WebKit-https.git
10 years ago2011-03-10 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org>
commit-queue@webkit.org [Fri, 11 Mar 2011 00:56:37 +0000 (00:56 +0000)]
2011-03-10  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>

        Reviewed by Adam Roben.

        Formatted Diff for attachment 23920 is mangled
        https://bugs.webkit.org/show_bug.cgi?id=21222

        The mangled diff files mentioned in the bug were using Mac line
        ending and this was causing problems for String#each_line. Now we
        normalize the line endings in patch data before processing it.

        * PrettyPatch/PrettyPatch.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Gyuyoung Kim <gyuyoung.kim@samsung.com>
gyuyoung.kim@samsung.com [Fri, 11 Mar 2011 00:56:09 +0000 (00:56 +0000)]
2011-03-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>

        Unreviewed build fix.

        [EFL] Fix build break when CROSS_PLATFORM_CONTEXT_MENUS is disabled.
        https://bugs.webkit.org/show_bug.cgi?id=56005

        There are build breaks when CROSS_PLATFORM_CONTEXT_MENUS is disabled.

        * platform/efl/ContextMenuEfl.cpp:
        (WebCore::ContextMenu::ContextMenu):
        * platform/efl/ContextMenuItemEfl.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80794 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Fri, 11 Mar 2011 00:52:53 +0000 (00:52 +0000)]
2011-03-10  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        startOfBlock and endOfBlock may return a position inside hr
        https://bugs.webkit.org/show_bug.cgi?id=56025

        Replaced calls to enclosingBlockFlowElement in startOfBlock and endOfBlock by
        calls to enclosingBlock. Added EditingBoundaryCrossingRule to the argument lists
        of startOfBlock, endOfBlock, and enclosingBlock. Also replaced the last boolean
        argument variable of enclosingNodeOfType by EditingBoundaryCrossingRule.

        Also replaced calls to enclosingBlockFlowElement in inSameBlock by calls to
        enclosingBlock to be consitent with startOfBlock and endOfBlock.

        This patch also replaced calls to deprecatedNode in startOfBlock, endOfBlock,
        and inSameBlock by calls to containerNode because the enclosing block of a position
        should never be before or after the position.

        No tests are added because this change only affects WebCore internally.

        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::initializePositionData): Calls enclosingNodeOfType.
        Pass CanCrossEditingBoundary instead of false.
        * editing/htmlediting.cpp:
        (WebCore::unsplittableElementForPosition): Ditto.
        (WebCore::enclosingBlock): Takes EditingBoundaryCrossingRule and passes it to
        enclosingNodeOfType.
        (WebCore::enclosingNodeOfType): Takes EditingBoundaryCrossingRule instead of boolean.
        Fixed a bug that it stops walking the tree when it reached the root editable node
        even when the editing boundary crossing rule is CanCrossEditingBoundary.
        * editing/htmlediting.h: Prototype changes.
        * editing/visible_units.cpp:
        (WebCore::startOfBlock): Calls enclosingBlock instead of enclosingBlockFlowElement.
        Also added an early exit when there's no enclosing block.
        (WebCore::endOfBlock): Ditto. The early exist in this case prevents crash in
        lastPositionInNode.
        (WebCore::inSameBlock): Calls enclosingBlock instead of enclosingBlockFlowElement.
        (WebCore::isStartOfBlock): Calls startOfBlock with CanCrossEditingBoundary because
        we don't care where the start of block is when we're comparing against the given position.
        (WebCore::isEndOfBlock): Ditto.
        * editing/visible_units.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80793 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Fri, 11 Mar 2011 00:49:01 +0000 (00:49 +0000)]
2011-03-10  Martin Robinson  <mrobinson@igalia.com>

        Build for WebKit2 GTK+ after r80714.

        * GNUmakefile.am: Update source list.
        * WebProcess/Plugins/Netscape/x11/NetscapePluginX11.cpp:
        (WebKit::x11Display): Add missing semicolon.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80792 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Alexey Proskuryakov <ap@apple.com>
ap@apple.com [Fri, 11 Mar 2011 00:48:25 +0000 (00:48 +0000)]
2011-03-10  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Dan Bernstein.

        Backspace deletes all non-BMP symbols at once, and then some
        https://bugs.webkit.org/show_bug.cgi?id=55971
        <rdar://problem/8725312>

        * platform/mac/editing/input/insert-delete-smp-symbol-expected.txt: Added.
        * platform/mac/editing/input/insert-delete-smp-symbol.html: Added.
2011-03-10  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Dan Bernstein.

        Backspace deletes all non-BMP symbols at once, and then some
        https://bugs.webkit.org/show_bug.cgi?id=55971
        <rdar://problem/8725312>

        Test: platform/mac/editing/input/insert-delete-smp-symbol.html

        * rendering/RenderText.cpp:
        (WebCore::isMark): This matches what Core Foundation does for all characters that I tested.
        (WebCore::RenderText::previousOffsetForBackwardDeletion): Changed to use isMark().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoREGRESSION (WebKit2): VoiceOver focus no longer follows keyboard focus
cfleizach@apple.com [Fri, 11 Mar 2011 00:40:05 +0000 (00:40 +0000)]
REGRESSION (WebKit2): VoiceOver focus no longer follows keyboard focus
https://bugs.webkit.org/show_bug.cgi?id=55959

Reviewed by Darin Adler.

NSApplication has to return the focused element within the app in order
for NSAccessibilityHandleFocusChanged() to work. The default implementation
relies on keyWindow. Consequently, in WK2, the WebProcess must override that method
to return the appropriate focused element within the WebProcess, since
there is no keyWindow.

* WebKit2.xcodeproj/project.pbxproj:
* WebProcess/WebPage/mac/AccessibilityWebPageObject.mm:
(-[AccessibilityWebPageObject accessibilityFocusedUIElement]):
* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::focusedWebPage):
* WebProcess/WebProcess.h:
* WebProcess/mac/NSApplicationOverride.mm: Added.
(-[NSApplication accessibilityFocusedUIElement]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Chris Evans <cevans@chromium.org>
inferno@chromium.org [Fri, 11 Mar 2011 00:36:29 +0000 (00:36 +0000)]
2011-03-10  Chris Evans  <cevans@chromium.org>

        Reviewed by Adam Barth.

        Error in StyleElement::process with large nodesets
        https://bugs.webkit.org/show_bug.cgi?id=56150

        Test: none due to excessive runtime and CRASH() vs. real crash.

        * dom/StyleElement.cpp:
        (WebCore::StyleElement::process): Handle large node sets better.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=47151
hyatt@apple.com [Fri, 11 Mar 2011 00:34:06 +0000 (00:34 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=47151

Reviewed by Simon Fraser.

Percentage heights should skip anonymous containing blocks when computing the used value.

Added fast/block/basic/percentage-height-inside-anonymous-block.html

Source/WebCore:

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computePercentageLogicalHeight):

LayoutTests:

* fast/block/basic/percent-height-inside-anonymous-block.html: Added.
* platform/mac/fast/block/basic/percent-height-inside-anonymous-block-expected.checksum: Added.
* platform/mac/fast/block/basic/percent-height-inside-anonymous-block-expected.png: Added.
* platform/mac/fast/block/basic/percent-height-inside-anonymous-block-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Nat Duca <nduca@chromium.org>
commit-queue@webkit.org [Fri, 11 Mar 2011 00:32:13 +0000 (00:32 +0000)]
2011-03-10  Nat Duca  <nduca@chromium.org>

        Reviewed by James Robinson.

        [chromium] Make updateAndDrawLayers argumentless.
        https://bugs.webkit.org/show_bug.cgi?id=55985

        Made the TilePainters and viewport parameters for
        LayerRendererChromium member variables instead of arguments on
        updateAndDrawLayers. In a future change, this will allows us to
        draw the layer tree without the WebView's assistance.

        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::create):
        (WebCore::LayerRendererChromium::LayerRendererChromium):
        (WebCore::LayerRendererChromium::verticalScrollbarRect):
        (WebCore::LayerRendererChromium::horizontalScrollbarRect):
        (WebCore::LayerRendererChromium::invalidateRootLayerRect):
        (WebCore::LayerRendererChromium::updateRootLayerContents):
        (WebCore::LayerRendererChromium::updateRootLayerScrollbars):
        (WebCore::LayerRendererChromium::drawRootLayer):
        (WebCore::LayerRendererChromium::setViewport):
        (WebCore::LayerRendererChromium::updateAndDrawLayers):
        (WebCore::LayerRendererChromium::updateLayers):
        (WebCore::LayerRendererChromium::drawLayers):
        (WebCore::LayerRendererChromium::setRootLayer):
        (WebCore::LayerRendererChromium::getFramebufferPixels):
        (WebCore::LayerRendererChromium::cleanupSharedObjects):
        * platform/graphics/chromium/LayerRendererChromium.h:
        (WebCore::LayerRendererChromium::viewportSize):
        * platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
        (WebCore::CCHeadsUpDisplay::draw):
2011-03-10  Nat Duca  <nduca@chromium.org>

        Reviewed by James Robinson.

        [chromium] Make updateAndDrawLayers argumentless.
        https://bugs.webkit.org/show_bug.cgi?id=55985

        Made the TilePainters and viewport parameters for
        LayerRendererChromium member variables instead of arguments on
        updateAndDrawLayers. In a future change, this will allows us to
        draw the layer tree without the WebView's assistance.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::resize):
        (WebKit::WebViewImpl::paint):
        (WebKit::WebViewImpl::scrollRootLayerRect):
        (WebKit::WebViewImpl::invalidateRootLayerRect):
        (WebKit::WebViewImplContentPainter::create):
        (WebKit::WebViewImplContentPainter::paint):
        (WebKit::WebViewImplContentPainter::WebViewImplContentPainter):
        (WebKit::WebViewImplScrollbarPainter::create):
        (WebKit::WebViewImplScrollbarPainter::paint):
        (WebKit::WebViewImplScrollbarPainter::WebViewImplScrollbarPainter):
        (WebKit::WebViewImpl::setIsAcceleratedCompositingActive):
        (WebKit::WebViewImpl::doComposite):
        (WebKit::WebViewImpl::reallocateRenderer):
        (WebKit::WebViewImpl::updateLayerRendererViewport):
        * src/WebViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Oleg Romashin <romaxa@gmail.com>
commit-queue@webkit.org [Fri, 11 Mar 2011 00:17:12 +0000 (00:17 +0000)]
2011-03-10  Oleg Romashin  <romaxa@gmail.com>

        Reviewed by Andreas Kling.

        Make Qt port compiling with ENABLE_PLUGIN_PROCESS=1, part3
        Solving conflict of Qt and Xlib includes

        * Shared/Plugins/NPIdentifierData.cpp:
        * Shared/Plugins/NPIdentifierData.h:
        * Shared/Plugins/NPObjectProxy.cpp:
        * Shared/Plugins/NPVariantData.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-01-25 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Fri, 11 Mar 2011 00:12:35 +0000 (00:12 +0000)]
2011-01-25  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Implement spin buttons for GTK+ 2.x
        https://bugs.webkit.org/show_bug.cgi?id=53098

        Unskip tests which are now passing with the spin button implementation.
        Add results for one test.

        * platform/gtk/Skipped: Unskip passing tests.
        * platform/gtk/fast/forms/input-appearance-spinbutton-disabled-readonly-expected.checksum: Added.
        * platform/gtk/fast/forms/input-appearance-spinbutton-disabled-readonly-expected.png: Added.
        * platform/gtk/fast/forms/input-appearance-spinbutton-disabled-readonly-expected.txt: Added.
2011-01-25  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Implement spin buttons for GTK+ 2.x
        https://bugs.webkit.org/show_bug.cgi?id=53098

        Implement spin buttons for GTK+ 2.x. This allows proper render and functioning
        for input type=number. The implementation is based on the one in gtkspinbutton.c
        from the GTK+ sources.

        * platform/gtk/RenderThemeGtk.h: Added new members and getters.
        * platform/gtk/RenderThemeGtk2.cpp:
        (WebCore::RenderThemeGtk::platformInit): Initialize spin button widget.
        (WebCore::RenderThemeGtk::adjustRepaintRect): We need to draw outside the
        spin button area to have proper rendering. There doesn't seem to be any other
        clean way of having spin buttons that are the size of the text input frame.
        (WebCore::RenderThemeGtk::adjustInnerSpinButtonStyle): Added implementation.
        (WebCore::RenderThemeGtk::paintInnerSpinButton): Ditto.
        (WebCore::RenderThemeGtk::gtkSpinButton): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org>
commit-queue@webkit.org [Fri, 11 Mar 2011 00:05:57 +0000 (00:05 +0000)]
2011-03-10  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>

        Reviewed by Adam Roben.

        PrettyPatch displays last two lines of a git-format-patch diff strangely
        https://bugs.webkit.org/show_bug.cgi?id=29317

        If the diff section has full range information, we parse only enough to cover the
        range. This avoids incorrectly showing trailing lines (like git signature) as part
        of the patch.

        * PrettyPatch/PrettyPatch.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Fri, 11 Mar 2011 00:05:14 +0000 (00:05 +0000)]
2011-03-10  Andreas Kling  <kling@webkit.org>

        Unreviewed build fix after r80774.

