WebKit-https.git
11 years ago Fix for WebKit bug 29968 - Selecting text with text-overflow ellipsis
jberlin@webkit.org [Thu, 24 Dec 2009 13:46:30 +0000 (13:46 +0000)]
    Fix for WebKit bug 29968 - Selecting text with text-overflow ellipsis
        should not show cut off text
        https://bugs.webkit.org/show_bug.cgi?id=29968

        Reviewed by Dan Bernstein.

        Also fixes the issue on both Mac OS X and Windows where the highlight
        would extend only partway into the ellipsis based on the size of the
        characters being truncated.

        Test: editing/selection/select-text-overflow-ellipsis.html

        WebCore:

        * rendering/EllipsisBox.cpp:
        (WebCore::EllipsisBox::paint):
        When the ellipsis have been selected, paints them with the color for
        selected text.
        (WebCore::EllipsisBox::selectionRect):
        Returns the selection rect for the ellipsis text.
        (WebCore::EllipsisBox::paintSelection):
        Paints the selection highlight around the ellipsis text.

        * rendering/EllipsisBox.h:
        (WebCore::EllipsisBox::EllipsisBox):
        Default the selection state of the EllipsisBox to SelectionNone.
        (WebCore::EllipsisBox::setSelectionState):
        Allow the selection state to be explicitly set by the InlineTextBox
        that it is associated with, since the selection state depends on the
        position of the selection within the truncated text.
        (WebCore::EllipsisBox::selectionState):
        Return the cached selection state.

        * rendering/InlineTextBox.cpp:
        (WebCore::InlineTextBox::selectionState):
        If the text is truncated and the selection extends into the truncation
        where there are ellipsis, set the selection state on the EllipsisBox.
        (WebCore::paintTextWithShadows):
        Allow for an upper bound to be specified on the maximum length of the
        text to be painted.
        (WebCore::InlineTextBox::paint):
        Make sure the text that is drawn gets appropriately truncated. This was
        not an issue before on Mac OS X because it wasn't necessary to draw the
        selected text separately (the selected text had the same color as the
        non-selected text).
        (WebCore::InlineTextBox::paintSelection):
        If the text is being truncated, let whatever is being painted in the
        space after the truncated text paint its own highlight.

        * rendering/InlineTextBox.h:
        (WebCore::InlineTextBox::truncation):
        Allow RenderText to access the truncation status.

        * rendering/RenderText.cpp:
        (WebCore::RenderText::selectionRectForRepaint):
        Make sure that any ellipsis box that falls within the selection also
        gets repainted.

        Places the expected results under the mac and win platforms separately
        because they are pixel test results and also the size of the
        RenderBlocks are showing up as slightly different between mac and win.

        LayoutTests:

        * editing/selection/select-text-overflow-ellipsis.html: Added.
        * platform/mac/editing/selection/select-text-overflow-ellipsis-expected.checksum: Added.
        * platform/mac/editing/selection/select-text-overflow-ellipsis-expected.png: Added.
        * platform/mac/editing/selection/select-text-overflow-ellipsis-expected.txt: Added.
        * platform/win/editing/selection/select-text-overflow-ellipsis-expected.checksum: Added.
        * platform/win/editing/selection/select-text-overflow-ellipsis-expected.png: Added.
        * platform/win/editing/selection/select-text-overflow-ellipsis-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-24 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 24 Dec 2009 12:33:39 +0000 (12:33 +0000)]
2009-12-24  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: add Settings.js class that would manage
        frontend settings.

        This change:
        - Splits preferences and settings
        - Pushes settings to the frontend on load
        - Stores settings on each change
        - Stores showInheritedStyles as setting.

        https://bugs.webkit.org/show_bug.cgi?id=32832

        * WebCore.gypi:
        * WebCore.vcproj/WebCore.vcproj:
        * inspector/InspectorBackend.cpp:
        (WebCore::InspectorBackend::saveFrontendSettings):
        * inspector/InspectorBackend.h:
        * inspector/InspectorBackend.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::populateScriptObjects):
        * inspector/InspectorController.h:
        * inspector/InspectorFrontend.cpp:
        (WebCore::InspectorFrontend::populateFrontendSettings):
        * inspector/InspectorFrontend.h:
        * inspector/InspectorFrontendHost.cpp:
        * inspector/InspectorFrontendHost.h:
        * inspector/InspectorFrontendHost.idl:
        * inspector/front-end/EventListenersSidebarPane.js:
        (WebInspector.EventListenersSidebarPane.prototype._settingsLoaded):
        (WebInspector.EventListenersSidebarPane.prototype):
        ():
        * inspector/front-end/InspectorBackendStub.js:
        (.WebInspector.InspectorBackendStub):
        (.WebInspector.InspectorBackendStub.prototype.saveFrontendSettings):
        (.WebInspector.InspectorBackendStub.prototype.releaseWrapperObjectGroup):
        * inspector/front-end/InspectorFrontendHostStub.js:
        (.WebInspector.InspectorFrontendHostStub):
        (.WebInspector.InspectorFrontendHostStub.prototype.hiddenPanels):
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel.prototype._createStatusbarButtons):
        (WebInspector.ResourcesPanel.prototype._settingsLoaded):
        (WebInspector.ResourcesPanel.prototype._toggleLargerResources):
        (WebInspector.ResourcesPanel.prototype._setLargerResources):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype._showScriptOrResource):
        (WebInspector.ScriptsPanel.prototype._addScriptToFilesMenu):
        * inspector/front-end/Settings.js: Added.
        (WebInspector.populateFrontendSettings):
        (WebInspector.Settings):
        (WebInspector.Settings.prototype._load):
        (WebInspector.Settings.prototype._installSetting):
        (WebInspector.Settings.prototype._get):
        (WebInspector.Settings.prototype._set):
        * inspector/front-end/StylesSidebarPane.js:
        (WebInspector.StylesSidebarPane.prototype._settingsLoaded):
        (WebInspector.StylesSidebarPane.prototype.update):
        (WebInspector.StylesSidebarPane.prototype._changeSetting):
        (WebInspector.StylesSidebarPane.prototype._changeColorFormat):
        (WebInspector.StylePropertiesSection.showInheritedToggleFunction):
        (WebInspector.StylePropertiesSection):
        * inspector/front-end/WatchExpressionsSidebarPane.js:
        (WebInspector.WatchExpressionsSidebarPane):
        (WebInspector.WatchExpressionsSidebarPane.prototype._settingsLoaded):
        (WebInspector.WatchExpressionsSidebarPane.prototype.refreshExpressions):
        (WebInspector.WatchExpressionsSection):
        (WebInspector.WatchExpressionsSection.prototype.saveExpressions):
        * inspector/front-end/WebKit.qrc:
        * inspector/front-end/inspector.html:
        * inspector/front-end/inspector.js:
        (WebInspector.loaded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-24 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 24 Dec 2009 11:12:45 +0000 (11:12 +0000)]
2009-12-24  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: follow up to r52545, adding missing tests.

        * inspector/console-clear-expected.txt: Added.
        * inspector/console-clear.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 24 Dec 2009 11:05:58 +0000 (11:05 +0000)]
2009-12-23  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: [REGRESSION] console's clear command does not work.

        https://bugs.webkit.org/show_bug.cgi?id=32897

        Test: inspector/console-clear.html

WebCore:
        * inspector/InjectedScriptHost.cpp:
        (WebCore::InjectedScriptHost::clearConsoleMessages):
        * inspector/InjectedScriptHost.h:
        * inspector/InjectedScriptHost.idl:
        * inspector/InspectorBackend.cpp:
        * inspector/InspectorBackend.h:
        * inspector/InspectorBackend.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::clearConsoleMessages):
        (WebCore::InspectorController::didCommitLoad):
        * inspector/InspectorController.h:
        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleView):
        (WebInspector.ConsoleView.prototype.requestClearConsoleMessages):
        (WebInspector.ConsoleView.prototype.clearConsoleMessages):
        (WebInspector.ConsoleView.prototype._clearButtonClicked):
        * inspector/front-end/InjectedScript.js:
        (InjectedScript.clearConsoleMessages):
        (InjectedScript._ensureCommandLineAPIInstalled):
        * inspector/front-end/InjectedScriptAccess.js:
        * inspector/front-end/inspector.js:
        (WebInspector.clearConsoleMessages):
        (WebInspector.reset):
LayoutTests:
        * inspector/console-clear-expected.txt: Added.
        * inspector/console-clear.html: Added.
        * inspector/console-tests.js:
        (dumpConsoleMessages.callback):
        (dumpConsoleMessages):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 24 Dec 2009 10:38:27 +0000 (10:38 +0000)]
2009-12-23  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Console dumps strings in escaped form.

        https://bugs.webkit.org/show_bug.cgi?id=32488

        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleMessage.prototype._formatIndividualValue):
        * inspector/console-format-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Eric Seidel <eric@webkit.org>
eric@webkit.org [Thu, 24 Dec 2009 07:18:27 +0000 (07:18 +0000)]
2009-12-23  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        bugzilla-tool should accept global options anywhere
        https://bugs.webkit.org/show_bug.cgi?id=26912

        * Scripts/bugzilla-tool:
         - Use the new global_options class property.
         - Add a handle_global_options callback to avoid needing callbacks for each global option.
        * Scripts/modules/multicommandtool.py:
         - Make the code use one combined option parser.
           This allows us to accept global options anywhere and
           individual command options before commands.
         - Add a handle_global_options callback to avoid needing callbacks for each global option.
         - Make the Command hold the option parser, but allow the tool to override it.
         - The default option parser is used for help printing and when Commands are run stand alone,
           but are otherwise not used.
         - Add Command.main to codify the idea that Commands should support being run stand-alone.
         - Change _split_args to _split_command_name_from_args now that args are unified.
        * Scripts/modules/multicommandtool_unittest.py:
         - Test that "tool" and "tool help" show the same help.
         - Test that args are accepted before commands

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler.
ap@apple.com [Thu, 24 Dec 2009 05:49:15 +0000 (05:49 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=32905
        With Pinyin Simplified IM, a wrong character is deleted from google.com suggestion

        Test: platform/mac/editing/input/selection-change-closes-typing.html

        * editing/Editor.cpp: (WebCore::Editor::confirmComposition): If this function wasn't allowed
        to change selection, it needs to close the current typing command - otherwise, its idea of
        selection would be used for future typing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-20 Chris Jerdonek <chris.jerdonek@gmail.com>
hamaji@chromium.org [Thu, 24 Dec 2009 05:31:33 +0000 (05:31 +0000)]
2009-12-20  Chris Jerdonek  <chris.jerdonek@gmail.com>

        Reviewed by David Levin.

