2010-03-19  Dirk Pranke <dpranke@chromium.org>

        Reviewed by Dimitri Glazkov.

        Land the chromium-mac LayoutTest baselines (file list ommitted due to
        length).

        https://bugs.webkit.org/show_bug.cgi?id=28397

        * platform/chromium-mac: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56251 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by David Kilzer.

        Assertion failure in media/video-controls-with-mutation-event-handler.html
        https://bugs.webkit.org/show_bug.cgi?id=36376

        Fix crashing test. Underlying the media element's volume slider is a
        range input, which sanitizes values before expected on the media control.

        * rendering/RenderMedia.cpp:
        (WebCore::RenderMedia::createVolumeSlider):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56250 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 of the new HTML5 loading events need to be asynchronous.

Reviewed by Darin Adler.

Laying the groundwork for:
https://bugs.webkit.org/show_bug.cgi?id=36201
https://bugs.webkit.org/show_bug.cgi?id=36202
https://bugs.webkit.org/show_bug.cgi?id=36334
https://bugs.webkit.org/show_bug.cgi?id=36335

Document already had an asynchronous event delivery mechanism for storage events, so
we can repurpose that for all async events.

No new tests. (No change in behavior)

* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::implicitClose): Use Document::schedule* for the related events.
(WebCore::Document::enqueueEvent): Renamed from enqueueStorageEvent
(WebCore::Document::pendingEventTimerFired): Renamed from "storageEventTimerFired"
(WebCore::Document::statePopped): Use Document::schedulePopstateEvent
(WebCore::Document::enqueuePageshowEvent): All Pageshow events are piped through here.
  This will be made asynchronous in a separate patch.
(WebCore::Document::enqueueHashchangeEvent): All Hashchange events are piped through here.
  This will be made asynchronous in a separate patch.
(WebCore::Document::enqueuePopstateEvent): All Popstate events are piped through here.
  This will be made asynchronous in a separate patch.
* dom/Document.h:
(WebCore::):

* history/CachedFrame.cpp:
(WebCore::CachedFrameBase::restore): Use Document::enqueuePageshowEvent

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadInSameDocument): Use Document::enqueueHashchangeEvent

* storage/StorageEventDispatcher.cpp:
(WebCore::StorageEventDispatcher::dispatch): Use Document::enqueueEvent

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56249 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        Help sheriff-bot avoid warning about flaky tests (and add more unit testing)
        https://bugs.webkit.org/show_bug.cgi?id=36354

        * Scripts/webkitpy/buildbot.py:
         - Make Build creation easier to Mock and test
        * Scripts/webkitpy/buildbot_unittest.py:
         - Test finding green to red transitions and suspect revisions
        * Scripts/webkitpy/commands/queries.py:
         - Make what-broke note when builders have only failed once.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56248 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=36337
        Log an error when an OOP plug-in sends an unknown object id

        Making these LOG_ERROR and not ASSERTs, because I don't want early returns to look
        temporary or redundant.

        * Plugins/Hosted/NetscapePluginInstanceProxy.mm:
        (WebKit::NetscapePluginInstanceProxy::evaluate):
        (WebKit::NetscapePluginInstanceProxy::invoke):
        (WebKit::NetscapePluginInstanceProxy::invokeDefault):
        (WebKit::NetscapePluginInstanceProxy::construct):
        (WebKit::NetscapePluginInstanceProxy::getProperty):
        (WebKit::NetscapePluginInstanceProxy::setProperty):
        (WebKit::NetscapePluginInstanceProxy::removeProperty):
        (WebKit::NetscapePluginInstanceProxy::hasProperty):
        (WebKit::NetscapePluginInstanceProxy::hasMethod):
        (WebKit::NetscapePluginInstanceProxy::enumerate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56247 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Darin Adler and Brady Eidson.

        First step toward:
        https://bugs.webkit.org/show_bug.cgi?id=36328

        This is the first step torward eliminating the WebKit WebNullPluginView class. The responsibility for this will soon
        be in platform-independent code in WebCore. The plan is to change RenderEmbeddedObject and give it the capability of
        drawing the missing plug-in text.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::requestObject): Renamed the renderer parameter to requestEmbeddedObject and made it
        a RenderEmbeddedObject.
        (WebCore::FrameLoader::loadPlugin): Likewise.
        * loader/FrameLoader.h: Updated loadPlugin() and requestObject() method signatures accordingly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Dimitri Glazkov.

        Fix the rebaselining tool, which was broken by r36324 when I
        added the concept of overridding expectations.

        https://bugs.webkit.org/show_bug.cgi?id=36374

        * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Darin Fisher  <darin@chromium.org>

        Reviewed by Dimitri Glazkov.

        [chromium] querying the current history item during a back/forward
        navigation should not clobber the scroll offset, etc.
        https://bugs.webkit.org/show_bug.cgi?id=36347

        We should only allow the current history item to be modified if we are
        no longer loading or if the load is a new navigation (i.e., not a
        back/forward/reload variant).

        * src/WebFrameImpl.cpp:
        (WebKit::WebFrameImpl::currentHistoryItem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Simon Hausmann.

        [Qt] Disable TILED_BACKING_STORE if Qt version is earlier than Qt4.6
        https://bugs.webkit.org/show_bug.cgi?id=36348

        Backing store implementation for QtWebKit requires at least Qt
        version 4.6.

        No new tests as there is no new functionality.

        * WebCore.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by David Kilzer.

        <input type=range> does not validate correctly without a renderer and the tests are incorrect
        https://bugs.webkit.org/show_bug.cgi?id=36259

        Setting value attribute on an <input type=range> to an out-of-range value fires oninput
        https://bugs.webkit.org/show_bug.cgi?id=16990

        Part 2 of 2: When setting the range element's value, overflows and underflows
        are automatically sanitized to valid values. Moved the general case
        sanitization code out of the Renderer into HTMLInputElement::sanitizeValue.

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::value): when getting a default value on reset() ensure the defaultValue is provided
        (WebCore::HTMLInputElement::sanitizeValue): clamp the value within the max/min/step range constraints
        * html/StepRange.cpp:
        (WebCore::StepRange::StepRange): allow const element in the constructor
        (WebCore::StepRange::clampValue): clamp from a String value
        * html/StepRange.h:
        (WebCore::StepRange::defaultValue): easy calculation of the default value for max/min/step range constraints
        * rendering/RenderSlider.cpp:
        (WebCore::RenderSlider::updateFromElement): no longer handle the general case sanitization in the renderer

2010-03-19  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by David Kilzer.

        <input type=range> does not validate correctly without a renderer and the tests are incorrect
        https://bugs.webkit.org/show_bug.cgi?id=36259

        Setting value attribute on an <input type=range> to an out-of-range value fires oninput
        https://bugs.webkit.org/show_bug.cgi?id=16990

        When setting the range element's value, overflows and underflows are
        automatically sanitized to valid values. Fixed the behavior and removed
        tests that were testing for improper behavior.

        * fast/forms/ValidityState-rangeOverflow-expected.txt:
        * fast/forms/ValidityState-rangeUnderflow-expected.txt:
        * fast/forms/input-stepup-stepdown-expected.txt:
        * fast/forms/script-tests/ValidityState-rangeOverflow.js:
        * fast/forms/script-tests/ValidityState-rangeUnderflow.js:
        * fast/forms/script-tests/input-stepup-stepdown.js:
        * fast/forms/script-tests/validationMessage.js:
        * fast/forms/validationMessage-expected.txt:

          Added test verifying that bug 16990 was fixed as well.

        * fast/forms/range-input-dynamic-oninput-expected.txt: Added.
        * fast/forms/range-input-dynamic-oninput.html Added.

          Improved test because handling of this case changed.

        * fast/forms/range-reset-expected.txt:
        * fast/forms/range-reset.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56242 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by David Kilzer.

