentials attribute

        Tests: http/tests/xmlhttprequest/cross-origin-authorization.html
               http/tests/xmlhttprequest/cross-origin-no-authorization.html

        * WebCore.xcodeproj/project.pbxproj: Made ThreadableLoader.h private, as enum definitions
        from it are now used in otehr private headers.

        * xml/XMLHttpRequest.h:
        * xml/XMLHttpRequest.idl:
        Added withCredentials attribute. When it is false (default), neither credentials nor cookies
        are sent with cross origin requests, When it is true, those are sent, but the server needs
        to allow handling results via Access-Control-Allow-Credentials header. It was always possible
        to send a cross-site request with credentials via IFRAME or IMG, so this just adds a way to
        read results, as long as the server reports that it's allowed.
        Having the default set to false ensures that requests won't fail unexpectedly because of
        stored credentials and cookies for other resources in the target protection space.

        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::loadRequestSynchronously): Allow stored credentials for same origin
        requests, and for cross origin ones that have withCredentials attribute set. Such code already
        existed for cookies (but it's simpler, because we could just set a ResourceRequest flag).
        (WebCore::XMLHttpRequest::loadRequestAsynchronously): Ditto.

        * platform/network/ResourceHandle.h: Added willSendRequest() - just like for other callbacks,
        is is easier to have code in the class. Also, loadResourceSynchronously() now takes a
        StoredCredentials option, matching async case.

        * platform/network/ResourceHandleClient.h:
        (WebCore::ResourceHandleClient::receivedCredential): Removed. This method could never be
        called, and no client used it.
        (WebCore::ResourceHandleClient::receivedRequestToContinueWithoutCredential): Ditto.

        * platform/network/ResourceHandleInternal.h:
        (WebCore::ResourceHandleInternal::ResourceHandleInternal): Split username and password out
        of request URL. We want to always get a callback for credentials to manage them in WebCore,
        so network back-end shouldn't see them too early.

        * platform/network/ResourceRequestBase.cpp:
        (WebCore::ResourceRequestBase::removeCredentials):
        * platform/network/ResourceRequestBase.h:
        Added a removeCredentials() method that removes login and password parts from request URL.

        * platform/network/mac/ResourceHandleMac.mm:
        (WebCoreCredentialStorage): Added a simple storage class for per-session credentials.
        (WebCore::ResourceHandle::loadResourceSynchronously): Pass allowStoredCredentials through.
        (WebCore::ResourceHandle::willSendRequest): On a redirect, credentials should be replaced.
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge): Try credentials from the URL
        and per-session credentials. Code was partially moved from Obj-C callback.
        (WebCore::ResourceHandle::receivedCredential): Intercept per-session credentials and store
        them in WebCore storage.
        (-[WebCoreResourceHandleAsDelegate connection:willSendRequest:redirectResponse:]): Don't
        store the redirected URL - we only needed credentials, which are now stored separately.
        (-[WebCoreResourceHandleAsDelegate connection:didReceiveAuthenticationChallenge:]): Removed
        code that was setting credentials from URL. First, the code is now in ResourceHandle, and
        also, it wasn't actually needed in Leopard release before this patch, see <rdar://problem/5298142>.
        (-[WebCoreSynchronousLoader dealloc]): Release credentials. Note that unlike ResourceHandle,
        this class still needs to track URL for checking whether a redirect is allowed. This is
        not a great solution, and we should unify client code to use the same checks in sync and
        async cases.
        (-[WebCoreSynchronousLoader connection:willSendRequest:redirectResponse:]): Just like in
        async case, put credentials aside to ensure that network back-end asks for them.
        (-[WebCoreSynchronousLoader connection:didReceiveAuthenticationChallenge:]): Use credentials
        from URL, or from WebCore storage.
        (-[WebCoreSynchronousLoader connectionShouldUseCredentialStorage:]): Don't use stored
        credentials when not allowed to.
        (+[WebCoreSynchronousLoader loadRequest:allowStoredCredentials:returningResponse:error:]):
        Put credentials aside to ensure that network back-end asks for them.

        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::WebCoreCredentialStorage::set):
        (WebCore::WebCoreCredentialStorage::get):
        (WebCore::willSendRequest):
        (WebCore::ResourceHandle::start):
        (WebCore::ResourceHandle::willSendRequest):
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
        (WebCore::ResourceHandle::receivedCredential):
        (WebCore::ResourceHandle::loadResourceSynchronously):
        (WebCore::WebCoreSynchronousLoader::willSendRequest):
        (WebCore::WebCoreSynchronousLoader::didReceiveChallenge):
        (WebCore::WebCoreSynchronousLoader::shouldUseCredentialStorage):
        (WebCore::WebCoreSynchronousLoader::load):
        Same changes as in Mac case.

        * platform/network/curl/ResourceHandleCurl.cpp:
        (WebCore::ResourceHandle::loadResourceSynchronously):
        * platform/network/qt/ResourceHandleQt.cpp:
        (WebCore::ResourceHandle::loadResourceSynchronously):
        * platform/network/soup/ResourceHandleSoup.cpp:
        (WebCore::ResourceHandle::loadResourceSynchronously):
        Trying not to break the build.

        * dom/XMLTokenizerLibxml2.cpp: (WebCore::openFunc):
        * xml/XSLTProcessor.cpp: (WebCore::docLoaderFunc):
        Unconditionally allow stored credentials for these, as they only support same origin loads.

        * workers/WorkerContext.cpp: (WebCore::WorkerContext::importScripts):
        WorkerContext.importScripts() can be cross-origin, but sending credentials with it is no worse
        than sending them with <script src=...>, so this is also unconditionally allowed.

        * loader/DocumentThreadableLoader.cpp:
        (WebCore::DocumentThreadableLoader::loadResourceSynchronously): Pass through storedCredentials.
        (WebCore::DocumentThreadableLoader::create): Ditto.
        (WebCore::DocumentThreadableLoader::DocumentThreadableLoader): Save storedCredentials and
        sameOrigin flags foruse in callbacks.
        (WebCore::DocumentThreadableLoader::willSendRequest): Assert that loaders aren't all confused.
        (WebCore::DocumentThreadableLoader::didSendData): Ditto.
        (WebCore::DocumentThreadableLoader::didReceiveResponse): Ditto.
        (WebCore::DocumentThreadableLoader::didReceiveData): Ditto.
        (WebCore::DocumentThreadableLoader::didFinishLoading): Ditto.
        (WebCore::DocumentThreadableLoader::didFail): Ditto.
        (WebCore::DocumentThreadableLoader::getShouldUseCredentialStorage): Don't use credential
        storage if that's not allowed by the code that invoked DocumentThreadableLoader.
        (WebCore::DocumentThreadableLoader::didReceiveAuthenticationChallenge): Simulate a failure
        and cancel the request if we are about to ask the user for credentials for a cross-origin
        request, which is forbidden by CORS (and would have been very confusing if allowed).
        (WebCore::DocumentThreadableLoader::receivedCancellation): Assert that loaders aren't all confused.