        Moved some sections of code in preparation to refactor
        check-webkit-style's argument parser to avoid setting
        global variables.

        https://bugs.webkit.org/show_bug.cgi?id=32592

        * Scripts/check-webkit-style:
          - Moved _USAGE string to style.py.
          - Addressed FIXME by eliminating dependencies on cpp_style.py.

        * Scripts/modules/cpp_style.py:
          - Moved default arguments and style categories to style.py.
          - Moved exit_with_usage(), exit_with_categories(), and
            parse_arguments() to style.py.
          - Removed references in _CppStyleState to the global
            variables now in style.py.

        * Scripts/modules/cpp_style_unittest.py:
          - Moved parse_arguments() unit tests to style_unittest.py.

        * Scripts/modules/style.py:
          - Added _USAGE string from check-webkit-style.
          - Added default arguments and style categories from cpp_style.py.
          - Added exit_with_usage(), exit_with_categories(), and
            parse_arguments() from cpp_sstyle.py.

        * Scripts/modules/style_unittest.py: Added.
          - Added parse_arguments() unit tests from cpp_style_unittest.py.

        * Scripts/run-webkit-unittests:
          - Added unit tests from style_unittest.py.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Eric Seidel <eric@webkit.org>
eric@webkit.org [Thu, 24 Dec 2009 02:31:27 +0000 (02:31 +0000)]
2009-12-23  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        commit-queue should mention bug 30084 when rejecting patches until it can be fixed
        https://bugs.webkit.org/show_bug.cgi?id=32911

        * Scripts/modules/bugzilla.py: Make the message even more fancy.
        * Scripts/modules/bugzilla_unittest.py: Test our new fancy message.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Yuta Kitamura <yutak@chromium.org>
eric@webkit.org [Thu, 24 Dec 2009 02:23:38 +0000 (02:23 +0000)]
2009-12-23  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Eric Seidel.

        PlatformContextSkia::beginLayerClippedToImage() needs to clip outside of the image.

        This method should limit the current painting window to the bounds of the given
        image, but it did not. Due to this bug, unnecessary contents were rendered in
        some (limited) situations.

        [Skia] PlatformContextSkia::beginLayerClippedToImage does not clip outside of the image
        https://bugs.webkit.org/show_bug.cgi?id=32817

        No new tests, because this is a fix of an existing test which is currently failing in
        Chromium.

        * platform/graphics/skia/PlatformContextSkia.cpp:
        (PlatformContextSkia::beginLayerClippedToImage): Add clipRect() call.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Thu, 24 Dec 2009 02:15:44 +0000 (02:15 +0000)]
2009-12-23  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Laszlo Gombos.

        Include stddef.h for ptrdiff_t
        https://bugs.webkit.org/show_bug.cgi?id=32891

        ptrdiff_t is typedef-ed in stddef.h.
        Include stddef.h in jit/ExecutableAllocator.h.

        * jit/ExecutableAllocator.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Thu, 24 Dec 2009 01:45:01 +0000 (01:45 +0000)]
2009-12-23  Nikolas Zimmermann  <nzimmermann@rim.com>

        Rubber-stamped by Eric Seidel.

        Remove all references to "GenerateConstructor", as this is the default behaviour now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoA new database should be registered with the Document,
dumi@chromium.org [Thu, 24 Dec 2009 01:38:53 +0000 (01:38 +0000)]
A new database should be registered with the Document,
DatabaseTracker and DatabaseThread only after it was successfully
opened and its version was successfully verified.

Reviewed by Eric Seidel.

Fixes a regression introduced by
http://trac.webkit.org/changeset/52530.

LayoutTests/storage/open-database-while-transaction-in-progress.html
should pass again (or rather, it should pass again when run after
open-database-set-empty-version.html).

https://bugs.webkit.org/show_bug.cgi?id=32913

* storage/Database.cpp:
(WebCore::Database::openDatabase):
(WebCore::Database::Database):
(WebCore::Database::performOpenAndVerify):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 David Levin <levin@chromium.org>
levin@chromium.org [Thu, 24 Dec 2009 01:15:10 +0000 (01:15 +0000)]
2009-12-23  David Levin  <levin@chromium.org>

        REGRESSION (r52494): Assertion failure in Frame::caretBlinkTimerFired() (selection()->isCaret())
        https://bugs.webkit.org/show_bug.cgi?id=32903

        No review but ok'ed by Darin Adler. Rolling out r52494 due to above regression.

LayoutTests:
        * fast/forms/selection-layout-reentry-strange-case-expected.txt: Removed.
        * fast/forms/selection-layout-reentry-strange-case.html: Removed.
        * platform/mac/accessibility/frame-with-title-expected.txt:

WebCore:
        * editing/SelectionController.cpp:
        (WebCore::SelectionController::SelectionController):
        (WebCore::SelectionController::setSelection):
        * editing/SelectionController.h:
        * page/Frame.cpp:
        (WebCore::Frame::setCaretVisible):
        (WebCore::Frame::selectionLayoutChanged):
        * page/FrameView.cpp:
        (WebCore::FrameView::needsLayout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Thu, 24 Dec 2009 00:55:01 +0000 (00:55 +0000)]
2009-12-23  Nikolas Zimmermann  <nzimmermann@rim.com>

        Reviewed by Eric Seidel.

        Reverse JS GenerateConstructor logic
        https://bugs.webkit.org/show_bug.cgi?id=32910

        Make 'GenerateConstructor' the default setting for all classes, defined in IDL files. The flag is now obsolete and
        can be removed from all IDL files in a follow-up patch. Add new 'OmitConstructor' flag, that allows to restore the
        old default behaviour: do not generate JSFoobarConstructor class.

        No change in functionality, despite the fact that we're generating a lot more constructors now, as they have to
        be exposed through DOMWindow.idl -- this can be done per affected class in follow-up patches. Especially the SVG
        classes have to be exposed, most of them are missing JS constructors so far.

        As side effect HTMLOptionsCollection is now correctly exposing its constructor, thus fixing a test in fast/dom/wrapper-classes.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Eric Seidel.
bweinstein@apple.com [Thu, 24 Dec 2009 00:34:22 +0000 (00:34 +0000)]
Rubber-stamped by Eric Seidel.

Added Windows specific results for a test that was committed in r52505.

* platform/win/fast/forms/basic-textareas-quirks-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 23 Dec 2009 23:22:34 +0000 (23:22 +0000)]
2009-12-23  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        "Refused to execute a JavaScript script" error when embedding SWF with
        a URL that is also a query parameter
        https://bugs.webkit.org/show_bug.cgi?id=32908

        Update expected results to show that we don't raise an alarm in this case.

        * http/tests/security/xssAuditor/object-src-inject-expected.txt:
2009-12-23  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        "Refused to execute a JavaScript script" error when embedding SWF with
        a URL that is also a query parameter
        https://bugs.webkit.org/show_bug.cgi?id=32908

        Don't block direct injections into the object src attribute unless
        there's an illegal character (like < or ") in the URL.  This change
        lets some very unusual vulnerabilities through the filter but removes a
        false positive that we've seen several times.

        * page/XSSAuditor.cpp:
        (WebCore::XSSAuditor::canLoadObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Laszlo Gombos <laszlo.1.gombos@nokia.com>
laszlo.1.gombos@nokia.com [Wed, 23 Dec 2009 22:34:28 +0000 (22:34 +0000)]
2009-12-23  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Not reviewed, adding myself to the reviewers list.

        * Scripts/modules/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemove pointers to a database that failed to open from the
dumi@chromium.org [Wed, 23 Dec 2009 22:33:21 +0000 (22:33 +0000)]
Remove pointers to a database that failed to open from the
DatabaseTracker and Document.

Reviewed by Dmitry Titov.

https://bugs.webkit.org/show_bug.cgi?id=32885

* storage/Database.cpp:
(WebCore::Database::openDatabase):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Patrick Gansterer <paroga@paroga.com>
eric@webkit.org [Wed, 23 Dec 2009 21:03:52 +0000 (21:03 +0000)]
2009-12-23  Patrick Gansterer  <paroga@paroga.com>

        Reviewed by Eric Seidel.

        Buildfix after r47092.

        * wtf/wince/MemoryManager.cpp:
        (WTF::tryFastMalloc):
        (WTF::tryFastZeroedMalloc):
        (WTF::tryFastCalloc):
        (WTF::tryFastRealloc):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 23 Dec 2009 20:40:03 +0000 (20:40 +0000)]
2009-12-23  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Create a prepare-diff command that creates a bug and a ChangeLog
        https://bugs.webkit.org/show_bug.cgi?id=32895

        The workflow Maciej and I discussed is as follows:

        1) Write code.
        2) bugzilla-tool prepare-diff
        3) Edit ChangeLogs
        4) bugzilla-tool post-diff

        We might want to experimenting with combining 2-4 into a single
        command, but that might be stressful to edit the ChangeLogs modally.

