WebKit-https.git
12 years agoUpdate build-webkit to handle a new version of WebKitSystemInterface.
mrowe@apple.com [Thu, 28 May 2009 06:31:53 +0000 (06:31 +0000)]
Update build-webkit to handle a new version of WebKitSystemInterface.

Rubber-stamped by Dan Bernstein.

* Scripts/build-webkit:

Update WebKitSystemInterface.

* libWebKitSystemInterfaceLeopard.a:
* libWebKitSystemInterfaceSnowLeopard.a: Added.
* libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
abarth@webkit.org [Thu, 28 May 2009 06:25:02 +0000 (06:25 +0000)]
WebCore:

2009-05-27  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Clean up window.open()'s use of lexical and dynamic scope.

        Test: http/tests/security/frameNavigation/context-for-window-open.html

        * bindings/js/JSDOMBinding.cpp:
        (WebCore::toDynamicFrame):
        (WebCore::processingUserGesture):
        (WebCore::completeURL):
        * bindings/js/JSDOMBinding.h:
        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::createWindow):
        (WebCore::JSDOMWindow::open):
        (WebCore::JSDOMWindow::showModalDialog):
        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::createWindow):

LayoutTests:

2009-05-27  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Test whether lexical or dynamic scope is used for window.open().

        * http/tests/security/frameNavigation/context-for-window-open-expected.txt: Added.
        * http/tests/security/frameNavigation/context-for-window-open.html: Added.
        * http/tests/security/frameNavigation/resources/middle-frame-for-location.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Oliver Hunt.
ap@webkit.org [Thu, 28 May 2009 06:02:24 +0000 (06:02 +0000)]
    Reviewed by Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=26056
        XPath string() function can be very slow

        * xml/XPathUtil.cpp: (WebCore::XPath::stringValue): Use an intermediate Vector when appending.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Thu, 28 May 2009 05:30:46 +0000 (05:30 +0000)]
2009-05-27  Maciej Stachowiak  <mjs@apple.com>

        Not reviewed, demo content.

        - Add doctype to Calendar demo.

        * demos/calendar/Calendar.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
mitz@apple.com [Thu, 28 May 2009 05:25:53 +0000 (05:25 +0000)]
WebCore:

        Reviewed by Sam Weinig.

        - fix <rdar://problem/6927761> <isindex> placeholder text is unstylable
          and initially not visible

        Test: fast/forms/isindex-placeholder.html

        * css/html4.css: Added a default style for <isindex> placeholder text.
        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::parseMappedAttribute): Changed to call
        the new protected function updatePlaceholderVisibility().
        * html/HTMLInputElement.h:
        (WebCore::HTMLInputElement::updatePlaceholderVisibility): Added this
        protected function to allow HTMLIsIndexElement::parseMappedAttribute()
        to invoke InputElement::updatePlaceholderVisibility().
        * html/HTMLIsIndexElement.cpp:
        (WebCore::HTMLIsIndexElement::parseMappedAttribute): Call
        updatePlaceholderVisibility() when parsing the placeholder attribute.
        * rendering/RenderTextControlSingleLine.cpp:
        (WebCore::RenderTextControlSingleLine::createInnerTextStyle): If there
        is no placeholder pseudo-element style, use the normal style.

LayoutTests:

        Reviewed by Sam Weinig.

        - test for <rdar://problem/6927761> <isindex> placeholder text is
          unstylable and initially not visible

        * fast/forms/isindex-placeholder.html: Added.
        * platform/mac/fast/forms/isindex-placeholder-expected.checksum: Added.
        * platform/mac/fast/forms/isindex-placeholder-expected.png: Added.
        * platform/mac/fast/forms/isindex-placeholder-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Sam Weinig <sam@webkit.org>
weinig@apple.com [Thu, 28 May 2009 05:06:10 +0000 (05:06 +0000)]
2009-05-27  Sam Weinig  <sam@webkit.org>

        Reviewed by Maciej Stachowiak.

        More cleanup of DOMWindow related functions.

        * bindings/js/JSDOMWindowBase.cpp:
        (WebCore::JSDOMWindowBase::willRemoveFromWindowShell): Renamed from JSDOMWindowBase::clear()
        * bindings/js/JSDOMWindowBase.h:

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::showModalDialog): Update Comment.

        * bindings/js/JSDOMWindowShell.cpp:
        * bindings/js/JSDOMWindowShell.h:
        Remove JSDOMWindowShell::clear().  It was unused.

        * bindings/js/ScriptController.cpp:
        (WebCore::ScriptController::clearWindowShell):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago<rdar://problem/6928025> Stack overflow in JSC::stringProtoFuncReplace() running...
oliver@apple.com [Thu, 28 May 2009 03:36:03 +0000 (03:36 +0000)]
<rdar://problem/6928025> Stack overflow in JSC::stringProtoFuncReplace() running jsFunFuzz

Reviewed by Mark Rowe

Add an additional exception check for stack overflow in String.replace

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Sam Weinig <sam@webkit.org>
weinig@apple.com [Thu, 28 May 2009 03:17:30 +0000 (03:17 +0000)]
2009-05-27  Sam Weinig  <sam@webkit.org>

        Reviewed by David Hyatt.

        Miscellaneous cleanup of DOMWindow related functions.

        * bindings/js/JSDOMGlobalObject.cpp:
        (WebCore::JSDOMGlobalObject::~JSDOMGlobalObject): Fix poorly named variables.
        * bindings/js/JSDOMWindowBase.cpp:
        (WebCore::JSDOMWindowBase::supportsProfiling): Don't jump through hoops
        checking for null frames if the build does not support profiling.
        (WebCore::JSDOMWindowBase::clear): Use setCurrentEvent(0) instead of calling
        clearHelperObjectProperties().  It is clearer.
        * bindings/js/JSDOMWindowBase.h:
        Removed now unused clearHelperObjectProperties() function, empty disconnectFrame()
        function, and the empty destructor.
        * bindings/js/JSDOMWindowShell.cpp:
        * bindings/js/JSDOMWindowShell.h:
        Remove disconnectFrame() which only called JSDOMWindowBase::disconnectFrame(), which
        is a no-op.
        * page/Frame.cpp:
        (WebCore::Frame::~Frame):
        (WebCore::Frame::pageDestroyed):
        Remove calls to JSDOMWindowShell::disconnectFrame() which only called
        JSDOMWindowBase::disconnectFrame(), which is a no-op.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Gustavo Noronha Silva <gns@gnome.org>
kov@webkit.org [Thu, 28 May 2009 00:13:22 +0000 (00:13 +0000)]
2009-05-27  Gustavo Noronha Silva  <gns@gnome.org>

        Reviewed by Xan Lopez.

        Add comment to clarify why scrolling is done as it is and not with
        gdk_window_scroll.

        * WebCoreSupport/ChromeClientGtk.cpp:
        (WebKit::ChromeClient::scroll):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 May 2009 22:16:23 +0000 (22:16 +0000)]
2009-05-27  Dimitri Glazkov  <dglazkov@chromium.org>

        Unreviewed, build fix.

        https://bugs.webkit.org/show_bug.cgi?id=25974
        Remove extra qualifiers from the ScheduledAction decl.

        * bindings/v8/ScheduledAction.h: Removed extraneous class qualifiers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Wed, 27 May 2009 21:23:30 +0000 (21:23 +0000)]
2009-05-27  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed make distcheck fix. One more missing file.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Antonio Gomes <antonio.gomes@openbossa.org>
hausmann@webkit.org [Wed, 27 May 2009 21:09:55 +0000 (21:09 +0000)]
2009-05-27  Antonio Gomes   <antonio.gomes@openbossa.org>

        Reviewed by Simon Hausmann.

        [Qt] Fixed trunk build on Linux after r44126.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Wed, 27 May 2009 20:46:54 +0000 (20:46 +0000)]
2009-05-27  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Oliver Hunt.

        <rdar://problem/6926046> REGRESSION (r43972): http://www.youtube.com/html5 crashes on open

        * platform/graphics/MediaPlayer.cpp:
        (WebCore::MediaPlayer::load): Don't look for a media engine based on an empty MIME type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Wed, 27 May 2009 20:04:40 +0000 (20:04 +0000)]
2009-05-27  Maciej Stachowiak  <mjs@apple.com>

        Not reviewed, demo content.

        - add Calendar demo.

        * demos/calendar: Added.
        * demos/calendar/Boom.aiff: Added.
        * demos/calendar/Calendar.css: Added.
        * demos/calendar/Calendar.html: Added.
        * demos/calendar/Calendar.js: Added.
        * demos/calendar/Calendar.manifest: Added.
        * demos/calendar/CalendarApp.icns: Added.
        * demos/calendar/Images: Added.
        * demos/calendar/Images/AirPort4.png: Added.
        * demos/calendar/Images/AirPortError.png: Added.
        * demos/calendar/Images/disclosureTriangleSmallDown.png: Added.
        * demos/calendar/Images/disclosureTriangleSmallRight.png: Added.
        * demos/calendar/Images/statusbarBackground.png: Added.
        * demos/calendar/Images/statusbarResizerVertical.png: Added.
        * demos/calendar/LocationImage.js: Added.
        * demos/calendar/Utilities.js: Added.
        * demos/calendar/favicon.ico: Added.
        * demos/calendar/mime.types: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Wed, 27 May 2009 19:20:42 +0000 (19:20 +0000)]
2009-05-27  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed (make distcheck) build fix; adding missing headers.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Jessie Berlin <jberlin@apple.com>
jberlin@apple.com [Wed, 27 May 2009 19:14:12 +0000 (19:14 +0000)]
2009-05-27  Jessie Berlin  <jberlin@apple.com>

        Reviewed by aroben

        Fix the Windows build.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 David Levin <levin@chromium.org>
levin@chromium.org [Wed, 27 May 2009 19:03:30 +0000 (19:03 +0000)]
2009-05-27  David Levin  <levin@chromium.org>

        Reviewed by Maciej Stachowiak.