        * loader/DocumentThreadableLoader.h: Updated for the new flags (storedCredentials and
        sameOrigin) that affect the loader. Eventually, we need to move all CORS logic from XHR here.

        * loader/ThreadableLoader.h: (StoredCredentials): Added another flag that affects loader
        behavior. We should combine all of these into a structure, and use it for sync requests, too.

        * loader/FrameLoader.cpp: (WebCore::FrameLoader::loadResourceSynchronously):
        * loader/FrameLoader.h:
        * loader/ThreadableLoader.cpp:
        (WebCore::ThreadableLoader::create):
        (WebCore::ThreadableLoader::loadResourceSynchronously):
        * loader/WorkerThreadableLoader.cpp:
        (WebCore::WorkerThreadableLoader::WorkerThreadableLoader):
        (WebCore::WorkerThreadableLoader::loadResourceSynchronously):
        (WebCore::WorkerThreadableLoader::MainThreadBridge::MainThreadBridge):
        (WebCore::WorkerThreadableLoader::MainThreadBridge::mainThreadCreateLoader):
        * loader/WorkerThreadableLoader.h:
        (WebCore::WorkerThreadableLoader::create):
        Pass through storedCredentials.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agospeculative 'doze build fix
barraclough@apple.com [Tue, 14 Apr 2009 07:19:34 +0000 (07:19 +0000)]
speculative 'doze build fix

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Gavin Barraclough <barraclough@apple.com>
barraclough@apple.com [Tue, 14 Apr 2009 07:06:41 +0000 (07:06 +0000)]
2009-04-13  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Cap'n Geoff Garen.

        Yarr!
        (Yet another regex runtime).

        Currently disabled by default since the interpreter, whilst awesomely
        functional, has not been optimized and is likely slower than PCRE, and
        the JIT, whilst faster than WREC, is presently incomplete and does not
        fallback to using an interpreter for the cases it cannot handle.