        QML property versioning is introduced in Qt 4.7.3, not 4.7.2.
        See also: http://bugreports.qt.nokia.com/browse/QTBUG-13451

        * declarative/plugin.cpp:
        (WebKitQmlPlugin::registerTypes):
        * declarative/qdeclarativewebview.cpp:
        * declarative/qdeclarativewebview_p.h:
        * tests/qdeclarativewebview/tst_qdeclarativewebview.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80781 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Levi Weintraub <leviw@chromium.org>
leviw@chromium.org [Thu, 10 Mar 2011 23:50:56 +0000 (23:50 +0000)]
2011-03-10  Levi Weintraub  <leviw@chromium.org>

        Reviewed by Ryosuke Niwa.

        InsertUnorderedList over a non-editable region and multiple lines enters an infinite loop
        https://bugs.webkit.org/show_bug.cgi?id=53409

        Avoiding crashes and infinite loops when listifying content with mixed-editability

        * editing/execCommand/insert-list-with-noneditable-content-expected.txt: Added.
        * editing/execCommand/insert-list-with-noneditable-content.html: Added.
2011-03-10  Levi Weintraub  <leviw@chromium.org>

        Reviewed by Ryosuke Niwa.

        InsertUnorderedList over a non-editable region and multiple lines enters an infinite loop
        https://bugs.webkit.org/show_bug.cgi?id=53409

        Fixing broken handling of mixed-editability content for InsertListCommand. Previously, if the selection
        spanned non-contenteditable regions, it would get stuck endlessly iterating the same region as the algorithm
        didn't skip the editable boundary.

        Test: editing/execCommand/insert-list-with-noneditable-content.html

        * editing/CompositeEditCommand.cpp:
        (WebCore::CompositeEditCommand::cleanupAfterDeletion): Changed signature to take the destination
        position for the active editing command. Without this, there are cases when the destination happens
        to be a placeholder, and we remove it.
        (WebCore::CompositeEditCommand::moveParagraphs):
        * editing/CompositeEditCommand.h:
        * editing/InsertListCommand.cpp:
        (WebCore::InsertListCommand::doApply): Added logic to the paragraph iteration loop to handle pockets of
        non-editable content in an editable context. Previously, this could cause an infinite loop.
        * editing/visible_units.cpp:
        (WebCore::startOfParagraph): Added a mode of operation where we'll jump across non-editable
        content in the same paragraph to reach the actual editable paragraph start.
        (WebCore::endOfParagraph): Ditto.
        (WebCore::startOfNextParagraph): Now uses the aforementioned non-editable content skipping mode of
        endOfParagraph.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Berend-Jan Wever <skylined@chromium.org>
commit-queue@webkit.org [Thu, 10 Mar 2011 23:49:24 +0000 (23:49 +0000)]
2011-03-10  Berend-Jan Wever  <skylined@chromium.org>

        Reviewed by Darin Adler.

        Calling focus() on an area element not in a document should not cause a NULL ptr crash
        https://bugs.webkit.org/show_bug.cgi?id=54877

        * fast/dom/HTMLAreaElement: Added.
        * fast/dom/HTMLAreaElement/area-islink-focus-null-ptr-crash-expected.txt: Added.
        * fast/dom/HTMLAreaElement/area-islink-focus-null-ptr-crash.html: Added.
2011-03-10  Berend-Jan Wever  <skylined@chromium.org>

        Reviewed by Darin Adler.

        Calling focus() on an area element not in a document should not cause a NULL ptr crash
        https://bugs.webkit.org/show_bug.cgi?id=54877

        Test: fast/dom/HTMLAreaElement/area-islink-focus-null-ptr-crash.html

        * dom/Element.cpp:
        (WebCore::Element::focus): Check element is in the document before allowing focus
        * html/HTMLAreaElement.cpp:
        (WebCore::HTMLAreaElement::imageElement): Check element has a parent before checking if its parent is a map

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Thu, 10 Mar 2011 23:49:01 +0000 (23:49 +0000)]
2011-03-10  Ryosuke Niwa  <rniwa@webkit.org>

        Windows rebaselines for r80755.

        * platform/win/editing/selection/caret-ltr-right-expected.txt:
        * platform/win/editing/selection/caret-rtl-2-left-expected.txt:
        * platform/win/editing/selection/caret-rtl-right-expected.txt:
        * platform/win/fast/forms/file-input-disabled-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Xiyuan Xia <xiyuan@chromium.org>
tony@chromium.org [Thu, 10 Mar 2011 23:46:23 +0000 (23:46 +0000)]
2011-03-10  Xiyuan Xia  <xiyuan@chromium.org>

        Reviewed by Tony Chang.

        [Chromium] Fix default single selection select's popup background on chromium/linux
        https://bugs.webkit.org/show_bug.cgi?id=56023

        * fast/html/select-dropdown-consistent-background-color-expected.txt: Added.
        * fast/html/select-dropdown-consistent-background-color.html: Added.
2011-03-10  Xiyuan Xia  <xiyuan@chromium.org>

        Reviewed by Tony Chang.

        [Chromium] Fix default single selection select's popup background on chromium/linux
        https://bugs.webkit.org/show_bug.cgi?id=56023

        Test: fast/html/select-dropdown-consistent-background-color.html

        * css/themeChromiumLinux.css:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Kris Jordan <krisjordan@gmail.com>
commit-queue@webkit.org [Thu, 10 Mar 2011 23:43:48 +0000 (23:43 +0000)]
2011-03-10  Kris Jordan  <krisjordan@gmail.com>

        Reviewed by Alexey Proskuryakov.

        Testing the improved Accept header as per bug 27267.
        https://bugs.webkit.org/show_bug.cgi?id=27267

        * http/tests/misc/xhtml-expected.txt: Tests against improved Accept header.
2011-03-10  Kris Jordan  <krisjordan@gmail.com>

        Reviewed by Alexey Proskuryakov.

        Improve default Accept header to give preference to HTML over XML.
        https://bugs.webkit.org/show_bug.cgi?id=27267

        * loader/FrameLoader.cpp:Changed default accept header to match
        FireFox' as per bug 27267.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Thu, 10 Mar 2011 23:41:20 +0000 (23:41 +0000)]
2011-03-10  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] [Webkit2] There are no scrollbars visible in the MiniBrowser
        https://bugs.webkit.org/show_bug.cgi?id=56125

        No new tests. This functionality will be tested once we have TestRunner
        implementation for WebKit2 with pixel dumping support.

        * GNUmakefile.am: Add a new define with tracks whether or not we are using
        our specialized GTK+ ScrollView.
        * platform/ScrollView.cpp: Only use the GTK+-specific ScrollView if we are compiling
        WebKit1.
        (WebCore::ScrollView::wheelEvent):
        * platform/gtk/ScrollViewGtk.cpp: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Alexis Menard <alexis.menard@openbossa.org>
commit-queue@webkit.org [Thu, 10 Mar 2011 23:36:38 +0000 (23:36 +0000)]
2011-03-10  Alexis Menard  <alexis.menard@openbossa.org>

        Reviewed by Andreas Kling.

        [Qt] QtDeclarative Webview element has a fixed white background
        https://bugs.webkit.org/show_bug.cgi?id=40918

        Implement a way to change the background color of the WebView QML element.
        This feature is activated for QtWebKit 1.1 version of the plugin.

        * declarative/plugin.cpp:
        (WebKitQmlPlugin::registerTypes):
        * declarative/qdeclarativewebview.cpp:
        (QDeclarativeWebView::backgroundColor):
        (QDeclarativeWebView::setBackgroundColor):
        * declarative/qdeclarativewebview_p.h:
        * tests/qdeclarativewebview/resources/webviewbackgroundcolor.qml: Added.
        * tests/qdeclarativewebview/tst_qdeclarativewebview.cpp:
        (tst_QDeclarativeWebView::backgroundColor):
        * tests/qdeclarativewebview/tst_qdeclarativewebview.qrc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Thu, 10 Mar 2011 23:32:54 +0000 (23:32 +0000)]
2011-03-10  Kent Tamura  <tkent@chromium.org>

        Reviewed by Dimitri Glazkov.

        Assertion fails by validating a form twice very quickly
        https://bugs.webkit.org/show_bug.cgi?id=56069

        * fast/forms/interactive-validation-assertion-by-validate-twice-expected.txt: Added.
        * fast/forms/interactive-validation-assertion-by-validate-twice.html: Added.
        * fast/forms/interactive-validation-attach-assertion.html: Remove unnecessary offsetLeft trick.
        * platform/gtk/Skipped:
        * platform/mac/Skipped:
        * platform/qt/Skipped:
        * platform/win/Skipped:
2011-03-10  Kent Tamura  <tkent@chromium.org>

        Reviewed by Dimitri Glazkov.

        Assertion fails by validating a form twice very quickly
        https://bugs.webkit.org/show_bug.cgi?id=56069

        If the interactive validation is invoked when a form control is in
        needsLayout() state, an assertion in isFocusable() fails. To avoid it,
        Add a call to updateLayoutIgnorePendingStylesheets() before isFocusable().

        Test: fast/forms/interactive-validation-assertion-by-validate-twice.html

        * html/HTMLFormElement.cpp:
        (WebCore::HTMLFormElement::validateInteractively):
          Calls updateLayoutIgnorePendingStylesheets().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Thu, 10 Mar 2011 23:29:50 +0000 (23:29 +0000)]
2011-03-10  Martin Robinson  <mrobinson@igalia.com>

        Unskipping some crashers on GTK+. These no longer crash locally and we'd
        like to figure out if they still crash on the bots. Move one flaky test
        to the appropriate section.

        * platform/gtk/Skipped: Unskip tests passing locally.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ryuan Choi <ryuan.choi@samsung.com>
commit-queue@webkit.org [Thu, 10 Mar 2011 23:23:10 +0000 (23:23 +0000)]
2011-03-10  Ryuan Choi  <ryuan.choi@samsung.com>

        Reviewed by Antonio Gomes.

        [EFL] Limit touch events when only page required touch events.
        https://bugs.webkit.org/show_bug.cgi?id=54857

        Block touch events when ChromeClient::needTouchEvents got false.

        * WebCoreSupport/ChromeClientEfl.cpp:
        (WebCore::ChromeClientEfl::needTouchEvents):
        * WebCoreSupport/ChromeClientEfl.h:
        * ewk/ewk_frame.cpp:
        (ewk_frame_feed_touch_event):
        * ewk/ewk_private.h:
        * ewk/ewk_view.cpp:
        (ewk_view_need_touch_events_set):
        (ewk_view_need_touch_events_get):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=47143
hyatt@apple.com [Thu, 10 Mar 2011 23:11:32 +0000 (23:11 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=47143

Reviewed by Beth Dakin.

Static distance computation is wrong when an ancestor is also positioned.  We need to go up the container()
chain (after initially starting with the parent()) when doing this computation, so that we properly skip
intermediate boxes between two positioned blocks.

Added fast/block/positioning/static-distance-with-positioned-ancestor.html.

Source/WebCore:

* rendering/RenderBox.cpp:
(WebCore::computeInlineStaticDistance):
(WebCore::computeBlockStaticDistance):

LayoutTests:

* fast/block/positioning/static-distance-with-positioned-ancestor.html: Added.
* platform/mac/fast/block/positioning/static-distance-with-positioned-ancestor-expected.checksum: Added.
* platform/mac/fast/block/positioning/static-distance-with-positioned-ancestor-expected.png: Added.
* platform/mac/fast/block/positioning/static-distance-with-positioned-ancestor-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Stanislav Paltis <Stanislav.Paltis@nokia.com>
commit-queue@webkit.org [Thu, 10 Mar 2011 23:10:58 +0000 (23:10 +0000)]
2011-03-10  Stanislav Paltis  <Stanislav.Paltis@nokia.com>

        Reviewed by Laszlo Gombos.

        [Qt] MemoryCache deadDecodedDataDeletionInterval is not exposed for client's usage
        https://bugs.webkit.org/show_bug.cgi?id=55945

        Added handling of dynamic/runtime property _q_deadDecodedDataDeletionInterval to
        set interval used to trigger when decoded data in dead list of object cache will
        be purged from object cache.

        * Api/qwebpage.cpp:
        (QWebPagePrivate::dynamicPropertyChangeEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org>
commit-queue@webkit.org [Thu, 10 Mar 2011 22:54:13 +0000 (22:54 +0000)]
2011-03-10  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>

        Reviewed by Antonio Gomes.

        Simplify how QWebFrame::requestedUrl() is obtained
        https://bugs.webkit.org/show_bug.cgi?id=55842

        When a load starts, store the requested URL until we know that it'll be
        available for us in the document loader -- after load finished.

        The existing auto tests cover the three different code paths in
        requestedUrl() and the new code passes the autotests. In each of those
        cases, we looked for the information in a different place, but in all
        of them, dispatchDidStartProvisionalLoad was called.

        This simplification will be useful to fix bug 32723. The way requestedUrl()
        is implementent, we can't use it as a fallback for url() when the setUrl()
        was called with an invalid URL.

        * Api/qwebframe.cpp:
        (QWebFrame::requestedUrl):
        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::dispatchDidStartProvisionalLoad):
        (WebCore::FrameLoaderClientQt::dispatchDidFinishLoad):
        * WebCoreSupport/FrameLoaderClientQt.h:
        (WebCore::FrameLoaderClientQt::lastRequestedUrl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Adrienne Walker <enne@google.com>
enne@google.com [Thu, 10 Mar 2011 22:47:40 +0000 (22:47 +0000)]
2011-03-10  Adrienne Walker  <enne@google.com>

        Reviewed by Kenneth Russell.