        <input type=range> does not validate correctly without a renderer and the tests are incorrect
        https://bugs.webkit.org/show_bug.cgi?id=36259

        Part 1 of 2: Refactoring the SliderRange struct out of RenderSlider
        into a more appropriate place. Changed the named to StepRange. Changed
        from a struct to a class.

          Added new files to the build.

        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:

          Renamed and moved class SliderRange to StepRange.

        * html/StepRange.cpp: Added.
        (WebCore::StepRange::StepRange):
        (WebCore::StepRange::clampValue):
        (WebCore::StepRange::valueFromElement):
        (WebCore::sliderPosition):
        * html/StepRange.h: Added.
        (WebCore::StepRange::proportionFromValue):
        (WebCore::StepRange::valueFromProportion):
        * rendering/RenderSlider.cpp:
        (WebCore::RenderSlider::updateFromElement): updated to use StepRange
        (WebCore::RenderSlider::setValueForPosition): updated to use StepRange

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Darin Fisher  <darin@chromium.org>

        Fix expected results for fast/loader/frame-src-change-added-to-history.html

        * fast/loader/frame-src-change-added-to-history-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36366
Repaint loop when painting using scaled CTMs with different translations

Patch by Antti Koivisto <koivisto@iki.fi> on 2010-03-19
Reviewed by Simon Hausmann.

The code in RenderBoxModelScaleObserver::shouldPaintBackgroundAtLowQuality
tests if the scaling has changed from the previous value by comparing the
transformations. The test fails if the scale is the same but the translation
changes. This can lead to infinite repaint loop if the document is painted
in pieces using different translations (for example for tiling).

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelScaleObserver::shouldPaintBackgroundAtLowQuality):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        Need a layout test for finding yensign from documents with Japanese encodings
        https://bugs.webkit.org/show_bug.cgi?id=36288

        * editing/selection/find-yensign-and-backslash-expected.txt: Added.
        * editing/selection/find-yensign-and-backslash.html: Added.
        * platform/gtk/editing/selection/find-yensign-and-backslash-expected.txt: Added.
        * platform/qt/editing/selection/find-yensign-and-backslash-expected.txt: Added.
        * platform/win/editing/selection/find-yensign-and-backslash-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: editing shorthands does not always work.

        https://bugs.webkit.org/show_bug.cgi?id=36362

        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::applyStyleText):
        (WebCore::InspectorDOMAgent::populateObjectWithStyleProperties):
        (WebCore::InspectorDOMAgent::shorthandValue):
        * inspector/InspectorDOMAgent.h:
        * inspector/front-end/StylesSidebarPane.js:
        (WebInspector.StylePropertyTreeElement.prototype):
        * inspector/front-end/inspector.css:
        (.section .properties li.disabled .enabled-button):
        * inspector/front-end/inspector.js:
        (WebInspector.startEditing.editingCommitted):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Don't replace clip when drawing complex text

Reviewed by Simon Hausmann.

* platform/graphics/qt/FontQt.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56236 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Misc bug fixes to make the SheriffBot actually work
        https://bugs.webkit.org/show_bug.cgi?id=36355

        With these changes, I can actually run the sheriff-bot from start to
        finish.

        * Scripts/webkitpy/irc/ircproxy.py:
        * Scripts/webkitpy/patch/patcher.py:
        * Scripts/webkitpy/statusserver.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56235 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed.

        Actually import the sheriff-bot command so we can run it.  Also, move
        the bot to #webkit-test so it doesn't cause a ruckus while we test it.

        * Scripts/webkitpy/patch/patcher.py:
        * Scripts/webkitpy/irc/ircbot.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56234 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Second cut at SheriffBot
        https://bugs.webkit.org/show_bug.cgi?id=36353

        This patch should contain a complete SheriffBot that's capable of
        saying reasonable things on IRC.  I had to refactor the use of
        CommitInfo to make the SheriffBot testable, but I did the minimum
        necessary.  We should grow webkitcheckout over time to contain the
        knowledge of ChangeLogs from scm.

        * Scripts/webkitpy/commands/sheriffbot.py:
        * Scripts/webkitpy/commands/sheriffbot_unittest.py:
        * Scripts/webkitpy/mock_bugzillatool.py:
        * Scripts/webkitpy/patch/patcher.py:
        * Scripts/webkitpy/webkitcheckout.py: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed.

        Fix Hyatt's IRC nickname.

        * Scripts/webkitpy/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed.

        Add IRC nicknames for the non-reviewer committers.

        * Scripts/webkitpy/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56231 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed.

        Add IRC nicknames for the non-reviewer committers.

        * Scripts/webkitpy/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed.

        More reviewer IRC nicknames.

        * Scripts/webkitpy/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed.

        Add a bunch of IRC nicknames for reviewers.

        * Scripts/webkitpy/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56228 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Zoltan Horvath  <zoltan@webkit.org>

        Reviewed by Oliver Hunt.

        Added USE_SYSTEM_MALLOC flag to build-webkit
        https://bugs.webkit.org/show_bug.cgi?id=21272

        Add system-alloc flag to build-webkit. It makes easy to switch
        between system allocator and TCmalloc.

        * Scripts/build-webkit:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56227 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Unreviewed "build" fix.

        Of course, I caused a regression in the file that isn't tested.  :(

        * Scripts/webkitpy/statusserver.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56226 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        Simplify BuildBot core builder code for easier re-use
        https://bugs.webkit.org/show_bug.cgi?id=36350

        I simply couldn't see anything through all this Yak-hair.

        * Scripts/webkitpy/buildbot.py:
        * Scripts/webkitpy/commands/queries.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56225 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        committers.py should know IRC nicknames
        https://bugs.webkit.org/show_bug.cgi?id=36349

        I'll add the actual nicknames in another patch.

        * Scripts/webkitpy/committers.py:
        * Scripts/webkitpy/committers_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56224 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Darin Fisher  <darin@chromium.org>

        Reviewed by Adam Barth.

        If a frame is already in the document, then setting its "src" attribute
        should add the resultant navigation to history.

        https://bugs.webkit.org/show_bug.cgi?id=9166

        This behavior is specified in section 4.8.2 of the HTML5 spec.

        Test: fast/loader/frame-src-change-added-to-history.html

        * html/HTMLFrameElementBase.cpp:
        (WebCore::HTMLFrameElementBase::openURL):
        (WebCore::HTMLFrameElementBase::setLocation):
        * html/HTMLFrameElementBase.h:
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::requestFrame):
        * loader/FrameLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56223 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36318

Patch by Anders Bakken  <Anders Bakken <agbakken@gmail.com>> on 2010-03-18
Reviewed by Simon Hausmann.

QtLauncher runs as a GuiClient by default in Qt Embedded which will
make it try to connect to an existing GuiServer. This patch makes it
run like a stand-alone app.

* QtLauncher/main.cpp:
(LauncherApplication::LauncherApplication):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed.

[Qt] Platform specific expected file updated after r56195

* platform/qt/fast/dom/Window/window-properties-expected.txt: updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        Move find_green_to_red_transition out of "what-broke" onto Builder for easier re-use
        https://bugs.webkit.org/show_bug.cgi?id=36345

        * Scripts/webkitpy/buildbot.py:
        * Scripts/webkitpy/commands/queries.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-19  Adam Barth  <abarth@webkit.org>

        Reviewed by Adam Barth.

        Actually pass the IRC password to the IRC object
        https://bugs.webkit.org/show_bug.cgi?id=36346

        I wanted to do this before, but both patches were in flight.  This
        patch finally closes the loop and makes the IRCProxy system complete.

        * Scripts/webkitpy/patch/patcher.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Add a StatusServer front end to the SVNRevision table on QueueStatusServer
        https://bugs.webkit.org/show_bug.cgi?id=36344

        No test because Browser was too hard to mock.  :(  I couldn't figure
        out how to make Mock be a dictionary as well as an object.

        * Scripts/webkitpy/statusserver.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Philip Chimento  <philip.chimento@gmail.com>

        Reviewed by Oliver Hunt.