        Bug 26029: FrameLoader::canLoad should allow calls with just a security origin
        https://bugs.webkit.org/show_bug.cgi?id=26029

        No functional changes, so no tests.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::canLoad):
        * loader/FrameLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler.
ap@webkit.org [Wed, 27 May 2009 18:01:34 +0000 (18:01 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=26009
        <rdar://problem/6925656> REGRESSION (r43973): Problems While Working With OpenVanilla

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _updateSelectionForInputManager]): Don't call updateWindows if the selection
        is None. This routinely happens during editing, and it doesn't mean that we left an editable
        area (in which case the selection changes to a non-editable one).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Wed, 27 May 2009 17:57:25 +0000 (17:57 +0000)]
2009-05-27  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed build fix.

        Remove file that does not exist from the build.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44193 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoReviewed by Darin Adler.
zimmermann@webkit.org [Wed, 27 May 2009 17:33:09 +0000 (17:33 +0000)]
Reviewed by Darin Adler.

Remove two unneeded protected constructors, these classes have pure-virtual functions, and can't be constructed anyway.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Wed, 27 May 2009 17:27:37 +0000 (17:27 +0000)]
2009-05-27  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Unreviewed. Adding new symbols index for 1.1.8.

        * docs/webkitgtk-docs.sgml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44191 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler.
ap@webkit.org [Wed, 27 May 2009 17:21:02 +0000 (17:21 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=17725
        XPath should be case insensitive for HTML

        Test: fast/xpath/node-name-case-sensitivity.html

        * xml/XPathStep.cpp: (WebCore::XPath::Step::nodeMatches): Made node name check case insensitive
        for HTML elements in HTML documents.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44190 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 John Sullivan <sullivan@apple.com>
sullivan@apple.com [Wed, 27 May 2009 17:00:40 +0000 (17:00 +0000)]
2009-05-27  John Sullivan  <sullivan@apple.com>

        fixed <rdar://problem/6925482> repro crash in WebCore::DragController::dragExited dropping
        bookmarks (at least) over Top Sites (at least)

        Reviewed by Kevin Decker

        * page/DragController.cpp:
        (WebCore::DragController::dragExited):
        nil check m_documentUnderMouse and take the "local file" case if it's nil

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44189 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore and JavaScriptCore
kov@webkit.org [Wed, 27 May 2009 14:02:31 +0000 (14:02 +0000)]
WebCore and JavaScriptCore

2009-05-27  Fridrich Strba  <fridrich.strba@bluewin.ch>

        Reviewed by Gustavo Noronha.

        When building on Windows, consider Windows specific files.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoReviewed by George Staikos.
zimmermann@webkit.org [Wed, 27 May 2009 12:25:19 +0000 (12:25 +0000)]
Reviewed by George Staikos.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=26031
InputElementData is too heavy, save storing two pointers by passing them
directly to the static helper functions (the only place which needs them).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoReviewed by George Staikos.
zimmermann@webkit.org [Wed, 27 May 2009 12:23:54 +0000 (12:23 +0000)]
Reviewed by George Staikos.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=26033
OptionElementData saves an Element pointer, which is unnessary.
Just pass it to the static function calls directly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Wed, 27 May 2009 12:17:00 +0000 (12:17 +0000)]
2009-05-27  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Disabling new tests to make the bot green.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-27 Fridrich Strba <fridrich.strba@bluewin.ch>
kov@webkit.org [Wed, 27 May 2009 12:04:33 +0000 (12:04 +0000)]
2009-05-27  Fridrich Strba  <fridrich.strba@bluewin.ch>

        Reviewed by Maciej Stachowiak.

        When building with MinGW, don't use the __declspec(dl{import,export})
        decorations and rely on the linker to use its nifty auto-import feature.
        It is extremely hard to get the decorations right with MinGW in general
        and impossible in WebKit, where the resulting shared library is linking
        together some static libraries.

        * JavaScriptCore/config.h
        * WebCore/config.h
        * DumpRenderTree/config.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScripCore:
zecke@webkit.org [Wed, 27 May 2009 10:21:49 +0000 (10:21 +0000)]
JavaScripCore:
2009-05-26  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=25613

        Be able to use GOwnPtr for GHashTable as well. The assumption
        is that the hash table has been created with g_hash_table_new_full
        and has proper destruction functions.

        * wtf/GOwnPtr.cpp:
        (WTF::GHashTable):
        * wtf/GOwnPtr.h:

WebCore:
2009-05-26  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=25613

        Add a WebCore::Widget that can embed GtkWidget with and
        without a GdkWindow. This can be used to embed any GtkWidget.

        Some bits are directly copied from the Scrollbar implementation
        but can not be easily shared here.

        * GNUmakefile.am:
        * platform/gtk/GtkPluginWidget.cpp: Added.
        (WebCore::GtkPluginWidget::GtkPluginWidget):
        (WebCore::GtkPluginWidget::invalidateRect):
        (WebCore::GtkPluginWidget::frameRectsChanged):
        (WebCore::GtkPluginWidget::paint):
        * platform/gtk/GtkPluginWidget.h: Added.

WebKit:
2009-05-26  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=25613

        Add a new signal called "create-plugin-widget" and emit it from
        within the FrameLoaderClientGtk::createPlugin method. If a
        GtkWidget gets returned it will be embedded as a WebCore::Widget
        using the new GtkPluginWidget.

        * WebCoreSupport/FrameLoaderClientGtk.cpp:
        (WebKit::FrameLoaderClient::createPlugin):
        * webkit/webkitwebview.cpp:
        (webkit_signal_accumulator_object_handled):
        (webkit_web_view_class_init):
        * webkitmarshal.list:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Holger Hans Peter Freyther <zecke@selfish.org>
zecke@webkit.org [Wed, 27 May 2009 10:21:10 +0000 (10:21 +0000)]
2009-05-26  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Gustavo Noronha.

        https://bugs.webkit.org/show_bug.cgi?id=25613

        Implement Widget::setIsSelected for Gtk+ by searching
        for a property of the name "webkit-widget-is-selected" and if
        such property exists we are going to set it. We expect
        the property to take a boolean.

        * platform/gtk/WidgetGtk.cpp:
        (WebCore::Widget::setIsSelected):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
weinig@apple.com [Wed, 27 May 2009 06:53:38 +0000 (06:53 +0000)]
WebCore:

2009-05-26  Sam Weinig  <sam@webkit.org>

        Reviewed by Brady Eidson.

        Fix for <rdar://problem/6916371>
        iWeb 'Announce' button does nothing after publishing to MobileMe

        Add ability to force content sniffing for all urls (including file: urls)

        * WebCore.base.exp:
        * platform/network/ResourceHandle.cpp:
        (WebCore::ResourceHandle::shouldContentSniffURL):
        (WebCore::ResourceHandle::forceContentSniffing):
        * platform/network/ResourceHandle.h:

WebKit/mac:

2009-05-26  Sam Weinig  <sam@webkit.org>

        Reviewed by Brady Eidson.

        Fix for <rdar://problem/6916371>
        iWeb 'Announce' button does nothing after publishing to MobileMe

        Add linked-on-or-after check to allow older WebKit apps to use the
        old behavior of sniffing everything (including file: urls)

        * Misc/WebKitVersionChecks.h:
        * WebView/WebView.mm:
        (-[WebView _commonInitializationWithFrameName:groupName:usesDocumentViews:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Benjamin C Meyer <benjamin.meyer@torchmobile.com>
zecke@webkit.org [Wed, 27 May 2009 06:35:59 +0000 (06:35 +0000)]
2009-05-26  Benjamin C Meyer  <benjamin.meyer@torchmobile.com>

        Reviewed by Holger Freyther.

        https://bugs.webkit.org/show_bug.cgi?id=25823

        Add an API to generate a QUrl out of a string correcting for errors and missing information.
        See the API documentation for more details about the function.

        Autotests: included

        * Api/qwebview.cpp:
        (QWebView::guessUrlFromString):
        * Api/qwebview.h:
        * QtLauncher/main.cpp:
        (MainWindow::MainWindow):
        (MainWindow::changeLocation):
        * tests/qwebview/.gitignore: Added.
        * tests/qwebview/qwebview.pro:
        * tests/qwebview/tst_qwebview.cpp:
        (tst_QWebView::initTestCase):
        (tst_QWebView::cleanupTestCase):
        (tst_QWebView::init):
        (tst_QWebView::cleanup):
        (tst_QWebView::guessUrlFromString_data):
        (tst_QWebView::guessUrlFromString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Laszlo Gombos <laszlo.1.gombos@nokia.com>
zecke@webkit.org [Wed, 27 May 2009 06:35:45 +0000 (06:35 +0000)]
2009-05-26  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Holger Freyther.

        https://bugs.webkit.org/show_bug.cgi?id=25996

        [Qt] Dead-code stripping for unix

        * WebCore.pro: Turn on GCC dead-code stripping flags for Unix

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago[GTK] Send onscroll events for the main FrameView
zecke@webkit.org [Wed, 27 May 2009 05:25:48 +0000 (05:25 +0000)]
[GTK] Send onscroll events for the main FrameView

2009-05-10  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Dave Hyatt.

        https://bugs.webkit.org/show_bug.cgi?id=25646

        [GTK] Send onscroll events for the main FrameView

        WebKit/GTK+ is currently not sending any onscroll
        events for a frame with external adjustments. This is
        due the fact that the value-changed signal of the GtkAdjustment
        is handled by WebCore::ScrollView directly and is not going through
        the WebCore::Scrollbar -> WebCore::ScrollbarClient ->
        WebCore::FrameView::valueChanged -> WebCore::ScrollView::valueChanged
        path.