        Removed submit-patch since it has the modal ChangeLog editing but none
        oof the bug creating fun.

        * Scripts/modules/bugzilla.py:
        * Scripts/modules/buildsteps.py:
        * Scripts/modules/commands/upload.py:
        * Scripts/modules/commands/upload_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/7487164> First line of text cannot be selected
mitz@apple.com [Wed, 23 Dec 2009 19:49:25 +0000 (19:49 +0000)]
<rdar://problem/7487164> First line of text cannot be selected
https://bugs.webkit.org/show_bug.cgi?id=32749

Reviewed by Darin Adler.

WebCore:

Test: fast/text/remove-zero-length-run.html

* rendering/RenderText.cpp:
(WebCore::RenderText::positionLineBox): Changed code that assumed that if a box was being
    removed, it was the only box in the RenderText. Instead, correctly preserve the list of
    text boxes.
(WebCore::RenderText::checkConsistency): Updated for earlier rename.

LayoutTests:

* fast/text/remove-zero-length-run-expected.txt: Added.
* fast/text/remove-zero-length-run.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed; added myself to the committers list.
loki@webkit.org [Wed, 23 Dec 2009 13:32:06 +0000 (13:32 +0000)]
Unreviewed; added myself to the committers list.

* Scripts/modules/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix for newlines turning into missing-gyphs when using SVG fonts
vestbo@webkit.org [Wed, 23 Dec 2009 12:52:15 +0000 (12:52 +0000)]
Fix for newlines turning into missing-gyphs when using SVG fonts

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=32710

Font::drawText() has three code paths:

  - drawTextUsingSVGFont()
  - drawSimpleText()
  - drawComplexText()

The latter two both have logic to replace newlines and tabs, as well
as a few other control characters that are considered spaces, with
spaces, before actually drawing the text. Previously the SVG font
path did not have this kind of logic.

This change introduces two helper functions in Font to do this space
replacement, which are then used for the simple path, the SVG path,
and the Qt implementation of the complex path.

Test: svg/custom/svg-fonts-in-html-linebreaks.html

WebCore:

* platform/graphics/Font.cpp:
(WebCore::Font::normalizeSpaces): Helper function
* platform/graphics/Font.h:
* platform/graphics/FontFastPath.cpp:
(WebCore::Font::glyphDataForCharacter): Use helper
* platform/graphics/qt/FontQt.cpp:
(WebCore::fromRawDataWithoutRef): Rename from "qstring"
(WebCore::Font::drawComplexText): Use helper
(WebCore::Font::floatWidthForComplexText): Use helper
(WebCore::Font::offsetForPositionForComplexText): Use helper
(WebCore::Font::selectionRectForComplexText): Use helper
* platform/text/CharacterNames.h: Add enum for 0x0020
* svg/SVGFont.cpp: Use helper to fix bug
(WebCore::SVGTextRunWalker::walk):

LayoutTests:

* svg/custom/resources/ABCFont.svg: Added. Font for test case
* svg/custom/svg-fonts-in-html-linebreaks-expected.txt: Added.
* svg/custom/svg-fonts-in-html-linebreaks.html: New test case

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-23 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 23 Dec 2009 11:05:04 +0000 (11:05 +0000)]
2009-12-23  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Adler.

        HTMLInputElement::valueAsDate getter support.
        https://bugs.webkit.org/show_bug.cgi?id=32876

        To implement the valueAsDate getter,
        - Add a method to calculate milliseconds from 1970-01-01 to ISODateTime.
        - Introduce m_type field to ISODateTime.

        Tests: fast/forms/input-valueasdate-date.html
               fast/forms/input-valueasdate-datetime.html
               fast/forms/input-valueasdate-datetimelocal.html
               fast/forms/input-valueasdate-month.html
               fast/forms/input-valueasdate-time.html
               fast/forms/input-valueasdate-week.html

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::valueAsDate):
          Implement it with ISODateTime::millisecondsSinceEpoch().
        (WebCore::HTMLInputElement::formStringToISODateTime):
          Early exit for a null String. This avoids assertion failures in ISODateTime::parse*().
        * html/ISODateTime.cpp:
        (WebCore::ISODateTime::parseMonth): Sets m_type.
        (WebCore::ISODateTime::parseDate): ditto.
        (WebCore::ISODateTime::parseWeek): ditto.
        (WebCore::ISODateTime::parseTime): ditto.
        (WebCore::ISODateTime::parseDateTimeLocal): ditto.
        (WebCore::ISODateTime::parseDateTime): ditto.
        (WebCore::ISODateTime::millisecondsSinceEpochForTime): Added.
        (WebCore::ISODateTime::millisecondsSinceEpoch): Added.
        * html/ISODateTime.h:
        (WebCore::ISODateTime::ISODateTime):
        (WebCore::ISODateTime::invalidMilliseconds):

2009-12-23  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Adler.

        HTMLInputElement::valueAsDate getter support.
        https://bugs.webkit.org/show_bug.cgi?id=32876

        Tests for the valueAsDate getter with various types.

        * fast/forms/input-valueasdate-date-expected.txt: Added.
        * fast/forms/input-valueasdate-date.html: Added.
        * fast/forms/input-valueasdate-datetime-expected.txt: Added.
        * fast/forms/input-valueasdate-datetime.html: Added.
        * fast/forms/input-valueasdate-datetimelocal-expected.txt: Added.
        * fast/forms/input-valueasdate-datetimelocal.html: Added.
        * fast/forms/input-valueasdate-month-expected.txt: Added.
        * fast/forms/input-valueasdate-month.html: Added.
        * fast/forms/input-valueasdate-time-expected.txt: Added.
        * fast/forms/input-valueasdate-time.html: Added.
        * fast/forms/input-valueasdate-week-expected.txt: Added.
        * fast/forms/input-valueasdate-week.html: Added.
        * fast/forms/script-tests/input-valueasdate-date.js: Added.
        * fast/forms/script-tests/input-valueasdate-datetime.js: Added.
        * fast/forms/script-tests/input-valueasdate-datetimelocal.js: Added.
        * fast/forms/script-tests/input-valueasdate-month.js: Added.
        * fast/forms/script-tests/input-valueasdate-time.js: Added.
        * fast/forms/script-tests/input-valueasdate-week.js: Added.

2009-12-23  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Adler.

        HTMLInputElement::valueAsDate getter support.
        https://bugs.webkit.org/show_bug.cgi?id=32876

        Expose dateToDaysFrom1970().

        * JavaScriptCore.exp:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * wtf/DateMath.cpp:
        (WTF::dateToDaysFrom1970):
        * wtf/DateMath.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52524 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 23 Dec 2009 08:00:52 +0000 (08:00 +0000)]
2009-12-22  Adam Barth  <abarth@webkit.org>

        Reviewed by Maciej Stachowiak.

        [bzt] Don't call create-patch twice during post-diff
        https://bugs.webkit.org/show_bug.cgi?id=32893

        If we call create-patch to get the bug number, cache the diff in the
        state so we don't need to call it again.

        * Scripts/modules/buildsteps.py:
        * Scripts/modules/commands/upload.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52523 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Daniel Bates <dbates@rim.com>
dbates@webkit.org [Wed, 23 Dec 2009 07:54:31 +0000 (07:54 +0000)]
2009-12-22  Daniel Bates  <dbates@rim.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=32680

        Fixes an issue in builtDylibPathForName so that it returns the correct
        path to the built QtWebKit library on the Qt Windows port. Currently,
        it returns the path to the file named QtWebKit.dll but this file does
        not exist on the Qt Windows build. Instead, the file is named
        QtWebKit4.dll.

        * Scripts/webkitdirs.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Daniel Bates <dbates@webkit.org>
dbates@webkit.org [Wed, 23 Dec 2009 07:28:54 +0000 (07:28 +0000)]
2009-12-22  Daniel Bates  <dbates@webkit.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=32799

        Renames enums in RenderStyleConstants::EListStyleType to conform to
        the WebKit Code Style Guidelines.

        No functionality was changed. So, no new tests.

        * css/CSSPrimitiveValueMappings.h:
        (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
        (WebCore::CSSPrimitiveValue::operator EListStyleType):
        * rendering/RenderListItem.cpp:
        (WebCore::RenderListItem::styleDidChange):
        * rendering/RenderListMarker.cpp:
        (WebCore::listMarkerText):
        (WebCore::RenderListMarker::paint):
        (WebCore::RenderListMarker::calcPrefWidths):
        (WebCore::RenderListMarker::updateMargins):
        (WebCore::RenderListMarker::getRelativeMarkerRect):
        * rendering/style/RenderStyle.h:
        (WebCore::InheritedFlags::initialListStyleType):
        * rendering/style/RenderStyleConstants.h:
        (WebCore::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Marc-Antoine Ruel <maruel@chromium.org>
eric@webkit.org [Wed, 23 Dec 2009 06:08:45 +0000 (06:08 +0000)]
2009-12-22  Marc-Antoine Ruel  <maruel@chromium.org>

        Reviewed by Adam Roben.