        Setting the GObject WebKitWebView property 'window-features' to NULL
        causes a crash.
        https://bugs.webkit.org/show_bug.cgi?id=36144

        * tests/testwebview.c: Add unit test for this bug.
        * webkit/webkitwebview.cpp: Don't allow the 'window-features' property
        to be set to NULL.
        * webkit/webkitwebwindowfeatures.cpp:
        (webkit_web_window_features_equal): Don't examine the members of either
        web_window_features argument if either is NULL, just return that they
        are not equal. Additionally, if they are the same object, return that
        they are equal.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        Split out CommitInfo class and add unit tests
        https://bugs.webkit.org/show_bug.cgi?id=36343

        Move more logic out of "what-broke" into a shared CommitInfo
        class so that it can be used by other commands and unit tested.

        * Scripts/webkitpy/commands/queries.py:
        * Scripts/webkitpy/commitinfo.py: Added.
        * Scripts/webkitpy/commitinfo_unittest.py: Added.
        * Scripts/webkitpy/unittests.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Tom Callaway  <tcallawa@redhat.com>

        Reviewed by Darin Adler.

        Bug 35429: Fix compile on SPARC64
        https://bugs.webkit.org/show_bug.cgi?id=35429

        * wtf/Platform.h: Set WTF_USE_JSVALUE64 for SPARC64

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore: Fix a minor crash with mismatched array sizes in SVG animation
elements.

Patch by Chris Evans <cevans@chromium.org> on 2010-03-18
Reviewed by Nikolas Zimmermann.

https://bugs.webkit.org/show_bug.cgi?id=35606

Test: svg/animations/keypoints-mismatch.svg

* svg/SVGAnimationElement.cpp:
(WebCore::SVGAnimationElement::startedActiveInterval):
Globally apply validations relevant to all animation types.

LayoutTests: Add test for minor SVG animation crashers.
See https://bugs.webkit.org/show_bug.cgi?id=35606

Patch by Chris Evans <cevans@chromium.org> on 2010-03-18
Reviewed by Nikolas Zimmermann.

* svg/animations/keypoints-mismatch.svg: Added.
* svg/animations/keypoints-mismatch-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Eric Seidel.

        add new results for failing win/linux test after http://trac.webkit.org/changeset/56175
        https://bugs.webkit.org/show_bug.cgi?id=36329

        Missed this test in r56175 since it does a drag on win/linux, but not a mac.
        The new results correctly do the smart-drag behavior of doing a smart-delete
        on drag regardless of whether the selection granularity is word granularity.

        * platform/win/editing/selection/drag-text-delay-expected.checksum: Added.
        * platform/win/editing/selection/drag-text-delay-expected.png: Added.
        * platform/win/editing/selection/drag-text-delay-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Vangelis Kokkevis  <vangelis@chromium.org>

        Reviewed by Adam Barth.

        Implement software composited graphics layers in Chromium using the Skia library.
        https://bugs.webkit.org/show_bug.cgi?id=35557
        This is an initial step in the implementation. Layer compositing is functioning
        but not optimized in any way. Subsesquent check-ins will be necessary to fine tune
        it.

        Test: No new exposed functionality so no new tests.

        * WebCore.gypi:
          Added new source files to the chromium build
        * platform/graphics/GraphicsLayer.h:
          Added necessary typedef's and forward declarations for Chromium.
        * platform/graphics/skia/GraphicsLayerSkia.cpp: Added.
        * platform/graphics/skia/GraphicsLayerSkia.h: Added.
          Declaration and implementation of the platform-specific GraphicsLayer class.
        * platform/graphics/skia/LayerRendererSkia.cpp: Added.
        * platform/graphics/skia/LayerRendererSkia.h: Added.
          Declaration and implementation of the Skia-based software compositor.
        * platform/graphics/skia/LayerSkia.cpp: Added.
        * platform/graphics/skia/LayerSkia.h: Added.
          Declaration and implementation of a compositable layer that uses a Skia canvas
          for backing store.
2010-03-18  Vangelis Kokkevis  <vangelis@chromium.org>

        Reviewed by Adam Barth.

        Adding support for the ACCELERATED_COMPOSITING render path to Chromium.
        https://bugs.webkit.org/show_bug.cgi?id=35557
        Currently compositing of layers is performed via s/w compositor which relies on Skia. This is an initial check-in
        and it's only been tested on Windows. Compiling the code requires seting "use_accelerated_compositing=1"
        to GYP_DEFINES. The update of layer contents and compositing is fairly inefficient but this will be fixed in
        subsequent check-ins.

        * src/ChromeClientImpl.cpp:
        (WebKit::ChromeClientImpl::attachRootGraphicsLayer):
        (WebKit::ChromeClientImpl::scheduleCompositingLayerSync):
        * src/ChromeClientImpl.h:
        (WebKit::ChromeClientImpl::setNeedsOneShotDrawingSynchronization):
         Added methods required by the RenderLayerCompositor

        * src/WebFrameImpl.cpp:
        (WebKit::WebFrameImpl::paintWithContext):
        (WebKit::WebFrameImpl::paint):
        * src/WebFrameImpl.h:
        Split WebFrameImpl::paint() into two methods to make it possible to call the paint
        routine with an existing GraphicsContext, which is necessary for painting the root layer into
        its own backing surface.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::WebViewImpl):
        (WebKit::WebViewImpl::paint):
        Modified method to handle the accelerated compositing path. Now, when doing accelerated compositing,
        paint() results in repainting the contents of the root layer and then doing a composite operation.
        (WebKit::WebViewImpl::setRootGraphicsLayer):
        (WebKit::WebViewImpl::setAcceleratedCompositing):
        (WebKit::WebViewImpl::updateRootLayerContents):
        (WebKit::WebViewImpl::setRootLayerNeedsDisplay):
        * src/WebViewImpl.h:
        (WebKit::WebViewImpl::isAcceleratedCompositing):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Fix LayoutTests/http/tests/appcache/max-size.html
        https://bugs.webkit.org/show_bug.cgi?id=36207

        * platform/qt/Skipped:

2010-03-18  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Fix LayoutTests/http/tests/appcache/max-size.html
        https://bugs.webkit.org/show_bug.cgi?id=36207

        Implement setAppCacheMaximumSize() for Qt.

        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::setAppCacheMaximumSize):
        * DumpRenderTree/qt/LayoutTestControllerQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        QueueStatusServer should be able to log broken bots
        https://bugs.webkit.org/show_bug.cgi?id=36341

        We need to add a new table to the QueueStatusServer to store persistent
        information for the SheriffBot.  The new table will keep track of which
        bots each SVN revision broke.

        * QueueStatusServer/handlers/__init__.py:
        * QueueStatusServer/handlers/svnrevision.py: Added.
        * QueueStatusServer/handlers/updatebase.py: Added.
        * QueueStatusServer/handlers/updatestatus.py:
        * QueueStatusServer/handlers/updatesvnrevision.py: Added.
        * QueueStatusServer/index.yaml:
        * QueueStatusServer/main.py:
        * QueueStatusServer/model/__init__.py:
        * QueueStatusServer/model/svnrevision.py: Added.
        * QueueStatusServer/templates/updatesvnrevision.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Darin Adler.

        Add a new test for testing whether the media type is restores
        properly after printing a page (changing media type to 'print').

        Add expected result and skip the test for platform that do not
        implement setMediaType().

        * fast/media/print-restores-previous-mediatype-expected.txt: Added.
        * fast/media/print-restores-previous-mediatype.html: Added.
        * platform/gtk/Skipped:
        * platform/mac/Skipped:
        * platform/win/Skipped:
2010-03-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Darin Adler.

        Make setPrinting() with printing=false, restore the previous
        media type in use.

        Test: fast/media/print-restores-previous-mediatype.html

        * page/Frame.cpp:
        (WebCore::Frame::setPrinting):
        * page/FrameView.cpp:
        (WebCore::FrameView::adjustMediaTypeForPrinting):
        * page/FrameView.h:
2010-03-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Darin Adler.