        [chromium] Make tiled compositor data structure more efficient.
        https://bugs.webkit.org/show_bug.cgi?id=54133

        Previously, the compositor had a sparse 2D array of tiles for the
        whole page, most of which were null.  The tiles were implicitly
        located based on their position in the array. This was inefficient
        when the page grew (e.g. infinite scrolling) and caused some bugs
        (e.g. width * height > MAX_INT).  This change modifies tiles to have
        explicit positions so they can be stored in a hash map.

        Tests: LayoutTests/compositing/

        * platform/graphics/chromium/LayerTilerChromium.cpp:
        (WebCore::LayerTilerChromium::reset):
        (WebCore::LayerTilerChromium::createTile):
        (WebCore::LayerTilerChromium::invalidateTiles):
        (WebCore::LayerTilerChromium::contentRectToTileIndices):
        (WebCore::LayerTilerChromium::tileAt):
        (WebCore::LayerTilerChromium::tileContentRect):
        (WebCore::LayerTilerChromium::tileLayerRect):
        (WebCore::LayerTilerChromium::invalidateRect):
        (WebCore::LayerTilerChromium::invalidateEntireLayer):
        (WebCore::LayerTilerChromium::update):
        (WebCore::LayerTilerChromium::updateFromPixels):
        (WebCore::LayerTilerChromium::draw):
        (WebCore::LayerTilerChromium::growLayerToContain):
        * platform/graphics/chromium/LayerTilerChromium.h:
        (WebCore::LayerTilerChromium::Tile::Tile):
        (WebCore::LayerTilerChromium::Tile::i):
        (WebCore::LayerTilerChromium::Tile::j):
        (WebCore::LayerTilerChromium::Tile::moveTo):
        (WebCore::LayerTilerChromium::TileMapKeyTraits::emptyValue):
        (WebCore::LayerTilerChromium::TileMapKeyTraits::constructDeletedValue):
        (WebCore::LayerTilerChromium::TileMapKeyTraits::isDeletedValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Thu, 10 Mar 2011 22:39:10 +0000 (22:39 +0000)]
2011-03-10  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed build fix for r80752; Negated the boolean logic.

        * editing/VisiblePosition.cpp:
        (WebCore::VisiblePosition::next):
        (WebCore::VisiblePosition::previous):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago- Fix plugins in release and production builds.
weinig@apple.com [Thu, 10 Mar 2011 22:28:02 +0000 (22:28 +0000)]
- Fix plugins in release and production builds.

Reviewed by Mark Rowe.

* WebKit2.xcodeproj/project.pbxproj:
Make sure to base the PluginProcess target on the correct xcconfig file
in all configurations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMake the same change to xheight on Windows for CG.
hyatt@apple.com [Thu, 10 Mar 2011 22:24:41 +0000 (22:24 +0000)]
Make the same change to xheight on Windows for CG.

Reviewed by Adam Roben.

* platform/graphics/win/SimpleFontDataCGWin.cpp:
(WebCore::SimpleFontData::platformInit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/9117047>
andersca@apple.com [Thu, 10 Mar 2011 22:17:12 +0000 (22:17 +0000)]
<rdar://problem/9117047>
Don't paint more than 60 times per second even when scrolling

Reviewed by Sam Weinig.

* WebView/WebClipView.mm:
(-[WebClipView _immediateScrollToPoint:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Tony Chang <tony@chromium.org>
tony@chromium.org [Thu, 10 Mar 2011 22:16:35 +0000 (22:16 +0000)]
2011-03-10  Tony Chang  <tony@chromium.org>

        Unreviewed, roll chromium version to r77259 to pick up
        webfontinfo api change.

        * DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=47147
hyatt@apple.com [Thu, 10 Mar 2011 22:13:25 +0000 (22:13 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=47147

Reviewed by Simon Fraser.

Fix for repaint issues when the root element is a table and needs to paint a background that
covers the entire canvas.  Remove paintRootBoxDecorations and replace with paintRootBoxFillLayers,
so that it can be called only for fill layer painting.  Make tables and normal boxes both
call this new method.  Fix the dirty check at the start of RenderTable paint to not happen if
the table is the root of the document (this same check exists in RenderBlock painting already).

Source/WebCore:

* rendering/RenderBox.cpp:
(WebCore::RenderBox::paintRootBoxFillLayers):
(WebCore::RenderBox::paintBoxDecorations):
(WebCore::RenderBox::paintBoxDecorationsWithSize):
* rendering/RenderBox.h:
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::calculateCompositedBounds):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::paint):
(WebCore::RenderTable::paintBoxDecorations):

LayoutTests:

* fast/backgrounds/positioned-root.html: Added.
* fast/backgrounds/table-root.html: Added.
* platform/mac/fast/backgrounds/positioned-root-expected.checksum: Added.
* platform/mac/fast/backgrounds/positioned-root-expected.png: Added.
* platform/mac/fast/backgrounds/positioned-root-expected.txt: Added.
* platform/mac/fast/backgrounds/table-root-expected.checksum: Added.
* platform/mac/fast/backgrounds/table-root-expected.png: Added.
* platform/mac/fast/backgrounds/table-root-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Thu, 10 Mar 2011 22:06:09 +0000 (22:06 +0000)]
2011-03-10  Mihai Parparita  <mihaip@chromium.org>

        Update chromium Snow Leopard expectations.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Sheriff Bot <webkit.review.bot@gmail.com>
tonyg@chromium.org [Thu, 10 Mar 2011 22:00:05 +0000 (22:00 +0000)]
2011-03-10  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r80749.
        http://trac.webkit.org/changeset/80749
        https://bugs.webkit.org/show_bug.cgi?id=56143

        "Caused at least 3 media tests to be flaky" (Requested by
        tonyg-cr on #webkit).

        * html/parser/HTMLDocumentParser.cpp:
        (WebCore::HTMLDocumentParser::canTakeNextToken):
        (WebCore::HTMLDocumentParser::pumpTokenizer):
        * html/parser/HTMLParserScheduler.cpp:
        * html/parser/HTMLParserScheduler.h:
        (WebCore::HTMLParserScheduler::checkForYieldBeforeToken):
        * page/FrameView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Alejandro G. Castro <alex@igalia.com>
alex@webkit.org [Thu, 10 Mar 2011 21:44:56 +0000 (21:44 +0000)]
2011-03-10  Alejandro G. Castro  <alex@igalia.com>

        Unreviewed, fixed Efl compilation after r80744.

        * Source/cmake/OptionsEfl.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRegression: Content not drawn when scrolling horizontally in an RTL page
weinig@apple.com [Thu, 10 Mar 2011 21:43:37 +0000 (21:43 +0000)]
Regression: Content not drawn when scrolling horizontally in an RTL page
https://bugs.webkit.org/show_bug.cgi?id=55077

Reviewed by David Hyatt.

* platform/ScrollView.cpp:
(WebCore::ScrollView::overhangAmount):
(WebCore::ScrollView::calculateOverhangAreasForPainting):
Take the scroll origin into account when calculating overhang.

* platform/ScrollView.h:
* rendering/RenderLayer.h:
* platform/ScrollableArea.h:
(WebCore::ScrollableArea::scrollOrigin):
Move identical scroll origin member from ScrollView and RenderLayer
to shared base ScrollableArea. This is also needed so that the animator
can access it.

* platform/mac/ScrollAnimatorMac.mm:
(WebCore::ScrollAnimatorMac::pinnedInDirection):
(WebCore::ScrollAnimatorMac::smoothScrollWithEvent):
(WebCore::ScrollAnimatorMac::snapRubberBandTimerFired):
Account for a scroll origin when doing calculating scroll offsets.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-09 Matthew Delaney <mdelaney@apple.com>
mdelaney@apple.com [Thu, 10 Mar 2011 21:41:00 +0000 (21:41 +0000)]
2011-03-09  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Simon Fraser.

        Plumb through settings for accelerated drawing for canvas
        https://bugs.webkit.org/show_bug.cgi?id=56039

        * WebView/WebPreferenceKeysPrivate.h:
        * WebView/WebPreferences.mm:
        * WebView/WebPreferencesPrivate.h:
2011-03-09  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Simon Fraser.

        Plumb through settings for accelerated drawing for canvas
        https://bugs.webkit.org/show_bug.cgi?id=56039

        No new tests. Doesn't affect behavior, just adding a switch to toggle canvas backends.

        * WebCore.exp.in:
        * html/HTMLCanvasElement.cpp:
        * html/canvas/CanvasRenderingContext2D.cpp:
        * page/Settings.cpp:
        * page/Settings.h:
2011-03-09  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Simon Fraser.

        Plumb through settings for accelerated drawing for canvas
        https://bugs.webkit.org/show_bug.cgi?id=56039

        * Shared/WebPreferencesStore.h:
        * UIProcess/API/C/WKPreferences.cpp:
        * UIProcess/API/C/WKPreferencesPrivate.h:
        * WebProcess/WebPage/WebPage.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=47157
hyatt@apple.com [Thu, 10 Mar 2011 21:27:40 +0000 (21:27 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=47157

Reviewed by Simon Fraser.

CSS2.1 test suite failures because the ex unit is broken with the Ahem font.

Remove the code that tries to also include the maxX of the glyph bounds for the
'x' glyph, since it just causes the x-height to be way too large in cases where the 'x' glyph extends
below the baseline.

Remove the Apple Symbol hack for ex units, since the person who added that was confused by another
issue, namely that CGFontGetXHeight wasn't being properly multiplied by the pointSize.  That's why
the value was too small.  Patched the code to just multiply by pointSize and took out the hack.

Fix causes many tests in the css2.1 directory to progress, so no new tests required.  Many other
layout tests change because the xHeight for Lucida Grande gets smaller by a little bit, and radio
buttons use ex horizontal margins by default.

Source/WebCore:

* platform/graphics/mac/SimpleFontDataMac.mm:
(WebCore::SimpleFontData::platformInit):

LayoutTests:

* platform/mac/accessibility/radio-button-group-members-expected.txt:
* platform/mac/css1/box_properties/acid_test-expected.txt:
* platform/mac/css1/font_properties/font_size-expected.txt:
* platform/mac/css1/text_properties/line_height-expected.txt:
* platform/mac/css1/text_properties/text_indent-expected.txt:
* platform/mac/css1/units/length_units-expected.txt:
* platform/mac/css2.1/t040302-c61-ex-len-00-b-a-expected.txt:
* platform/mac/css2.1/t040302-c61-rel-len-00-b-ag-expected.txt:
* platform/mac/css2.1/t09-c5526c-display-00-e-expected.txt:
* platform/mac/css2.1/t1008-c44-ln-box-02-d-ag-expected.txt:
* platform/mac/css2.1/t100801-c544-valgn-01-d-ag-expected.txt:
* platform/mac/css2.1/t100801-c544-valgn-03-d-agi-expected.txt:
* platform/mac/css2.1/t100801-c548-ln-ht-02-b-ag-expected.txt:
* platform/mac/css2.1/t1507-c526-font-sz-02-b-a-expected.txt:
* platform/mac/css2.1/t1601-c547-indent-00-b-a-expected.txt:
* platform/mac/css2.1/t1604-c541-word-sp-00-b-a-expected.txt:
* platform/mac/css2.1/t1604-c542-letter-sp-00-b-a-expected.txt:
* platform/mac/editing/selection/caret-ltr-2-expected.txt:
* platform/mac/editing/selection/caret-ltr-2-left-expected.txt:
* platform/mac/editing/selection/caret-ltr-expected.txt:
* platform/mac/editing/selection/caret-ltr-right-expected.txt:
* platform/mac/editing/selection/caret-rtl-2-expected.txt:
* platform/mac/editing/selection/caret-rtl-2-left-expected.txt:
* platform/mac/editing/selection/caret-rtl-expected.txt:
* platform/mac/editing/selection/caret-rtl-right-expected.txt:
* platform/mac/fast/block/basic/011-expected.txt:
* platform/mac/fast/block/margin-collapse/103-expected.txt:
* platform/mac/fast/css/ex-after-font-variant-expected.txt:
* platform/mac/fast/css/non-standard-checkbox-size-expected.txt:
* platform/mac/fast/forms/001-expected.txt:
* platform/mac/fast/forms/basic-inputs-expected.txt:
* platform/mac/fast/forms/box-shadow-override-expected.txt:
* platform/mac/fast/forms/checkbox-radio-onchange-expected.txt:
* platform/mac/fast/forms/file-input-disabled-expected.txt:
* platform/mac/fast/forms/form-element-geometry-expected.txt:
* platform/mac/fast/forms/formmove-expected.txt:
* platform/mac/fast/forms/formmove2-expected.txt:
* platform/mac/fast/forms/indeterminate-expected.txt:
* platform/mac/fast/forms/input-appearance-height-expected.txt:
* platform/mac/fast/forms/input-value-expected.txt:
* platform/mac/fast/forms/minWidthPercent-expected.txt:
* platform/mac/fast/forms/radio-attr-order-expected.txt:
* platform/mac/fast/forms/radio-nested-labels-expected.txt:
* platform/mac/fast/forms/radio_checked-expected.txt:
* platform/mac/fast/forms/radio_checked_dynamic-expected.txt:
* platform/mac/fast/inline/positionedLifetime-expected.txt:
* platform/mac/fast/lists/008-expected.txt:
* platform/mac/fast/lists/008-vertical-expected.txt:
* platform/mac/fast/media/mq-relative-constraints-08-expected.txt:
* platform/mac/fast/overflow/overflow-rtl-expected.txt:
* platform/mac/fast/overflow/overflow-rtl-vertical-expected.txt:
* platform/mac/fast/parser/bad-xml-slash-expected.txt:
* platform/mac/fast/replaced/replaced-breaking-expected.txt:
* platform/mac/fast/replaced/width100percent-checkbox-expected.txt:
* platform/mac/fast/replaced/width100percent-radio-expected.txt:
* platform/mac/fast/text/textIteratorNilRenderer-expected.txt:
* platform/mac/fast/text/whitespace/normal-after-nowrap-breaking-expected.txt:
* platform/mac/fast/text/whitespace/pre-wrap-spaces-after-newline-expected.txt:
* platform/mac/http/tests/navigation/javascriptlink-frames-expected.txt:
* platform/mac/svg/custom/alignment-baseline-modes-expected.txt:
* platform/mac/svg/custom/dominant-baseline-modes-expected.txt:
* platform/mac/tables/mozilla/bugs/bug1318-expected.txt:
* platform/mac/tables/mozilla/bugs/bug4527-expected.txt:
* platform/mac/tables/mozilla_expected_failures/core/captions1-expected.txt:
* platform/mac/tables/mozilla_expected_failures/core/captions2-expected.txt:
* svg/custom/invalid-text-content-expected.png:
* svg/custom/text-zoom-expected.png:
* svg/custom/use-clipped-transform-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSkip a test that sometimes times out in Windows Debug builds
aroben@apple.com [Thu, 10 Mar 2011 21:17:53 +0000 (21:17 +0000)]
Skip a test that sometimes times out in Windows Debug builds

<http://webkit.org/b/56080> tracks the failure.