        Fix the above problem by wrapping the GtkAdjustment we get
        assigned from GTK+ in a ScrollbarGtk that will not have any
        visual appearance. Remove code from ScrollView that knows
        about adjustments and create a special case for
        WebCore::ScrollView::createScrollbar that will create such
        a special WebCore::ScrollbarGtk.

        * platform/ScrollView.cpp: Remove adjustment code
        (WebCore::ScrollView::setHasHorizontalScrollbar):
        (WebCore::ScrollView::setHasVerticalScrollbar):
        (WebCore::ScrollView::updateScrollbars):
        (WebCore::ScrollView::wheelEvent):
        * platform/ScrollView.h: Remove adjustment code
        * platform/gtk/ScrollViewGtk.cpp:
        (WebCore::ScrollView::platformDestroy):
        (WebCore::ScrollView::createScrollbar):
        (WebCore::ScrollView::setGtkAdjustments):
        * platform/gtk/ScrollbarGtk.cpp:
        (ScrollbarGtk::createScrollbar): Special case.
        (ScrollbarGtk::ScrollbarGtk):  New ctor and work on the adjustment
        (ScrollbarGtk::~ScrollbarGtk): Disconnect signal
        (ScrollbarGtk::frameRectsChanged): Do nothing when we lack a platformWidget
        * platform/gtk/ScrollbarGtk.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Cameron Zwarich <zwarich@apple.com>
cwzwarich@webkit.org [Wed, 27 May 2009 05:20:08 +0000 (05:20 +0000)]
2009-05-26  Cameron Zwarich  <zwarich@apple.com>

        Reviewed by Maciej Stachowiak.

        Bug 26035: Make ImageSource::frameHasAlphaAtIndex() return false for JPEGs with CG
        <https://bugs.webkit.org/show_bug.cgi?id=26035>
        <rdar://problem/6924087>

        * platform/graphics/cg/ImageSourceCG.cpp:
        (WebCore::ImageSource::frameHasAlphaAtIndex): return false if the image
        is JPEG, there is no image type, or m_decoder is null.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix build breakage of JPEGImageDecoder.cpp by including stdio.h again.
zecke@webkit.org [Wed, 27 May 2009 04:30:32 +0000 (04:30 +0000)]
Fix build breakage of JPEGImageDecoder.cpp by including stdio.h again.

2009-05-26  Holger Hans Peter Freyther  <zecke@selfish.org>

        Unreviewed build fix for the JPEGImageDecoder.

        Revision 44167 removed stdio.h from the included headers and
        this will lead to an error that FILE is not known inside the
        jpeglib.h Put back the stdio.h include.

        * platform/image-decoders/jpeg/JPEGImageDecoder.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago<rdar://problem/6924033> REGRESSION: Assertion failure due to forward references
oliver@apple.com [Wed, 27 May 2009 03:22:47 +0000 (03:22 +0000)]
<rdar://problem/6924033> REGRESSION: Assertion failure due to forward references

Reviewed by Gavin Barraclough.

Add a pattern type for forward references to ensure that we don't confuse the
quantifier alternatives assertion.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Yichao Yin <yichao.yin@torchmobile.com.cn>
staikos@webkit.org [Wed, 27 May 2009 02:54:09 +0000 (02:54 +0000)]
2009-05-26  Yichao Yin  <yichao.yin@torchmobile.com.cn>

        Reviewed by Maciej Stachowiak.

        Fix the crash issue while running layout tests after enabled XHTMLMP
        https://bugs.webkit.org/show_bug.cgi?id=26017

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Stephanie Lewis <slewis@apple.com>
slewis@apple.com [Wed, 27 May 2009 02:47:40 +0000 (02:47 +0000)]
2009-05-26  Stephanie Lewis  <slewis@apple.com>

        Reviewed by Ada Chan and Oliver Hunt.

        Remove WebView observers in during fast teardown.
        <rdar://problem/6922619> REGRESSION (Tiger-only?): After restoring windows from the previous session, a crash occurs while attempting to quit Safari

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44172 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Gavin Barraclough <barraclough@apple.com>
barraclough@apple.com [Wed, 27 May 2009 02:47:35 +0000 (02:47 +0000)]
2009-05-26  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver Hunt.

        Fix for: <rdar://problem/6918095> REGRESSION: jQuery load() issue (25981),
        and also an ASSERT failure on http://ihasahotdog.com/.

        When overwriting a property on a dictionary with a cached specific value,
        clear the cache if new value being written is different.

        * JavaScriptCore.exp:
            Export the new symbols.
        * jit/JITStubs.cpp:
        (JSC::JITStubs::cti_op_get_by_id_method_check_second):
            Close dictionary prototypes upon caching a method access, as would happen when caching
            a regular get_by_id.
        * runtime/JSObject.h:
        (JSC::JSObject::propertyStorage):
        (JSC::JSObject::locationForOffset):
            Make these methods private.
        (JSC::JSObject::putDirectInternal):
            When overwriting a property on a dictionary with a cached specific value,
            clear the cache if new value being written is different.
        * runtime/Structure.cpp:
        (JSC::Structure::despecifyDictionaryFunction):
            Reset the specific value field for a given property in a dictionary.
        (JSC::Structure::despecifyFunctionTransition):
            Rename of 'changeFunctionTransition' (this was already internally refered to as a despecification).
        * runtime/Structure.h:
            Declare new method.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44171 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
andersca@apple.com [Wed, 27 May 2009 00:45:06 +0000 (00:45 +0000)]
WebCore:

2009-05-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6901751>
        REGRESSION (r35515): Tiger crash painting the selection on registration page of car2go.com

        Don't use the WKCGContextIsSafeToClip function; it wasn't working correctly. Instead, just disable
        the improved selection drawing on Tiger.

        * WebCore.Tiger.exp:
        * platform/graphics/cg/GraphicsContextCG.cpp:
        (WebCore::GraphicsContext::fillPath):
        (WebCore::GraphicsContext::strokePath):
        (WebCore::GraphicsContext::fillRect):
        (WebCore::GraphicsContext::clip):
        (WebCore::GraphicsContext::clipOut):
        (WebCore::GraphicsContext::clipOutEllipseInRect):
        (WebCore::GraphicsContext::clipPath):
        (WebCore::GraphicsContext::addInnerRoundedRectClip):
        (WebCore::GraphicsContext::strokeRect):
        * platform/mac/WebCoreSystemInterface.h:
        * platform/mac/WebCoreSystemInterface.mm:
        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::fillSelectionGaps):

WebKit/mac:

2009-05-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6901751>
        REGRESSION (r35515): Tiger crash painting the selection on registration page of car2go.com

        Remove WKCGContextIsSafeToClip.

        * WebCoreSupport/WebSystemInterface.m:
        (InitWebCoreSystemInterface):

WebKitLibraries:

2009-05-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6901751>
        REGRESSION (r35515): Tiger crash painting the selection on registration page of car2go.com

        Remove WKCGContextIsSafeToClip.

        * WebKitSystemInterface.h:
        * libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Gavin Barraclough <barraclough@apple.com>
barraclough@apple.com [Wed, 27 May 2009 00:33:13 +0000 (00:33 +0000)]
2009-05-26  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Oliver "pieces of eight" Hunt.

        When reseting RegexPattern class, should fully reset the class, not just bits of it.
        In particular, we delete the cached character classes (for wordchars, etc), but do
        not reset the set of pointers to the cached classes.  In the case of a repeated parse
        due to an illegal back-reference we will continue to use the deleted character class.

        * yarr/RegexPattern.h:
        (JSC::Yarr::RegexPattern::reset):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Olivier DOLE <odole@pleyo.com>
bfulgham@webkit.org [Wed, 27 May 2009 00:29:29 +0000 (00:29 +0000)]
2009-05-26  Olivier DOLE  <odole@pleyo.com>

        Reviewed by Maciej.

        Store sunspider test suite results in a specific directory for each
        different type of benchmark (sunspider, V8 or ubench).

        * sunspider:
        * sunspider-compare-results:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Peter Kasting <pkasting@google.com>
bfulgham@webkit.org [Wed, 27 May 2009 00:20:04 +0000 (00:20 +0000)]
2009-05-26  Peter Kasting  <pkasting@google.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=25709 part two
        Mostly cosmetic changes, mainly to get image decoders for Cairo build to
        comply with WebKit style guidelines:
        * Fix header guard names
        * Fix initializer list wrapping
        * Fix class/function "{" wrapping
        * Fix wrapping of long boolean expressions
        * Fix whitespace around operators
        * Eliminate whitespace on otherwise-blank lines
        * No {} on single-line conditional arms
        * Use 0 instead of NULL
        * Don't #include "config.h" in headers

        Also a few non-cosmetic changes:
        * Remove unneeded #ifdef protecting the contents of these files.  The
          platforms listed don't match who uses these files anymore (even
          without considering Skia), and platforms which don't use them at all
          (like CG) aren't harmed by the files, since they don't have
          equivalently-named classes.  It's simpler to just ditch these.
        * Convert a use of Vector<>& into SharedBuffer*, since that's how the
          data is really stored anyway.
        * Use size() instead of m_size, for consistency with Skia code, and
          future-proofing (if a particular decoder wants to specialize this).
        * Move an ASSERT that was slightly more aggressive than necessary.
        * Change variable names/types for clarity/accuracy and to match Skia.
        * Remove unnecessary macro magic to work around a problem which no
          longer exists in the third-party JPEG headers.
        * Stop silencing a portability warning about setjmp (the vcproj/vsprops
          are the right place for this)
        * In Skia code, don't explicitly check |m_failed| before calling
          isSizeAvailable(), which itself checks |m_failed|.