        * JavaScriptCore.xcodeproj/project.pbxproj:
        * assembler/MacroAssemblerX86Common.h:
        (JSC::MacroAssemblerX86Common::move):
        (JSC::MacroAssemblerX86Common::swap):
        (JSC::MacroAssemblerX86Common::signExtend32ToPtr):
        (JSC::MacroAssemblerX86Common::zeroExtend32ToPtr):
        (JSC::MacroAssemblerX86Common::branch32):
        (JSC::MacroAssemblerX86Common::branch16):
        * assembler/X86Assembler.h:
        (JSC::X86Assembler::cmpw_im):
        (JSC::X86Assembler::testw_rr):
        (JSC::X86Assembler::X86InstructionFormatter::immediate16):
        * runtime/RegExp.cpp:
        (JSC::RegExp::RegExp):
        (JSC::RegExp::~RegExp):
        (JSC::RegExp::create):
        (JSC::RegExp::compile):
        (JSC::RegExp::match):
        * runtime/RegExp.h:
        * wtf/Platform.h:
        * yarr: Added.
        * yarr/RegexCompiler.cpp: Added.
        (JSC::Yarr::CharacterClassConstructor::CharacterClassConstructor):
        (JSC::Yarr::CharacterClassConstructor::reset):
        (JSC::Yarr::CharacterClassConstructor::append):
        (JSC::Yarr::CharacterClassConstructor::putChar):
        (JSC::Yarr::CharacterClassConstructor::isUnicodeUpper):
        (JSC::Yarr::CharacterClassConstructor::isUnicodeLower):
        (JSC::Yarr::CharacterClassConstructor::putRange):
        (JSC::Yarr::CharacterClassConstructor::charClass):
        (JSC::Yarr::CharacterClassConstructor::addSorted):
        (JSC::Yarr::CharacterClassConstructor::addSortedRange):
        (JSC::Yarr::newlineCreate):
        (JSC::Yarr::digitsCreate):
        (JSC::Yarr::spacesCreate):
        (JSC::Yarr::wordcharCreate):
        (JSC::Yarr::nondigitsCreate):
        (JSC::Yarr::nonspacesCreate):
        (JSC::Yarr::nonwordcharCreate):
        (JSC::Yarr::RegexPatternConstructor::RegexPatternConstructor):
        (JSC::Yarr::RegexPatternConstructor::~RegexPatternConstructor):
        (JSC::Yarr::RegexPatternConstructor::reset):
        (JSC::Yarr::RegexPatternConstructor::assertionBOL):
        (JSC::Yarr::RegexPatternConstructor::assertionEOL):
        (JSC::Yarr::RegexPatternConstructor::assertionWordBoundary):
        (JSC::Yarr::RegexPatternConstructor::atomPatternCharacter):
        (JSC::Yarr::RegexPatternConstructor::atomBuiltInCharacterClass):
        (JSC::Yarr::RegexPatternConstructor::atomCharacterClassBegin):
        (JSC::Yarr::RegexPatternConstructor::atomCharacterClassAtom):
        (JSC::Yarr::RegexPatternConstructor::atomCharacterClassRange):
        (JSC::Yarr::RegexPatternConstructor::atomCharacterClassBuiltIn):
        (JSC::Yarr::RegexPatternConstructor::atomCharacterClassEnd):
        (JSC::Yarr::RegexPatternConstructor::atomParenthesesSubpatternBegin):
        (JSC::Yarr::RegexPatternConstructor::atomParentheticalAssertionBegin):
        (JSC::Yarr::RegexPatternConstructor::atomParenthesesEnd):
        (JSC::Yarr::RegexPatternConstructor::atomBackReference):
        (JSC::Yarr::RegexPatternConstructor::copyDisjunction):
        (JSC::Yarr::RegexPatternConstructor::copyTerm):
        (JSC::Yarr::RegexPatternConstructor::quantifyAtom):
        (JSC::Yarr::RegexPatternConstructor::disjunction):
        (JSC::Yarr::RegexPatternConstructor::regexBegin):
        (JSC::Yarr::RegexPatternConstructor::regexEnd):
        (JSC::Yarr::RegexPatternConstructor::regexError):
        (JSC::Yarr::RegexPatternConstructor::setupAlternativeOffsets):
        (JSC::Yarr::RegexPatternConstructor::setupDisjunctionOffsets):
        (JSC::Yarr::RegexPatternConstructor::setupOffsets):
        (JSC::Yarr::compileRegex):
        * yarr/RegexCompiler.h: Added.
        * yarr/RegexInterpreter.cpp: Added.
        (JSC::Yarr::Interpreter::appendParenthesesDisjunctionContext):
        (JSC::Yarr::Interpreter::popParenthesesDisjunctionContext):
        (JSC::Yarr::Interpreter::DisjunctionContext::DisjunctionContext):
        (JSC::Yarr::Interpreter::DisjunctionContext::operator new):
        (JSC::Yarr::Interpreter::allocDisjunctionContext):
        (JSC::Yarr::Interpreter::freeDisjunctionContext):
        (JSC::Yarr::Interpreter::ParenthesesDisjunctionContext::ParenthesesDisjunctionContext):
        (JSC::Yarr::Interpreter::ParenthesesDisjunctionContext::operator new):
        (JSC::Yarr::Interpreter::ParenthesesDisjunctionContext::restoreOutput):
        (JSC::Yarr::Interpreter::ParenthesesDisjunctionContext::getDisjunctionContext):
        (JSC::Yarr::Interpreter::allocParenthesesDisjunctionContext):
        (JSC::Yarr::Interpreter::freeParenthesesDisjunctionContext):
        (JSC::Yarr::Interpreter::InputStream::InputStream):
        (JSC::Yarr::Interpreter::InputStream::next):
        (JSC::Yarr::Interpreter::InputStream::rewind):
        (JSC::Yarr::Interpreter::InputStream::read):
        (JSC::Yarr::Interpreter::InputStream::readChecked):
        (JSC::Yarr::Interpreter::InputStream::reread):
        (JSC::Yarr::Interpreter::InputStream::prev):
        (JSC::Yarr::Interpreter::InputStream::getPos):
        (JSC::Yarr::Interpreter::InputStream::setPos):
        (JSC::Yarr::Interpreter::InputStream::atStart):
        (JSC::Yarr::Interpreter::InputStream::atEnd):
        (JSC::Yarr::Interpreter::InputStream::checkInput):
        (JSC::Yarr::Interpreter::InputStream::uncheckInput):
        (JSC::Yarr::Interpreter::testCharacterClass):
        (JSC::Yarr::Interpreter::tryConsumeCharacter):
        (JSC::Yarr::Interpreter::checkCharacter):
        (JSC::Yarr::Interpreter::tryConsumeCharacterClass):
        (JSC::Yarr::Interpreter::checkCharacterClass):
        (JSC::Yarr::Interpreter::tryConsumeBackReference):
        (JSC::Yarr::Interpreter::matchAssertionBOL):
        (JSC::Yarr::Interpreter::matchAssertionEOL):
        (JSC::Yarr::Interpreter::matchAssertionWordBoundary):
        (JSC::Yarr::Interpreter::matchPatternCharacter):
        (JSC::Yarr::Interpreter::backtrackPatternCharacter):
        (JSC::Yarr::Interpreter::matchCharacterClass):
        (JSC::Yarr::Interpreter::backtrackCharacterClass):
        (JSC::Yarr::Interpreter::matchBackReference):
        (JSC::Yarr::Interpreter::backtrackBackReference):
        (JSC::Yarr::Interpreter::recordParenthesesMatch):
        (JSC::Yarr::Interpreter::resetMatches):
        (JSC::Yarr::Interpreter::resetAssertionMatches):
        (JSC::Yarr::Interpreter::parenthesesDoBacktrack):
        (JSC::Yarr::Interpreter::matchParenthesesOnceBegin):
        (JSC::Yarr::Interpreter::matchParenthesesOnceEnd):
        (JSC::Yarr::Interpreter::backtrackParenthesesOnceBegin):
        (JSC::Yarr::Interpreter::backtrackParenthesesOnceEnd):
        (JSC::Yarr::Interpreter::matchParentheticalAssertionOnceBegin):
        (JSC::Yarr::Interpreter::matchParentheticalAssertionOnceEnd):
        (JSC::Yarr::Interpreter::backtrackParentheticalAssertionOnceBegin):
        (JSC::Yarr::Interpreter::backtrackParentheticalAssertionOnceEnd):
        (JSC::Yarr::Interpreter::matchParentheses):
        (JSC::Yarr::Interpreter::backtrackParentheses):
        (JSC::Yarr::Interpreter::matchTerm):
        (JSC::Yarr::Interpreter::backtrackTerm):
        (JSC::Yarr::Interpreter::matchAlternative):
        (JSC::Yarr::Interpreter::matchDisjunction):
        (JSC::Yarr::Interpreter::matchNonZeroDisjunction):
        (JSC::Yarr::Interpreter::interpret):
        (JSC::Yarr::Interpreter::Interpreter):
        (JSC::Yarr::ByteCompiler::ParenthesesStackEntry::ParenthesesStackEntry):
        (JSC::Yarr::ByteCompiler::ByteCompiler):
        (JSC::Yarr::ByteCompiler::compile):
        (JSC::Yarr::ByteCompiler::checkInput):
        (JSC::Yarr::ByteCompiler::assertionBOL):
        (JSC::Yarr::ByteCompiler::assertionEOL):
        (JSC::Yarr::ByteCompiler::assertionWordBoundary):
        (JSC::Yarr::ByteCompiler::atomPatternCharacter):
        (JSC::Yarr::ByteCompiler::atomCharacterClass):
        (JSC::Yarr::ByteCompiler::atomBackReference):
        (JSC::Yarr::ByteCompiler::atomParenthesesSubpatternBegin):