* platform/win-xp/Skipped: Added jquery/effects.html. (Ideally we'd only skip this in Debug
builds regardless of whether it's XP or Vista or 7, but doing that requires fixing
<http://webkit.org/b/55254>.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebKit2: Need a way to clear only the in-memory resource caches
jberlin@webkit.org [Thu, 10 Mar 2011 21:17:46 +0000 (21:17 +0000)]
WebKit2: Need a way to clear only the in-memory resource caches
https://bugs.webkit.org/show_bug.cgi?id=56022

Reviewed by Oliver Hunt and Brian Weinstein.

Add a parameter to WKContextClearResourceCaches to specify whether to clear all the caches
or just the memory caches.

* Shared/ResourceCachesToClear.h: Added.
* UIProcess/API/C/WKAPICast.h:
(WebKit::toResourceCachesToClear):

* UIProcess/API/C/WKContext.cpp:
(WKContextClearResourceCaches):
Add the new parameter.
* UIProcess/API/C/WKContext.h:
* UIProcess/WebContext.cpp:
(WebKit::WebContext::clearResourceCaches):
Ditto.
* UIProcess/WebContext.h:

* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::clearResourceCaches):
Pass the new parameter through to the platform-specific functions.
* WebProcess/WebProcess.h:
Add the new parameter and give it a default of AllResourceCaches.
* WebProcess/WebProcess.messages.in:
Ditto.
* WebProcess/gtk/WebProcessGtk.cpp:
(WebKit::WebProcess::platformClearResourceCaches):
Ditto.
* WebProcess/qt/WebProcessQt.cpp:
(WebKit::WebProcess::platformClearResourceCaches):
Ditto.
* WebProcess/win/WebProcessWin.cpp:
(WebKit::WebProcess::platformClearResourceCaches):
When told to clear only the in-memory caches, return early so as not to clear the CFNetwork
disk cache.
* WebProcess/mac/WebProcessMac.mm:
(WebKit::WebProcess::platformClearResourceCaches):
Ditto.

* WebKit2.xcodeproj/project.pbxproj:
Add ResourceCachesToClear.h
* win/WebKit2.vcproj:
Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Thu, 10 Mar 2011 21:15:45 +0000 (21:15 +0000)]
2011-03-10  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        VisiblePosition's next and previous should take an enum instead of a boolean
        https://bugs.webkit.org/show_bug.cgi?id=56135

        SelectionController::isAll now takes EditingBoundaryCrossingRule instead of
        StayInEditableContent.

        * WebView/WebView.mm:
        (-[WebView _selectionIsAll]):
2011-03-10  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        VisiblePosition's next and previous should take an enum instead of a boolean
        https://bugs.webkit.org/show_bug.cgi?id=56135

        Changed the argument type of VisiblePosition::next and VisiblePosition::previous
        from bool to EditingBoundaryCrossingRule. Also got rid of StayInEditableContent enum
        in VisibleSelection and replaced it by EditingBoundaryCrossingRule because the enum
        was only used in VisiblePosition::isAll and subsequently in SelectionController::isAll
        to call VisiblePosition::next and VisiblePosition::previous.

        * WebCore.exp.in:
        * dom/Position.cpp:
        (WebCore::Position::trailingWhitespacePosition):
        * editing/ApplyBlockElementCommand.cpp:
        (WebCore::ApplyBlockElementCommand::doApply):
        * editing/CompositeEditCommand.cpp:
        (WebCore::CompositeEditCommand::breakOutOfEmptyMailBlockquotedParagraph):
        * editing/InsertListCommand.cpp:
        (WebCore::InsertListCommand::doApply):
        (WebCore::InsertListCommand::listifyParagraph):
        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::shouldMergeStart):
        (WebCore::ReplaceSelectionCommand::shouldMergeEnd):
        (WebCore::ReplaceSelectionCommand::doApply):
        * editing/SelectionController.cpp:
        (WebCore::SelectionController::modifyExtendingRight):
        (WebCore::SelectionController::modifyExtendingForward):
        (WebCore::SelectionController::modifyMovingForward):
        (WebCore::SelectionController::modifyExtendingLeft):
        (WebCore::SelectionController::modifyExtendingBackward):
        (WebCore::SelectionController::modifyMovingBackward):
        * editing/SelectionController.h:
        (WebCore::SelectionController::isAll):
        * editing/TypingCommand.cpp:
        (WebCore::TypingCommand::deleteKeyPressed):
        (WebCore::TypingCommand::forwardDeleteKeyPressed):
        * editing/VisiblePosition.cpp:
        (WebCore::VisiblePosition::next):
        (WebCore::VisiblePosition::previous):
        * editing/VisiblePosition.h:
        * editing/VisibleSelection.cpp:
        (WebCore::VisibleSelection::isAll):
        (WebCore::VisibleSelection::setStartAndEndFromBaseAndExtentRespectingGranularity):
        * editing/VisibleSelection.h:
        * editing/htmlediting.cpp:
        (WebCore::selectionForParagraphIteration):
        * editing/visible_units.cpp:
        (WebCore::startOfNextParagraph):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Thu, 10 Mar 2011 21:12:54 +0000 (21:12 +0000)]
2011-03-10  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        Fix allocation of native function with a cached thunk
        https://bugs.webkit.org/show_bug.cgi?id=56127

        Fix this race condition found while fixing zombies.

        * collector/handles/HandleHeap.cpp:
        (JSC::HandleHeap::clearWeakPointers):
        * runtime/Heap.cpp:
        (JSC::Heap::reset):
        * runtime/JSFunction.cpp:
        (JSC::JSFunction::JSFunction):
        (JSC::JSFunction::markChildren):
        * runtime/JSValue.h:
        (JSC::JSValue::decode):
        * runtime/JSZombie.cpp:
        (JSC::JSZombie::leakedZombieStructure):
        * runtime/JSZombie.h:
        (JSC::JSZombie::createStructure):
        * runtime/MarkedBlock.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Thu, 10 Mar 2011 21:08:27 +0000 (21:08 +0000)]
2011-03-10  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Tony Gentilcore.

        Remove CRASH() calls added to track down bug 53045
        https://bugs.webkit.org/show_bug.cgi?id=56137

        Remove CRASH() calls added by r76575 and re-label ones added by r80155
        and r80269 as being associated with bug 56124 (which may still be
        happening).

        * css/CSSImageValue.cpp:
        (WebCore::CSSImageValue::cachedImage):
        * css/CSSSelector.h:
        * css/CSSSelectorList.cpp:
        (WebCore::CSSSelectorList::deleteSelectors):
        * loader/cache/CachedResource.cpp:
        (WebCore::CachedResource::CachedResource):
        (WebCore::CachedResource::~CachedResource):
        * loader/cache/CachedResource.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Tony Gentilcore <tonyg@chromium.org>
tonyg@chromium.org [Thu, 10 Mar 2011 20:57:15 +0000 (20:57 +0000)]
2011-03-10  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Eric Seidel.

        Let the parser yield for layout before running scripts
        https://bugs.webkit.org/show_bug.cgi?id=54355

        Prior to this patch, the parser would yield to perform a layout/paint before running a
        script only if the script or a stylesheet blocking the script is not loaded yet. Since we
        don't preload scan into the body while parsing the head, typically we'll block on a script
        early in the body that causes us to yield to do the first paint within a reasonable time.

        However, I'm planning to change the PreloadScanner to scan into the body from the head.
        That significantly improves overall load time, but would hurt first paint time because
        fewer scripts would be blocked during parsing and thus wouldn't yield.

        This change causes us to yield before running scripts if we haven't painted yet (regardless
        of whether or not the script is loaded). In addition to allowing the above mentioned
        PreloadScanner change to be implemented without regressing first paint time, this also
        improves first paint time by itself.

        I tested Alexa's top 45 websites using Web Page Replay to control the content and simulate
        bandwidth. This patch improved average first paint time by 1% over an unlimited connection,
        6% over a 1Mbps connection and 11% over a 5Mbps connection. There was no statistically
        signifcant change in page load time.

        Within the pages tested, 33 had no statistically significant change in time to first paint,
        12 improved, and none regressed. Of the improved, some of the standouts from the 1Mbps set
        are: 20% on youtube, 37% on wiki, 27% on ebay, 13% on cnn, 16% on espn, 74% on sohu.

        * html/parser/HTMLDocumentParser.cpp:
        (WebCore::HTMLDocumentParser::canTakeNextToken): This is the new yield point.
        (WebCore::HTMLDocumentParser::pumpTokenizer): Remove ASSERT that we are not paused. isPaused
        means that we are waiting for a script. Bug 54574 changed pumpTokenizer() so that it does
        the right thing whether we are just before a token or waiting for a script. Now that we may
        yield before a token or before a script, this may be called while paused.
        * html/parser/HTMLParserScheduler.cpp:
        (WebCore::HTMLParserScheduler::checkForYieldBeforeScript): Added.
        * page/FrameView.h:
        (WebCore::FrameView::hasEverPainted): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoIgnore errors from files being held open when creating a test result archive
aroben@apple.com [Thu, 10 Mar 2011 20:53:30 +0000 (20:53 +0000)]
Ignore errors from files being held open when creating a test result archive

Fixes <http://webkit.org/b/55581> Test results don't get uploaded if test-result-archive
fails to delete the results directory

Reviewed by Mark Rowe.

* BuildSlaveSupport/test-result-archive:
(archiveTestResults): Ignore errno 2 when deleting the results directory, since it can be
thrown when a process is holding a file open.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Thu, 10 Mar 2011 20:41:43 +0000 (20:41 +0000)]
2011-03-10  Martin Robinson  <mrobinson@igalia.com>

        Fixed the help message for WebGL support on GTK+ to reflect the fact that
        it's turned off by default and is experimental.

        * configure.ac: Fix WebGL help message.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] fast/workers/stress-js-execution.html is crashing on Qt bot (intermittently)
luiz@webkit.org [Thu, 10 Mar 2011 20:40:07 +0000 (20:40 +0000)]
[Qt] fast/workers/stress-js-execution.html is crashing on Qt bot (intermittently)
https://bugs.webkit.org/show_bug.cgi?id=33008

Reviewed by Andreas Kling.

Source/JavaScriptCore:

Defining WTF_USE_PTHREAD_BASED_QT=1 for platforms where QThread uses pthread internally.
Symbian is excluded because pthread_kill does not work on it. Mac is excluded because
it has its own ways to do JSC threading.

Defining WTF_USE_PTHREADS inside MachineStackMarker.cpp if USE(PTHREAD_BASED_QT) is true.

* runtime/MachineStackMarker.cpp:
* wtf/Platform.h:

LayoutTests:

Removing several tests from the Qt skip list. Those tests were flaky because of race conditions
related to web workers.

* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Jeff Miller <jeffm@apple.com>
jeffm@apple.com [Thu, 10 Mar 2011 20:39:05 +0000 (20:39 +0000)]
2011-03-10  Jeff Miller  <jeffm@apple.com>

        Reviewed by Adam Roben.

        WKViewRegisterEditCommandCallback should use WKViewUndoType
        https://bugs.webkit.org/show_bug.cgi?id=56120

        * UIProcess/API/C/win/WKView.h: Change undoOrRedo parameter in WKViewRegisterEditCommandCallback to be WKViewUndoType.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Alejandro G. Castro <alex@igalia.com>
alex@webkit.org [Thu, 10 Mar 2011 20:32:48 +0000 (20:32 +0000)]
2011-03-10  Alejandro G. Castro  <alex@igalia.com>

        Reviewed by Martin Robinson.