        Remove the only dependency on perl-libwin32 in cygwin

        * Scripts/num-cpus:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 James Su <suzhe@chromium.org>
eric@webkit.org [Wed, 23 Dec 2009 03:59:24 +0000 (03:59 +0000)]
2009-12-22  James Su  <suzhe@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] Keyboard shortcut in dropdown not working.
        https://bugs.webkit.org/show_bug.cgi?id=32008

        Consider PlatformKeyboardEvent::Char type event as character type
        event on all platforms. It fixes the "type ahead find" feature
        of the popup listbox on Linux and Mac platforms.

        Merge the case-sensitive type ahead find fix from
        WebCore/dom/SelectElement.cpp. See
        https://bugs.webkit.org/show_bug.cgi?id=29103

        * platform/chromium/PopupMenuChromium.cpp:
        (WebCore::isCharacterTypeEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Carol Szabo <carol.szabo@nokia.com>
eric@webkit.org [Wed, 23 Dec 2009 02:26:12 +0000 (02:26 +0000)]
2009-12-22  Carol Szabo  <carol.szabo@nokia.com>

        Reviewed by Eric Seidel.

        [Qt] The "Search" input type is not rendered correctly when styled.
        https://bugs.webkit.org/show_bug.cgi?id=32417

        There are no new tests because only pixel tests may be affected by
        this change and those changes can only be done on the build bot.

        * platform/qt/RenderThemeQt.cpp:
        (WebCore::RenderThemeQt::paintSearchField):
        Since currently we paint the Search control as a QLineEdit, I have
        deferred painting to WebCore, as in the case of the text input so
        that the painting is consistent.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Kenneth Russell <kbr@google.com>
eric@webkit.org [Wed, 23 Dec 2009 02:19:04 +0000 (02:19 +0000)]
2009-12-22  Kenneth Russell  <kbr@google.com>

        Reviewed by Eric Seidel.

        [Chromium] Garbage in transparent regions of images uploaded as textures
        https://bugs.webkit.org/show_bug.cgi?id=32888

        * fast/canvas/webgl/bug-32888-expected.txt: Added.
        * fast/canvas/webgl/bug-32888.html: Added.
        * fast/canvas/webgl/resources/bug-32888-texture.png: Added.
2009-12-22  Kenneth Russell  <kbr@google.com>

        Reviewed by Eric Seidel.

        [Chromium] Garbage in transparent regions of images uploaded as textures
        https://bugs.webkit.org/show_bug.cgi?id=32888

        Test: fast/canvas/webgl/bug-32888.html (works in Safari / DumpRenderTree)

        * src/GraphicsContext3D.cpp:
        (WebCore::GraphicsContext3D::texImage2D):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Yongjun Zhang <yongjun.zhang@nokia.com>
eric@webkit.org [Wed, 23 Dec 2009 02:11:52 +0000 (02:11 +0000)]
2009-12-22  Yongjun Zhang  <yongjun.zhang@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        https://bugs.webkit.org/show_bug.cgi?id=32121
        [Qt] implement ImageDecoderQt::clearFrameBufferCache().

        * platform/graphics/qt/ImageDecoderQt.cpp:
        (WebCore::ImageDecoderQt::initializeDecoder):
        (WebCore::ImageDecoderQt::setData):
        (WebCore::ImageDecoderQt::clearFrameBufferCache):
        * platform/graphics/qt/ImageDecoderQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Nayan Kumar K <nayankk@gmail.com>
eric@webkit.org [Wed, 23 Dec 2009 02:04:31 +0000 (02:04 +0000)]
2009-12-22  Nayan Kumar K  <nayankk@gmail.com>

        Reviewed by Eric Seidel.

        Fix assert failure with WML pages having 'Refresh' task in
        'onenterforward' event.

        https://bugs.webkit.org/show_bug.cgi?id=30989

        Writing Layout test is not possible. Please refer to the discussion
        in bug id #30989. A sample test case is attached in BugZilla.

        * wml/WMLCardElement.cpp:
        (WebCore::WMLCardElement::handleIntrinsicEventIfNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Adam Langley <agl@google.com>
agl@chromium.org [Wed, 23 Dec 2009 01:34:28 +0000 (01:34 +0000)]
2009-12-22  Adam Langley  <agl@google.com>

        Build fix: not reviewed.

        r52504 (https://bugs.webkit.org/show_bug.cgi?id=32845) broke the
        Chromium build on Windows and Mac due to a misplaced #endif. The
        buildbot didn't find the problem because it runs on Linux.

        Have verified this patch with the author of the above.

        * src/GraphicsContext3D.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemoved DataGrid from results.
darin@apple.com [Wed, 23 Dec 2009 01:20:29 +0000 (01:20 +0000)]
Removed DataGrid from results.

* fast/dom/Window/window-properties-expected.txt: Removed DataGrid classes.
* fast/dom/prototype-inheritance-2-expected.txt: Ditto.
* fast/dom/prototype-inheritance-expected.txt: Ditto.
* fast/js/global-constructors-expected.txt: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoDisabled DataGrid tests.
darin@apple.com [Wed, 23 Dec 2009 01:18:25 +0000 (01:18 +0000)]
Disabled DataGrid tests.

* fast/dom/HTMLDataGridElement/DataGridColumns-basic.html: Removed.
* fast/dom/HTMLDataGridElement/DataGridColumns-basic.html-disabled: Copied from fast/dom/HTMLDataGridElement/DataGridColumns-basic.html.
* fast/dom/HTMLDataGridElement/DataGridColumns-dom-attributes.html: Removed.
* fast/dom/HTMLDataGridElement/DataGridColumns-dom-attributes.html-disabled: Copied from fast/dom/HTMLDataGridElement/DataGridColumns-dom-attributes.html.
* fast/dom/HTMLDataGridElement/DataGridColumns-dom.html: Removed.
* fast/dom/HTMLDataGridElement/DataGridColumns-dom.html-disabled: Copied from fast/dom/HTMLDataGridElement/DataGridColumns-dom.html.
* fast/dom/HTMLDataGridElement/DataGridDataSource-basic.html: Removed.
* fast/dom/HTMLDataGridElement/DataGridDataSource-basic.html-disabled: Copied from fast/dom/HTMLDataGridElement/DataGridDataSource-basic.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoASSERTION FAILED: m_numNodeListCaches
darin@apple.com [Wed, 23 Dec 2009 00:49:46 +0000 (00:49 +0000)]
ASSERTION FAILED: m_numNodeListCaches
https://bugs.webkit.org/show_bug.cgi?id=19526
<rdar://problem/7431572>

Reviewed by Sam Weinig.

WebCore:

Test: fast/dom/NodeList/adoptNode-node-list-cache.html

* dom/Node.cpp:
(WebCore::Node::setDocument): Call removeNodeListCache on the old
document and addNodeListCache on the new one if moving a node that
has node lists.

LayoutTests:

* fast/dom/NodeList/adoptNode-node-list-cache-expected.txt: Added.
* fast/dom/NodeList/adoptNode-node-list-cache.html: Added.
* fast/dom/NodeList/resources: Added.
* fast/dom/NodeList/resources/adoptNode-node-list-cache-subframe.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAnother try at fixing Windows build.
darin@apple.com [Wed, 23 Dec 2009 00:48:27 +0000 (00:48 +0000)]
Another try at fixing Windows build.

* WebKitPrefix.cpp: Touch it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFirst cut at fixing Windows build.
darin@apple.com [Wed, 23 Dec 2009 00:26:25 +0000 (00:26 +0000)]
First cut at fixing Windows build.

* WebCorePrefix.cpp: Touch it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago* pywebsocket/mod_pywebsocket: Added property svn:ignore to ignore
darin@apple.com [Tue, 22 Dec 2009 23:59:21 +0000 (23:59 +0000)]
* pywebsocket/mod_pywebsocket: Added property svn:ignore to ignore
generated ".pyc" files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTurn off datagrid by default, at least for all platforms Apple ships.
darin@apple.com [Tue, 22 Dec 2009 23:53:34 +0000 (23:53 +0000)]
Turn off datagrid by default, at least for all platforms Apple ships.
The datagrid implementation isn't ready for general web use yet.

Reviewed by Mark Rowe.

JavaScriptCore:

* Configurations/FeatureDefines.xcconfig: Turn off datagrid by default.

WebCore:

* Configurations/FeatureDefines.xcconfig: Turn off datagrid by default.
* WebCore.vcproj/WebCoreCommon.vsprops: Ditto.
* WebCore.vcproj/build-generated-files.sh: Ditto.
* page/DOMWindow.idl: Sorted things here to touch the file to give
the Windows build a slightly better chance of success.

WebKit/mac:

* Configurations/FeatureDefines.xcconfig: Turn off datagrid by default.

WebKit/win:

* WebKit.vcproj/WebKit.vcproj: Turn off datagrid by default.

WebKitTools:

* Scripts/build-webkit: Turn off datagrid by default.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 23:50:05 +0000 (23:50 +0000)]
2009-12-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Eric Seidel.

        Upstream platform-specific variant of
        plugins/get-url-with-blank-target.html . There is some uncertainty
        over which behavior is correct - Chromium's and Safari/Win in this
        test, or Safari/Mac in the generic test. See
        https://bugs.webkit.org/show_bug.cgi?id=32886 for the bug to converge
        the two.
        This patch is bug https://bugs.webkit.org/show_bug.cgi?32762 .

        * platform/chromium/plugins/get-url-with-blank-target-expected.txt: Added.
        * platform/chromium/plugins/get-url-with-blank-target.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 23:12:39 +0000 (23:12 +0000)]
2009-12-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Eric Seidel.