        (JSC::Yarr::ByteCompiler::atomParentheticalAssertionBegin):
        (JSC::Yarr::ByteCompiler::popParenthesesStack):
        (JSC::Yarr::ByteCompiler::dumpDisjunction):
        (JSC::Yarr::ByteCompiler::closeAlternative):
        (JSC::Yarr::ByteCompiler::atomParenthesesEnd):
        (JSC::Yarr::ByteCompiler::regexBegin):
        (JSC::Yarr::ByteCompiler::regexEnd):
        (JSC::Yarr::ByteCompiler::alterantiveDisjunction):
        (JSC::Yarr::ByteCompiler::emitDisjunction):
        (JSC::Yarr::byteCompileRegex):
        (JSC::Yarr::interpretRegex):
        * yarr/RegexInterpreter.h: Added.
        (JSC::Yarr::ByteTerm::):
        (JSC::Yarr::ByteTerm::ByteTerm):
        (JSC::Yarr::ByteTerm::BOL):
        (JSC::Yarr::ByteTerm::CheckInput):
        (JSC::Yarr::ByteTerm::EOL):
        (JSC::Yarr::ByteTerm::WordBoundary):
        (JSC::Yarr::ByteTerm::BackReference):
        (JSC::Yarr::ByteTerm::AlternativeBegin):
        (JSC::Yarr::ByteTerm::AlternativeDisjunction):
        (JSC::Yarr::ByteTerm::AlternativeEnd):
        (JSC::Yarr::ByteTerm::PatternEnd):
        (JSC::Yarr::ByteTerm::invert):
        (JSC::Yarr::ByteTerm::capture):
        (JSC::Yarr::ByteDisjunction::ByteDisjunction):
        (JSC::Yarr::BytecodePattern::BytecodePattern):
        (JSC::Yarr::BytecodePattern::~BytecodePattern):
        * yarr/RegexJIT.cpp: Added.
        (JSC::Yarr::RegexGenerator::optimizeAlternative):
        (JSC::Yarr::RegexGenerator::matchCharacterClassRange):
        (JSC::Yarr::RegexGenerator::matchCharacterClass):
        (JSC::Yarr::RegexGenerator::jumpIfNoAvailableInput):
        (JSC::Yarr::RegexGenerator::jumpIfAvailableInput):
        (JSC::Yarr::RegexGenerator::checkInput):
        (JSC::Yarr::RegexGenerator::atEndOfInput):
        (JSC::Yarr::RegexGenerator::notAtEndOfInput):
        (JSC::Yarr::RegexGenerator::jumpIfCharEquals):
        (JSC::Yarr::RegexGenerator::jumpIfCharNotEquals):
        (JSC::Yarr::RegexGenerator::readCharacter):
        (JSC::Yarr::RegexGenerator::storeToFrame):
        (JSC::Yarr::RegexGenerator::loadFromFrame):
        (JSC::Yarr::RegexGenerator::TermGenerationState::TermGenerationState):
        (JSC::Yarr::RegexGenerator::TermGenerationState::resetAlternative):
        (JSC::Yarr::RegexGenerator::TermGenerationState::alternativeValid):
        (JSC::Yarr::RegexGenerator::TermGenerationState::nextAlternative):
        (JSC::Yarr::RegexGenerator::TermGenerationState::alternative):
        (JSC::Yarr::RegexGenerator::TermGenerationState::resetTerm):
        (JSC::Yarr::RegexGenerator::TermGenerationState::termValid):
        (JSC::Yarr::RegexGenerator::TermGenerationState::nextTerm):
        (JSC::Yarr::RegexGenerator::TermGenerationState::term):
        (JSC::Yarr::RegexGenerator::TermGenerationState::lookaheadTerm):
        (JSC::Yarr::RegexGenerator::TermGenerationState::isSinglePatternCharacterLookaheadTerm):
        (JSC::Yarr::RegexGenerator::TermGenerationState::inputOffset):
        (JSC::Yarr::RegexGenerator::TermGenerationState::jumpToBacktrack):
        (JSC::Yarr::RegexGenerator::TermGenerationState::setBacktrackGenerated):
        (JSC::Yarr::RegexGenerator::jumpToBacktrackCheckEmitPending):
        (JSC::Yarr::RegexGenerator::genertateAssertionBOL):
        (JSC::Yarr::RegexGenerator::genertateAssertionEOL):
        (JSC::Yarr::RegexGenerator::matchAssertionWordchar):
        (JSC::Yarr::RegexGenerator::genertateAssertionWordBoundary):
        (JSC::Yarr::RegexGenerator::genertatePatternCharacterSingle):
        (JSC::Yarr::RegexGenerator::genertatePatternCharacterPair):
        (JSC::Yarr::RegexGenerator::genertatePatternCharacterFixed):
        (JSC::Yarr::RegexGenerator::genertatePatternCharacterGreedy):
        (JSC::Yarr::RegexGenerator::genertatePatternCharacterNonGreedy):
        (JSC::Yarr::RegexGenerator::genertateCharacterClassSingle):
        (JSC::Yarr::RegexGenerator::genertateCharacterClassFixed):
        (JSC::Yarr::RegexGenerator::genertateCharacterClassGreedy):
        (JSC::Yarr::RegexGenerator::genertateCharacterClassNonGreedy):
        (JSC::Yarr::RegexGenerator::generateParenthesesSingleDisjunctionOneAlternative):
        (JSC::Yarr::RegexGenerator::generateParenthesesSingle):
        (JSC::Yarr::RegexGenerator::generateTerm):
        (JSC::Yarr::RegexGenerator::generateDisjunction):
        (JSC::Yarr::RegexGenerator::RegexGenerator):
        (JSC::Yarr::RegexGenerator::generate):
        (JSC::Yarr::jitCompileRegex):
        (JSC::Yarr::executeRegex):
        * yarr/RegexJIT.h: Added.
        (JSC::Yarr::RegexCodeBlock::RegexCodeBlock):
        * yarr/RegexParser.h: Added.
        (JSC::Yarr::):
        (JSC::Yarr::Parser::):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::CharacterClassParserDelegate):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::begin):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::atomPatternCharacterUnescaped):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::atomPatternCharacter):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::atomBuiltInCharacterClass):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::end):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::assertionWordBoundary):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::atomBackReference):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::flush):
        (JSC::Yarr::Parser::CharacterClassParserDelegate::):
        (JSC::Yarr::Parser::Parser):
        (JSC::Yarr::Parser::parseEscape):
        (JSC::Yarr::Parser::parseAtomEscape):
        (JSC::Yarr::Parser::parseCharacterClassEscape):
        (JSC::Yarr::Parser::parseCharacterClass):
        (JSC::Yarr::Parser::parseParenthesesBegin):
        (JSC::Yarr::Parser::parseParenthesesEnd):
        (JSC::Yarr::Parser::parseQuantifier):
        (JSC::Yarr::Parser::parseTokens):
        (JSC::Yarr::Parser::parse):
        (JSC::Yarr::Parser::saveState):
        (JSC::Yarr::Parser::restoreState):
        (JSC::Yarr::Parser::atEndOfPattern):
        (JSC::Yarr::Parser::peek):
        (JSC::Yarr::Parser::peekIsDigit):
        (JSC::Yarr::Parser::peekDigit):
        (JSC::Yarr::Parser::consume):
        (JSC::Yarr::Parser::consumeDigit):
        (JSC::Yarr::Parser::consumeNumber):
        (JSC::Yarr::Parser::consumeOctal):
        (JSC::Yarr::Parser::tryConsume):
        (JSC::Yarr::Parser::tryConsumeHex):
        (JSC::Yarr::parse):
        * yarr/RegexPattern.h: Added.
        (JSC::Yarr::CharacterRange::CharacterRange):
        (JSC::Yarr::):
        (JSC::Yarr::PatternTerm::):
        (JSC::Yarr::PatternTerm::PatternTerm):
        (JSC::Yarr::PatternTerm::BOL):
        (JSC::Yarr::PatternTerm::EOL):
        (JSC::Yarr::PatternTerm::WordBoundary):
        (JSC::Yarr::PatternTerm::invert):
        (JSC::Yarr::PatternTerm::capture):
        (JSC::Yarr::PatternTerm::quantify):
        (JSC::Yarr::PatternAlternative::PatternAlternative):
        (JSC::Yarr::PatternAlternative::lastTerm):
        (JSC::Yarr::PatternAlternative::removeLastTerm):
        (JSC::Yarr::PatternDisjunction::PatternDisjunction):
        (JSC::Yarr::PatternDisjunction::~PatternDisjunction):
        (JSC::Yarr::PatternDisjunction::addNewAlternative):
        (JSC::Yarr::RegexPattern::RegexPattern):
        (JSC::Yarr::RegexPattern::~RegexPattern):
        (JSC::Yarr::RegexPattern::reset):
        (JSC::Yarr::RegexPattern::containsIllegalBackReference):
        (JSC::Yarr::RegexPattern::newlineCharacterClass):
        (JSC::Yarr::RegexPattern::digitsCharacterClass):
        (JSC::Yarr::RegexPattern::spacesCharacterClass):
        (JSC::Yarr::RegexPattern::wordcharCharacterClass):
        (JSC::Yarr::RegexPattern::nondigitsCharacterClass):
        (JSC::Yarr::RegexPattern::nonspacesCharacterClass):
        (JSC::Yarr::RegexPattern::nonwordcharCharacterClass):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMissed code from last patch
oliver@apple.com [Tue, 14 Apr 2009 06:39:22 +0000 (06:39 +0000)]
Missed code from last patch