        Make it possible for the Qt DRT to set the media type from
        the LayoutTestController.

        * Api/qwebframe.cpp:
        (qt_drt_setMediaType):
2010-03-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Darin Adler.

        Add a new method to the Qt LayoutTestController for
        changing media type and make the DRT support dry-run printing.

        * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
        (WebCore::DumpRenderTree::DumpRenderTree):
        (WebCore::DumpRenderTree::dryRunPrint):
        * DumpRenderTree/qt/DumpRenderTreeQt.h:
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::setMediaType):
        * DumpRenderTree/qt/LayoutTestControllerQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Joe Ligman  <joseph.ligman@nokia.com>

        Reviewed by Simon Hausmann.

        [Qt] New API scrollRecursively has several problems.
        https://bugs.webkit.org/show_bug.cgi?id=35873

        Remove scrollRecursively from the Qt 4.7 API
        Update the internal API to accept a hit test position
        for nested scrolling

        * Api/qwebframe.cpp:
        (webframe_scrollOverflow):
        (qtwebkit_webframe_scrollRecursively):
        * Api/qwebframe.h:
        * Api/qwebframe_p.h:
        * tests/qwebframe/tst_qwebframe.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56208 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Update test results after r56182.  "background-color: rgba(0, 0, 0, 0)" is now reported as "background-color: transparent".

Rubber-stamped by Samuel Weinig.

* fast/css/getComputedStyle/computed-style-expected.txt:
* platform/mac/fast/css/getComputedStyle/computed-style-expected.txt:
* platform/qt/fast/css/getComputedStyle/computed-style-expected.txt:
* platform/win/fast/css/getComputedStyle/computed-style-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<webkit.org/b/36338> Remove unused RenderReplaced::adjustOverflowForBoxShadowAndReflect() declaration

Rubber-stamped by Darin Adler.

This declaration should have been removed in r47440.

* rendering/RenderReplaced.h:
(WebCore::RenderReplaced::adjustOverflowForBoxShadowAndReflect): Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56206 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Dragging a PDF image triggers assertion in DragController::startDrag()
https://bugs.webkit.org/show_bug.cgi?id=36247

Reviewed by Dmitry Titov.

WebCore:

Test: fast/images/drag-pdf-as-image.html

* platform/graphics/cg/PDFDocumentImage.cpp:
(WebCore::PDFDocumentImage::filenameExtension):
* platform/graphics/cg/PDFDocumentImage.h:

LayoutTests:

* fast/images/drag-pdf-as-image-expected.txt: Added.
* fast/images/drag-pdf-as-image.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56205 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        Avoid taking a reference to a String owned by a temporary AtomicString.

        * css/CSSSelector.cpp:
        (WebCore::CSSSelector::RareData::parseNth):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56204 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Land new expected results.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add missing file from r56186.

* fast/dynamic/float-remove-above-line-2-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Chromium interface change to support Blob.slice.
https://bugs.webkit.org/show_bug.cgi?id=35318

Reviewed by Darin Fisher.

* features.gypi:
* public/WebHTTPBody.h:
* public/WebKitClient.h:
(WebKit::WebKitClient::getFileModificationTime):
* src/ChromiumBridge.cpp:
(WebCore::ChromiumBridge::getFileModificationTime):
* src/WebHTTPBody.cpp:
(WebKit::WebHTTPBody::elementAt):
(WebKit::WebHTTPBody::appendFile):
(WebKit::WebHTTPBody::appendFileRange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Fixed parenting issues with HW layers on Windows
        https://bugs.webkit.org/show_bug.cgi?id=32449

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Dimitri Glazkov.

        Add upstream LayoutTests/platform/chromium* directories to the
        baseline search path for new-run-webkit-tests in preparation for
        upstreaming all of the Chromium baselines. Note that this does
        not actually create the directories themselves, but that's okay.

        https://bugs.webkit.org/show_bug.cgi?id=36324

        * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
        * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
        * Scripts/webkitpy/layout_tests/port/chromium_win.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Anders Carlsson.

        https://bugs.webkit.org/show_bug.cgi?id=36327
        Test that a plug-in can override Node methods of its element

        * plugins/override-node-method-expected.txt: Added.
        * plugins/override-node-method.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, build fix.

[Chromium] Change the order of mac/ and cg/ include dirs to fix compile break
due to http://trac.webkit.org/changeset/56192,

* WebCore.gyp/WebCore.gyp: Changed the order.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Simon Fraser  <simon.fraser@apple.com>

        No review.

        Fix brace style after r56170.

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::SelectorChecker::checkOneSelector):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36325

Reviewed by Oliver Hunt.

Add support for stopImmediatePropagation from DOM Level 3 Events.

Added fast/events/stop-immediate-propagation.html.

* dom/Event.cpp:
(WebCore::Event::Event):
* dom/Event.h:
(WebCore::Event::timeStamp):
(WebCore::Event::stopPropagation):
(WebCore::Event::stopImmediatePropagation):
(WebCore::Event::propagationStopped):
(WebCore::Event::immediatePropagationStopped):
* dom/Event.idl:
* dom/EventTarget.cpp:
(WebCore::EventTarget::fireEventListeners):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, build fix.

Return frame accessor back to Geolocation, it's used in GeolocationServiceChromium.
The accessor was removed in http://trac.webkit.org/changeset/56188.

* page/Geolocation.h:
(WebCore::Geolocation::frame): Added back.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-17  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Dimitri Glazkov.

        Add the concept of an "overrides" file for expectations so that we
        can store test_expectations both upstream and downstream for a port
        that runs both in webkit.org and in a separate repository (like
        Chromium). Also add some unit tests for the expectations module.

        https://bugs.webkit.org/show_bug.cgi?id=36249

        * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_expectations_test.py: Added.
        * Scripts/webkitpy/layout_tests/port/base.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56193 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Help reduce build problems due to font declarations.
https://bugs.webkit.org/show_bug.cgi?id=36190

Reviewed by Adam Roben.

Functions unchanged, no new tests.

* WebCore.vcproj/WebCore.vcproj: Add new file, and show
  movement of FontPlatformData.h to cg/ directory.
* platform/graphics/cairo/FontPlatformData.h: Merge in
  WinCairo-specific declarations.  Correct constructor
  declaration to use type float for font size.
(WebCore::FontPlatformData::FontPlatformData):
(WebCore::FontPlatformData::hfont):
(WebCore::FontPlatformData::useGDI):
(WebCore::FontPlatformData::fontFace):
(WebCore::FontPlatformData::setSize):
(WebCore::FontPlatformData::hash):
(WebCore::FontPlatformData::isHashTableDeletedValue):
* platform/graphics/cairo/FontPlatformDataCairo.cpp:
(WebCore::FontPlatformData::FontPlatformData): Correct
  constructor to accept type float for font size.
* platform/graphics/cg/FontPlatformData.h: Copied from WebCore/platform/graphics/win/FontPlatformData.h.
(WebCore::FontPlatformData::FontPlatformData):
(WebCore::FontPlatformData::hfont):
(WebCore::FontPlatformData::cgFont):
(WebCore::FontPlatformData::operator==):
* platform/graphics/win/FontPlatformData.h: Removed.
* platform/graphics/win/FontPlatformDataCairoWin.cpp:
(WebCore::FontPlatformData::operator==): Move implementation
  here (previously in header) to match other cairo-based ports.
* platform/graphics/win/RefCountedHFONT.h: Added.  This
  was extracted from win/FontPlatformData.h, and is now shared
  by the cg/FontPlatformData.h and cairo/FontPlatformData.h
(WebCore::RefCountedHFONT::create):
(WebCore::RefCountedHFONT::createDeleted):
(WebCore::RefCountedHFONT::~RefCountedHFONT):
(WebCore::RefCountedHFONT::hfont):
(WebCore::RefCountedHFONT::hash):
(WebCore::RefCountedHFONT::RefCountedHFONT):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dave Hyatt.

        https://bugs.webkit.org/show_bug.cgi?id=36274
        CSS3 :nth-child selector 'odd' keyword case sensitivity test fails

        The tests for "odd", "even" etc. for nth-child selectors should be case insensitive.