        Upstream a bunch of forms tests from chromium.org.
        https://bugs.webkit.org/show_bug.cgi?id=32755

        * fast/forms/basic-selects.html: Added.
        * fast/forms/basic-textareas-quirks.html: Added.
        * fast/forms/document-write-empty-expected.txt: Added.
        * fast/forms/document-write-empty.html: Added.
        * fast/forms/form-action-expected.txt: Added.
        * fast/forms/form-action.html: Added.
        * fast/forms/lazy-event-listener-scope-chain-expected.txt: Added.
        * fast/forms/lazy-event-listener-scope-chain.html: Added.
        * fast/forms/resources/logo.gif: Added.
        * fast/forms/selected-index-value-expected.txt: Added.
        * fast/forms/selected-index-value.html: Added.
        * fast/forms/style-display-none-expected.txt: Added.
        * fast/forms/style-display-none.html: Added.
        * platform/mac/fast/forms/basic-selects-expected.checksum: Added.
        * platform/mac/fast/forms/basic-selects-expected.png: Added.
        * platform/mac/fast/forms/basic-selects-expected.txt: Added.
        * platform/mac/fast/forms/basic-textareas-quirks-expected.checksum:
        * platform/mac/fast/forms/basic-textareas-quirks-expected.png:
        * platform/mac/fast/forms/basic-textareas-quirks-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Kenneth Russell <kbr@google.com>
eric@webkit.org [Tue, 22 Dec 2009 23:00:20 +0000 (23:00 +0000)]
2009-12-22  Kenneth Russell  <kbr@google.com>

        Reviewed by Eric Seidel.

        [Chromium] WebGL crashes intermittently on Linux
        https://bugs.webkit.org/show_bug.cgi?id=32845

        The dlclose'ing of libGL.so.1 and dlopen'ing of it each time a
        GraphicsContext3D was created was occasionally causing it to be
        re-mapped at a different base address. Since GLEW is not
        re-initialized every time, primarily for performance reasons, its
        cached function pointers were pointing to garbage. Stopped closing
        and re-opening libGL.so.1 each time; now it is loaded lazily, when
        the first 3D context is created. Also reused the X display
        connection since the GLX routines' correctness might hinge upon it
        not resulting in a change of GL implementation.

        * src/GraphicsContext3D.cpp:
        (WebCore::GraphicsContext3DInternal::GLConnection::chooseFBConfig):
        (WebCore::GraphicsContext3DInternal::GLConnection::createNewContext):
        (WebCore::GraphicsContext3DInternal::GLConnection::createPbuffer):
        (WebCore::GraphicsContext3DInternal::GLConnection::destroyPbuffer):
        (WebCore::GraphicsContext3DInternal::GLConnection::makeCurrent):
        (WebCore::GraphicsContext3DInternal::GLConnection::destroyContext):
        (WebCore::GraphicsContext3DInternal::GLConnection::getCurrentContext):
        (WebCore::GraphicsContext3DInternal::GLConnection::GLConnection):
        (WebCore::GraphicsContext3DInternal::GLConnection::tryLoad):
        (WebCore::GraphicsContext3DInternal::GLConnection::create):
        (WebCore::GraphicsContext3DInternal::GLConnection::~GLConnection):
        (WebCore::GraphicsContext3DInternal::GraphicsContext3DInternal):
        (WebCore::GraphicsContext3DInternal::~GraphicsContext3DInternal):
        (WebCore::GraphicsContext3DInternal::makeContextCurrent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Steve Block <steveblock@google.com>
eric@webkit.org [Tue, 22 Dec 2009 22:33:31 +0000 (22:33 +0000)]
2009-12-22  Steve Block  <steveblock@google.com>

        Reviewed by David Levin.

        Updates Android's scheduleDispatchFunctionsOnMainThread() to use new
        AndroidThreading class, rather than using JavaSharedClient directly.
        This fixes the current layering violation.
        https://bugs.webkit.org/show_bug.cgi?id=32651

        The pattern is copied from Chromium, which uses the ChromiumThreading
        class. This patch also fixes the style in ChromiumThreading.h.

        * wtf/android/AndroidThreading.h: Added. Declares AndroidThreading.
        * wtf/android/MainThreadAndroid.cpp: Modified
        (WTF::scheduleDispatchFunctionsOnMainThread): Uses AndroidThreading.
        * wtf/chromium/ChromiumThreading.h: Modified. Fixes style.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 22:26:05 +0000 (22:26 +0000)]
2009-12-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Eric Seidel.

        Add platform-specific chromium test.
        https://bugs.webkit.org/show_bug.cgi?id=32741

        * platform/chromium/fast/dom/interval-expected.txt: Added.
        * platform/chromium/fast/dom/interval.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Eric Seidel <eric@webkit.org>
eric@webkit.org [Tue, 22 Dec 2009 22:18:43 +0000 (22:18 +0000)]
2009-12-22  Eric Seidel  <eric@webkit.org>

        Reviewed by Darin Adler.

        validate-committer-lists fails when run from an SVN checkout
        https://bugs.webkit.org/show_bug.cgi?id=31974

        * Scripts/validate-committer-lists:
         - Make it print a warning message instead of throwing
           an exception when running from an SVN checkout.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix a couple of problems with UntypedPtrAndBitfield.
barraclough@apple.com [Tue, 22 Dec 2009 22:05:17 +0000 (22:05 +0000)]
Fix a couple of problems with UntypedPtrAndBitfield.

Reviewed by Sam Weinig.

Add a m_leaksPtr to reduce false positives from leaks in debug builds
(this isn't perfect because we'd like a solution for release builds,
but this is now at least as good as a PtrAndFlags would be).

Switch SmallStringsto use a regular string for the base, rather than
a static one.  UntypedPtrAndBitfield assumes all strings are at least
8 byte aligned; this migt not be true of static strings.  Shared buffers
are heap allocated, as are all UStringImpls other than static strings.
Static strings cannot end up being the owner string of substrings,
since the only static strings are length 0.

* runtime/SmallStrings.cpp:
(JSC::SmallStringsStorage::SmallStringsStorage):
* runtime/UStringImpl.h:
(JSC::UntypedPtrAndBitfield::UntypedPtrAndBitfield):
(JSC::UStringImpl::UStringImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber stamped by Dan Bernstein.
darin@apple.com [Tue, 22 Dec 2009 21:59:15 +0000 (21:59 +0000)]
Rubber stamped by Dan Bernstein.

* page/FrameView.h: Removed incorrect comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 21:44:00 +0000 (21:44 +0000)]
2009-12-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Eric Seidel.

        Add platform-specific chromium test.
        https://bugs.webkit.org/show_bug.cgi?id=32741

        * platform/chromium/fast/dom/interval-expected.txt: Added.
        * platform/chromium/fast/dom/interval.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 21:36:51 +0000 (21:36 +0000)]
2009-12-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Adam Barth.

        Attempting to upstream xss-inactive-closure test from the Chromium
        test suite. https://bugs.webkit.org/show_bug.cgi?id=32691 .

        * http/tests/security/listener/resources/xss-inactive-closure-child-2.html: Added.
        * http/tests/security/listener/resources/xss-inactive-closure-child.html: Added.
        * http/tests/security/listener/xss-inactive-closure-expected.txt: Added.
        * http/tests/security/listener/xss-inactive-closure.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Philippe Normand <pnormand@igalia.com>
eric@webkit.org [Tue, 22 Dec 2009 21:18:50 +0000 (21:18 +0000)]
2009-12-22  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Adam Barth.

        check-webkit-style should not warn about NULL usage in g_object_{get,set}
        https://bugs.webkit.org/show_bug.cgi?id=32858

        * Scripts/modules/cpp_style.py:
        * Scripts/modules/cpp_style_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Tue, 22 Dec 2009 20:09:42 +0000 (20:09 +0000)]
2009-12-22  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: simplify highlighter definition.

        https://bugs.webkit.org/show_bug.cgi?id=32869
WebCore:
        * inspector/front-end/CSSSourceSyntaxHighlighter.js:
        (WebInspector.CSSSourceSyntaxHighlighter):
        * inspector/front-end/JavaScriptSourceSyntaxHighlighter.js:
        (WebInspector.JavaScriptSourceSyntaxHighlighter.):
        (WebInspector.JavaScriptSourceSyntaxHighlighter):
        * inspector/front-end/SourceSyntaxHighlighter.js:
        (WebInspector.SourceSyntaxHighlighter.prototype.lex):
        * inspector/front-end/utilities.js:
        ():
LayoutTests:
        * inspector/syntax-highlight-css-expected.txt:
        * inspector/syntax-highlight-javascript-expected.txt:
        * inspector/syntax-highlight.js:
        (frontend_dumpSyntaxHighlight):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoReentrancy problem with selection in some edge cases.
darin@apple.com [Tue, 22 Dec 2009 19:26:29 +0000 (19:26 +0000)]
Reentrancy problem with selection in some edge cases.
https://bugs.webkit.org/show_bug.cgi?id=32842
rdar://problem/7449974

Reviewed by Sam Weinig.

WebCore:

Test: fast/forms/selection-layout-reentry-strange-case.html

Move the selection display update process done in the
selectionLayoutChanged function into the layout timer
instead of doing it immediately when selection changes occur.

* editing/SelectionController.cpp:
(WebCore::SelectionController::SelectionController):
Initialize m_needsDisplayUpdate to false.
(WebCore::SelectionController::setSelection): Call
the new setNeedsDisplayUpdate function instead of the old
badly named Frame::selectionLayoutChanged function.
(WebCore::SelectionController::setNeedsDisplayUpdate):
Set m_needsDisplayUpdate. If it is just becoming true, then
call FrameView::scheduleRelayout.

* editing/SelectionController.h: Added setNeedsDisplayUpdate,
needsDisplayUpdate, and m_needsDisplayUpdate.