        Some Gtk code uses defined(USE_FREETYPE) instead of just USE(FREETYPE)
        https://bugs.webkit.org/show_bug.cgi?id=55996

        Use the macre USE instead of defined fro WTF_USE_FREETYPE and
        WTF_USE_PANGO.

        * GNUmakefile.am:
        * platform/graphics/cairo/OwnPtrCairo.cpp:
        * platform/graphics/cairo/OwnPtrCairo.h:
        * platform/graphics/cairo/RefPtrCairo.cpp:
        * platform/graphics/cairo/RefPtrCairo.h:
        * platform/graphics/gtk/FontGtk.cpp:
        (WebCore::setPangoAttributes):
        (WebCore::Font::drawComplexText):
        (WebCore::Font::floatWidthForComplexText):
        (WebCore::Font::offsetForPositionForComplexText):
        (WebCore::Font::selectionRectForComplexText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoBug 56077 - ES5 conformance issues with RegExp.prototype
barraclough@apple.com [Thu, 10 Mar 2011 20:31:21 +0000 (20:31 +0000)]
Bug 56077 - ES5 conformance issues with RegExp.prototype

Reviewed by Oliver Hunt.

There are three issues causing test failures in sputnik.

(1) lastIndex should be converted at the point it is used, not the point it is set (this is visible if valueOf is overridden).
(2) The 'length' property of the test/exec functions should be 1.
(3) If no input is specified, the input to test()/exec() is "undefined" (i.e. ToString(undefined)) - not RegExp.input.

Source/JavaScriptCore:

* runtime/RegExpObject.cpp:
(JSC::RegExpObject::markChildren):
    - Added to mark lastIndex
(JSC::regExpObjectLastIndex):
(JSC::setRegExpObjectLastIndex):
    - lastIndex is now stored as a JSValue.
(JSC::RegExpObject::match):
    - Use accessor methods to get/set lastIndex, add fast case for isUInt32 (don't convert to double).
* runtime/RegExpObject.h:
(JSC::RegExpObject::setLastIndex):
(JSC::RegExpObject::setLastIndex):
    - Set lastIndex, either from a size_t or a JSValue.
(JSC::RegExpObject::getLastIndex):
    - Get lastIndex.
(JSC::RegExpObject::RegExpObjectData::RegExpObjectData):
    - Initialize as a JSValue.
* runtime/RegExpPrototype.cpp:
(JSC::RegExpPrototype::RegExpPrototype):
    - Add test/exec properties with length 1.
* runtime/StringPrototype.cpp:
(JSC::stringProtoFuncMatch):
(JSC::stringProtoFuncSearch):
    - Do not read RegExp.input if none is provided.
* tests/mozilla/js1_2/regexp/RegExp_input.js:
* tests/mozilla/js1_2/regexp/RegExp_input_as_array.js:
    - Update these tests (they relied on non-ES5 behaviour).

LayoutTests:

* fast/js/kde/function_length-expected.txt:
* fast/js/kde/script-tests/function_length.js:
    - 'length' of RexExp.test/RexExp.exec is 1
* fast/js/regexp-caching-expected.txt:
* fast/js/regexp-test-null-string-expected.txt:
* fast/js/regexp-test-null-string.html:
    - test() is equivalent to test("undefined")
* sputnik/Conformance/15_Native_Objects/15.10_RegExp/15.10.6/15.10.6.2_RegExp.prototype.exec/S15.10.6.2_A11-expected.txt:
* sputnik/Conformance/15_Native_Objects/15.10_RegExp/15.10.6/15.10.6.2_RegExp.prototype.exec/S15.10.6.2_A1_T16-expected.txt:
* sputnik/Conformance/15_Native_Objects/15.10_RegExp/15.10.6/15.10.6.2_RegExp.prototype.exec/S15.10.6.2_A4_T11-expected.txt:
* sputnik/Conformance/15_Native_Objects/15.10_RegExp/15.10.6/15.10.6.3_RegExp.prototype.test/S15.10.6.3_A11-expected.txt:
* sputnik/Conformance/15_Native_Objects/15.10_RegExp/15.10.6/15.10.6.3_RegExp.prototype.test/S15.10.6.3_A1_T16-expected.txt:
    - These tests now pass!

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Thu, 10 Mar 2011 20:25:42 +0000 (20:25 +0000)]
2011-03-10  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Rolled back in 80277 and 80280 with event handler layout test failures fixed.
        https://bugs.webkit.org/show_bug.cgi?id=55653

        The failures were caused by a last minute typo: assigning to currentEvent
        instead of m_currentEvent.

        * JavaScriptCore.exp:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * JavaScriptCore.xcodeproj/project.pbxproj:
        * bytecompiler/BytecodeGenerator.cpp:
        * jit/JITOpcodes.cpp:
        * jit/JITOpcodes32_64.cpp:
        * runtime/Arguments.h:
        * runtime/JSActivation.cpp:
        * runtime/JSActivation.h:
        * runtime/JSCell.h:
        * runtime/JSGlobalObject.cpp:
        * runtime/JSGlobalObject.h:
        * runtime/JSObject.cpp:
        * runtime/JSStaticScopeObject.cpp:
        * runtime/JSStaticScopeObject.h:
        * runtime/JSVariableObject.h:
        * runtime/MarkedSpace.cpp:
        * runtime/MarkedSpace.h:
2011-03-10  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Rolled back in 80277 and 80280 with event handler layout test failures fixed.
        https://bugs.webkit.org/show_bug.cgi?id=55653

        The failures were caused by a last minute typo: assigning to currentEvent
        instead of m_currentEvent.

        * JSRun.cpp:
        * JSRun.h:
2011-03-10  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Rolled back in 80277 and 80280 with event handler layout test failures fixed.
        https://bugs.webkit.org/show_bug.cgi?id=55653

        The failures were caused by a last minute typo: assigning to currentEvent
        instead of m_currentEvent.

        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/JSDOMGlobalObject.cpp:
        * bindings/js/JSDOMGlobalObject.h:
        * bindings/js/JSDOMWindowBase.cpp:
        * bindings/js/JSDOMWindowBase.h:
        * bindings/js/JSDOMWindowCustom.h:
        * bindings/js/JSWorkerContextBase.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoDon't search for intra-line diffs in really long lines
aroben@apple.com [Thu, 10 Mar 2011 20:22:53 +0000 (20:22 +0000)]
Don't search for intra-line diffs in really long lines

Doing so can lead to hangs (or at least really slow execution).

Fixes <http://webkit.org/b/56109> run-webkit-tests sometimes times out on Windows XP Debug
(Tests) after fast/text/large-text-composed-char-dos.html fails (due to PrettyPatch
hanging?)

Reviewed by David Kilzer.

* PrettyPatch/PrettyPatch.rb:
(PrettyPatch.MAXIMUM_INTRALINE_DIFF_LINE_LENGTH): Added this new constant.
(PrettyPatch.DiffSection.initialize): Don't bother looking for intra-line diffs in lines
longer than the maximum length.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=56129, vertical text broken on Lion and Leopard.
hyatt@apple.com [Thu, 10 Mar 2011 20:15:39 +0000 (20:15 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=56129, vertical text broken on Lion and Leopard.

Reviewed by Dan Bernstein.

Add Snow Leopard ifdefs for the scaling by point size and then the division by unitsPerEm to the
translationsTransform applied to the results from CTFontGetVerticalTranslationsForGlyphs, since
this is done already on Lion and Leopard.

* platform/graphics/mac/FontMac.mm:
(WebCore::showGlyphsWithAdvances):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Sheriff Bot <webkit.review.bot@gmail.com>
mihaip@chromium.org [Thu, 10 Mar 2011 19:33:48 +0000 (19:33 +0000)]
2011-03-10  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r80701.
        http://trac.webkit.org/changeset/80701
        https://bugs.webkit.org/show_bug.cgi?id=56126

        fast/dom/Geolocation/window-close-crash.html fails on Snow
        Leopard release builds (Requested by mihaip on #webkit).

        * fast/dom/Geolocation/page-reload-cancel-permission-requests-expected.txt: Removed.
        * fast/dom/Geolocation/page-reload-cancel-permission-requests.html: Removed.
        * fast/dom/Geolocation/resources/page-reload-cancel-permission-requests-inner.html: Removed.
        * fast/dom/Geolocation/script-tests/page-reload-cancel-permission-requests.js: Removed.
        * platform/gtk/Skipped:
        * platform/mac/Skipped:
        * platform/qt-wk2/Skipped:
2011-03-10  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r80701.
        http://trac.webkit.org/changeset/80701
        https://bugs.webkit.org/show_bug.cgi?id=56126

        fast/dom/Geolocation/window-close-crash.html fails on Snow
        Leopard release builds (Requested by mihaip on #webkit).

        * page/DOMWindow.cpp:
        * page/DOMWindow.h:
        * page/Frame.cpp:
        (WebCore::Frame::pageDestroyed):
        (WebCore::Frame::transferChildFrameToNewDocument):
        * page/Geolocation.cpp:
        (WebCore::Geolocation::~Geolocation):
        (WebCore::Geolocation::disconnectFrame):
        (WebCore::Geolocation::lastPosition):
        (WebCore::Geolocation::requestPermission):
        (WebCore::Geolocation::startUpdating):
        (WebCore::Geolocation::stopUpdating):
        * page/Geolocation.h:
        * page/GeolocationController.cpp:
        (WebCore::GeolocationController::~GeolocationController):
        * page/Navigator.cpp:
        * page/Navigator.h:
        * platform/mock/GeolocationClientMock.cpp:
        * platform/mock/GeolocationClientMock.h:
2011-03-10  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r80701.
        http://trac.webkit.org/changeset/80701
        https://bugs.webkit.org/show_bug.cgi?id=56126

        fast/dom/Geolocation/window-close-crash.html fails on Snow
        Leopard release builds (Requested by mihaip on #webkit).

        * public/WebGeolocationClientMock.h:
        * src/WebGeolocationClientMock.cpp:
2011-03-10  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r80701.
        http://trac.webkit.org/changeset/80701
        https://bugs.webkit.org/show_bug.cgi?id=56126

        fast/dom/Geolocation/window-close-crash.html fails on Snow
        Leopard release builds (Requested by mihaip on #webkit).

        * DumpRenderTree/LayoutTestController.cpp:
        (LayoutTestController::staticFunctions):
        * DumpRenderTree/LayoutTestController.h:
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::LayoutTestController):
        * DumpRenderTree/chromium/LayoutTestController.h:
        * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
        * DumpRenderTree/mac/LayoutTestControllerMac.mm:
        * DumpRenderTree/mac/UIDelegate.h:
        * DumpRenderTree/mac/UIDelegate.mm:
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        * DumpRenderTree/qt/LayoutTestControllerQt.h:
        * DumpRenderTree/win/LayoutTestControllerWin.cpp:
        * DumpRenderTree/wx/LayoutTestControllerWx.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Thu, 10 Mar 2011 19:27:49 +0000 (19:27 +0000)]
2011-03-10  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        jquery/manipulation.html fails after r80598
        https://bugs.webkit.org/show_bug.cgi?id=56019

        Removed jQuery tests from skiplists

        * platform/gtk/Skipped:
        * platform/qt/Skipped:
2011-03-09  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        jquery/manipulation.html fails after r80598
        https://bugs.webkit.org/show_bug.cgi?id=56019

        When linking a call, codeblock now takes ownership of the linked function
        This removes the need for unlinking, and thus the incorrectness that was
        showing up in these tests.

        * bytecode/CodeBlock.cpp:
        (JSC::CodeBlock::~CodeBlock):
        (JSC::CodeBlock::markAggregate):
        * bytecode/CodeBlock.h:
        (JSC::CallLinkInfo::CallLinkInfo):
        (JSC::CallLinkInfo::setUnlinked):
        (JSC::CodeBlock::addCaller):
        * jit/JIT.cpp:
        (JSC::JIT::privateCompile):
        (JSC::JIT::linkCall):
        (JSC::JIT::linkConstruct):
        * jit/JIT.h:
        * runtime/Executable.cpp:
        * runtime/Executable.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Thu, 10 Mar 2011 19:22:42 +0000 (19:22 +0000)]
2011-03-10  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Ojan Vafai.

        Unable to paste in input[type="text"] with last-child selector
        https://bugs.webkit.org/show_bug.cgi?id=50736

        Added regressions tests since this bug no longer reproduces on TOT WebKit.

        * editing/input/div-first-child-rule-input-expected.txt: Added.
        * editing/input/div-first-child-rule-input.html: Added.
        * editing/input/div-first-child-rule-textarea-expected.txt: Added.
        * editing/input/div-first-child-rule-textarea.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Thu, 10 Mar 2011 18:44:18 +0000 (18:44 +0000)]
2011-03-10  Martin Robinson  <mrobinson@igalia.com>

        [GTK] plugins/return-negative-one-from-write.html crashes
        https://bugs.webkit.org/show_bug.cgi?id=41903

        * platform/gtk/Skipped: Unskipping test which appears to be passing locally.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Thu, 10 Mar 2011 18:20:36 +0000 (18:20 +0000)]
2011-03-10  Adam Barth  <abarth@webkit.org>

        Update test expectation after my partial revert of r80670.