        * platform/image-decoders/ImageDecoder.h: Fix header guards, "{" wrapping.
        (WebCore::RGBA32Buffer::):
        (WebCore::RGBA32Buffer::RGBA32Buffer): Fix initializer list wrapping.
        (WebCore::ImageDecoder::ImageDecoder): Fix initializer list wrapping.
        * platform/image-decoders/bmp/BMPImageDecoder.cpp: Remove unneeded #ifdef.
        * platform/image-decoders/bmp/BMPImageDecoder.h: Fix header guards, "{" wrapping.
        * platform/image-decoders/gif/GIFImageDecoder.cpp: Remove unneeded #ifdef, fix "{" wrapping.
        (WebCore::GIFImageDecoderPrivate::GIFImageDecoderPrivate): Use an initializer instead of a block-level statement.
        (WebCore::GIFImageDecoderPrivate::decode): Expect a SharedBuffer.
        (WebCore::GIFImageDecoderPrivate::getColorMap):
        (WebCore::GIFImageDecoder::GIFImageDecoder): Fix initializer list wrapping.
        (WebCore::GIFImageDecoder::frameCount): Add comment.
        (WebCore::GIFImageDecoder::frameBufferAtIndex): Explicitly cast, fix whitespace around operators.
        (WebCore::GIFImageDecoder::decode): Pass a SharedBuffer.
        (WebCore::GIFImageDecoder::initFrameBuffer): Use size() instead of m_size, move ASSERT, fix boolean wrapping, fix indenting.
        (WebCore::GIFImageDecoder::prepEmptyFrameBuffer): Use size() instead of m_size.
        (WebCore::GIFImageDecoder::haveDecodedRow): Use size() instead of m_size, eliminate unneeded whitespace, change variable name.
        (WebCore::GIFImageDecoder::frameComplete): Use size() instead of m_size, no {} on single-line conditional arms, fix boolean wrapping.
        * platform/image-decoders/gif/GIFImageDecoder.h: Fix header guards, "{" wrapping.
        * platform/image-decoders/gif/GIFImageReader.cpp: Remove unneeded #ifdef.
        (GIFImageReader::read):
        * platform/image-decoders/gif/GIFImageReader.h: Fix header guards, change variable type.
        * platform/image-decoders/ico/ICOImageDecoder.cpp: Remove unneeded #ifdef.
        * platform/image-decoders/ico/ICOImageDecoder.h: Fix header guards, "{" wrapping.
        * platform/image-decoders/jpeg/JPEGImageDecoder.cpp: Remove unneeded #ifdef, macro magic, silencing of warning.
        (WebCore::JPEGImageDecoder::outputScanlines): Use size() instead of m_size, change variable name.
        * platform/image-decoders/jpeg/JPEGImageDecoder.h: Fix header guards, "{" wrapping.
        * platform/image-decoders/png/PNGImageDecoder.cpp: Remove unneeded #ifdef, silencing of warning, change variable type.
        (WebCore::PNGImageReader::PNGImageReader): Fix initializer list wrapping.
        (WebCore::PNGImageReader::close): Add comment, zero another member for consistency.
        (WebCore::PNGImageDecoder::PNGImageDecoder): Fix indenting.
        (WebCore::PNGImageDecoder::decodingFailed): Define in .cpp, not .h.
        (WebCore::PNGImageDecoder::rowAvailable):
        * platform/image-decoders/png/PNGImageDecoder.h: Fix header guards, "{" wrapping.
        * platform/image-decoders/skia/GIFImageDecoder.cpp:
        (WebCore::GIFImageDecoder::isSizeAvailable): Don't check m_failed unnecessarily.
        (WebCore::GIFImageDecoder::frameBufferAtIndex): Fix whitespace around operators.
        * platform/image-decoders/skia/GIFImageReader.h: "unsigned" is sufficient to convey "unsigned int".
        * platform/image-decoders/skia/ImageDecoder.h: Remove unnecessary #includes.
        (WebCore::ImageDecoder::ImageDecoder): Fix initializer list wrapping.
        * platform/image-decoders/skia/JPEGImageDecoder.cpp:
        (WebCore::JPEGImageReader::JPEGImageReader): Use 0 instead of NULL.
        (WebCore::JPEGImageDecoder::isSizeAvailable): Don't check m_failed unnecessarily.
        * platform/image-decoders/skia/PNGImageDecoder.cpp:
        (WebCore::PNGImageReader::PNGImageReader): Fix initializer list wrapping.
        (WebCore::PNGImageDecoder::PNGImageDecoder): Fix indenting.
        (WebCore::PNGImageDecoder::isSizeAvailable): Don't check m_failed unnecessarily.
        (WebCore::rowAvailable):
        * platform/image-decoders/skia/XBMImageDecoder.h: Don't #include config.h.
        * platform/image-decoders/xbm/XBMImageDecoder.cpp: Remove unneeded #ifdef.
        * platform/image-decoders/xbm/XBMImageDecoder.h: Fix header guards, "{" wrapping.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44167 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoVersioning.
mrowe@apple.com [Wed, 27 May 2009 00:13:32 +0000 (00:13 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler.
mitz@apple.com [Tue, 26 May 2009 23:57:23 +0000 (23:57 +0000)]
    Reviewed by Darin Adler.

        - fix <rdar://problem/6922371> REGRESSION (r43973): Can't type first
          character with an umlaut, circumflex, or accent in Mail or in a wiki

        * WebView/WebHTMLView.mm:
        (isTextInput): Moved here.
        (isInPasswordField): Moved here.
        (-[WebHTMLView becomeFirstResponder]): Update the exposeInputContext
        flag and let NSApplication update the input manager with the new input
        context if necessary.
        (-[WebHTMLView _updateSelectionForInputManager]): Changed to use the
        NSApp global instead of +[NSApplication sharedApplication].

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Brent Fulgham <bfulgham@webkit.org>
bfulgham@webkit.org [Tue, 26 May 2009 23:06:42 +0000 (23:06 +0000)]
2009-05-26  Brent Fulgham  <bfulgham@webkit.org>

        Build fix to correct r44161.

        * wtf/FastAllocBase.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Jian Li <jianli@chromium.org>
jianli@chromium.org [Tue, 26 May 2009 22:58:23 +0000 (22:58 +0000)]
2009-05-26  Jian Li  <jianli@chromium.org>

        Reviewed by Dimitri Glazkov.

        https://bugs.webkit.org/show_bug.cgi?id=25944
        Remove the uses of V8 Locker in worker execution code. No additional test is needed.
        The normal WebKit layout tests should cover it. However, layout tests that start
        multiple workers will fail to pass due to test shell limitation in Chromium. To cover
        this, UI tests will be added (http://code.google.com/p/chromium/issues/detail?id=12554).

        * bindings/v8/V8WorkerContextEventListener.cpp:
        (WebCore::V8WorkerContextEventListener::handleEvent):
        * bindings/v8/WorkerContextExecutionProxy.cpp:
        (WebCore::WorkerContextExecutionProxy::dispose):
        (WebCore::WorkerContextExecutionProxy::initV8IfNeeded):
        (WebCore::WorkerContextExecutionProxy::evaluate):
        (WebCore::WorkerContextExecutionProxy::findOrCreateEventListenerHelper):
        (WebCore::WorkerContextExecutionProxy::RemoveEventListener):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Zoltan Horvath <horvath.zoltan.6@stud.u-szeged.hu>
bfulgham@webkit.org [Tue, 26 May 2009 22:25:56 +0000 (22:25 +0000)]
2009-05-26  Zoltan Horvath  <horvath.zoltan.6@stud.u-szeged.hu>

        Reviewed by Maciej Stachowiak.

        Inherite HashTable from FastAllocBase, because it has been instantiated by
        'new' in JavaScriptCore/runtime/JSGlobalData.cpp.

        * wtf/HashTable.h:
        * wtf/FastAllocBase.h: Remove 'wtf' path from TypeTraits.h to allow use outside of wtf.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Dmitry Titov <dimich@chromium.org>
dimich@chromium.org [Tue, 26 May 2009 21:41:34 +0000 (21:41 +0000)]
2009-05-26  Dmitry Titov  <dimich@chromium.org>

        Reviewed by Dimitri Glazkov.

        https://bugs.webkit.org/show_bug.cgi?id=25974
        Enable timers in Chromuim workers.

        * bindings/v8/ScheduledAction.cpp:
        (WebCore::ScheduledAction::execute): Added function which can execute a callback using WorkerContext.
        * bindings/v8/ScheduledAction.h:
        (WebCore::ScheduledAction::ScheduledAction): added url parameter to capture the worker's location.
        * bindings/v8/custom/V8WorkerContextCustom.cpp:
        (WebCore::SetTimeoutOrInterval): replaced NotImplemented with code to create ScheduledAction and DOMTimer.
        Also, removed declarations for clearTimeout/clearInterval callbacks since they are now directly generated from IDL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 26 May 2009 21:19:06 +0000 (21:19 +0000)]
2009-05-26  Adam Barth  <abarth@webkit.org>

        https://bugs.webkit.org/show_bug.cgi?id=26011

        Unreviewed change to disable flaky test as requested by Sam Weinig.

        * http/tests/security/postMessage/origin-follows-lexical-scope-expected.txt: Removed.
        * http/tests/security/postMessage/origin-follows-lexical-scope.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix the Tiger build.
mrowe@apple.com [Tue, 26 May 2009 20:43:23 +0000 (20:43 +0000)]
Fix the Tiger build.