        Test: fast/css/nth-child-odd-case-insensitive.html

        * css/CSSSelector.cpp:
        (WebCore::CSSSelector::RareData::parseNth):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56191 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36265

Reviewed by Oliver Hunt.

Add support for event.defaultPrevented from DOM level 3.

Added fast/events/defaultprevented.html

* dom/Event.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56190 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Oliver Hunt  <oliver@apple.com>

        Reviewed by Sam Weinig.

        Add API to directly expose JSON parsing
        https://bugs.webkit.org/show_bug.cgi?id=34887

        Add API to expose JSON parsing directly, and add tests to testapi

        * API/JSValueRef.cpp:
        (JSValueMakeFromJSONString):
        (JSValueCreateJSONString):
        * API/tests/testapi.c:
        (main):
        * JavaScriptCore.exp:
        * runtime/JSONObject.cpp:
        (JSC::JSONStringify):
        * runtime/JSONObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56189 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove Geolocation.lastPosition, no longer in the spec.
https://bugs.webkit.org/show_bug.cgi?id=36255
rdar://problem/7746357

Reviewed by Kenneth Rohde Christiansen.

* WebCore.base.exp: Updated since Geolocation's destructor is now non-virtual.

* page/Geolocation.cpp:
(WebCore::Geolocation::lastPosition): Add an assertion; it's only legal to
call this if access to the location is allowed.

* page/Geolocation.h: Removed unneeded includes. Made destructor non-virtual,
although it will still be virtual if any of the base classes have a virtual
destructor. Made lastPosition, isAllowed, and isDenied functions private.
Removed unused suspend, resume, setShouldClearCache, shouldClearCache,
and frame functions.

* page/Geolocation.idl: Removed lastPosition read-only attribute. No longer in
the Geolocation specification.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36275

Reviewed by Simon Fraser.

Make sure :not does not allow pseudo-elements inside it. (Hooray for another pointless
restriction on :not.)

Added fast/css/invalid-not-with-pseudo-element.html

* css/CSSGrammar.y:
* css/CSSSelector.h:
(WebCore::CSSSelector::matchesPseudoElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore: <rdar://problem/7761400> Rework the fix for
https://bugs.webkit.org/show_bug.cgi?id=18722

Reviewed by Darin Adler.

Test: fast/dynamic/float-remove-above-line-2.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::removeFloatingObject): Treat 0- and less-than-0-height floats
as having a height of 1 so that they intersect with the line they originate on.
(WebCore::RenderBlock::clearFloats): Use numeric_limits.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::layoutInlineChildren): Removed the intersection checks here,
so that a float is always included in the float list of the line it originates on, even
if it does not intersect that line. This ensures that every float is accounted for, which
is necessary during incremental layout when adding floats from clean lines.

LayoutTests: Additional test for <rdar://problem/7761400>

Reviewed by Darin Adler.

* fast/dynamic/float-remove-above-line-2.html: Copied from LayoutTests/fast/dynamic/float-remove-above-line.html.
* fast/dynamic/float-remove-above-line-expected.txt: Updated.
* fast/dynamic/float-remove-above-line.html: Fixed inconsequential typo in end tag.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add expected results.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Antti Koivisto  <koivisto@iki.fi>

        Reviewed by Kenneth Rohde Christiansen.

        https://bugs.webkit.org/show_bug.cgi?id=36102
        [Qt] Scaling control API for tiled backing store

        Commit the new scale synchronously after unfreeze to avoid ugliness.

        * platform/graphics/TiledBackingStore.cpp:
        (WebCore::TiledBackingStore::TiledBackingStore):
        (WebCore::TiledBackingStore::setContentsScale):
        (WebCore::TiledBackingStore::commitScaleChange):
        (WebCore::TiledBackingStore::setContentsFrozen):
        * platform/graphics/TiledBackingStore.h:
2010-03-18  Antti Koivisto  <koivisto@iki.fi>

        Reviewed by Kenneth Rohde Christiansen.

        https://bugs.webkit.org/show_bug.cgi?id=36102
        [Qt] Scaling control API for tiled backing store

        The scale is set by passing the QGraphicsWebView scale to the backing store. The
        only new API is the tiledBackingStoreFrozen property which allows disabling
        all updates (for example during zoom animation).

        * Api/qgraphicswebview.cpp:
        (QGraphicsWebViewPrivate::_q_scaleChanged):
        (QGraphicsWebViewPrivate::updateTiledBackingStoreScale):
        (QGraphicsWebView::QGraphicsWebView):
        (QGraphicsWebView::isTiledBackingStoreFrozen):
        (QGraphicsWebView::setTiledBackingStoreFrozen):
        * Api/qgraphicswebview.h:
        * Api/qwebframe.cpp:
        * Api/qwebframe_p.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>

        Reviewed by Kenneth Rohde Christiansen.

        Add an overlay QGraphicsTextItem to QtLauncher so we can display FPS info
        on the launcher and not on the terminal anymore.

        [Qt] QtLauncher's FPS info should be displayed on an overlay text item
        https://bugs.webkit.org/show_bug.cgi?id=36244

        * QtLauncher/webview.cpp:
        (WebViewGraphicsBased::WebViewGraphicsBased):
        (WebViewGraphicsBased::setFrameRateMeasurementEnabled):
        (WebViewGraphicsBased::updateFrameRate):
        * QtLauncher/webview.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36281

Reviewed by Simon Fraser.

Make sure an exception is raised if an @import rule is inserted in the wrong place.

Also make sure (so that this particular test case passes) that rgba(0, 0, 0, 0) is dumped
as transparent by getComputedStyle, since it most commonly occurs in background-color and
that default makes more sense than dumping rgba values.

Added fast/css/invalid-import-insertion.html

* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::createColor):
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::insertRule):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        First cut at SheriffBot
        https://bugs.webkit.org/show_bug.cgi?id=36253

        This patch contains a first attempt at writing a sheriff bot.
        Currently, we're missing the logic that actually finds the SVN revision
        numbers to complain about, but once we have that, we'll have the rest
        of the infrustructure to ping IRC and to file bugs.

        There's a lot to fill in for the SheriffBot, but this patch give us the
        framework in which to do it.

        This patch required a bit of refactoring of AbstractQueue because
        SheriffBot is the first bot that doesn't process patches (it processes
        SVN revisions).  Accordingly, I've factored out AbstractPatchQueue to
        hold the parts of AbstractQueue that are specific to dealing with
        patches.  Some of the choices here might not be obvious yet, but we can
        tweak them as our needs become clearer.

        * Scripts/webkitpy/commands/queues.py:
        * Scripts/webkitpy/commands/queues_unittest.py:
        * Scripts/webkitpy/commands/sheriffbot.py: Added.
        * Scripts/webkitpy/commands/sheriffbot_unittest.py: Added.
        * Scripts/webkitpy/mock_bugzillatool.py:
            Added a MockIRC object to the mock tool.
        * Scripts/webkitpy/multicommandtool.py:
            Added a finalize method so the tool can disconnect from IRC
            cleanly instead of just droping the socket.
        * Scripts/webkitpy/multicommandtool_unittest.py:
        * Scripts/webkitpy/patch/patcher.py:
            Added support for talking to IRC.
        * Scripts/webkitpy/unittests.py:
            We should add a commands/unittests.py file at some point to make
            the commands module more self-contained.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Luiz Agostini  <luiz.agostini@openbossa.org>

        Reviewed by Kenneth Rohde Christiansen.

        Multiselect Popup - Listbox click simulation
        https://bugs.webkit.org/show_bug.cgi?id=36177

        Listbox popups will need to notify the corresponding select elements that a
        selection change happened. The current HTMLSelectElement interface does not
        allow multiple selections.