        * platform/mac/accessibility/aria-expanded-notifications-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Pratik Solanki <psolanki@apple.com>
psolanki@apple.com [Thu, 10 Mar 2011 18:14:44 +0000 (18:14 +0000)]
2011-03-10  Pratik Solanki  <psolanki@apple.com>

        Reviewed by Alexey Proskuryakov.

        ASSERT_NOT_REACHED triggered in WebCore::mapHTTPPipeliningPriorityToResourceLoadPriority
        https://bugs.webkit.org/show_bug.cgi?id=56075

        Map priority value 3, which means no priority, to ResourceLoadPriorityUnresolved.

        * platform/network/cf/ResourceRequestCFNet.h:
        (WebCore::mapHTTPPipeliningPriorityToResourceLoadPriority):
        (WebCore::mapResourceLoadPriorityToHTTPPipeliningPriority):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrei Popescu <andreip@google.com>
andreip@google.com [Thu, 10 Mar 2011 17:52:25 +0000 (17:52 +0000)]
2011-03-10  Andrei Popescu  <andreip@google.com>

        Unreviewed, build fix.

        [Chromium] fast/canvas/webgl/uninitialized-test.html times out on all platforms.
        https://bugs.webkit.org/show_bug.cgi?id=56113

        It seems this was introduced in http://src.chromium.org/viewvc/chrome?view=rev&revision=77637
        Updating test expectations for now.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrei Popescu <andreip@google.com>
andreip@google.com [Thu, 10 Mar 2011 17:06:39 +0000 (17:06 +0000)]
2011-03-10  Andrei Popescu  <andreip@google.com>

        Unreviewed, build fix.

        [Chromium] svg/W3C-SVG-1.1/text-text-03-b.svg is sometimes crashing on Linux
        https://bugs.webkit.org/show_bug.cgi?id=56110

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd Windows expected failure results for fast/forms/onchange-setvalueforuser.html
aroben@apple.com [Thu, 10 Mar 2011 17:05:44 +0000 (17:05 +0000)]
Add Windows expected failure results for fast/forms/onchange-setvalueforuser.html

The failure is tracked by <http://webkit.org/b/55834>.

* platform/win/fast/forms/onchange-setvalueforuser-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd Windows expected failure results for fast/lists/003-vertical.html
aroben@apple.com [Thu, 10 Mar 2011 17:03:55 +0000 (17:03 +0000)]
Add Windows expected failure results for fast/lists/003-vertical.html

The Windows results match the pre-r80648 Mac results. The failure is tracked by
<http://webkit.org/b/56101>.

* platform/win/fast/lists/003-vertical-expected.checksum: Added.
* platform/win/fast/lists/003-vertical-expected.png: Added.
* platform/win/fast/lists/003-vertical-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrey Adaikin <aandrey@google.com>
podivilov@chromium.org [Thu, 10 Mar 2011 16:46:02 +0000 (16:46 +0000)]
2011-03-10  Andrey Adaikin  <aandrey@google.com>

        Reviewed by Pavel Feldman.

        Web Inspector: [Text editor] Substitute live-edit mode activation by double-click
        https://bugs.webkit.org/show_bug.cgi?id=56084

        Also fixed a regression: console was not opening/closing on ESC key press because tabIndex="0" attribute was preserved for a readOnly viewer.

        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame):
        (WebInspector.SourceFrame.prototype._startEditing):
        (WebInspector.SourceFrame.prototype._registerShortcuts):
        (WebInspector.SourceFrame.prototype._handleKeyDown):
        (WebInspector.SourceFrame.prototype._handleSave):
        (WebInspector.SourceFrame.prototype._handleRevertEditing):
        (WebInspector.SourceFrame.prototype._doubleClick):
        * inspector/front-end/TextEditorModel.js:
        (WebInspector.TextEditorModel.prototype.get text):
        * inspector/front-end/TextViewer.js:
        (WebInspector.TextViewer.prototype.get readOnly):
        (WebInspector.TextEditorMainPanel):
        (WebInspector.TextEditorMainPanel.prototype.set readOnly):
        (WebInspector.TextEditorMainPanel.prototype.get readOnly):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrei Popescu <andreip@google.com>
andreip@google.com [Thu, 10 Mar 2011 16:43:59 +0000 (16:43 +0000)]
2011-03-10  Andrei Popescu  <andreip@google.com>

        Unreviewed, build fix.

        [Chromium] fast/text/international/vertical-text-glyph-test.html fails on OSX 10.6
        https://bugs.webkit.org/show_bug.cgi?id=56108

        Update Chromium test expectations.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSkip another spinbutton test on Windows
aroben@apple.com [Thu, 10 Mar 2011 16:42:21 +0000 (16:42 +0000)]
Skip another spinbutton test on Windows

The missing functionality is tracked by <http://webkit.org/b/38381>.

* platform/win/Skipped: Added fast/forms/input-appearance-spinbutton-visibility.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd Windows expected failure results for fast/text/hyphenate-limit-before-after.html
aroben@apple.com [Thu, 10 Mar 2011 16:40:37 +0000 (16:40 +0000)]
Add Windows expected failure results for fast/text/hyphenate-limit-before-after.html

The failure is tracked by <http://webkit.org/b/56102>.

* platform/win/fast/text/hyphenate-limit-before-after-expected.txt: Added. Copied this from
mac-snowleopard.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd Windows expected failure results for fast/blockflow/text-orientation-basic.html
aroben@apple.com [Thu, 10 Mar 2011 16:38:34 +0000 (16:38 +0000)]
Add Windows expected failure results for fast/blockflow/text-orientation-basic.html

The failure is tracked by <http://webkit.org/b/56101>.

* platform/win/fast/blockflow/text-orientation-basic-expected.checksum: Added.
* platform/win/fast/blockflow/text-orientation-basic-expected.png: Added.
* platform/win/fast/blockflow/text-orientation-basic-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Greg Simon <gregsimon@chromium.org>
pfeldman@chromium.org [Thu, 10 Mar 2011 16:36:53 +0000 (16:36 +0000)]
2011-03-10  Greg Simon  <gregsimon@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Need new graphic icon for garbage collect button.
        https://bugs.webkit.org/show_bug.cgi?id=55794

        No new tests: gc tests are flaky due to non-determinisic
        behavior of collection APIs (more notes in bug)

        * English.lproj/localizedStrings.js:
        * WebCore.gypi:
        * bindings/js/ScriptProfiler.cpp:
        (WebCore::ScriptProfiler::collectGarbage):
        * bindings/js/ScriptProfiler.h:
        * bindings/v8/ScriptProfiler.cpp:
        (WebCore::ScriptProfiler::collectGarbage):
        * bindings/v8/ScriptProfiler.h:
        * inspector/Inspector.idl:
        * inspector/InspectorProfilerAgent.cpp:
        (WebCore::InspectorProfilerAgent::collectGarbage):
        * inspector/InspectorProfilerAgent.h:
        * inspector/front-end/Images/garbageCollectButtonGlyph.png: Added.
        * inspector/front-end/TimelinePanel.js:
        (WebInspector.TimelinePanel.prototype.get statusBarItems):
        (WebInspector.TimelinePanel.prototype._createStatusbarButtons):
        (WebInspector.TimelinePanel.prototype._garbageCollectButtonClicked):
        * inspector/front-end/inspector.css:
        (.garbage-collect-status-bar-item .glyph):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrei Popescu <andreip@google.com>
andreip@google.com [Thu, 10 Mar 2011 16:32:57 +0000 (16:32 +0000)]
2011-03-10  Andrei Popescu  <andreip@google.com>

        Unreviewed, build fix.

        [Chromium] fast/text/justify-ideograph-vertical.html fails on OSX 10.5
        https://bugs.webkit.org/show_bug.cgi?id=56106

        Update Chromium test expectations.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrey Kosyakov <caseq@chromium.org>
caseq@chromium.org [Thu, 10 Mar 2011 16:14:53 +0000 (16:14 +0000)]
2011-03-10  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: better names for HAR export commands
        https://bugs.webkit.org/show_bug.cgi?id=56097

        Rename Export to HAR to Copy entry/network log as HAR
        Expose Copy as HAR unconditionally, remove related settings entry.

        * English.lproj/localizedStrings.js:
        * inspector/front-end/NetworkPanel.js:
        (WebInspector.NetworkPanel.prototype._contextMenu):
        * inspector/front-end/Settings.js:

2011-03-10  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: better names for HAR export commands
        https://bugs.webkit.org/show_bug.cgi?id=56097

        * src/js/DevTools.js: removed override for resourceExportEnabled

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Qi Zhang <qi.2.zhang@nokia.com>
laszlo.1.gombos@nokia.com [Thu, 10 Mar 2011 15:57:37 +0000 (15:57 +0000)]
2011-03-10  Qi Zhang  <qi.2.zhang@nokia.com>

        Reviewed by Laszlo Gombos.

        [Qt] Mobile Devices should include Model and Firmware Version in Webkit Generated User Agent String
        https://bugs.webkit.org/show_bug.cgi?id=48636

        Fix a typo in features.pri that turns on this feature on mobile
        devices by default.

        * features.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Thu, 10 Mar 2011 15:38:50 +0000 (15:38 +0000)]
2011-03-10  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Invalid expected parameter list in CSSAgent.setSelectorText() callback (CSSStyleModel.js)
        https://bugs.webkit.org/show_bug.cgi?id=56092

        * inspector/front-end/CSSStyleModel.js:
        (WebInspector.CSSStyleModel.prototype.setRuleSelector):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 10 Mar 2011 15:17:57 +0000 (15:17 +0000)]
2011-03-10  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: rolling out r80478 and its follow ups for breaking
        inspector and engadget.com.
        https://bugs.webkit.org/show_bug.cgi?id=49401

        * dom/Document.cpp:
        (WebCore::Document::Document):
        (WebCore::performTask):
        (WebCore::Document::postTask):
        * dom/Document.h:
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::setDefersLoading):
        * manual-tests/database-callback-deferred.html: Removed.
        * page/PageGroupLoadDeferrer.cpp:
        (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
        (WebCore::PageGroupLoadDeferrer::~PageGroupLoadDeferrer):

2011-03-10  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: rolling out r80478 and its follow ups for breaking
        inspector and engadget.com.
        https://bugs.webkit.org/show_bug.cgi?id=49401

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 15:11:05 +0000 (15:11 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip 1 more svg test failing by one pixel in 32-bits
        Debug, see bug 39022.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Thu, 10 Mar 2011 14:52:44 +0000 (14:52 +0000)]
2011-03-10  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Kenneth Rohde Christiansen.

        Tiled backing store's delegated scroll request uses incorrect convention
        https://bugs.webkit.org/show_bug.cgi?id=56011

        Use a point instead of delta, when relaying the scroll request from
        ScrollView::setScrollPosition.

        * loader/EmptyClients.h:
        (WebCore::EmptyChromeClient::delegatedScrollRequested):
        * page/Chrome.cpp:
        (WebCore::Chrome::delegatedScrollRequested):
        * page/Chrome.h:
        * page/ChromeClient.h:
        * platform/HostWindow.h:
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::setScrollPosition):
2011-03-10  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Kenneth Rohde Christiansen.

        Tiled backing store's delegated scroll request uses incorrect convention
        https://bugs.webkit.org/show_bug.cgi?id=56011

        Adapt internal API to match the change from delta to point on the
        WebCore side, and convert the point to a delta for the public API.

        * WebCoreSupport/ChromeClientQt.cpp:
        (WebCore::ChromeClientQt::delegatedScrollRequested):
        * WebCoreSupport/ChromeClientQt.h:
2011-03-10  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Kenneth Rohde Christiansen.

        Tiled backing store's delegated scroll request uses incorrect convention
        https://bugs.webkit.org/show_bug.cgi?id=56011

        Adapt all of the internal API to match the delta to point
        conversion at the WebCore side.

        * UIProcess/API/qt/qwkpage.cpp:
        (QWKPagePrivate::pageDidRequestScroll):
        * UIProcess/API/qt/qwkpage_p.h:
        * UIProcess/PageClient.h:
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::pageDidRequestScroll):
        * UIProcess/WebPageProxy.h:
        * UIProcess/WebPageProxy.messages.in:
        * WebProcess/WebCoreSupport/WebChromeClient.cpp:
        (WebKit::WebChromeClient::delegatedScrollRequested):
        * WebProcess/WebCoreSupport/WebChromeClient.h:
        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::pageDidRequestScroll):
        * WebProcess/WebPage/WebPage.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 14:51:41 +0000 (14:51 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] unittests/testapplicationcache crashes
        https://bugs.webkit.org/show_bug.cgi?id=56083

        Removed the webkit_application_cache_get_database_directory_path
        API as the underlying ApplicationCacheStorage doesn't allow
        setting the cache path multiple times.

        * tests/testapplicationcache.c:
        (main):
        * webkit/webkitapplicationcache.cpp:
        (webkit_application_cache_get_database_directory_path):
        * webkit/webkitapplicationcache.h:
        * webkit/webkitglobals.cpp:
        (webkitInit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Thu, 10 Mar 2011 14:34:44 +0000 (14:34 +0000)]
2011-03-10  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Andreas Kling.