* Misc/WebNSObjectExtras.mm:
(-[NSInvocation _webkit_invokeAndHandleException:]): Rename the local variable so that
it doesn't conflict with a member variable on Tiger.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 26 May 2009 20:28:48 +0000 (20:28 +0000)]
2009-05-26  Dimitri Glazkov  <dglazkov@chromium.org>

        Unreviewed, build fix.

        https://bugs.webkit.org/show_bug.cgi?id=23649
        Update V8 bindings to match SQLTransactionErrorCallback change.

        * bindings/v8/custom/V8CustomSQLTransactionErrorCallback.cpp:
        (WebCore::V8CustomSQLTransactionErrorCallback::handleEvent): Changed to return void.
        * bindings/v8/custom/V8CustomSQLTransactionErrorCallback.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Tue, 26 May 2009 20:05:35 +0000 (20:05 +0000)]
2009-05-26  Dirk Schulze  <krit@webkit.org>

        Reviewed by Nikolas Zimmermann.

        Move FilterEffect to platform/graphics/filters. First step to get an
        SVG independent filter system.

        * GNUmakefile.am:
        * WebCore.pro:
        * WebCore.xcodeproj/project.pbxproj:
        * platform/graphics/filters/FilterEffect.cpp: Added.
        (WebCore::FilterEffect::FilterEffect):
        (WebCore::FilterEffect::~FilterEffect):
        (WebCore::FilterEffect::externalRepresentation):
        * platform/graphics/filters/FilterEffect.h: Added.
        (WebCore::FilterEffect::xBoundingBoxMode):
        (WebCore::FilterEffect::setXBoundingBoxMode):
        (WebCore::FilterEffect::yBoundingBoxMode):
        (WebCore::FilterEffect::setYBoundingBoxMode):
        (WebCore::FilterEffect::widthBoundingBoxMode):
        (WebCore::FilterEffect::setWidthBoundingBoxMode):
        (WebCore::FilterEffect::heightBoundingBoxMode):
        (WebCore::FilterEffect::setHeightBoundingBoxMode):
        (WebCore::FilterEffect::subRegion):
        (WebCore::FilterEffect::setSubRegion):
        (WebCore::FilterEffect::resultImage):
        (WebCore::FilterEffect::setEffectBuffer):
        * svg/FilterEffect.cpp: Removed.
        * svg/FilterEffect.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
dglazkov@chromium.org [Tue, 26 May 2009 19:37:14 +0000 (19:37 +0000)]
WebCore:

2009-05-26  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=23649
        Update SQLTransactionErrorCallback to not return a value per changes in
        HTML5 spec.

        Test: storage/transaction-error-callback.html

        * bindings/js/JSCustomSQLTransactionErrorCallback.cpp:
        (WebCore::JSCustomSQLTransactionErrorCallback::handleEvent):
        * bindings/js/JSCustomSQLTransactionErrorCallback.h:
        * storage/SQLTransaction.cpp:
        (WebCore::SQLTransaction::SQLTransaction):
        (WebCore::SQLTransaction::postflightAndCommit):
        (WebCore::SQLTransaction::handleTransactionError):
        (WebCore::SQLTransaction::deliverTransactionErrorCallback):
        (WebCore::SQLTransaction::cleanupAfterTransactionErrorCallback):
        * storage/SQLTransaction.h:
        * storage/SQLTransactionErrorCallback.h:

LayouTests:

2009-05-26  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=23649
        Add layout test for various cases of SQLTransactionErrorCallback
        invocation.

        * storage/transaction-error-callback-expected.txt: Added.
        * storage/transaction-error-callback.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Darin Adler <darin@apple.com>
darin@apple.com [Tue, 26 May 2009 18:29:06 +0000 (18:29 +0000)]
2009-05-26  Darin Adler  <darin@apple.com>

        Reviewed by Mark Rowe.

        <rdar://problem/6921835> REGRESSION (r44115): Crash in Leopard Mail viewing message

        Fix case where we autorelease an object on the main thread that we are returning
        to another thread when using our _webkit_invokeOnMainThread method.

        * Misc/WebNSObjectExtras.mm:
        (returnTypeIsObject): Added.
        (-[WebMainThreadInvoker forwardInvocation:]): Autorelease the object on the calling
        thread, balancing a retain done on the main thread.
        (-[NSInvocation _webkit_invokeAndHandleException:]): Retain the object on the main
        thread.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Tue, 26 May 2009 18:26:16 +0000 (18:26 +0000)]
2009-05-26  David Hyatt  <hyatt@apple.com>

        Back out the workaround for Mail crashing.  Darin is working on the real fix.

        * Misc/WebNSObjectExtras.mm:
        (-[WebMainThreadInvoker forwardInvocation:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
hyatt@apple.com [Tue, 26 May 2009 18:16:58 +0000 (18:16 +0000)]
WebCore:

2009-05-26  David Hyatt  <hyatt@apple.com>

        Reviewed by Darin Adler.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=25969

        Detect the case where we can fit without scrollbars when the view shrinks, so that we don't mistakenly
        continue to show scrollbars.

        Added two tests in fast/dynamic.

        * platform/ScrollView.cpp:
        (WebCore::ScrollView::updateScrollbars):

WebKit/mac:

2009-05-26  David Hyatt  <hyatt@apple.com>

        Reviewed by Darin Adler.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=25969.

        Stop using notifications for boundsChanged, since the notifications are being sent too late.
        Since subviews get resized before parents do, the notification comes in telling the WebHTMLView
        that its size has changed *before* we've done setNeedsLayout in the WebView size change callback.

        Become more like the Windows platform and just do the bounds changed immediately as our size is
        being altered by subclassing setFrameSize in WebView.

        Also patch WebDynamicScrollbarsView to detect the case where neither axis is really spilling out
        past the viewport, so that in shrinking situations we figure out that we don't need scrollbars
        any more.

        * WebView/WebDynamicScrollBarsView.m:
        (-[WebDynamicScrollBarsView updateScrollers]):
        * WebView/WebView.mm:
        (-[WebView _boundsChangedToNewSize:]):
        (-[WebView setFrameSize:]):
        (-[WebView viewWillMoveToWindow:]):
        (-[WebView viewDidMoveToWindow]):

LayoutTests:

2009-05-26  David Hyatt  <hyatt@apple.com>

        Reviewed by Darin Adler.

        Fix for bug 25969, scrollbars remain after document shrinks.  Add a couple of layout tests
        written by Dan that show various problems with scrollbars following resizes.

        * fast/dynamic/window-resize-scrollbars-test.html: Added.
        * fast/dynamic/window-scrollbars-test.html: Added.
        * platform/mac/fast/dynamic/window-resize-scrollbars-test-expected.checksum: Added.
        * platform/mac/fast/dynamic/window-resize-scrollbars-test-expected.png: Added.
        * platform/mac/fast/dynamic/window-resize-scrollbars-test-expected.txt: Added.
        * platform/mac/fast/dynamic/window-scrollbars-test-expected.checksum: Added.
        * platform/mac/fast/dynamic/window-scrollbars-test-expected.png: Added.
        * platform/mac/fast/dynamic/window-scrollbars-test-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoBug 25801: change AccessibilityAria -> AccessibilityARIA
cfleizach@apple.com [Tue, 26 May 2009 17:49:32 +0000 (17:49 +0000)]
Bug 25801: change AccessibilityAria -> AccessibilityARIA
https://bugs.webkit.org/show_bug.cgi?id=25801

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44151 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Stephen White <senorblanco@chromium.org>
dglazkov@chromium.org [Tue, 26 May 2009 16:00:29 +0000 (16:00 +0000)]
2009-05-26  Stephen White  <senorblanco@chromium.org>

        Reviewed by Dimitri Glazkov.

        https://bugs.webkit.org/show_bug.cgi?id=26012

        Fix the Skia path to normalize the given source rect when
        drawing patterns, so it accepts negative width/height (as CG does).
        Fixes Chromium bug http://www.crbug.com/6167.
        * platform/graphics/skia/ImageSkia.cpp:
        (WebCore::Image::drawPattern):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-26 Pavel Feldman <pfeldman@chromium.org>
ap@webkit.org [Tue, 26 May 2009 12:23:10 +0000 (12:23 +0000)]
2009-05-26  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Serialize calls to WebInspector. This is the first step where method
        calls are being serialized to arrays (not yet JSON stringified ones) and
        are being dispatched on the client. This change also allows client to override
        InspectorFrontend, so that the serialized calls can be made on a given object
        instead of in-process WebInspector global. This will be the main control flow
        when InspectorController is decoupled from the in-process frontend.

        https://bugs.webkit.org/show_bug.cgi?id=26010

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::scriptObjectReady):
        (WebCore::InspectorController::overrideFrontendObject):
        (WebCore::InspectorController::show):
        (WebCore::InspectorController::populateScriptObjects):
        * inspector/InspectorController.h:
        * inspector/InspectorFrontend.cpp:
        (WebCore::InspectorFrontend::addMessageToConsole):
        (WebCore::InspectorFrontend::addResource):
        (WebCore::InspectorFrontend::updateResource):
        (WebCore::InspectorFrontend::removeResource):
        (WebCore::InspectorFrontend::updateFocusedNode):
        (WebCore::InspectorFrontend::setAttachedWindow):
        (WebCore::InspectorFrontend::inspectedWindowScriptObjectCleared):
        (WebCore::InspectorFrontend::showPanel):
        (WebCore::InspectorFrontend::populateInterface):
        (WebCore::InspectorFrontend::reset):
        (WebCore::InspectorFrontend::debuggerWasEnabled):
        (WebCore::InspectorFrontend::debuggerWasDisabled):
        (WebCore::InspectorFrontend::profilerWasEnabled):
        (WebCore::InspectorFrontend::profilerWasDisabled):
        (WebCore::InspectorFrontend::parsedScriptSource):
        (WebCore::InspectorFrontend::failedToParseScriptSource):
        (WebCore::InspectorFrontend::addProfile):
        (WebCore::InspectorFrontend::setRecordingProfile):
        (WebCore::InspectorFrontend::pausedScript):
        (WebCore::InspectorFrontend::resumedScript):
        (WebCore::InspectorFrontend::addDatabase):
        (WebCore::InspectorFrontend::addDOMStorage):
        (WebCore::InspectorFrontend::newFunctionCall):
        (WebCore::InspectorFrontend::callSimpleFunction):
        * inspector/InspectorFrontend.h:
        * inspector/front-end/inspector.js:
        (WebInspector.dispatch):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler.
ap@webkit.org [Tue, 26 May 2009 06:20:53 +0000 (06:20 +0000)]
    Reviewed by Darin Adler.