        The new method  listBoxSelectItem will be used for that. I have refactored part
        of the mouse handling code in bug 36124 and I am now reusing it here for
        <select multiple> popups. All the other cases will handled as they were before to be
        sure that no side effects will show up.

        * dom/SelectElement.cpp:
        (WebCore::SelectElement::updateListBoxSelection):
        (WebCore::SelectElement::listBoxOnChange):
        * dom/SelectElement.h:
        * html/HTMLSelectElement.cpp:
        (WebCore::HTMLSelectElement::listBoxPopupClick):
        * html/HTMLSelectElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Antti Koivisto  <koivisto@iki.fi>

        Reviewed by Kenneth Rohde Christiansen.

        https://bugs.webkit.org/show_bug.cgi?id=36102
        [Qt] Scaling control API for tiled backing store

        Add animated smooth zooming to Qt launcher when in tiled mode.

        * QtLauncher/main.cpp:
        (LauncherWindow::LauncherWindow):
        (LauncherWindow::zoomAnimationFinished):
        (LauncherWindow::applyZoom):
        (LauncherWindow::zoomIn):
        (LauncherWindow::zoomOut):
        * QtLauncher/webview.h:
        (WebViewGraphicsBased::graphicsWebView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Support using IRC accounts with a password
        https://bugs.webkit.org/show_bug.cgi?id=36287

        Add a global option to specify an IRC password so we can use the
        sheriffbot account (which needs a password).

        * Scripts/webkitpy/irc/ircbot.py:
        * Scripts/webkitpy/irc/ircproxy.py:
        * Scripts/webkitpy/patch/patcher.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Not reviewed.

        https://bugs.webkit.org/show_bug.cgi?id=36210
        plugins/resize-from-plugin.html fails on some platforms

        Landing new results for Qt and Gtk, taken from buildbot.

        Gtk results are a pass, although the difference in clip rect may indicate some issue to look into.
        Qt results are a failure.

        * platform/gtk/plugins/resize-from-plugin-expected.txt: Copied from LayoutTests/plugins/resize-from-plugin-expected.txt.
        * platform/qt/plugins/resize-from-plugin-expected.txt: Copied from LayoutTests/plugins/resize-from-plugin-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Eric Seidel  <eric@webkit.org>

        Just fixing missing parenthesis typo, no review.

        * Scripts/webkitpy/commands/queries.py: '%' has higher precedence than 'or', use parentheses.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-12  Ojan Vafai  <ojan@chromium.org>

        Reviewed by David Levin.

        smartdelete should only occur after double-click
        https://bugs.webkit.org/show_bug.cgi?id=35314

        * editing/deleting/non-smart-delete.html:
        * editing/pasteboard/drag-drop-modifies-page.html:
        * editing/pasteboard/page-zoom-expected.txt:
        Space is removed because we used to incorrectly smartpaste here.
        New behavior matches NSTextView.
        * editing/style/apply-through-end-of-document-expected.txt:
        The change here is that selectall + smartpaste doesn't put a BR,
        but selectall + normal-paste does. Not sure if that's a bug or not,
        but it's not a side effect of this change.
        * editing/style/style-boundary-005.html:
        * platform/mac/editing/deleting/non-smart-delete-expected.checksum:
        * platform/mac/editing/deleting/non-smart-delete-expected.png:
        * platform/mac/editing/deleting/non-smart-delete-expected.txt:
        * platform/mac/editing/execCommand/paste-1-expected.txt:
        * platform/mac/editing/execCommand/paste-2-expected.txt:
        * platform/mac/editing/pasteboard/drag-drop-modifies-page-expected.checksum:
        * platform/mac/editing/pasteboard/drag-drop-modifies-page-expected.png:
        * platform/mac/editing/pasteboard/drag-drop-modifies-page-expected.txt:

        * platform/mac/editing/selection/expanding-selections-expected.txt:
        * platform/mac/editing/selection/expanding-selections2-expected.txt:
        * platform/mac/editing/selection/triple-click-in-pre-expected.txt:
        The above three results change because granularity is no longer part of the seleciton,
        so changes in granularity don't fire selection change notifications.

        * platform/mac/editing/style/style-boundary-005-expected.checksum:
        * platform/mac/editing/style/style-boundary-005-expected.png:
        * platform/mac/editing/style/style-boundary-005-expected.txt:
2010-03-12  Ojan Vafai  <ojan@chromium.org>

        Reviewed by David Levin.

        smartdelete should only occur after double-click
        https://bugs.webkit.org/show_bug.cgi?id=35314

        1. Consolidate all notions of selection-granularity into SelectionController.
        2. Now only mouse-based selections store a selection-granularity. This matches NSTextView.

        New tests were added in http://trac.webkit.org/changeset/55913.

        * WebCore.base.exp:
        * editing/MoveSelectionCommand.cpp:
        (WebCore::MoveSelectionCommand::MoveSelectionCommand):
        (WebCore::MoveSelectionCommand::doApply):
        * editing/MoveSelectionCommand.h:
        (WebCore::MoveSelectionCommand::create):
        * editing/SelectionController.cpp:
        (WebCore::SelectionController::SelectionController):
        (WebCore::SelectionController::setSelection):
        (WebCore::SelectionController::modify):
        (WebCore::SelectionController::clear):
        * editing/SelectionController.h:
        (WebCore::SelectionController::setSelection):
        (WebCore::SelectionController::granularity):
        * editing/VisibleSelection.cpp:
        (WebCore::VisibleSelection::VisibleSelection):
        (WebCore::VisibleSelection::expandUsingGranularity):
        (WebCore::VisibleSelection::setStartAndEndFromBaseAndExtentRespectingGranularity):
        (WebCore::VisibleSelection::validate):
        * editing/VisibleSelection.h:
        (WebCore::operator==):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::clear):
        * page/DOMSelection.cpp:
        (WebCore::DOMSelection::extend):
        * page/DragController.cpp:
        (WebCore::DragController::concludeEditDrag):
        * page/EventHandler.cpp:
        (WebCore::EventHandler::selectClosestWordFromMouseEvent):
        (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent):
        (WebCore::EventHandler::handleMousePressEventTripleClick):
        (WebCore::EventHandler::handleMousePressEventSingleClick):
        (WebCore::EventHandler::updateSelectionForMouseDrag):
        * page/Frame.cpp:
        (WebCore::Frame::Frame):
        (WebCore::Frame::selectionGranularity):
        * page/Frame.h:
        * rendering/RenderTextControl.cpp:
        (WebCore::RenderTextControl::setSelectionRange):
2010-03-12  Ojan Vafai  <ojan@chromium.org>

        Reviewed by David Levin.

        smartdelete should only occur after double-click
        https://bugs.webkit.org/show_bug.cgi?id=35314

        * src/WebFrameImpl.cpp:
        (WebKit::WebFrameImpl::selectWordAroundPosition):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dave Hyatt.

        https://bugs.webkit.org/show_bug.cgi?id=36273
        CSS3 :indeterminate and input type=radio test fails

        The :indeterminate pseudo-class should apply to both radio buttons and checkboxes.

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::setIndeterminate): Use allowsIndeterminate() to determine if the input
        supports the indeterminate state.
        (WebCore::HTMLInputElement::preDispatchEventHandler): Use allowsIndeterminate() to determine whether
        to handle the indeterminate state.
        (WebCore::HTMLInputElement::postDispatchEventHandler): Ditto.
        * html/HTMLInputElement.h:
        (WebCore::HTMLInputElement::allowsIndeterminate): New utility method, returns true for radio buttons
        and checkboxes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, build fix.

[V8] Turn npObjectTypeInfo into a function, make everyone happy.