        [WK2] Make non-transparent windowless plugins paint on X11
        https://bugs.webkit.org/show_bug.cgi?id=55660

        Implement painting for non-transparent windowless X11 plugins.
        The essence of the logic has been adapted from WebCore.
        The implementation is stubbed for GTK and working only on Qt
        but it should be really easy to finish it for GTK.

        * WebKit2.pro:
        * WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
        (WebKit::NPN_GetValue): Handle X11 specific values that are necessary.
        * WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
        (WebKit::NetscapePlugin::NetscapePlugin): Initialize the X11 specific
        members that has been added.
        (WebKit::NetscapePlugin::callSetWindow): Always set the x and y coordinates
        of the NP_Window to 0 on X11 since we are using a backings store as the
        painting area for the plugin.
        * WebProcess/Plugins/Netscape/NetscapePlugin.h:
        * WebProcess/Plugins/Netscape/gtk/NetscapePluginGtk.cpp: Removed in favour
        of a common implementation for X11.
        * WebProcess/Plugins/Netscape/qt/NetscapePluginQt.cpp: Ditto.
        * WebProcess/Plugins/Netscape/x11/NetscapePluginX11.cpp: Added.
        (WebKit::getPluginDisplay):

        Platform specific static getters.
        (WebKit::x11Display):
        (WebKit::displayDepth):
        (WebKit::rootWindowID):
        (WebKit::x11Screen):

        (WebKit::NetscapePlugin::platformPostInitialize): Set up the visual
        settings and the colormap for the plugin.
        (WebKit::NetscapePlugin::platformDestroy):
        (WebKit::NetscapePlugin::platformInvalidate): Remained stub.
        (WebKit::NetscapePlugin::platformGeometryDidChange): Create a new
        pixmap that fits the geometry.
        (WebKit::NetscapePlugin::platformPaint): Propagate a paint event
        to the plugin, sync with it if necessary and draw the pixmap to
        the screen.
        (WebKit::toNP): Remained stub.
        (WebKit::NetscapePlugin::platformHandleMouseEvent): Remained stub.
        (WebKit::NetscapePlugin::platformHandleWheelEvent): Remained stub.
        (WebKit::NetscapePlugin::platformSetFocus): Remained stub.
        (WebKit::NetscapePlugin::platformHandleMouseEnterEvent): Remained stub.
        (WebKit::NetscapePlugin::platformHandleMouseLeaveEvent): Remained stub.
        (WebKit::NetscapePlugin::platformHandleKeyboardEvent): Remained stub.
        * WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
        (WebKit::WebFrameLoaderClient::createPlugin): Hack. Inject wmode=opaque
        key-value pair to the plugin parameters to force Flash to act in
        non-transparent windowless mode. Qt also doing this in WebCore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 10 Mar 2011 14:26:25 +0000 (14:26 +0000)]
2011-03-10  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: brush up DOM agent API.
        https://bugs.webkit.org/show_bug.cgi?id=56093

        * http/tests/inspector/elements-test.js:
        (initialize_ElementTest.InspectorTest.nodeWithId.processChildren):
        (initialize_ElementTest.InspectorTest.nodeWithId.documentRequested):
        (initialize_ElementTest.InspectorTest.nodeWithId):
        (initialize_ElementTest.InspectorTest.dumpDOMAgentTree.dump):
        (initialize_ElementTest.InspectorTest.dumpDOMAgentTree):
        (initialize_ElementTest):
        * inspector/elements/elements-panel-rewrite-href.html:
        * inspector/elements/elements-panel-search.html:
        * inspector/elements/elements-panel-selection-on-refresh.html:
        * inspector/elements/mutate-unknown-node.html:
2011-03-10  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: brush up DOM agent API.
        https://bugs.webkit.org/show_bug.cgi?id=56093

        * inspector/Inspector.idl:
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::childNodes):
        (WebCore::InspectorDOMAgent::setNodeName):
        (WebCore::InspectorDOMAgent::outerHTML):
        (WebCore::InspectorDOMAgent::setNodeValue):
        * inspector/InspectorDOMAgent.h:
        * inspector/front-end/DOMAgent.js:
        (WebInspector.DOMNode):
        (WebInspector.DOMNode.prototype.hasAttributes):
        (WebInspector.DOMNode.prototype.nodeType):
        (WebInspector.DOMNode.prototype.nodeName):
        (WebInspector.DOMNode.prototype.setNodeName):
        (WebInspector.DOMNode.prototype.localName):
        (WebInspector.DOMNode.prototype.nodeValue):
        (WebInspector.DOMNode.prototype.setNodeValue):
        (WebInspector.DOMNode.prototype.setAttribute):
        (WebInspector.DOMNode.prototype.attributes):
        (WebInspector.DOMNode.prototype.removeAttribute):
        (WebInspector.DOMNode.prototype.childNodes.mycallback):
        (WebInspector.DOMNode.prototype.childNodes):
        (WebInspector.DOMNode.prototype.outerHTML):
        (WebInspector.DOMNode.prototype.setOuterHTML):
        (WebInspector.DOMNode.prototype.removeNode):
        (WebInspector.DOMNode.prototype.copyNode):
        (WebInspector.DOMNode.prototype.path):
        (WebInspector.DOMNode.prototype._setAttributesPayload):
        (WebInspector.DOMNode.prototype._addAttribute):
        (WebInspector.DOMAgent.prototype._characterDataModified):
        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel.prototype.updateBreadcrumb):
        (WebInspector.ElementsPanel.prototype.decorateNodeLabel):
        (WebInspector.ElementsPanel.prototype.handleCopyEvent):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeOutline.prototype.findTreeElement):
        (WebInspector.ElementsTreeElement):
        (WebInspector.ElementsTreeElement.prototype.updateChildren):
        (WebInspector.ElementsTreeElement.prototype._updateChildren):
        (WebInspector.ElementsTreeElement.prototype._startEditingTarget):
        (WebInspector.ElementsTreeElement.prototype._startEditing):
        (WebInspector.ElementsTreeElement.prototype._attributeEditingCommitted):
        (WebInspector.ElementsTreeElement.prototype._tagNameEditingCommitted.moveToNextAttributeIfNeeded):
        (WebInspector.ElementsTreeElement.prototype._tagNameEditingCommitted):
        (WebInspector.ElementsTreeElement.prototype._textNodeEditingCommitted):
        (WebInspector.ElementsTreeElement.prototype._attributeHTML):
        ():
        * inspector/front-end/EventListenersSidebarPane.js:
        ():
        * inspector/front-end/MetricsSidebarPane.js:
        * inspector/front-end/StylesSidebarPane.js:
        * inspector/front-end/utilities.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 14:09:02 +0000 (14:09 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Gustavo Noronha Silva.

        [GTK] run-gtk-tests doesn't fail if a test crashes
        https://bugs.webkit.org/show_bug.cgi?id=56089

        * Scripts/run-gtk-tests: gtester -k exits with a 0 exit status
        even if a test crashes. So run the tests individually and if a
        test fails remember its exit code so run-gtk-test fails too, if
        required.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 13:44:04 +0000 (13:44 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip 2 more flacky tests on GTK 32-bits Release.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-09 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Thu, 10 Mar 2011 13:28:32 +0000 (13:28 +0000)]
2011-03-09  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: [STYLES] Editing a property value adds a word for any property value that uses a paren
        https://bugs.webkit.org/show_bug.cgi?id=56002

        * inspector/front-end/StylesSidebarPane.js: Introduce an additional check.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-09 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Thu, 10 Mar 2011 13:05:59 +0000 (13:05 +0000)]
2011-03-09  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: use DebuggerPresentation instead of DebuggerModel in source frame delegate.
        https://bugs.webkit.org/show_bug.cgi?id=56034

        * inspector/front-end/DebuggerPresentationModel.js:
        (WebInspector.DebuggerPresentationModel.prototype.continueToLine):
        (WebInspector.DebuggerPresentationModel.prototype.setBreakpoint):
        (WebInspector.DebuggerPresentationModel.prototype.updateBreakpoint):
        (WebInspector.DebuggerPresentationModel.prototype.findBreakpoint):
        (WebInspector.DebuggerPresentationModel.prototype._sourceLocationToActualLocation):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype._createSourceFrame):
        (WebInspector.SourceFrameDelegateForScriptsPanel):
        (WebInspector.SourceFrameDelegateForScriptsPanel.prototype.setBreakpoint):
        (WebInspector.SourceFrameDelegateForScriptsPanel.prototype.updateBreakpoint):
        (WebInspector.SourceFrameDelegateForScriptsPanel.prototype.removeBreakpoint):
        (WebInspector.SourceFrameDelegateForScriptsPanel.prototype.findBreakpoint):
        (WebInspector.SourceFrameDelegateForScriptsPanel.prototype.continueToLine):
        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype._contextMenu.else.editBreakpointCondition.didEditBreakpointCondition):
        (WebInspector.SourceFrame.prototype._contextMenu.else.editBreakpointCondition):
        (WebInspector.SourceFrame.prototype._contextMenu.else.setBreakpointEnabled):
        (WebInspector.SourceFrame.prototype._contextMenu):
        (WebInspector.SourceFrame.prototype._mouseDown):
        (WebInspector.SourceFrameDelegate.prototype.removeBreakpoint):
        (WebInspector.SourceFrameDelegate.prototype.updateBreakpoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-09 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Thu, 10 Mar 2011 12:42:39 +0000 (12:42 +0000)]
2011-03-09  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: elements dom tree - word wrap toggle
        https://bugs.webkit.org/show_bug.cgi?id=44311

        A "Word Wrap" option is now shown in a context menu for the entire DOM tree content area
        and persisted into application settings.

        * English.lproj/localizedStrings.js:
        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel.prototype._contextMenuEventFired):
        (WebInspector.ElementsPanel.prototype._contextMenuEventFired.toggleWordWrap):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeOutline):
        (WebInspector.ElementsTreeOutline.prototype._treeElementFromEvent):
        (WebInspector.ElementsTreeOutline.prototype.populateContextMenu):
        (WebInspector.ElementsTreeElement.prototype.onreveal):
        * inspector/front-end/Settings.js:
        (WebInspector.Settings):
        * inspector/front-end/inspector.css:
        (#elements-content.nowrap):
        (#elements-content > ol):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 12:28:29 +0000 (12:28 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip 1 more svg test failing by one pixel in 32-bits
        Debug, see bug 39022.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 12:26:34 +0000 (12:26 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip 2 more svg tests failing by one pixel in 32-bits
        Release, see bug 39022.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-09 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Thu, 10 Mar 2011 12:21:22 +0000 (12:21 +0000)]
2011-03-09  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: re-implement breakpoints sidebar pane based on debugger presentation model.
        https://bugs.webkit.org/show_bug.cgi?id=55823

        * inspector/front-end/BreakpointsSidebarPane.js:
        (WebInspector.JavaScriptBreakpointsSidebarPane):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.addBreakpoint.didLoadSnippet):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.addBreakpoint):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.removeBreakpoint):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.highlightBreakpoint):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.clearBreakpointHighlight):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._createBreakpointItemId):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._breakpointClicked):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._breakpointCheckboxClicked):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype._contextMenu):
        (WebInspector.JavaScriptBreakpointsSidebarPane.prototype.reset):
        * inspector/front-end/DebuggerPresentationModel.js:
        (WebInspector.DebuggerPresentationModel):
        (WebInspector.DebuggerPresentationModel.prototype._parsedScriptSource):
        (WebInspector.DebuggerPresentationModel.prototype._failedToParseScriptSource):
        (WebInspector.DebuggerPresentationModel.prototype._revealHiddenBreakpoints):
        (WebInspector.DebuggerPresentationModel.prototype.breakpointsForSourceFileId):
        (WebInspector.DebuggerPresentationModel.prototype.setBreakpointEnabled):
        (WebInspector.DebuggerPresentationModel.prototype.removeBreakpoint):
        (WebInspector.DebuggerPresentationModel.prototype._breakpointAdded):
        (WebInspector.DebuggerPresentationModel.prototype.set selectedCallFrame):
        (WebInspector.DebuggerPresentationModel.prototype._actualLocationToSourceLocation):
        (WebInspector.DebuggerPresentationModel.prototype.reset):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel):
        (WebInspector.ScriptsPanel.prototype._breakpointAdded):
        (WebInspector.ScriptsPanel.prototype._breakpointRemoved):
        (WebInspector.ScriptsPanel.prototype._debuggerPaused):
        (WebInspector.ScriptsPanel.prototype.reset):
        (WebInspector.ScriptsPanel.prototype._clearInterface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Andrey Adaikin <aandrey@google.com>
podivilov@chromium.org [Thu, 10 Mar 2011 12:11:00 +0000 (12:11 +0000)]
2011-03-10  Andrey Adaikin  <aandrey@google.com>

        Reviewed by Pavel Feldman.

        Web Inspector: Highlighter refactoring
        https://bugs.webkit.org/show_bug.cgi?id=56015

        * inspector/editor/highlighter-basics-expected.txt:
        * inspector/editor/highlighter-basics.html:
        * inspector/editor/highlighter-long-line.html:
        * inspector/editor/highlighter-paste-in-comment-expected.txt:
        * inspector/editor/highlighter-paste-in-comment.html:
2011-03-10  Andrey Adaikin  <aandrey@google.com>

        Reviewed by Pavel Feldman.