* page/Frame.cpp:
(WebCore::Frame::setCaretVisible): Call setNeedsDisplayUpdate.
(WebCore::Frame::selectionLayoutChanged): Call
setNeedsDisplayUpdate to set it to false, since this is the
function that performs "selection display update". Later I want
to rename this function.

* page/FrameView.cpp:
(WebCore::FrameView::needsLayout): Add a new clause, since
we need a call to layout if needsDisplayUpdate is true.

LayoutTests:

* fast/forms/selection-layout-reentry-strange-case-expected.txt: Added.
* fast/forms/selection-layout-reentry-strange-case.html: Added.

* platform/mac/accessibility/frame-with-title-expected.txt: Updated since
the number of layouts is now different.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Tue, 22 Dec 2009 18:58:34 +0000 (18:58 +0000)]
2009-12-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        RVCT (__ARMCC_VERSION < 400000) does not provide strcasecmp and strncasecmp
        https://bugs.webkit.org/show_bug.cgi?id=32857

        Add implementation of strcasecmp and strncasecmp for RVCT < 4.0
        because earlier versions of RVCT 4.0 does not provide these functions.

        * wtf/StringExtras.cpp: Added.
        (strcasecmp):
        (strncasecmp):
        * wtf/StringExtras.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Tue, 22 Dec 2009 18:51:00 +0000 (18:51 +0000)]
2009-12-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        Allocate RemoteFontStream on the heap
        https://bugs.webkit.org/show_bug.cgi?id=32850

        RemoteFontStream is allocated on the stack, so its memory is freed
        immediately when FontCustomPlatformData::createFontCustomPlatformData
        returns.

        SkTypeface::CreateFromStream increments the reference count, but it keeps
        the pointer to the memory allocated on the stack which is not valid
        anymore.

        RemoteFontStream is a descendant of SkRefCount and SkRefCount::unref
        invokes SkDELETE(this) internally once the reference count reaches zero.
        This means that SkRefCount-ed instances must be allocated on the heap.

        * platform/graphics/chromium/FontCustomPlatformData.cpp:
        (WebCore::createFontCustomPlatformData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Tue, 22 Dec 2009 18:43:55 +0000 (18:43 +0000)]
2009-12-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        when condition COMPILER(RVCT) is fulfilled, we are trying to free a pointer to array allocated on the stack
        https://bugs.webkit.org/show_bug.cgi?id=32861

        Don't free the pointer for COMPILER(RVCT).

        * dom/XMLTokenizerLibxml2.cpp:
        (WebCore::XMLTokenizer::error):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Tue, 22 Dec 2009 18:36:41 +0000 (18:36 +0000)]
2009-12-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        Define ALWAYS_INLINE and WTF_PRIVATE_INLINE to __forceinline for RVCT
        https://bugs.webkit.org/show_bug.cgi?id=32853

        Use __forceinline forces RVCT to compile a C or C++ function
        inline. The compiler attempts to inline the function, regardless of
        the characteristics of the function.

        * wtf/AlwaysInline.h:
        * wtf/FastMalloc.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Dan Bernstein.
weinig@apple.com [Tue, 22 Dec 2009 17:39:43 +0000 (17:39 +0000)]
Rubber-stamped by Dan Bernstein.

Remove no longer necessary experimental single view #ifdefs from Widget and
ScrollView.

* page/Chrome.h:
* page/mac/ChromeMac.mm:
* platform/ScrollView.cpp:
(WebCore::ScrollView::wheelEvent):
* platform/Widget.cpp:
* platform/Widget.h:
* platform/mac/ScrollViewMac.mm:
* platform/mac/WidgetMac.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-22 Yaar Schnitman <yaar@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 17:33:26 +0000 (17:33 +0000)]
2009-12-22  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Darin Fisher.

        autoComplete and getElementById methods for WebKit API

        https://bugs.webkit.org/show_bug.cgi?id=32866

        * public/WebDocument.h:
        * public/WebInputElement.h:
        * src/WebDocument.cpp:
        (WebKit::WebDocument::getElementById):
        * src/WebInputElement.cpp:
        (WebKit::WebInputElement::autoComplete):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Tue, 22 Dec 2009 16:21:05 +0000 (16:21 +0000)]
2009-12-21  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Do not lose original request for non-main resource redirects.

        https://bugs.webkit.org/show_bug.cgi?id=32678

        * inspector/InspectorResource.cpp:
        (WebCore::InspectorResource::appendRedirect):
        * inspector/front-end/Resource.js:
        (WebInspector.Resource):
        * inspector/front-end/inspector.js:
        (WebInspector.updateResource):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Holger Freyther.
hausmann@webkit.org [Tue, 22 Dec 2009 13:13:23 +0000 (13:13 +0000)]
Rubber-stamped by Holger Freyther.

.:

Adjusted path to QtLauncher.

* WebKit.pro:

WebKit/qt:

Moved QtLauncher to WebKitTools/

* QtLauncher: Removed.
* QtLauncher/QtLauncher.pro: Removed.
* QtLauncher/main.cpp: Removed.

WebKitTools:

Moved QtLauncher from WebKit/qt.

* QtLauncher: Copied from WebKit/qt/QtLauncher.
* QtLauncher/QtLauncher.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Skip layoutTestController.notifyDone() related test until fix.
ossy@webkit.org [Tue, 22 Dec 2009 07:46:33 +0000 (07:46 +0000)]
[Qt] Skip layoutTestController.notifyDone() related test until fix.
https://bugs.webkit.org/show_bug.cgi?id=31626

* platform/qt/Skipped: http/tests/misc/set-window-opener-to-null.html added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 07:04:43 +0000 (07:04 +0000)]
2009-12-21  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Eric Seidel.

        Upstream chrome/fast/events/iframe-onload-remove-self-no-crash.html .
        https://bugs.webkit.org/show_bug.cgi?id=32742 .

        * fast/frames/iframe-onload-remove-self-no-crash-expected.txt: Added.
        * fast/frames/iframe-onload-remove-self-no-crash.html: Added.
        * fast/frames/resources/iframe-onload-remove-self-no-crash-child.html: Added.
        * fast/frames/resources/iframe-onload-remove-self-no-crash.js: Added.
        (onWindowLoad):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Tue, 22 Dec 2009 06:57:16 +0000 (06:57 +0000)]
2009-12-21  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Eric Seidel.

        Use ASSERT instead of assert in image-decoders
        https://bugs.webkit.org/show_bug.cgi?id=32852

        Other decoders use WTF's ASSERT macro.
        Follow this convention.

        * platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
        (WebCore::JPEGImageReader::decode):
        * platform/image-decoders/png/PNGImageDecoder.cpp:
        (WebCore::PNGImageDecoder::headerAvailable):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoProspective GTK build fix: Add UStringImpl.cpp/h to the build.
hausmann@webkit.org [Tue, 22 Dec 2009 06:54:19 +0000 (06:54 +0000)]
Prospective GTK build fix: Add UStringImpl.cpp/h to the build.

* GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix the Qt build, add UStringImpl.cpp to the build.
hausmann@webkit.org [Tue, 22 Dec 2009 06:53:15 +0000 (06:53 +0000)]
Fix the Qt build, add UStringImpl.cpp to the build.

* JavaScriptCore.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 22 Dec 2009 06:46:57 +0000 (06:46 +0000)]
2009-12-21  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Automate the process of calling prepare-ChangeLog
        https://bugs.webkit.org/show_bug.cgi?id=32816

        This patch automates the process of creating a bug and patch and
        uploading it to bugzilla.  The first cut just calls
        prepare-ChangeLog.

        This patch required some refactoring of upload.py to the Step
        model, but that's worth doing anyway.

        * Scripts/bugzilla-tool:
        * Scripts/modules/buildsteps.py:
        * Scripts/modules/commands/download.py:
        * Scripts/modules/commands/upload.py:
        * Scripts/modules/commands/upload_unittest.py:
        * Scripts/modules/mock_bugzillatool.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Tue, 22 Dec 2009 05:44:22 +0000 (05:44 +0000)]
2009-12-21  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Eric Seidel.

        Use vsnprintf instead of vasprintf for RVCT.
        https://bugs.webkit.org/show_bug.cgi?id=32851

        RVCT does not support vasprint, so use vsnprintf instead.

        * dom/XMLTokenizerLibxml2.cpp:
        (WebCore::XMLTokenizer::error):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Tue, 22 Dec 2009 04:47:46 +0000 (04:47 +0000)]
2009-12-21  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Invalid url should raise SYNTAX_ERR exception.
        https://bugs.webkit.org/show_bug.cgi?id=32700

        * websocket/tests/bad-sub-protocol-expected.txt: add CONSOLE MESSAGEs
        * websocket/tests/script-tests/url-parsing.js: add invalid url tests.
          Also changed url from ws://127.0.0.1/ to ws://127.0.0.1:8880/websocket/tests/simple to make it sure no errors on console message for these tests by not receiving unexpected response from 127.0.0.1:80
        * websocket/tests/url-parsing-expected.txt:
2009-12-21  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Invalid url should raise SYNTAX_ERR exception.
        https://bugs.webkit.org/show_bug.cgi?id=32700

        Check url is valid in WebSocket::connect.
        Also log the detailed reason of websocket failures to console.

        * websockets/WebSocket.cpp:
        (WebCore::encodeProtocolString):
        (WebCore::WebSocket::connect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Tue, 22 Dec 2009 04:31:13 +0000 (04:31 +0000)]
2009-12-21  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Adam Barth.