* bindings/v8/NPV8Object.cpp:
(WebCore::npObjectTypeInfo): Added new function.
(npCreateV8ScriptObject):
* bindings/v8/NPV8Object.h: Turned npObjectTypeInfo into function.
* bindings/v8/V8NPObject.cpp: Replaced references to function

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Adam Roben and Anders Carlsson.

        https://bugs.webkit.org/show_bug.cgi?id=36210
        plugins/resize-from-plugin.html fails on some platforms

WebKitTools:
        Turns out that most platforms don't use "cross-platform" main.cpp. Copied code added for
        the test to their versions of the file.

        * DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
        (webkit_test_plugin_set_window):
        * DumpRenderTree/win/TestNetscapePlugin/main.cpp:
        (NPP_SetWindow):

LayoutTests:
        * platform/win/plugins/resize-from-plugin-expected.txt: Copied from LayoutTests/plugins/resize-from-plugin-expected.txt.
        Windows results are different, probably because windowed plug-ins just get the whole window
        as clip rect.

        * platform/gtk/Skipped:
        * platform/qt/Skipped:
        Unskipped the tests to see how they work with TestNetscapePlugin fixes from this patch.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56172 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=36283

Reviewed by Simon Fraser.

The wrong constants were being used for the code that attempted to compact the background-repeat shorthand to
a single value.

Added fast/backgrounds/background-repeat-computed-style.html

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::fillRepeatToCSSValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56171 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dave Hyatt.

        https://bugs.webkit.org/show_bug.cgi?id=36271
        CSS3 :enabled on an input type=hidden element fails

        The CSS3 selectors spec now makes it clear that hidden inputs should respect
        the :enabled and :disabled pseudo-classes.

        Test: fast/forms/hidden-input-enabled.html

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::SelectorChecker::checkOneSelector): Remove checks for hidden inputs.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore: Intro text at Star Wars demo is clipped.
<rdar://problem/7560979>
https://bugs.webkit.org/show_bug.cgi?id=33909

Reviewed by Simon Fraser.

Test: compositing/repaint/layer-repaint.html

We are flipping the coordinates for drawing, therefore
we must flip them also when we invalidate the rectangles.

* platform/graphics/win/GraphicsLayerCACF.cpp:
(WebCore::WebLayer::setNeedsDisplay):

LayoutTests: Intro text at Star Wars demo is clipped.
<rdar://problem/7560979>
https://bugs.webkit.org/show_bug.cgi?id=33909

Reviewed by Simon Fraser.

We are flipping the coordinates for drawing, therefore
we must flip them also when we invalidate the rectangles.

* compositing/repaint/layer-repaint.html: Added.
* platform/mac/compositing/repaint/layer-repaint-expected.checksum: Added.
* platform/mac/compositing/repaint/layer-repaint-expected.png: Added.
* platform/mac/compositing/repaint/layer-repaint-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Nate Chapin  <japhet@chromium.org>

        Unreviewed, Chromium build fix.

        Missed a #include in V8Collection.h.

        * bindings/v8/V8Collection.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix for https://bugs.webkit.org/show_bug.cgi?id=36284.

Reviewed by Simon Fraser.

Make sure the CSS parser properly rejects negative border widths.

Added fast/borders/negative-border-width.html

* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56167 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Nate Chapin  <japhet@chromium.org>

        Reviewed by Dimitri Glazkov.

        [V8] Remove V8Index.h and all references to V8ClassIndex.

        https://bugs.webkit.org/show_bug.cgi?id=33477

        * WebCore.gypi:
        * bindings/scripts/CodeGeneratorV8.pm:
        * bindings/v8/NPV8Object.cpp:
        * bindings/v8/NPV8Object.h:
        * bindings/v8/V8Collection.h:
        * bindings/v8/V8DOMWindowShell.cpp:
        * bindings/v8/V8DOMWindowShell.h:
        * bindings/v8/V8DOMWrapper.cpp:
        * bindings/v8/V8DOMWrapper.h:
        * bindings/v8/V8GCController.cpp:
        * bindings/v8/V8Helpers.cpp:
        * bindings/v8/V8Index.h: Removed.
        * bindings/v8/V8IsolatedContext.h:
        * bindings/v8/V8NPObject.cpp:
        * bindings/v8/V8Proxy.cpp:
        * bindings/v8/V8Proxy.h:
        * bindings/v8/V8SVGPODTypeWrapper.h:
        * bindings/v8/WorkerContextExecutionProxy.cpp:
        * bindings/v8/WorkerContextExecutionProxy.h:
        * bindings/v8/WrapperTypeInfo.h: Move WrapperTypeInfo struct into its own file from V8Index.h.
        * bindings/v8/custom/V8CanvasRenderingContext2DCustom.cpp:
        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        * bindings/v8/custom/V8EventSourceConstructor.cpp:
        * bindings/v8/custom/V8HTMLAudioElementConstructor.cpp:
        * bindings/v8/custom/V8HTMLAudioElementConstructor.h:
        * bindings/v8/custom/V8HTMLImageElementConstructor.cpp:
        * bindings/v8/custom/V8HTMLImageElementConstructor.h:
        * bindings/v8/custom/V8HTMLOptionElementConstructor.cpp:
        * bindings/v8/custom/V8HTMLOptionElementConstructor.h:
        * bindings/v8/custom/V8HistoryCustom.cpp:
        * bindings/v8/custom/V8LocationCustom.cpp:
        * bindings/v8/custom/V8WebGLArrayCustom.h:
        * bindings/v8/custom/V8WebGLByteArrayCustom.cpp:
        * bindings/v8/custom/V8WebGLFloatArrayCustom.cpp:
        * bindings/v8/custom/V8WebGLIntArrayCustom.cpp:
        * bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
        * bindings/v8/custom/V8WebGLShortArrayCustom.cpp:
        * bindings/v8/custom/V8WebGLUnsignedByteArrayCustom.cpp:
        * bindings/v8/custom/V8WebGLUnsignedIntArrayCustom.cpp:
        * bindings/v8/custom/V8WebGLUnsignedShortArrayCustom.cpp:
        * bindings/v8/custom/V8WebKitPointConstructor.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Preserve console history between debugging sessions.

        https://bugs.webkit.org/show_bug.cgi?id=36223

        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleView.prototype._settingsLoaded):
        (WebInspector.ConsoleView.prototype._enterKeyPressed.printResult):
        (WebInspector.ConsoleView.prototype._enterKeyPressed):
        * inspector/front-end/Settings.js:
        (WebInspector.Settings.prototype._load):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Added reviewer name for r56155

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Kim Grönholm  <kim.gronholm@nomovok.com>

        Reviewed by Simon Hausmann.

        [Qt] GraphicsLayer: Pausing and resuming of animations don't work as expected
        https://bugs.webkit.org/show_bug.cgi?id=36219

        The play-state layout test now works, and can be removed from the skip list.

        * platform/qt/Skipped: animations/play-state.html is removed from the skipped list
2010-03-18  Kim Grönholm  <kim.gronholm@nomovok.com>

        Reviewed by Simon Hausmann.

        [Qt] GraphicsLayer: Pausing and resuming of animations don't work as expected
        https://bugs.webkit.org/show_bug.cgi?id=36219

        addAnimation and pauseAnimation were handling animation state changes in
        the wrong way.
        Apparently WebCore calls addAnimation for resuming, so we have to make
        sure to resume an existing animation if that happens. Also, timeOffset
        is now used for synchronization of the Qt animation clock with the
        WebCore clock, as opposed to using it as a delay timer.
        Both those fixes were necessary to get the layout test to work.

        LayoutTests/animations/play-state.html now passed on Qt.

        * platform/graphics/qt/GraphicsLayerQt.cpp:
        (WebCore::AnimationQtBase::AnimationQtBase):
        (WebCore::TransformAnimationQt::updateState):
        (WebCore::GraphicsLayerQt::addAnimation):
        (WebCore::GraphicsLayerQt::pauseAnimation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Adam Langley  <agl@chromium.org>

        Reviewed by David Levin.

        [chromium] Fix inverted logic in per-strike renderer patch.