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 14 Apr 2009 06:33:26 +0000 (06:33 +0000)]
2009-04-13  David Hyatt  <hyatt@apple.com>

        Reviewed by Cameron Zwarich.

        Remove an outdated comment in ScrollView.h

        * platform/ScrollView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBUG 25171: It should be possible to manually set the name of an anonymous function
oliver@apple.com [Tue, 14 Apr 2009 06:24:43 +0000 (06:24 +0000)]
BUG 25171: It should be possible to manually set the name of an anonymous function
<https://bugs.webkit.org/show_bug.cgi?id=25171>

Submitted by Francisco Tolmasky <francisco@280north.com>
Reviewed by Oliver Hunt.

This change adds the displayName property to functions, which when set overrides the
normal name when appearing in the console.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Tue, 14 Apr 2009 05:03:55 +0000 (05:03 +0000)]
2009-04-13  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Mark Rowe.

        https://bugs.webkit.org/show_bug.cgi?id=25177

        QTTrack's -media method always return NULL in 64-bit QTKit, so
        MediaPlayerPrivate::disableUnsupportedTracks should use
        QTTrackMediaTypeAttribute to get a track's type.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivate::disableUnsupportedTracks):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 14 Apr 2009 04:52:51 +0000 (04:52 +0000)]
2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Cameron Zwarich.

        More fix for https://bugs.webkit.org/show_bug.cgi?id=21260
        Unbounded memory growth when churning elements with anonymous event handler functions

        Stop using protected event listeners on the window object.

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::mark): Mark our event listeners, since they're
        no longer protected.

        (WebCore::JSDOMWindow::addEventListener):
        (WebCore::JSDOMWindow::removeEventListener): Create unprotected event
        listeners, since we mark them now.

        * bindings/js/JSEventListener.h: Made some functions public so
        DOMWindow could call them.

        * bindings/js/JSNodeCustom.cpp: Moved markEventListeners to a header,
        so it could be shared.

        * bindings/scripts/CodeGeneratorJS.pm: Generate event listener marking
        and invalidating code for the DOMWindow.

        * dom/RegisteredEventListener.h:
        (WebCore::markEventListeners):
        (WebCore::invalidateEventListeners): Added helper functions.

        * page/DOMWindow.idl: Make the window's event listener attributes not
        protected.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Tue, 14 Apr 2009 04:13:58 +0000 (04:13 +0000)]
2009-04-13  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Mark Rowe.

        OS X standalone media documents should not switch from <video> to <embed> when
        QTKit load state is QTMovieLoadStateLoading as that happens frequently when a loading
        with a slow network connection or when a movie is not saved fast-start.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivate::updateStates): Don't open movie with plug-in when QTKit's
        loadstate is QTMovieLoadStateLoading, wait until it drops below QTMovieLoadStateError.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Dmitry Titov <dimich@chromium.org>
dimich@chromium.org [Tue, 14 Apr 2009 01:03:14 +0000 (01:03 +0000)]
2009-04-13  Dmitry Titov  <dimich@chromium.org>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=24453
        REGRESSION: If setTimeout is called on a iframe's window, the DOM changes to
        the main document that timer callback makes are not flushed.

        * bindings/js/ScheduledAction.cpp:
        (WebCore::ScheduledAction::execute):
        Use Document::updateStyleForAllDocuments() instead of document->updateStyleIfNeeded()
        since timers may affect documents other then their own.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Tue, 14 Apr 2009 00:21:32 +0000 (00:21 +0000)]
2009-04-13  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by David Hyatt.

        Check document for nullity to avoid crashing on GTK+, when
        creating the WebView.

        * page/FrameView.cpp:
        (WebCore::FrameView::needsLayout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Forgot to capitalize the "US".
adachan@apple.com [Tue, 14 Apr 2009 00:19:46 +0000 (00:19 +0000)]
    Forgot to capitalize the "US".

        * platform/text/win/TextBreakIteratorInternalICUWin.cpp:
        (WebCore::currentTextBreakLocaleID):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago <rdar://problem/6564956> URL field double-click selects entire domain, rather...
adachan@apple.com [Tue, 14 Apr 2009 00:10:14 +0000 (00:10 +0000)]
    <rdar://problem/6564956> URL field double-click selects entire domain, rather than single word or entire URL
        Revert back to the old word breaking behavior by using en_US_POSIX locale.

        Reviewed by Steve Falkenburg.

        * platform/text/win/TextBreakIteratorInternalICUWin.cpp:
        (WebCore::currentTextBreakLocaleID):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin.
kdecker@apple.com [Mon, 13 Apr 2009 23:42:35 +0000 (23:42 +0000)]
    Reviewed by Darin.

        <rdar://problem/6784955> REGRESSION: closing a tab containing a PDF causes world leaks

        Simplify the _trackFirstResponder method by just caching the value instead of retaining
        a Cocoa object.

        * WebView/WebPDFView.h: Eliminated trackedFirstResponder object and replaced it with
        a firstResponderIsPDFDocumentView boolean.
        * WebView/WebPDFView.mm:
        (-[WebPDFView dealloc]): Removed no longer necessary ASSERT.
        (-[WebPDFView viewWillMoveToWindow:]): Removed call to release and nil-out trackedFirstResponder,
        which no longer exists. In the new code all we do now is set firstResponderIsPDFDocumentView to NO.
        (-[WebPDFView _trackFirstResponder]): Rewrote this method to just cache the value instead
        of retaining an object.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Darin Adler <darin@apple.com>
darin@apple.com [Mon, 13 Apr 2009 22:36:43 +0000 (22:36 +0000)]
2009-04-13  Darin Adler  <darin@apple.com>

        Reviewed by Mark Rowe.

        * Scripts/check-for-exit-time-destructors: Add ERROR: and WARNING: prefixes to
        make these scripts work better with the build window in new versions of Xcode.
        * Scripts/check-for-global-initializers: Ditto.
        * Scripts/check-for-weak-vtables: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Darin Adler <darin@apple.com>
darin@apple.com [Mon, 13 Apr 2009 22:18:32 +0000 (22:18 +0000)]
2009-04-13  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        * platform/cf/BinaryPropertyList.cpp: Added comment.
        * platform/cf/BinaryPropertyList.h: Fixed mistakes in comments.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
darin@apple.com [Mon, 13 Apr 2009 22:16:57 +0000 (22:16 +0000)]
WebCore:

2009-04-13  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Bug 25158: hit testing doesn't work right in some cases involving anonymous blocks
        https://bugs.webkit.org/show_bug.cgi?id=25158
        rdar://problem/6769693

        Test: editing/selection/click-after-nested-block.html

        * rendering/RenderBlock.cpp:
        (WebCore::isChildHitTestCandidate): Added. Helper so the two loops in positionForPoint
        can share code.
        (WebCore::RenderBlock::positionForPoint): Removed unneeded special case for first
        child box. Changed special case for last child box to propertly handle the various
        types of child box that are not hit test candidates (the same ones already handled
        by the hit test loop below), adding in a loop so we find the last candidate.