        <rdar://problem/6918671> REGRESSION (r41700): Crashes in debug DumpRenderTree at RefCountedBase::ref().

        * dom/QualifiedName.cpp: (WebCore::QualifiedName::QualifiedName):
        * dom/QualifiedName.h: (WebCore::QualifiedName::QualifiedNameImpl::QualifiedNameImpl):
        Adjust empty namespace before QNameComponentsTranslator black magic, not after.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoJavaScriptCore:
levin@chromium.org [Tue, 26 May 2009 04:21:32 +0000 (04:21 +0000)]
JavaScriptCore:

2009-05-25  David Levin  <levin@chromium.org>

        Reviewed by Maciej Stachowiak and Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=25126
        Allow the buffer underlying UString to be shared.

        In order to not grow the underlying size of any structure,
        there is a union in the Rep string which holds
         + m_sharedBuffer -- a pointer to the shared ref counted buffer
           if the class is BaseString and the buffer is being shared OR
         + m_baseString -- the BaseString if the class is only UString::Rep
           but not a UString::BaseString

        Ideally, m_sharedBuffer would be a RefPtr, but it cannot be because
        it is in a union.

        No change in sunspider perf.

        * JavaScriptCore.vcproj/WTF/WTF.vcproj:
        * JavaScriptCore.xcodeproj/project.pbxproj:
        * runtime/UString.cpp:
        (JSC::UString::Rep::share):
        (JSC::UString::Rep::destroy):
        (JSC::UString::BaseString::sharedBuffer):
        (JSC::UString::BaseString::setSharedBuffer):
        (JSC::UString::BaseString::slowIsBufferReadOnly):
        (JSC::expandCapacity):
        (JSC::UString::Rep::reserveCapacity):
        (JSC::UString::expandPreCapacity):
        (JSC::concatenate):
        (JSC::UString::append):
        * runtime/UString.h:
        (JSC::UString::Rep::Rep):
        (JSC::UString::Rep::):
        (JSC::UString::BaseString::isShared):
        (JSC::UString::BaseString::isBufferReadOnly):
        (JSC::UString::Rep::baseString):
        * wtf/CrossThreadRefCounted.h:
        (WTF::CrossThreadRefCounted::isShared):
        * wtf/OwnFastMallocPtr.h: Added.
        (WTF::OwnFastMallocPtr::OwnFastMallocPtr):
        (WTF::OwnFastMallocPtr::~OwnFastMallocPtr):
        (WTF::OwnFastMallocPtr::get):
        (WTF::OwnFastMallocPtr::release):

JavaScriptGlue:

2009-05-25  David Levin  <levin@chromium.org>

        Reviewed by Maciej Stachowiak and Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=25126

        Added forwarding headers.

        * ForwardingHeaders/wtf/CrossThreadRefCounted.h: Added.
        * ForwardingHeaders/wtf/OwnFastMallocPtr.h: Added.

WebCore:

2009-05-25  David Levin  <levin@chromium.org>

        Reviewed by Maciej Stachowiak and Oliver Hunt.

        Added forwarding headers.

        * ForwardingHeaders/wtf/CrossThreadRefCounted.h: Added.
        * ForwardingHeaders/wtf/OwnFastMallocPtr.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRe-add interpreter logic to jit-enabled builds as GCC mysteriously regresses without it
oliver@apple.com [Tue, 26 May 2009 00:20:17 +0000 (00:20 +0000)]
Re-add interpreter logic to jit-enabled builds as GCC mysteriously regresses without it

Reviewed by Maciej Stachowiak

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44143 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
andersca@apple.com [Mon, 25 May 2009 22:48:48 +0000 (22:48 +0000)]
WebCore:

2009-05-25  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        - WebCore side of <rdar://problem/6914001>.

        Add a new m_allowPopupsFromPlugin flag.

        * bindings/js/ScriptController.cpp:
        (WebCore::ScriptController::ScriptController):
        (WebCore::ScriptController::processingUserGesture):
        * bindings/js/ScriptController.h:
        (WebCore::ScriptController::setAllowPopupsFromPlugin):
        (WebCore::ScriptController::allowPopupsFromPlugin):

WebKit/mac:

2009-05-25  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        - WebKit side of <rdar://problem/6914001>.

        * Plugins/Hosted/NetscapePluginHostManager.h:
        * Plugins/Hosted/NetscapePluginHostManager.mm:
        (WebKit::NetscapePluginHostManager::didCreateWindow):
        Iterate over all plug-in hosts. If one is in full-screen mode, make sure to activate the
        WebKit app instead.

        * Plugins/Hosted/NetscapePluginHostProxy.h:
        (WebKit::NetscapePluginHostProxy::isMenuBarVisible):
        Add getter.

        * Plugins/Hosted/NetscapePluginHostProxy.mm:
        (WKPCEvaluate):
        Add allowPopups flag.

        * Plugins/Hosted/NetscapePluginInstanceProxy.h:
        * Plugins/Hosted/NetscapePluginInstanceProxy.mm:
        (WebKit::NetscapePluginInstanceProxy::PluginRequest::PluginRequest):
        (WebKit::NetscapePluginInstanceProxy::PluginRequest::allowPopups):
        (WebKit::NetscapePluginInstanceProxy::loadURL):
        (WebKit::NetscapePluginInstanceProxy::evaluateJavaScript):
        (WebKit::NetscapePluginInstanceProxy::loadRequest):
        (WebKit::NetscapePluginInstanceProxy::evaluate):
        Use "allowPopups" instead of "userGesture".

        * Plugins/Hosted/WebKitPluginClient.defs:
        Add allowPopups argument to Evaluate.

        * Plugins/Hosted/WebKitPluginHostTypes.h:
        Use "allowPopups" instead of "userGesture".

        * WebCoreSupport/WebChromeClient.mm:
        (WebChromeClient::createWindow):
        Call didCreateWindow here.

        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::dispatchCreatePage):
        Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
abarth@webkit.org [Mon, 25 May 2009 21:33:33 +0000 (21:33 +0000)]
WebCore:

2009-05-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        https://bugs.webkit.org/show_bug.cgi?id=26006

        Sort out our use of dynamicGlobalObject and lexicalGlobalObject for
        window.location.  The correct use appears to be as follows:

        1) Use dynamicGlobalObject to find the user gesture.
        2) Use dynamicGlobalObject to complete URLs.
        3) Use lexicalGlobalObject to find the referrer.
        4) Use lexicalGlobalObject for the frame navigation checks.
        5) Use lexicalGlobalObject for the XSS checks.

        Tests: http/tests/security/frameNavigation/context-for-location-assign.html
               http/tests/security/frameNavigation/context-for-location-href.html
               http/tests/security/frameNavigation/context-for-location.html

        * bindings/js/JSDOMBinding.cpp:
        (WebCore::shouldAllowNavigation):
        (WebCore::toLexicalFrame):
        (WebCore::processingUserGesture):
        (WebCore::completeURL):
        * bindings/js/JSDOMBinding.h:
        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::setLocation):
        * bindings/js/JSLocationCustom.cpp:
        (WebCore::navigateIfAllowed):
        (WebCore::JSLocation::setHref):
        (WebCore::JSLocation::replace):
        (WebCore::JSLocation::reload):
        (WebCore::JSLocation::assign):
        (WebCore::JSLocation::toString):
        (WebCore::JSLocationPrototype::customPut):
        * bindings/v8/V8Utilities.cpp:
        (WebCore::processingUserGesture):
        (WebCore::shouldAllowNavigation):
        (WebCore::completeURL):
        (WebCore::navigateIfAllowed):
        * bindings/v8/V8Utilities.h:
        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::V8Custom::WindowSetLocation):
        * bindings/v8/custom/V8LocationCustom.cpp:
        (WebCore::ACCESSOR_SETTER):
        (WebCore::CALLBACK_FUNC_DECL):

LayoutTests:

2009-05-25  Adam Barth  <abarth@webkit.og>

        Reviewed by Sam Weinig.

        https://bugs.webkit.org/show_bug.cgi?id=26006

        Test our use of dynamicGlobalObject and lexicalGlobalObject for
        window.location.

        * http/tests/security/frameNavigation/context-for-location-assign-expected.txt: Added.
        * http/tests/security/frameNavigation/context-for-location-assign.html: Added.
        * http/tests/security/frameNavigation/context-for-location-expected.txt: Added.
        * http/tests/security/frameNavigation/context-for-location-href-expected.txt: Added.
        * http/tests/security/frameNavigation/context-for-location-href.html: Added.
        * http/tests/security/frameNavigation/context-for-location.html: Added.
        * http/tests/security/frameNavigation/resources/middle-frame-for-location.html: Added.
        * http/tests/security/frameNavigation/resources/target-for-location.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44135 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUnreviewed attempt to fix flaky origin-follows-lexical-scope test. (Test only fails...
abarth@webkit.org [Mon, 25 May 2009 21:11:55 +0000 (21:11 +0000)]
Unreviewed attempt to fix flaky origin-follows-lexical-scope test.  (Test only fails on buildbot.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Mon, 25 May 2009 19:33:32 +0000 (19:33 +0000)]
2009-05-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        https://bugs.webkit.org/show_bug.cgi?id=26011

        Try to make test less flaky.