        I screwed up in r55089 and got one of the conditions backwards. Thanks
        to Roman Tsisyk for pointing it out.

        https://bugs.webkit.org/show_bug.cgi?id=35495

        * platform/chromium/fast/text/chromium-linux-fontconfig-renderstyle.html: Added.
2010-03-18  Adam Langley  <agl@chromium.org>

        Reviewed by David Levin.

        [chromium] Fix inverted logic in per-strike renderer patch.

        I screwed up in r55089 and got one of the conditions backwards. Thanks
        to Roman Tsisyk for pointing it out.

        https://bugs.webkit.org/show_bug.cgi?id=35495

        Test: platform/chromium/fast/text/chromium-linux-fontconfig-renderstyle.html

        * platform/graphics/chromium/FontPlatformDataLinux.cpp:
        (WebCore::FontPlatformData::querySystemForRenderStyle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Show inherited styles for a selected DOM element

        https://bugs.webkit.org/show_bug.cgi?id=28039

        * English.lproj/localizedStrings.js:
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::getStyles):
        (WebCore::InspectorDOMAgent::populateObjectWithStyleProperties):
        * inspector/front-end/DOMAgent.js:
        (WebInspector.CSSStyleDeclaration):
        * inspector/front-end/StylesSidebarPane.js:
        (WebInspector.StylesSidebarPane.prototype._update):
        (WebInspector.StylesSidebarPane.prototype._containsInherited):
        (WebInspector.StylePropertiesSection):
        (WebInspector.StylePropertiesSection.prototype.isPropertyInherited):
        (WebInspector.StylePropertiesSection.prototype.onpopulate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Leandro Pereira  <leandro@profusion.mobi>

        Reviewed by Kenneth Rohde Christiansen.

        Add support to run-launcher to open the EFL example browser.
        http://webkit.org/b/36181

        * Scripts/webkitdirs.pm:
        * Scripts/run-launcher:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Spatial Navigation: Add isNull and document convenience methods to FocusCandidate
https://bugs.webkit.org/show_bug.cgi?id=36167

Reviewed by Gustavo Noronha.
Patch by Antonio Gomes <tonikitoo@webkit.org>

It turns out that Spatial Navigation related code (in FocusController.cpp for
instance) can be simplified and look better if FocusCandidate class offer some
convinience method. This patch introduces a couple of them (isNull and a Document
getter).

A followup refactoring patch will be making use of these helper methods.

* page/SpatialNavigation.h:
(WebCore::FocusCandidate::FocusCandidate): Added a "Node*" parameter
to FocusCandidate's constructor.
(WebCore::FocusCandidate::isNull): In the Spatial Navigation logic, a NULL
FocusCandidate object is one that does not hold a valid reference to a "Node*".
(WebCore::FocusCandidate::document): This method returns a pointer for the FocusCandidate
node's document, if |node| is a valid pointer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Noam Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Antti Koivisto.

        [Qt] [Performance] GraphicsLayerQt updates the scene too often
        https://bugs.webkit.org/show_bug.cgi?id=36158

        This fix makes sure that flushChanges is only called when necessary,
        by calling the notifySync function asynchronously, which makes sure flushChanges() is called
        after the WebCore compositor has made all its changes.

        This has shown a visual improvement on several test-cases.

        * Makefile:
        * platform/graphics/qt/GraphicsLayerQt.cpp:
        (WebCore::GraphicsLayerQtImpl::):
        (WebCore::GraphicsLayerQtImpl::notifySyncRequired):
        (WebCore::GraphicsLayerQtImpl::notifyChange):
        (WebCore::GraphicsLayerQtImpl::flushChanges):
        (WebCore::GraphicsLayerQt::setMaskLayer):
        (WebCore::GraphicsLayerQt::setPosition):
        (WebCore::GraphicsLayerQt::setAnchorPoint):
        (WebCore::GraphicsLayerQt::setSize):
        (WebCore::GraphicsLayerQt::setTransform):
        (WebCore::GraphicsLayerQt::setChildrenTransform):
        (WebCore::GraphicsLayerQt::setPreserves3D):
        (WebCore::GraphicsLayerQt::setMasksToBounds):
        (WebCore::GraphicsLayerQt::setDrawsContent):
        (WebCore::GraphicsLayerQt::setBackgroundColor):
        (WebCore::GraphicsLayerQt::clearBackgroundColor):
        (WebCore::GraphicsLayerQt::setContentsOpaque):
        (WebCore::GraphicsLayerQt::setBackfaceVisibility):
        (WebCore::GraphicsLayerQt::setOpacity):
        (WebCore::GraphicsLayerQt::setContentsRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Sergio Villar Senin  <svillar@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Failing tests http/tests/misc/image-blocked-src-change.html
        & http/tests/misc/image-blocked-src-no-change.html
        https://bugs.webkit.org/show_bug.cgi?id=36227

        Removed tests from the list of Skipped:
        http/tests/misc/image-blocked-src-no-change.html
        http/tests/misc/image-blocked-src-change.html
        http/tests/security/local-CSS-from-remote.html
        http/tests/security/local-video-poster-from-remote.html
        http/tests/security/frame-loading-via-document-write.html
        http/tests/security/local-iFrame-from-remote.html
        http/tests/security/local-video-src-from-remote.html
        http/tests/security/local-image-from-remote.html
        http/tests/security/local-JavaScript-from-remote.html

        * platform/gtk/Skipped:
2010-03-18  Sergio Villar Senin  <svillar@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Failing tests http/tests/misc/image-blocked-src-change.html
        & http/tests/misc/image-blocked-src-no-change.html
        https://bugs.webkit.org/show_bug.cgi?id=36227

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (webViewConsoleMessage): print only the file name instead of the
        whole URI when printing messages with local URI's

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        Teach what-broke how to look up reviewer and author Committer objects by name
        https://bugs.webkit.org/show_bug.cgi?id=36264

        * Scripts/webkitpy/commands/queries.py:
         - Add committer_by_name lookups for both reviewer and author
         - Improve printing in the cases where lookups fail.
        * Scripts/webkitpy/committers.py:
         - Add committer_by_name
        * Scripts/webkitpy/committers_unittest.py:
         - Test committer_by_name

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Stephen White  <senorblanco@chromium.org>

        Parsing of rgb() colors in HTML canvas is still fairly slow.
        findNamedColor() was showing up as hot, so this patch attempts to
        avoid calling it by putting the rgb() check ahead of the named
        colors.  It also removes a call to Color::Color() from
        CanvasStyle, which was doing a redundant check for the hex
        color and named colors, which CSSParser already does.  Finally, it
        changes the ad-hoc parser to check for negative values, and clamps them
        at zero.  This avoids the lex/yacc path for rgb() colors with
        negative color components.

        https://bugs.webkit.org/show_bug.cgi?id=36199

        Covered by fast/canvas/set-colors.html.

        * css/CSSParser.cpp:
        (WebCore::parseInt):
        (WebCore::CSSParser::parseColor):
        * html/canvas/CanvasStyle.cpp:
        (WebCore::CanvasStyle::applyFillColor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Chang Shu  <chang.shu@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        https://bugs.webkit.org/show_bug.cgi?id=36220

        [Qt] Unskip the tests that pass after the fix for bug 36139.
        http/tests/appcache/foreign-iframe-main.html
        http/tests/appcache/manifest-with-empty-file.html
        http/tests/appcache/resource-redirect.html

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-03-18  Leandro Pereira  <leandro@profusion.mobi>

        Reviewed by Kenneth Rohde Christiansen.

        Add EFL port files to efl/ewk.
        http://webkit.org/b/35931

        * efl/ewk/ewk_settings.cpp: Added.
        * efl/ewk/ewk_settings.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<webkit.org/b/36246> Node::removeAllEventListenersSlowCase is declared but not used

Reviewed by Sam Weinig.

* dom/Node.h:
(WebCore::Node::removeAllEventListenersSlowCase): Removed
declaration.  Implementation was removed in r48701.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@56152 268f45cc-cd09-0410-ab3c-d52691b4dbfc