LayoutTests:

2009-04-13  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein.

        Bug 25158: hit testing doesn't work right in some cases involving anonymous blocks
        https://bugs.webkit.org/show_bug.cgi?id=25158
        rdar://problem/6769693

        * editing/selection/click-after-nested-block-expected.txt: Added.
        * editing/selection/click-after-nested-block.html: Added.

        * fast/events/pointer-events-expected.txt: Updated since this is now a plain-text test.
        * fast/events/pointer-events.html: Changed this to a plain-text test, because the old
        results reflected a difference in hit testing behavior that this test was not intended
        to exercise and it's better to have the test be more cross-platform. Also removed
        unneeded use of waitUntilDone.

        * fast/js/finally-codegen-failure.html: Replaced with standard wrapper created by
        running the make-js-test-wrappers script. If we want to include type and charset
        attributes in the script elements, we can change make-js-test-wrappers to do that
        for all tests at some point.
        * fast/js/function-apply-aliased.html: Ditto.
        * fast/js/function-call-aliased.html: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoJavaScriptCore:
ggaren@apple.com [Mon, 13 Apr 2009 21:44:54 +0000 (21:44 +0000)]
JavaScriptCore:

2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Disabled another JavaScriptCore test because it fails on Windows but
        not Mac, so it makes the bots red.

        * tests/mozilla/expected.html:

WebKitTools:

2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Disabled another JavaScriptCore test because it fails on Windows but
        not Mac, so it makes the bots red.

        * Scripts/run-javascriptcore-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdate accessibility layout test from last checkin.
hyatt@apple.com [Mon, 13 Apr 2009 21:39:29 +0000 (21:39 +0000)]
Update accessibility layout test from last checkin.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebcore:
hyatt@apple.com [Mon, 13 Apr 2009 21:34:21 +0000 (21:34 +0000)]
Webcore:

2009-04-13  David Hyatt  <hyatt@apple.com>

        Reviewed by Sam Weinig.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=25125.  Rework scrolling so that a layout happens first when
        it's already needed so that the code doesn't end up making bad decisions based off invalid document sizes.

        * WebCore.base.exp:
        * page/FrameView.h:
        (WebCore::FrameView::visibleContentsResized):
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::ScrollView):
        (WebCore::ScrollView::updateScrollbars):
        * platform/ScrollView.h:

WebKit/mac:

2009-04-13  David Hyatt  <hyatt@apple.com>

        Reviewed by Sam Weinig.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=25125.  Rework scrolling so that a layout happens first when
        it's already needed so that the code doesn't end up making bad decisions based off invalid document sizes.

        This patch also eliminates WebHTMLView's separate notion of needing a layout and just consolidates it with
        WebCore's notion of needing layout.

        * WebView/WebDynamicScrollBarsView.m:
        (-[WebDynamicScrollBarsView updateScrollers]):
        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _frameOrBoundsChanged]):
        (-[WebHTMLView initWithFrame:]):
        (-[WebHTMLView layoutToMinimumPageWidth:maximumPageWidth:adjustingViewSize:]):
        (-[WebHTMLView setNeedsLayout:]):
        (-[WebHTMLView _layoutIfNeeded]):
        (-[WebHTMLView _needsLayout]):
        * WebView/WebHTMLViewInternal.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows part 2 of <rdar://5438063> Saving history containing 100,000 entries causes...
aroben@apple.com [Mon, 13 Apr 2009 21:31:02 +0000 (21:31 +0000)]
Windows part 2 of <rdar://5438063> Saving history containing 100,000 entries causes pauses of 2s while browsing

WebCore:

        Add SharedBuffer::wrapCFData

        This is the CF-equivalent of wrapNSData.

        Reviewed by Brady Eidson.

        * platform/SharedBuffer.h:
        * platform/cf/SharedBufferCF.cpp:
        (WebCore::SharedBuffer::wrapCFData):

WebKit/win:

        Windows part 2 of <rdar://5438063> Saving history containing 100,000
        entries causes pauses of 2s while browsing

        Reviewed by Brady Eidson.

        * Interfaces/IWebHistoryPrivate.idl: Added data, analagous to
        WebKit/mac's -[WebHistory _data] method.

        * WebHistory.cpp:
        (WebHistory::saveHistoryGuts): Changed to call data() to get the data for
        saving.
        (WebHistory::data): Added. Returns the data for saving as an IStream.
        (WebHistory::data): Added. Returns the data for saving as a CFDataRef.

        * WebHistory.h: Added data.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Mon, 13 Apr 2009 21:27:14 +0000 (21:27 +0000)]
2009-04-13  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Rubber-stamped by Holger Freyther.

        Further massage tests to disable tests that are failingto make the
        results more useful.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoJavaScriptCore:
ggaren@apple.com [Mon, 13 Apr 2009 21:09:43 +0000 (21:09 +0000)]
JavaScriptCore:

2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Disabled two JavaScriptCore tests because they fail on Window or Mac but
        not both, so they make the bots red.

        * tests/mozilla/expected.html: Updated expected results.

WebKitTools:

2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Disabled two JavaScriptCore tests because they fail on Window or Mac but
        not both, so they make the bots red.

        * Scripts/run-javascriptcore-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdding a comment to my previous code change.
justin.garcia@apple.com [Mon, 13 Apr 2009 20:57:04 +0000 (20:57 +0000)]
Adding a comment to my previous code change.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Darin Adler <darin@apple.com>
darin@apple.com [Mon, 13 Apr 2009 20:55:49 +0000 (20:55 +0000)]
2009-04-13  Darin Adler  <darin@apple.com>

        * WebView/WebViewPrivate.h: Updated comments.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago - build fix
mitz@apple.com [Mon, 13 Apr 2009 20:48:38 +0000 (20:48 +0000)]
    - build fix

        * page/EventHandler.cpp:
        (WebCore::documentPointForWindowPoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
antti@apple.com [Mon, 13 Apr 2009 20:46:34 +0000 (20:46 +0000)]
WebCore:

2009-04-13  Antti Koivisto  <antti@apple.com>

        Reviewed by Darin Adler.

        <rdar://problem/6740294> Increase the connection count per host

        Try to use six connections per host if possible.