        Upstream a test from chromium.org.
        https://bugs.webkit.org/show_bug.cgi?id=32761

        * http/tests/misc/resources/content-iframe.html: Added.
        * http/tests/misc/set-window-opener-to-null-expected.txt: Added.
        * http/tests/misc/set-window-opener-to-null.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix for Windows Debug, if this breaks Release I will roll it out.
bweinstein@apple.com [Tue, 22 Dec 2009 03:36:11 +0000 (03:36 +0000)]
Build fix for Windows Debug, if this breaks Release I will roll it out.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoShould be final part of the Windows build fix.
bweinstein@apple.com [Tue, 22 Dec 2009 03:01:42 +0000 (03:01 +0000)]
Should be final part of the Windows build fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoDon't allocate a WebGeolocationControllerClient unless that compile-time setting...
mrowe@apple.com [Tue, 22 Dec 2009 03:00:08 +0000 (03:00 +0000)]
Don't allocate a WebGeolocationControllerClient unless that compile-time setting is enabled.

Reviewed by Sam Weinig.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdate WebKitSystemInterface.
mrowe@apple.com [Tue, 22 Dec 2009 02:49:00 +0000 (02:49 +0000)]
Update WebKitSystemInterface.

Reviewed by Samuel Weinig.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoHopefully closer to a build fix. JavaScriptCore will build now.
bweinstein@apple.com [Tue, 22 Dec 2009 02:41:40 +0000 (02:41 +0000)]
Hopefully closer to a build fix. JavaScriptCore will build now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows Build fix part 5.
barraclough@apple.com [Tue, 22 Dec 2009 02:25:39 +0000 (02:25 +0000)]
Windows Build fix part 5.
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix breakage of world introduced in build fix to r52463.
barraclough@apple.com [Tue, 22 Dec 2009 02:04:55 +0000 (02:04 +0000)]
Fix breakage of world introduced in build fix to r52463.

Reviewed by NOBODY (build fix).

* runtime/UStringImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMake sure that we check for the right thread.
andersca@apple.com [Tue, 22 Dec 2009 02:03:05 +0000 (02:03 +0000)]
Make sure that we check for the right thread.

Reviewed by Sam Weinig.

* platform/mac/WebCoreObjCExtras.mm:
(WebCoreObjCScheduleDeallocateOnMainThread):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agounreviewed windows build fix #4
barraclough@apple.com [Tue, 22 Dec 2009 01:28:25 +0000 (01:28 +0000)]
unreviewed windows build fix #4

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agounreviewed windows build fix #3
barraclough@apple.com [Tue, 22 Dec 2009 01:21:18 +0000 (01:21 +0000)]
unreviewed windows build fix #3

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agounreviewed windows build fix #2
barraclough@apple.com [Tue, 22 Dec 2009 01:08:27 +0000 (01:08 +0000)]
unreviewed windows build fix #2

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore: REGRESSION(4.0.4-ToT): Indent deletes non highlighted text in gmail.
enrica@apple.com [Tue, 22 Dec 2009 00:57:59 +0000 (00:57 +0000)]
WebCore: REGRESSION(4.0.4-ToT): Indent deletes non highlighted text in gmail.
<rdar://problem/7489326>
https://bugs.webkit.org/show_bug.cgi?id=32843

Reviewed by Maciej Stachowiak.

The fix for 7442387 did not handle the case where the end of paragraph
is not a descendant of the computed outer block.
Updated editing/execCommand/indent-with-style2.html to cover this case too.

* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::cloneParagraphUnderNewElement):

LayoutTests: REGRESSION(4.0.4-ToT): Indent deletes non highlighted text in gmail.
<rdar://problem/7489326>
https://bugs.webkit.org/show_bug.cgi?id=32843

Reviewed by Maciej Stachowiak.

The fix for 7442387 did not handle the case where the end of paragraph
is not a descendant of the computed outer block.
Updated editing/execCommand/indent-with-style2.html to cover this case too.

* editing/execCommand/indent-with-style2-expected.txt:
* editing/execCommand/indent-with-style2.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agounreviewed windows build fix #1
barraclough@apple.com [Tue, 22 Dec 2009 00:53:03 +0000 (00:53 +0000)]
unreviewed windows build fix #1

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agohttps://bugs.webkit.org/show_bug.cgi?id=32831
barraclough@apple.com [Tue, 22 Dec 2009 00:38:24 +0000 (00:38 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=32831
Replace UString::Rep implementation, following introduction of ropes to JSC.

Reviewed by Darin Adler.

JavaScriptCore:

    * Remove redundant overcapacity mechanisms.
    * Reduce memory cost of Rep's.
    * Add an inline storage mechanism akin to that in WebCore's StringImpl.

~1% Sunspider progression.

* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj:
* runtime/JSString.cpp:
(JSC::JSString::resolveRope):
* runtime/SmallStrings.cpp:
(JSC::SmallStringsStorage::SmallStringsStorage):
* runtime/UString.cpp:
(JSC::initializeUString):
(JSC::createRep):
(JSC::UString::createFromUTF8):
(JSC::UString::createUninitialized):
(JSC::UString::spliceSubstringsWithSeparators):
(JSC::UString::replaceRange):
(JSC::UString::ascii):
(JSC::UString::operator=):
(JSC::UString::toStrictUInt32):
(JSC::equal):
* runtime/UString.h:
(JSC::UString::isEmpty):
(JSC::UString::cost):
(JSC::makeString):
* runtime/UStringImpl.cpp: Added.
(JSC::UStringImpl::baseSharedBuffer):
(JSC::UStringImpl::sharedBuffer):
(JSC::UStringImpl::destroy):
(JSC::UStringImpl::computeHash):
* runtime/UStringImpl.h: Added.
(JSC::UntypedPtrAndBitfield::UntypedPtrAndBitfield):
(JSC::UntypedPtrAndBitfield::asPtr):
(JSC::UntypedPtrAndBitfield::operator&=):
(JSC::UntypedPtrAndBitfield::operator|=):
(JSC::UntypedPtrAndBitfield::operator&):
(JSC::UStringImpl::create):
(JSC::UStringImpl::createCopying):
(JSC::UStringImpl::createUninitialized):
(JSC::UStringImpl::data):
(JSC::UStringImpl::size):
(JSC::UStringImpl::cost):
(JSC::UStringImpl::hash):
(JSC::UStringImpl::computedHash):
(JSC::UStringImpl::setHash):
(JSC::UStringImpl::identifierTable):
(JSC::UStringImpl::setIdentifierTable):
(JSC::UStringImpl::ref):
(JSC::UStringImpl::deref):
(JSC::UStringImpl::allocChars):
(JSC::UStringImpl::copyChars):
(JSC::UStringImpl::computeHash):
(JSC::UStringImpl::null):
(JSC::UStringImpl::empty):
(JSC::UStringImpl::checkConsistency):
(JSC::UStringImpl::):
(JSC::UStringImpl::UStringImpl):
(JSC::UStringImpl::operator new):
(JSC::UStringImpl::bufferOwnerString):
(JSC::UStringImpl::bufferOwnership):
(JSC::UStringImpl::isStatic):

JavaScriptGlue:

* ForwardingHeaders/wtf/PossiblyNull.h: Added.
    - add forwarding header.

WebCore:

* ForwardingHeaders/runtime/UStringImpl.h: Added.
    - add forwarding header.

* platform/text/StringImpl.cpp:
(WebCore::StringImpl::ustring):
    - order of arguments to UString::Rep constructor for shared strings changed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Mon, 21 Dec 2009 23:50:00 +0000 (23:50 +0000)]
2009-12-21  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Adam Barth.

        Upstream a bunch of platform-specific Chromium plugin tests.
        https://bugs.webkit.org/show_bug.cgi?id=32751

        * platform/chromium/plugins/call-as-function-expected.txt: Added.
        * platform/chromium/plugins/call-as-function.html: Added.
        * platform/chromium/plugins/get-url-with-iframe-target-no-crash-expected.txt: Added.
        * platform/chromium/plugins/get-url-with-iframe-target-no-crash.html: Added.
        * platform/chromium/plugins/multiple-plugins-expected.txt: Added.
        * platform/chromium/plugins/multiple-plugins.html: Added.
        * platform/chromium/plugins/nested-plugin-objects-expected.txt: Added.
        * platform/chromium/plugins/nested-plugin-objects.html: Added.
        * platform/chromium/plugins/refcount-leaks-expected.txt: Added.
        * platform/chromium/plugins/refcount-leaks.html: Added.
        * platform/chromium/plugins/return-npobject-expected.txt: Added.
        * platform/chromium/plugins/return-npobject.html: Added.
        * platform/chromium/plugins/script-object-invoke-expected.txt: Added.
        * platform/chromium/plugins/script-object-invoke.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Dirk Pranke <dpranke@chromium.org>
eric@webkit.org [Mon, 21 Dec 2009 23:42:11 +0000 (23:42 +0000)]
2009-12-21  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Darin Adler.

        Upstream a test to ensure we handle variations on "javascript:" URL
        correctly.

        https://bugs.webkit.org/show_bug.cgi?id=32740

        * fast/dom/xss-DENIED-javascript-variations-expected.txt: Added.
        * fast/dom/xss-DENIED-javascript-variations.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdate WebKitSystemInterface.
mrowe@apple.com [Mon, 21 Dec 2009 23:39:16 +0000 (23:39 +0000)]
Update WebKitSystemInterface.

Reviewed by Samuel Weinig.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-18 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 21 Dec 2009 23:15:50 +0000 (23:15 +0000)]
2009-12-18  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam barth.