        * http/tests/security/postMessage/origin-follows-lexical-scope-expected.txt:
        * http/tests/security/postMessage/origin-follows-lexical-scope.html:
        * http/tests/security/postMessage/resources/middle-frame-for-lexical.html:
        * http/tests/security/postMessage/resources/post-message-listener-with-ready.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 25 May 2009 18:48:12 +0000 (18:48 +0000)]
2009-05-25  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        - Fix <rdar://problem/6915849>.

        Release the placeholder window after -[NSApplication runModalForWindow] returns.

        * Plugins/Hosted/NetscapePluginHostProxy.mm:
        (WebKit::NetscapePluginHostProxy::beginModal):
        (WebKit::NetscapePluginHostProxy::endModal):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44132 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Fridrich Strba <fridrich.strba@bluewin.ch>
xan@webkit.org [Mon, 25 May 2009 15:41:33 +0000 (15:41 +0000)]
2009-05-25  Fridrich Strba  <fridrich.strba@bluewin.ch>

        Reviewed by Maciej Stachowiak.

        The functions written in assembly need to have a leading
        underscore on Windows too.

        * jit/JITStubs.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44131 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Fridrich Strba <fridrich.strba@bluewin.ch>
xan@webkit.org [Mon, 25 May 2009 15:41:21 +0000 (15:41 +0000)]
2009-05-25  Fridrich Strba  <fridrich.strba@bluewin.ch>

        Reviewed by Maciej Stachowiak.

        With Windows compilers older then MSVC 2005, the functions
        gmtime_s and localtime_s don't exist. The gmtime and localtime are
        on Windows always thread safe. So use them in the cases where
        gmtime_s and localtime_s are not present.

        * loader/FTPDirectoryDocument.cpp:
        * loader/FTPDirectoryParser.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44130 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Holger Hans Peter Freyther <zecke@selfish.org>
hausmann@webkit.org [Mon, 25 May 2009 12:55:39 +0000 (12:55 +0000)]
2009-05-25  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Simon Hausmann.

        Follow up on r44120. PluginPackageMac.cpp is used
        by Qt on Mac and contained a copy of the m_browserFunc
        assignments and ::equal, ::hash and ::compareFileVersion.
        Change it to use the ones from PluginPackage.cpp as well.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44129 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Simon Hausmann <simon.hausmann@nokia.com>
hausmann@webkit.org [Mon, 25 May 2009 12:39:15 +0000 (12:39 +0000)]
2009-05-25  Simon Hausmann  <simon.hausmann@nokia.com>

        Reviewed by Tor Arne Vestbø.

        https://bugs.webkit.org/show_bug.cgi?id=25612

        Support conditionals in the tag factories, by placing
        feature #ifdefs around the individual factory functions
        and the registration.

        Made the Audio and Video elements conditional in the tag factories

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44128 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Simon Hausmann <simon.hausmann@nokia.com>
hausmann@webkit.org [Mon, 25 May 2009 11:45:23 +0000 (11:45 +0000)]
2009-05-25  Simon Hausmann  <simon.hausmann@nokia.com>

        Fix the Gtk build when video is disabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44127 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-25 Simon Hausmann <simon.hausmann@nokia.com>
hausmann@webkit.org [Mon, 25 May 2009 11:35:05 +0000 (11:35 +0000)]
2009-05-25  Simon Hausmann  <simon.hausmann@nokia.com>

        Fix the Qt build when video is disabled.

        * WebCore.pro: Always generate the IDL files for the media elements. They
        contain proper ENABLE(VIDEO) #ifdefs.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44126 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix ChangeLog typo.
abarth@webkit.org [Mon, 25 May 2009 08:36:46 +0000 (08:36 +0000)]
Fix ChangeLog typo.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44125 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore:
abarth@webkit.org [Mon, 25 May 2009 08:29:17 +0000 (08:29 +0000)]
WebCore:

2009-05-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=26004

        The origin for postMessage should follow lexical scope, not dynamic
        scope.  Yes, this is a super-obscure conner case.

        Test: http/tests/security/postMessage/origin-follows-lexical-scope.html

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::postMessage):
        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::CALLBACK_FUNC_DECL):

LayoutTests:

2009-05-25  Adam Barth  <abarth@webkit.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=26004

        Test that the origin for postMessage follows the lexical scope.

        * http/tests/security/postMessage/origin-follows-lexical-scope-expected.txt: Added.
        * http/tests/security/postMessage/origin-follows-lexical-scope.html: Added.
        * http/tests/security/postMessage/resources/middle-frame-for-lexical.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44124 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Darin Fisher <darin@chromium.org>
darin@chromium.org [Mon, 25 May 2009 06:27:48 +0000 (06:27 +0000)]
2009-05-24  Darin Fisher  <darin@chromium.org>

        Reviewed by Mark Rowe.

        Fix build bustage related to PassOwnPtr changes.

        https://bugs.webkit.org/show_bug.cgi?id=25999

        * platform/graphics/chromium/TransparencyWin.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44123 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago - revert an accidental change from r43964.
mitz@apple.com [Mon, 25 May 2009 05:20:32 +0000 (05:20 +0000)]
    - revert an accidental change from r43964.

        * WebView/WebView.mm:
        (+[WebView _setShouldUseFontSmoothing:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44122 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Holger Hans Peter Freyther <zecke@selfish.org>
zecke@webkit.org [Mon, 25 May 2009 03:29:26 +0000 (03:29 +0000)]
2009-05-24  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=25988

        Minor cleanup in PluginPackage::hash. Let the compiler
        figure out the size of the array. Do not have it in the
        definition and specially not the in the call to
        StringImpl::hash.

        * plugins/PluginPackage.cpp:
        (WebCore::PluginPackage::hash):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44121 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Holger Hans Peter Freyther <zecke@selfish.org>
zecke@webkit.org [Mon, 25 May 2009 03:29:06 +0000 (03:29 +0000)]
2009-05-24  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=25988

        Move ::hash, ::equal and ::compareFileVersion from
        the duplicated copies in PluginPackageQt.cpp and
        PluginPackageGtk to PluginPackage.cpp. We need to #ifdef
        this for the PluginPackageWin version. Use the new style
        ENABLE() for this feature/policy and enable it in the Gtk+
        and Qt buildsystem.

        * GNUmakefile.am:
        * WebCore.pro:
        * plugins/PluginPackage.cpp:
        (WebCore::PluginPackage::hash):
        (WebCore::PluginPackage::equal):
        (WebCore::PluginPackage::compareFileVersion):
        * plugins/gtk/PluginPackageGtk.cpp:
        * plugins/qt/PluginPackageQt.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44120 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Holger Hans Peter Freyther <zecke@selfish.org>
zecke@webkit.org [Mon, 25 May 2009 03:28:41 +0000 (03:28 +0000)]
2009-05-24  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=25988

        So far the initialization of the browserfuncs struct
        is done in PluginPackageQt, PluginPackageGtk and
        PluginPackageWin there is no need for this duplication.

        The PluginPackageWin version got copied to PluginPackage

        * plugins/PluginPackage.cpp:
        (WebCore::PluginPackage::initializeBrowserFuncs):
        * plugins/PluginPackage.h:
        * plugins/gtk/PluginPackageGtk.cpp:
        (WebCore::PluginPackage::load):
        * plugins/qt/PluginPackageQt.cpp:
        (WebCore::PluginPackage::load):
        * plugins/win/PluginPackageWin.cpp:
        (WebCore::PluginPackage::load):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44119 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Sun, 24 May 2009 22:43:27 +0000 (22:43 +0000)]
2009-05-24  Dirk Schulze  <krit@webkit.org>

        Reviewed by Nikolas Zimmermann.

        WebKit needs cross-platform filter system
        [https://bugs.webkit.org/show_bug.cgi?id=19991]

        Added standard input for filter system.

        No testcases were added. Filter system is still off.

        * GNUmakefile.am:
        * WebCore.pro:
        * WebCore.xcodeproj/project.pbxproj:
        * platform/graphics/filters/SourceAlpha.cpp: Added.
        (WebCore::SourceAlpha::create):
        (WebCore::SourceAlpha::effectName):
        (WebCore::SourceAlpha::apply):
        (WebCore::SourceAlpha::dump):
        * platform/graphics/filters/SourceAlpha.h: Added.
        (WebCore::SourceAlpha::SourceAlpha):
        * platform/graphics/filters/SourceGraphic.cpp: Added.
        (WebCore::SourceGraphic::create):
        (WebCore::SourceGraphic::effectName):
        (WebCore::SourceGraphic::apply):
        (WebCore::SourceGraphic::dump):
        * platform/graphics/filters/SourceGraphic.h: Added.
        (WebCore::SourceGraphic::SourceGraphic):
        * svg/FilterBuilder.cpp: Added.
        (WebCore::FilterBuilder::FilterBuilder):
        (WebCore::FilterBuilder::add):
        (WebCore::FilterBuilder::getEffectById):
        (WebCore::FilterBuilder::clearEffects):
        * svg/FilterBuilder.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44118 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Steve Falkenburg <sfalken@apple.com>
sfalken@apple.com [Sun, 24 May 2009 18:13:14 +0000 (18:13 +0000)]
2009-05-24  Steve Falkenburg  <sfalken@apple.com>

        Build fix for experimental PGO Windows target.