        * WebCore.base.exp:
        * loader/loader.cpp:
        (WebCore::Loader::Loader):
        * platform/mac/WebCoreSystemInterface.h:
        * platform/mac/WebCoreSystemInterface.mm:
        * platform/network/ResourceRequestBase.cpp:
        (WebCore::initMaximumHTTPConnectionCountPerHost):
        * platform/network/ResourceRequestBase.h:
        * platform/network/mac/ResourceRequestMac.mm:
        (WebCore::initMaximumHTTPConnectionCountPerHost):

WebKit/mac:

2009-04-13  Antti Koivisto  <antti@apple.com>

        Reviewed by Darin Adler.

        <rdar://problem/6740294> Increase the connection count per host

        * WebCoreSupport/WebSystemInterface.m:
        (InitWebCoreSystemInterface):

WebKitLibraries:

2009-04-13  Antti Koivisto  <antti@apple.com>

        Reviewed by Darin Adler.

        <rdar://problem/6740294> Increase the connection count per host

        * WebKitSystemInterface.h:
        * libWebKitSystemInterfaceLeopard.a:
        * libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
justin.garcia@apple.com [Mon, 13 Apr 2009 20:38:03 +0000 (20:38 +0000)]
WebCore:

2009-04-13  Justin Garcia  <justin.garcia@apple.com>

        Reviewed by Adele Peterson.

        https://bugs.webkit.org/show_bug.cgi?id=25153
        Can't place the caret into text field that scrolls the window on focus

        We refetch the target node in the shadow node case, and when we do so the window coordinate for the mouse event may
        be invalid because of scrolling that the focus handler did.  Cache the document point (that we derived from the window
        coordinate) and use that any time we refetch.

        * page/EventHandler.cpp:
        (WebCore::documentPointForWindowPoint):
        (WebCore::EventHandler::handleMousePressEvent):
        (WebCore::EventHandler::prepareMouseEvent):

LayoutTests:

2009-04-13  Justin Garcia  <justin.garcia@apple.com>

        Reviewed by Adele Peterson.

        <rdar://problem/6748324> REGRESSION (Kirkwood): Can't type in Netflix search field

        * fast/forms/25153-expected.txt: Added.
        * fast/forms/25153.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Sam Weinig <sam@webkit.org>
weinig@apple.com [Mon, 13 Apr 2009 20:29:45 +0000 (20:29 +0000)]
2009-04-13  Sam Weinig  <sam@webkit.org>

        Fix bad expected result.

        * fast/frames/frame-unload-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Sam Weinig <sam@webkit.org>
weinig@apple.com [Mon, 13 Apr 2009 20:26:42 +0000 (20:26 +0000)]
2009-04-13  Sam Weinig  <sam@webkit.org>

        Reviewed by Geoffrey Garen.

        Rename JavaScriptCanOpenWindowsAutomatically to javaScriptCanOpenWindowsAutomatically
        to conform to our style guidelines.

        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::allowPopUp):
        * page/Settings.h:
        (WebCore::Settings::javaScriptCanOpenWindowsAutomatically):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Mon, 13 Apr 2009 20:24:43 +0000 (20:24 +0000)]
2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Mac build fix: Made another header private.

        * WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Jian Li <jianli@chromium.org>
dimich@chromium.org [Mon, 13 Apr 2009 20:20:53 +0000 (20:20 +0000)]
2009-04-13  Jian Li  <jianli@chromium.org>

        Reviewed by Darin Fisher.

        https://bugs.webkit.org/show_bug.cgi?id=25121
        Upstream changes to handle error messages from V8 when running worker script.

        * bindings/v8/WorkerContextExecutionProxy.cpp:
        (WebCore::reportFatalErrorInV8):
        (WebCore::handleConsoleMessage):
        (WebCore::WorkerContextExecutionProxy::WorkerContextExecutionProxy):
        (WebCore::WorkerContextExecutionProxy::dispose):
        (WebCore::WorkerContextExecutionProxy::initV8IfNeeded):
        (WebCore::WorkerContextExecutionProxy::initContextIfNeeded):
        (WebCore::WorkerContextExecutionProxy::evaluate):
        * bindings/v8/WorkerContextExecutionProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Brady Eidson <beidson@apple.com>
beidson@apple.com [Mon, 13 Apr 2009 20:18:06 +0000 (20:18 +0000)]
2009-04-13  Brady Eidson  <beidson@apple.com>

        Reviewed by Tim Hatcher

        Tweak my last check-in, moving the thread violation check up to the API-level calls so the logging is more
        useful to developers/users.

        * WebView/WebFrame.mm:
        (-[WebFrame _loadData:MIMEType:textEncodingName:baseURL:unreachableURL:]): If not on the main thread, only
          perform the "call on main thead" workaround, as the log/exception raising is now up at the API level.
        (-[WebFrame loadData:MIMEType:textEncodingName:baseURL:]): Perform a thread violation check here so logging
          is more meaningful.
        (-[WebFrame loadHTMLString:baseURL:]): Ditto.
        (-[WebFrame loadAlternateHTMLString:baseURL:forUnreachableURL:]): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Mon, 13 Apr 2009 20:16:36 +0000 (20:16 +0000)]
2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Build fix: Made setInlineEventListenerForType not inline, since Document
        calls it now. Made RegisteredEventListener.h a private header on Mac,
        so WebKit can use it.

        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::setInlineEventListenerForType):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Sam Weinig.
mitz@apple.com [Mon, 13 Apr 2009 20:14:47 +0000 (20:14 +0000)]
    Reviewed by Sam Weinig.

        - fix <rdar://problem/6404439> REGRESSION: Overlapping text and other
          layout issues in View Sample feature of MYOB FIrstEdge v3

        * page/Navigator.cpp:
        (WebCore::shouldHideFourDot): Added "tdqm_loader.js" to the list of
        JavaScript files this quirk applies to.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Brady Eidson <beidson@apple.com>
beidson@apple.com [Mon, 13 Apr 2009 20:03:18 +0000 (20:03 +0000)]
2009-04-13  Brady Eidson  <beidson@apple.com>

        Reviewed by Kevin Decker

        <rdar://problem/6712063> Garmin WebUpdater crashes

        * WebView/WebFrame.mm:
        (-[WebFrame _loadData:MIMEType:textEncodingName:baseURL:unreachableURL:]):  If not called on the main thread,
          add a ThreadViolationCheckRoundTwo() call to either log or raise an exception.  In the case where it's
          only a log, reschedule the _loadData call to occur on the main thread.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
weinig@apple.com [Mon, 13 Apr 2009 19:50:32 +0000 (19:50 +0000)]
WebCore:

2009-04-13  Sam Weinig  <sam@webkit.org>

        Reviewed by Oliver Hunt.

        Move open and showModalDialog functions out of JSDOMWindowBase
        and into JSDOMWindow by partially generating them.

        This slightly changes the behavior of getting window.showModalDialog
        when the embedding app does not support it to return undefined regardless
        of shadowing.  This should not be an issue in a practice, but will be addressed
        when we add a generic way to handle runtime specialization of property
        access.