        Generate V8 bindings non-constructor callback declarations
        in CodeGeneratorV8.pm instead of manually declaring them in
        V8CustomBinding.h.

        https://bugs.webkit.org/show_bug.cgi?id=32638

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 21 Dec 2009 23:00:17 +0000 (23:00 +0000)]
2009-12-21  Nate Chapin  <japhet@chromium.org>

        Unreviewed, build fix.

        Cleared instead of deleted some layout test files, making them fail. Delete them for real.

        * http/tests/navigation/image-load-in-beforeunload-handler-expected.txt: Removed.
        * http/tests/navigation/image-load-in-beforeunload-handler.html: Removed.
        * http/tests/navigation/image-load-in-unload-handler-expected.txt: Removed.
        * http/tests/navigation/image-load-in-unload-handler.html: Removed.
        * http/tests/navigation/resources/wait-then-notify-done.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago* Scripts/do-webcore-rename: Did a rename, so checking in the
darin@apple.com [Mon, 21 Dec 2009 22:52:14 +0000 (22:52 +0000)]
* Scripts/do-webcore-rename: Did a rename, so checking in the
version of the script I used.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 21 Dec 2009 22:39:46 +0000 (22:39 +0000)]
2009-12-21  Nate Chapin  <japhet@chromium.org>

        Rubber-stamped by Darin Adler.

        Revert r52446 due to crashiness.

        https://bugs.webkit.org/show_bug.cgi?id=32839

        * loader/DocumentThreadableLoader.cpp:
        (WebCore::DocumentThreadableLoader::loadRequest):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::FrameLoader):
        (WebCore::FrameLoader::stopLoading):
        (WebCore::FrameLoader::loadURL):
        (WebCore::FrameLoader::loadWithDocumentLoader):
        (WebCore::FrameLoader::stopAllLoaders):
        (WebCore::FrameLoader::continueLoadAfterNavigationPolicy):
        (WebCore::FrameLoader::pageHidden):
        * loader/FrameLoader.h:
        (WebCore::FrameLoader::suppressOpenerInNewFrame):
        * loader/Request.cpp:
        (WebCore::Request::Request):
        (WebCore::Request::~Request):
        * loader/Request.h:
        * loader/SubresourceLoader.cpp:
        (WebCore::SubresourceLoader::create):
        * loader/SubresourceLoader.h:
        * loader/loader.cpp:
        (WebCore::Loader::load):
        (WebCore::Loader::cancelRequests):
        (WebCore::Loader::Host::servePendingRequests):
        (WebCore::Loader::Host::cancelPendingRequests):
        (WebCore::Loader::Host::cancelRequests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Adam Barth <abarth@webkit.org>
eric@webkit.org [Mon, 21 Dec 2009 22:12:12 +0000 (22:12 +0000)]
2009-12-21  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Update status-bubble to show all the queues
        https://bugs.webkit.org/show_bug.cgi?id=32838

        Also, move statusbubble over to use memcache.

        * QueueStatusServer/handlers/statusbubble.py:
        * QueueStatusServer/templates/statusbubble.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdd support for V8 Date binding.
pkasting@chromium.org [Mon, 21 Dec 2009 21:26:43 +0000 (21:26 +0000)]
Add support for V8 Date binding.
https://bugs.webkit.org/show_bug.cgi?id=32699

Patch by Kent Tamura <tkent@chromium.org> on 2009-12-19
Reviewed by Adam Barth.

This implements the same behavior as the recent change of
CodeGeneratorJS.pm and JSDOMBinding.cpp.

* bindings/scripts/CodeGeneratorV8.pm:
  Produce toWebCoreDate() or v8DateOrNull() for Date type.
* bindings/v8/V8Binding.cpp:
(WebCore::toWebCoreDate):
  Converts a JavaScript object to a double representing Date.
(WebCore::v8DateOrNull):
  Converts a double representing Date to a JavaScript Date object or null.
* bindings/v8/V8Binding.h: Declare toWebCoreDate() and v8DateOrNull().
* html/HTMLInputElement.idl: Delete V8_BINGIN exclusion for valueAsDate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Adam Barth <abarth@webkit.org>
eric@webkit.org [Mon, 21 Dec 2009 20:19:08 +0000 (20:19 +0000)]
2009-12-21  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Delete the boring "Empty queue" status messages in QueueStatusServer
        https://bugs.webkit.org/show_bug.cgi?id=32818

        * QueueStatusServer/app.yaml:
        * QueueStatusServer/cron.yaml:
        * QueueStatusServer/handlers/gc.py:
        * QueueStatusServer/main.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoDoc: Minor fixes to language.
hausmann@webkit.org [Mon, 21 Dec 2009 20:18:35 +0000 (20:18 +0000)]
Doc: Minor fixes to language.

Patch by David Boddie <dboddie@trolltech.com> on 2009-12-21
Reviewed by Simon Hausmann.

* Api/qwebpage.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRenamed m_sel to m_selection. Used do-webcore-rename.
darin@apple.com [Mon, 21 Dec 2009 20:15:17 +0000 (20:15 +0000)]
Renamed m_sel to m_selection. Used do-webcore-rename.

Reviewed by Mark Rowe.

* editing/SelectionController.cpp:
(WebCore::SelectionController::setSelection):
(WebCore::SelectionController::nodeWillBeRemoved):
(WebCore::SelectionController::willBeModified):
(WebCore::SelectionController::directionOfEnclosingBlock):
(WebCore::SelectionController::modifyExtendingRight):
(WebCore::SelectionController::modifyExtendingForward):
(WebCore::SelectionController::modifyMovingRight):
(WebCore::SelectionController::modifyMovingForward):
(WebCore::SelectionController::modifyExtendingLeft):
(WebCore::SelectionController::modifyExtendingBackward):
(WebCore::SelectionController::modifyMovingLeft):
(WebCore::SelectionController::modifyMovingBackward):
(WebCore::SelectionController::modify):
(WebCore::SelectionController::expandUsingGranularity):
(WebCore::SelectionController::xPosForVerticalArrowNavigation):
(WebCore::SelectionController::setBase):
(WebCore::SelectionController::setExtent):
(WebCore::SelectionController::layout):
(WebCore::SelectionController::caretRenderer):
(WebCore::SelectionController::invalidateCaretRect):
(WebCore::SelectionController::paintCaret):
(WebCore::SelectionController::debugRenderer):
(WebCore::SelectionController::contains):
(WebCore::SelectionController::selectAll):
(WebCore::SelectionController::formatForDebugger):
(WebCore::SelectionController::showTreeForThis):
* editing/SelectionController.h:
* editing/gtk/SelectionControllerGtk.cpp:
(WebCore::SelectionController::notifyAccessibilityForSelectionChange):
* editing/mac/SelectionControllerMac.mm:
(WebCore::SelectionController::notifyAccessibilityForSelectionChange):
Let the script do its thing. Made sure no tests broke.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Carol Szabo <carol.szabo@nokia.com>
eric@webkit.org [Mon, 21 Dec 2009 20:11:47 +0000 (20:11 +0000)]
2009-12-21  Carol Szabo  <carol.szabo@nokia.com>

        Reviewed by Darin Adler.

        Inconsistent use of counterName and identifier in CSS counter code and loss of information about the counter type.
        https://bugs.webkit.org/show_bug.cgi?id=31814

        No new tests because there are no functional changes in this patch.

        * rendering/CounterNode.cpp:
        (WebCore::CounterNode::CounterNode):
        (WebCore::CounterNode::computeCountInParent):
        (WebCore::showTreeAndMark):
        * rendering/RenderCounter.cpp:
        (WebCore::planCounter):
        (WebCore::findPlaceForCounter):
        (WebCore::makeCounterNode):
        (WebCore::RenderCounter::originalText):
        Changed to use identifier instead of counterName and actsAsReset or hasResetType, as appropriate instead of isReset.
        * rendering/CounterNode.h:
        (WebCore::CounterNode::actsAsReset):
        (WebCore::CounterNode::hasResetType):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-21 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Mon, 21 Dec 2009 20:00:55 +0000 (20:00 +0000)]
2009-12-21  Dirk Schulze  <krit@webkit.org>

        Reviewed by Darin Adler and Nikolas Zimmermann.

        Speed-up SVG Masking
        https://bugs.webkit.org/show_bug.cgi?id=32738

        This patch makes SVG Masking faster. The luminance calculaton of the ImageBuffer,
        that is created by pixel manipulation, got optimized. The ImageBuffer and it's
        luminance is created once now, not on every call of applyMask.
        The size of the intermediate ImageBuffer depends on the visible area now and is
        clipped by the mask rect.

        The patch doesn't change functionality so no new tests needed.
        The new ImageBuffer handling in Mask causes two updates of test results. The pixel
        tests and LayoutTests pass, but the checksum doesn't match.

        * svg/SVGMaskElement.cpp:
        (WebCore::SVGMaskElement::drawMaskerContent):
        * svg/graphics/SVGResourceMasker.cpp:
        (WebCore::SVGResourceMasker::applyMask):

2009-12-21  Dirk Schulze  <krit@webkit.org>

        Reviewed by Darin Adler and Nikolas Zimmermann.

        Speed-up SVG Masking
        https://bugs.webkit.org/show_bug.cgi?id=32738

        Update pixel test results for two SVG masking operations. They don't fail
        and there is no noticable difference between the old and new results. But
        Masking uses smaller ImageBuffers now and that changed the checksum.

        * platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.checksum:
        * platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.png:
        * platform/mac/svg/batik/masking/maskRegions-expected.checksum:
        * platform/mac/svg/batik/masking/maskRegions-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52449 268f45cc-cd09-0410-ab3c-d52691b4dbfc