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44117 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Simon Hausmann <simon.hausmann@nokia.com>
hausmann@webkit.org [Sun, 24 May 2009 18:10:41 +0000 (18:10 +0000)]
2009-05-24  Simon Hausmann  <simon.hausmann@nokia.com>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=25612

        Add support conditionals in the JSC bindings generator code for attributes,
        by placing feature #ifdefs around the setters/getters.

        Made the audio and video element specific DOM attributes conditional
        in the generated code, by moving the #ifdefs from the IDL files into
        the generated code. This allows for re-using the same generated code
        with or without ENABLE(VIDEO).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44116 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Darin Adler <darin@apple.com>
darin@apple.com [Sun, 24 May 2009 16:34:52 +0000 (16:34 +0000)]
2009-05-24  Darin Adler  <darin@apple.com>

        Reviewed by Dan Bernstein and Tim Hatcher.

        <rdar://problem/6913045> New console message spewed by automator CLI after installing JetstreamLeo40B21a

        * Misc/WebNSObjectExtras.mm:
        (-[WebMainThreadInvoker initWithTarget:]): Tweaked argument name.
        (-[WebMainThreadInvoker forwardInvocation:]): Removed call to retainArguments.
        This was unneeded and in the case of a newly created but not yet fully initialized
        NSView object it caused the abovementioned bug.
        (-[WebMainThreadInvoker handleException:]): Tweaked argument name. Added assertion.
        (-[NSInvocation _webkit_invokeAndHandleException:]): Tweaked name of local variable
        that holds the exception.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44115 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Darin Adler <darin@apple.com>
darin@apple.com [Sun, 24 May 2009 15:34:24 +0000 (15:34 +0000)]
2009-05-24  Darin Adler  <darin@apple.com>

        * StringsNotToBeLocalized.txt: Updated for recent changes. Got rid of most full-file exceptions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44114 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-24 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Sun, 24 May 2009 07:48:50 +0000 (07:48 +0000)]
2009-05-24  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Mark Rowe.

        https://bugs.webkit.org/show_bug.cgi?id=25987
        [GTK] Clarify WebKitWebView::download-requested documentation

        Explain a bit why this signal is not emitted by default and offer
        a common solution/setup for that issue.

        * webkit/webkitwebview.cpp:
        (webkit_web_view_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44113 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUnreviewed build fix for Qt. Change signature to have PassOwnPtr.
zecke@webkit.org [Sun, 24 May 2009 06:38:03 +0000 (06:38 +0000)]
Unreviewed build fix for Qt. Change signature to have PassOwnPtr.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44112 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoLand the updated version of WebKitLauncher that the nightly builds have been using.
mrowe@apple.com [Sun, 24 May 2009 05:14:44 +0000 (05:14 +0000)]
Land the updated version of WebKitLauncher that the nightly builds have been using.

Reviewed by Oliver Hunt.

Changes include:
1) Support for self-updating via Sparkle.
2) Now respects the "Open using Rosetta" preference.
3) Passes environment variables through to the spawned Safari process.
4) Logs a list of extensions that are detected to the system console.
5) Disables extensions that are known to cause crashes too early in the
   launch process to be detected by the extension-detection code.

* WebKitLauncher/Configurations/Base.xcconfig: Added.
* WebKitLauncher/Configurations/WebKitLauncher.xcconfig: Added.
* WebKitLauncher/Configurations/WebKitNightlyEnabler.xcconfig: Added.
* WebKitLauncher/Info.plist:
* WebKitLauncher/WebKitLauncher.xcodeproj/project.pbxproj:
* WebKitLauncher/WebKitNightlyEnabler.h: Added.
* WebKitLauncher/WebKitNightlyEnabler.m:
(myBundleDidLoad):
(myApplicationWillFinishLaunching):
(poseAsWebKitApp):
(webKitLauncherBundle):
(enableWebKitNightlyBehaviour):
* WebKitLauncher/WebKitNightlyEnablerSparkle.h: Added.
* WebKitLauncher/WebKitNightlyEnablerSparkle.m: Added.
(initializeSparkle):
(updatePermissionPromptDescription):
(skipSignatureVerificationInDownloadDidFinish):
(updateAlertPanel):
(updateAlertInitForAlertPanel):
(-[NSAlert webKitHandleButtonPress:]):
(setMethodImplementation):
* WebKitLauncher/main.m:
(displayErrorAndQuit):
(getLastVersionShown):
(saveLastVersionShown):
(getPathForStartPage):
(getCurrentVersion):
(getShowStartPageVersion):
(startPageDisabled):
(addStartPageToArgumentsIfNeeded):
(preferredArchitecture):
(myExecve):
(locateSafariBundle):
(currentMacOSXVersion):
(checkFrameworkPath):
(main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44111 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix typo "am" -> "an" in the Changelog of my previous commit
jmalonzo@webkit.org [Sun, 24 May 2009 04:21:34 +0000 (04:21 +0000)]
Fix typo "am" -> "an" in the Changelog of my previous commit

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44110 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-23 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Sun, 24 May 2009 04:06:57 +0000 (04:06 +0000)]
2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Unreviewed build fix. Add gstreamer-video-0.10 libs to
        GSTREAMER_LIBS to resolve am undefined reference to gst_video_get_size
        - symbol used in MediaPlayerPrivateGstreamer.

        * configure.ac:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44109 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler.
mitz@apple.com [Sat, 23 May 2009 23:28:58 +0000 (23:28 +0000)]
    Reviewed by Darin Adler.

        - clean up ColorMac a little

        * platform/graphics/mac/ColorMac.h: Whitespace change.
        * platform/graphics/mac/ColorMac.mm: Removed redundant #import
        statements and sorted the remaining ones.
        (WebCore::makeRGBAFromNSColor): Replaced four Objective-C method calls
        with a single call and removed redundant casts to int.
        (WebCore::nsColor): Removed ".0f" from number literals used as CGFloats.
        In one case, this avoids an intermediate conversion to float before
        conversion to double on 64-bit.
        (WebCore::CGColorFromNSColor): Replaced four Objective-C method calls
        with a single call.
        (WebCore::focusRingColor):
        (+[WebCoreControlTintObserver controlTintDidChange]): Whitespace change.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44108 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-23 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Sat, 23 May 2009 22:39:33 +0000 (22:39 +0000)]
2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Rubber-stamped by Gustavo Noronha.

        Break the switch once private browsing is selected.

        * webkit/webkitwebsettings.cpp:
        (webkit_web_settings_set_property):
        (webkit_web_settings_get_property):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44106 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-23 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Sat, 23 May 2009 22:35:15 +0000 (22:35 +0000)]
2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Gustavo Noronha and Xan Lopez.

        [Gtk] Enable database and localStorage support
        https://bugs.webkit.org/show_bug.cgi?id=25629

        Enable storage/ tests for Gtk.

        * platform/gtk/Skipped:

2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Gustavo Noronha and Xan Lopez.

        [Gtk] Enable database and localStorage support
        https://bugs.webkit.org/show_bug.cgi?id=25629

        Add settings to enable or disable database and localStorage
        support and enable them by default. This allows databases to be
        shown in the WebInspector as well as allow us to run the storage
        layout tests.

        * webkit/webkitwebsettings.cpp:
        (webkit_web_settings_class_init):
        (webkit_web_settings_set_property):
        (webkit_web_settings_get_property):
        (webkit_web_settings_copy):
        * webkit/webkitwebview.cpp:
        (webkit_web_view_update_settings):
        (webkit_web_view_settings_notify):

2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Gustavo Noronha and Xan Lopez.

        [Gtk] Enable database and localStorage support
        https://bugs.webkit.org/show_bug.cgi?id=25629

        Enable database and localStorage for running the storage/ tests.

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (resetWebViewToConsistentStateBeforeTesting):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44105 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-23 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Sat, 23 May 2009 22:22:52 +0000 (22:22 +0000)]
2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Xan Lopez.

        [Gtk] ROLE_ROW_HEADER should not be used for list item bullets/numbers
        https://bugs.webkit.org/show_bug.cgi?id=25900

        Use ATK_ROLE_TEXT for ListMarkerRole.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (atkRole):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44104 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2009-05-23 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Sat, 23 May 2009 22:20:31 +0000 (22:20 +0000)]
2009-05-23  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Xan Lopez.

        [Gtk] Various autotools build refactoring and fixes
        https://bugs.webkit.org/show_bug.cgi?id=25286

        Refactor library LIBS. Move third-party libs in libwebkit instead
        of libWebCore.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44103 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reformat WebCore::requiresContextForWordBoundary()
ddkilzer@apple.com [Sat, 23 May 2009 22:08:53 +0000 (22:08 +0000)]
    Reformat WebCore::requiresContextForWordBoundary()

        * platform/text/TextBoundaries.h:
        (WebCore::requiresContextForWordBoundary):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44102 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoNot reviewed. Roll out ResourceHandle patch, that I accidently included in my last...
zimmermann@webkit.org [Sat, 23 May 2009 21:29:20 +0000 (21:29 +0000)]
Not reviewed. Roll out ResourceHandle patch, that I accidently included in my last commit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44101 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoReviewed by Darin Adler.
zimmermann@webkit.org [Sat, 23 May 2009 21:25:22 +0000 (21:25 +0000)]
Reviewed by Darin Adler.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=23808
After a long journey, add the last missing ABC to share form control element support with HTMLSelectElement.
RenderListBox/RenderMenuList operates on SelectElement instead of HTMLSelectElement now, querying selected index etc.
This makes it possible to use these render objects for ie. WML, as it's done before for input & option elements.

WMLSelectElement will be created soon, in a follow-up patch, adding more methods to SelectElement, sharing code between
HTMLSelectElement/WMLSelectElement.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@44100 268f45cc-cd09-0410-ab3c-d52691b4dbfc