        * DerivedSources.make: Remove JSDOMWindowBase.lut.h
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * bindings/js/JSDOMWindowBase.cpp:
        (WebCore::):
        (WebCore::JSDOMWindowBase::getOwnPropertySlot): Move canShowModalDialog check from
        here to JSDOMWindow::customGetOwnPropertySlot.
        (WebCore::JSDOMWindowBase::put): Remove dead code.
        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::createWindow):
        (WebCore::JSDOMWindow::open):
        (WebCore::JSDOMWindow::showModalDialog):
        * bindings/js/JSDOMWindowCustom.h:
        (WebCore::JSDOMWindow::customGetOwnPropertySlot):
        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::parseModalDialogFeatures): Move from JSDOMWindowBase.
        (WebCore::DOMWindow::allowPopUp): Ditto.
        (WebCore::DOMWindow::canShowModalDialog): Ditto.
        (WebCore::DOMWindow::canShowModalDialogNow): Ditto.
        * page/DOMWindow.h:
        * page/DOMWindow.idl:

LayoutTests:

2009-04-13  Sam Weinig  <sam@webkit.org>

        Reviewed by Oliver Hunt.

        Update tests after moving showModalDialog from JSDOMWindowBase to
        JSDOMWindow and slightly changing the behavior when showModalDialog
        is not available, as is the case for DumpRenderTree.

        * fast/dom/Window/window-function-frame-getter-precedence-expected.txt:
        * fast/dom/Window/window-properties-expected.txt:
        * platform/mac-leopard/fast/dom/Window/window-properties-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Mon, 13 Apr 2009 19:43:47 +0000 (19:43 +0000)]
2009-04-13  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Moved storage for window event listeners into the window object.

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::addEventListener):
        (WebCore::JSDOMWindow::removeEventListener): Add / remove window event
        listeners to / from the window object.

        * dom/Document.cpp:
        (WebCore::Document::removeAllEventListeners):
        (WebCore::Document::clear):
        (WebCore::Document::setWindowInlineEventListenerForTypeAndAttribute): Moved
        a bunch of code to the DOMWindow class, and left behind a few calls through
        to DOMWindow, to support legacy clients. (Eventually, these will go away, too.)

        * dom/Document.h: ditto.

        * dom/Node.cpp:
        (WebCore::Node::dispatchGenericEvent):
        (WebCore::Node::dispatchWindowEvent): Dipatch window events on the window
        object, not the document.

        * dom/Node.h:
        * dom/RegisteredEventListener.h: Moved RegisteredEventListenerVector
        declaration, since it's required in more places now.

        * history/CachedFrame.cpp:
        (WebCore::CachedFrame::clear): Updated for rename. Also, remove event
        listeners before detaching and clearing the document's window pointer,
        since the the document needs its window pointer in order to tell the window
        to clear its event listeners.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::stopLoading):
        (WebCore::FrameLoader::canCachePageContainingThisFrame):
        (WebCore::FrameLoader::logCanCacheFrameDecision): Updated for rename, and
        movement of window event listeners into the window.

        * loader/ImageDocument.cpp:
        (WebCore::ImageDocument::createDocumentStructure): Updated for movement
        of window event listeners into the window.

        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::handleEvent):
        (WebCore::DOMWindow::addEventListener):
        (WebCore::DOMWindow::removeEventListener):
        (WebCore::DOMWindow::removeAllEventListeners):
        (WebCore::DOMWindow::hasEventListener):
        (WebCore::DOMWindow::setInlineEventListenerForType):
        (WebCore::DOMWindow::removeInlineEventListenerForType):
        (WebCore::DOMWindow::inlineEventListenerForType):
        (WebCore::DOMWindow::addPendingFrameUnloadEventCount):
        (WebCore::DOMWindow::removePendingFrameUnloadEventCount):
        (WebCore::DOMWindow::addPendingFrameBeforeUnloadEventCount):
        (WebCore::DOMWindow::removePendingFrameBeforeUnloadEventCount):
        * page/DOMWindow.h: Moved a bunch of code to the DOMWindow class, from the Document.

        * page/Frame.cpp:
        (WebCore::Frame::shouldClose): Updated for movement of window event listeners into the window.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Greg Bolsinga <bolsinga@apple.com>
bolsinga@apple.com [Mon, 13 Apr 2009 19:05:51 +0000 (19:05 +0000)]
2009-04-13  Greg Bolsinga  <bolsinga@apple.com>

        Reviewed by Sam Weinig.

        https://bugs.webkit.org/show_bug.cgi?id=25149

        The Geolocation timer needs to start after the UA has granted permission.

        * bindings/js/JSGeolocationCustom.cpp: Use PassRefPtr<PositionOptions>
        (WebCore::JSGeolocation::getCurrentPosition):
        (WebCore::JSGeolocation::watchPosition):
        * page/Geolocation.cpp: Add a RefPtr<PositionOptions> so it can be used
        later when the timer is started. Change PositionOptions* to PassRefPtr<PositionOptions>
        where needed.
        (WebCore::Geolocation::GeoNotifier::GeoNotifier):
        (WebCore::Geolocation::GeoNotifier::startTimer):
        (WebCore::Geolocation::getCurrentPosition):
        (WebCore::Geolocation::watchPosition):
        (WebCore::Geolocation::setIsAllowed): Start the timer is the UA allows
        (WebCore::Geolocation::startTimer):
        (WebCore::Geolocation::startTimersForOneShots):
        (WebCore::Geolocation::startTimersForWatchers):
        (WebCore::Geolocation::startTimers):
        * page/Geolocation.h:
        (WebCore::Geolocation::GeoNotifier::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Pamela Greene <pam@chromium.org>
pam@chromium.org [Mon, 13 Apr 2009 18:01:58 +0000 (18:01 +0000)]
2009-04-13  Pamela Greene  <pam@chromium.org>

        Reviewed by Darin Fisher.

        Fix Chromium (V8) build by renaming methods to match changes
        in http://trac.webkit.org/changeset/42377.

        https://bugs.webkit.org/show_bug.cgi?id=25141

        * bindings/v8/ScheduledAction.cpp:
     (WebCore::ScheduledAction::execute): updateRendering() -> updateStyleIfNeeded()
        * bindings/v8/V8AbstractEventListener.cpp:
        (WebCore::V8AbstractEventListener::handleEvent): updateDocumentsRendering() -> updateStyleForAllDocuments()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Darin Adler <darin@apple.com>
darin@apple.com [Mon, 13 Apr 2009 17:52:42 +0000 (17:52 +0000)]
2009-04-13  Darin Adler  <darin@apple.com>

        * Scripts/make-js-test-wrappers: Added another exception.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-04-13 Darin Fisher <darin@chromium.org>
darin@chromium.org [Mon, 13 Apr 2009 17:47:33 +0000 (17:47 +0000)]
2009-04-13  Darin Fisher  <darin@chromium.org>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=25136
        CRASH in DocumentLoader::removeSubresourceLoader due to null m_frame

        Test: fast/frames/frame-unload-crash.html

        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::updateLoading): Added null check of m_frame.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@42442 268f45cc-cd09-0410-ab3c-d52691b4dbfc