        Web Inspector: Highlighter refactoring
        https://bugs.webkit.org/show_bug.cgi?id=56015

        * inspector/front-end/TextEditorHighlighter.js:
        (WebInspector.TextEditorHighlighter):
        (WebInspector.TextEditorHighlighter.prototype.set mimeType):
        (WebInspector.TextEditorHighlighter.prototype.highlight):
        (WebInspector.TextEditorHighlighter.prototype.updateHighlight):
        (WebInspector.TextEditorHighlighter.prototype._highlightInChunks):
        (WebInspector.TextEditorHighlighter.prototype._highlightLines):
        (WebInspector.TextEditorHighlighter.prototype._selectHighlightState):
        (WebInspector.TextEditorHighlighter.prototype._clearHighlightState):
        * inspector/front-end/TextViewer.js:
        (WebInspector.TextEditorMainPanel.prototype._buildChunks):
        (WebInspector.TextEditorMainPanel.prototype._updateHighlightsForRange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 John Knottenbelt <jknotten@chromium.org>
jknotten@chromium.org [Thu, 10 Mar 2011 11:28:08 +0000 (11:28 +0000)]
2011-03-10  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Andreas Kling.

        [Qt] Skip fast/dom/Geolocation/page-reload-cancel-permission-requests.html pending implementation of LayoutTestController::numberOfPendingGeolocationPermissionRequests
        https://bugs.webkit.org/show_bug.cgi?id=56087

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-05 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Thu, 10 Mar 2011 11:24:07 +0000 (11:24 +0000)]
2011-03-05  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: move breakpoint column adjustment to debugger model.
        https://bugs.webkit.org/show_bug.cgi?id=55821

        Test: inspector/debugger/debug-inlined-scripts.html

        * inspector/front-end/DebuggerModel.js:
        (WebInspector.DebuggerModel.prototype.setBreakpoint.didSetBreakpoint):
        (WebInspector.DebuggerModel.prototype.setBreakpoint):
        * inspector/front-end/SourceFrameContent.js:
        (WebInspector.SourceFrameContent.prototype.sourceFrameLineNumberToActualLocation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-02-16 John Knottenbelt <jknotten@chromium.org>
jknotten@chromium.org [Thu, 10 Mar 2011 10:15:49 +0000 (10:15 +0000)]
2011-02-16  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Dmitry Titov.

        Detach Geolocation from Frame when Page destroyed.
        https://bugs.webkit.org/show_bug.cgi?id=52877

        Ensure that all geolocation permission requests are cancelled
        when the page is detached from its frame.

        * fast/dom/Geolocation/page-reload-cancel-permission-requests-expected.txt: Added.
        * fast/dom/Geolocation/page-reload-cancel-permission-requests.html: Added.
        * fast/dom/Geolocation/resources/page-reload-cancel-permission-requests-inner.html: Added.
        * fast/dom/Geolocation/script-tests/page-reload-cancel-permission-requests.js: Added.
        * platform/gtk/Skipped:
        * platform/mac/Skipped:
        * platform/qt-wk2/Skipped:
2011-01-21  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Dmitry Titov.

        Detach Geolocation from Frame when Page destroyed.
        https://bugs.webkit.org/show_bug.cgi?id=52877

        On Page destruction, any outstanding Geolocation permission
        requests should be cancelled, because the Geolocation can only
        access the client indirectly via m_frame->page().

        Page destruction is signalled by a call to the
        Frame::pageDestroyed() method. This explictly calls
        DOMWindow::resetGeolocation which ultimately calls Geolocation::reset.

        Geolocation::reset() detaches from the GeolocationController,
        cancels requests, watches and single shots, and sets the
        permission state back to Unknown.

        Frame::pageDestroyed() is also called by FrameLoader even though
        the page is not destroyed. We should still cancel permission
        requests, because the GeolocationClient will become inaccessible
        to the Geolocation object after this call.

        Frame::transferChildFrameToNewDocument also indirectly calls
        Geolocation::reset when the frame is reparented between
        pages. Ideally we would like the Geolocation's activities to
        continue after reparenting, see bug
        https://bugs.webkit.org/show_bug.cgi?id=55577

        Since GeolocationController is owned by Page, and all Geolocation
        objects will now unsubscribe from the GeolocationController on
        pageDetached(), we no longer need to call stopUpdating() from the
        GeolocationController's destructor. Instead we can simply assert
        that there should be no no observers. See related bug
        https://bugs.webkit.org/show_bug.cgi?id=52216 .

        Introduced new method 'numberOfPendingPermissionRequests' on
        GeolocationClientMock to count the number of outstanding pending
        permission requests. This provides a reusable implementation for
        client-based implementations of the LayoutTestController's
        numberOfPendingGeolocationPermissionRequests method.

        Test: fast/dom/Geolocation/page-reload-cancel-permission-requests.html

        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::resetGeolocation):
        * page/DOMWindow.h:
        * page/Frame.cpp:
        (WebCore::Frame::pageDestroyed):
        (WebCore::Frame::transferChildFrameToNewDocument):
        * page/Geolocation.cpp:
        (WebCore::Geolocation::~Geolocation):
        (WebCore::Geolocation::page):
        (WebCore::Geolocation::reset):
        (WebCore::Geolocation::disconnectFrame):
        (WebCore::Geolocation::lastPosition):
        (WebCore::Geolocation::requestPermission):
        (WebCore::Geolocation::startUpdating):
        (WebCore::Geolocation::stopUpdating):
        * page/Geolocation.h:
        * page/GeolocationController.cpp:
        (WebCore::GeolocationController::~GeolocationController):
        * page/Navigator.cpp:
        (WebCore::Navigator::resetGeolocation):
        * page/Navigator.h:
        * platform/mock/GeolocationClientMock.cpp:
        (WebCore::GeolocationClientMock::numberOfPendingPermissionRequests):
        * platform/mock/GeolocationClientMock.h:
2011-01-26  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Dmitry Titov.

        Detach Geolocation from Frame when Page destroyed.
        https://bugs.webkit.org/show_bug.cgi?id=52877

        Add accessors to the WebGeolocationClientMock to allow the number of
        pending geolocation permission requests to be queried.

        * public/WebGeolocationClientMock.h:
        * src/WebGeolocationClientMock.cpp:
        (WebKit::WebGeolocationClientMock::numberOfPendingPermissionRequests):
2011-01-26  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Dmitry Titov.

        Detach Geolocation from Frame when Page destroyed.
        https://bugs.webkit.org/show_bug.cgi?id=52877

        Extend the layout test controller to expose the number of pending
        geolocation requests, so that we can test that the requests have
        been cancelled on page close.

        * DumpRenderTree/LayoutTestController.cpp:
        (numberOfPendingGeolocationPermissionRequestsCallback):
        (LayoutTestController::staticFunctions):
        * DumpRenderTree/LayoutTestController.h:
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::LayoutTestController):
        (LayoutTestController::numberOfPendingGeolocationPermissionRequests):
        * DumpRenderTree/chromium/LayoutTestController.h:
        * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
        (LayoutTestController::numberOfPendingGeolocationPermissionRequests):
        * DumpRenderTree/mac/LayoutTestControllerMac.mm:
        (LayoutTestController::numberOfPendingGeolocationPermissionRequests):
        * DumpRenderTree/mac/UIDelegate.h:
        * DumpRenderTree/mac/UIDelegate.mm:
        (-[UIDelegate numberOfPendingGeolocationPermissionRequests]):
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::numberOfPendingGeolocationPermissionRequests):
        * DumpRenderTree/qt/LayoutTestControllerQt.h:
        * DumpRenderTree/win/LayoutTestControllerWin.cpp:
        (LayoutTestController::numberOfPendingGeolocationPermissionRequests):
        * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
        (LayoutTestController::numberOfPendingGeolocationPermissionRequests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed.
ossy@webkit.org [Thu, 10 Mar 2011 09:57:37 +0000 (09:57 +0000)]
Unreviewed.

Add Qt specific expected files for tests introduced in r80622 and r80682.

* platform/qt/fast/dom/scroll-reveal-left-overflow-expected.checksum: Added.
* platform/qt/fast/dom/scroll-reveal-left-overflow-expected.png: Added.
* platform/qt/fast/dom/scroll-reveal-left-overflow-expected.txt: Added.
* platform/qt/fast/dom/scroll-reveal-top-overflow-expected.checksum: Added.
* platform/qt/fast/dom/scroll-reveal-top-overflow-expected.png: Added.
* platform/qt/fast/dom/scroll-reveal-top-overflow-expected.txt: Added.
* platform/qt/fast/dynamic/dirty-float-in-clean-line-expected.checksum: Added.
* platform/qt/fast/dynamic/dirty-float-in-clean-line-expected.png: Added.
* platform/qt/fast/dynamic/dirty-float-in-clean-line-expected.txt: Added.
* platform/qt/fast/dynamic/float-at-line-break-expected.checksum: Added.
* platform/qt/fast/dynamic/float-at-line-break-expected.png: Added.
* platform/qt/fast/dynamic/float-at-line-break-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 09:33:11 +0000 (09:33 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip fast/js/large-expressions.html on GTK.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 10 Mar 2011 09:23:17 +0000 (09:23 +0000)]
2011-03-10  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip jquery/effects.html on GTK.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Thu, 10 Mar 2011 09:11:30 +0000 (09:11 +0000)]
2011-03-10  Adam Barth  <abarth@webkit.org>

        Partial revert of http://trac.webkit.org/changeset/80670.  This test
        has been failing sporatically since it was changed in that revision.

        * platform/mac/accessibility/aria-expanded-notifications.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-10 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Thu, 10 Mar 2011 08:33:11 +0000 (08:33 +0000)]
2011-03-10  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Darin Adler.

        Able to move nodes across documents
        https://bugs.webkit.org/show_bug.cgi?id=19524

        Makes cross-document appendChild, insertBefore, Range.insertNode and Range.surroundContents work.
        This matches Gecko and the new Dom Core spec. There are a number of Range methods where we don't
        match Gecko or the spec that will need to be updated in a following patch.

        * dom/html/level1/core/hc_elementwrongdocumenterr-expected.txt:
        * dom/html/level1/core/hc_namednodemapwrongdocumenterr-expected.txt:
        * dom/xhtml/level1/core/hc_elementwrongdocumenterr-expected.txt:
        * dom/xhtml/level1/core/hc_namednodemapwrongdocumenterr-expected.txt:
        * fast/dom/move-nodes-across-documents-expected.txt: Added.
        * fast/dom/move-nodes-across-documents.html: Added.
2011-03-10  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Darin Adler.

        Able to move nodes across documents
        https://bugs.webkit.org/show_bug.cgi?id=19524

        Makes cross-document appendChild, insertBefore, Range.insertNode and Range.surroundContents work.
        This matches Gecko and the new Dom Core spec. There are a number of Range methods where we don't
        match Gecko or the spec that will need to be updated in a following patch.

        Test: fast/dom/move-nodes-across-documents.html

        * dom/Element.cpp:
        (WebCore::Element::removeAttributeNode):
        * dom/NamedNodeMap.cpp:
        (WebCore::NamedNodeMap::setNamedItem):
        * dom/Node.cpp:
        (WebCore::Node::setDocumentRecursively):
        (WebCore::checkAcceptChild):
        (WebCore::Node::checkReplaceChild):
        * dom/Range.cpp:
        (WebCore::Range::insertNode):
        (WebCore::Range::surroundContents):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNot reviewed.
antti@apple.com [Thu, 10 Mar 2011 07:31:13 +0000 (07:31 +0000)]
Not reviewed.

Reverting crash catching code, the bug being hunted was fixed by
http://trac.webkit.org/changeset/80686

* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::CachedResourceLoader):
(WebCore::CachedResourceLoader::~CachedResourceLoader):
(WebCore::CachedResourceLoader::requestImage):
(WebCore::CachedResourceLoader::requestUserCSSStyleSheet):
(WebCore::CachedResourceLoader::requestResource):
(WebCore::CachedResourceLoader::setAutoLoadImages):
(WebCore::CachedResourceLoader::load):
(WebCore::CachedResourceLoader::loadDone):
(WebCore::CachedResourceLoader::preload):
(WebCore::CachedResourceLoader::requestPreload):
* loader/cache/CachedResourceLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed, build fix.
pkasting@chromium.org [Thu, 10 Mar 2011 06:54:17 +0000 (06:54 +0000)]
Unreviewed, build fix.

* StringsNotToBeLocalized.txt: Add function names I forgot.
* platform/win/SystemInfo.cpp: #if out some uncalled functions on WinCE,
  since they don't compile anyway.  Use ZeroMemory() instead of "= {0}"
  since Qt is stupid and (sometimes?!) warns about it.
(WebCore::windowsVersion):
(WebCore::processorArchitecture):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd UA string tags for Windows 64.
pkasting@chromium.org [Thu, 10 Mar 2011 06:31:32 +0000 (06:31 +0000)]
Add UA string tags for Windows 64.
https://bugs.webkit.org/show_bug.cgi?id=55226

Reviewed by Ryosuke Niwa.

* StringsNotToBeLocalized.txt:
* platform/win/SystemInfo.cpp:
(WebCore::osVersionForUAString):
(WebCore::isWOW64):
(WebCore::processorArchitecture):
(WebCore::architectureTokenForUAString):
(WebCore::windowsVersionForUAString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@80693 268f45cc-cd09-0410-ab3c-d52691b4dbfc