WebKit-https.git
10 years ago2009-12-14 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Tue, 15 Dec 2009 06:44:34 +0000 (06:44 +0000)]
2009-12-14  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Darin Adler.

        Some Browser-hosted SunSpider files are not valid HTML5
        https://bugs.webkit.org/show_bug.cgi?id=32536

        Made various fixes to validate as HTML5 with no warnings (mainly removing bogus
        close tags and adding a meta charset declaration).

        * hosted/sunspider.html:
        * hosted/versions.html:
        * resources/TEMPLATE.html:
        * resources/driver-TEMPLATE.html:
        * resources/results-TEMPLATE.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52139 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 06:25:37 +0000 (06:25 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Fix failing unittest.

        * Scripts/modules/bugzilla_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52138 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Tue, 15 Dec 2009 06:24:53 +0000 (06:24 +0000)]
2009-12-14  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Eric Seidel.

        Make sunspider-0.9.1 the default content set (both command-line and hosted)
        https://bugs.webkit.org/show_bug.cgi?id=32537

        * hosted/sunspider.html: Change default suite link.
        * sunspider: Change default suite.
        * sunspider-compare-results: Likewise.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52137 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 06:16:25 +0000 (06:16 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Kill _create_step_sequence
        https://bugs.webkit.org/show_bug.cgi?id=32539

        It's cleaner to represent not having these sequences as an empty
        sequence instead of None.

        * Scripts/modules/commands/download.py:
        * Scripts/modules/stepsequence.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Eric Seidel <eric@webkit.org>
eric@webkit.org [Tue, 15 Dec 2009 06:00:55 +0000 (06:00 +0000)]
2009-12-14  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        Move Credential handling out into a separate module
        https://bugs.webkit.org/show_bug.cgi?id=32531

        * Scripts/modules/bugzilla.py:
        * Scripts/modules/credentials.py: Added.
        * Scripts/modules/credentials_unittest.py: Added.
        * Scripts/run-webkit-unittests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52135 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 05:53:04 +0000 (05:53 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Address reviewer comments from an earlier patch.  I didn't
        do this earlier because I was worried about conflicts in dependent
        patches.

        * Scripts/modules/commands/download.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 05:38:44 +0000 (05:38 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Make download commands declarative
        https://bugs.webkit.org/show_bug.cgi?id=32469

        This patch "properly" factors most of the download commands.  These
        commands are now largely declarative, which is the final step of this
        grand refactoring.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 05:38:11 +0000 (05:38 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Add AbstractPatchSequencingCommand to remove redundant code
        https://bugs.webkit.org/show_bug.cgi?id=32468

        Redundant code is bad.  This patch moves us towards more declarative
        commands.

        * Scripts/modules/commands/download.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52132 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 05:32:03 +0000 (05:32 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Kill WebKitApplyingScripts
        https://bugs.webkit.org/show_bug.cgi?id=32467

        Ah!  I've been wanting to do this for a long time.  This patch brings
        the applying commands into the patch processing fold.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52131 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 05:31:03 +0000 (05:31 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Convert rollout to StepSequence
        https://bugs.webkit.org/show_bug.cgi?id=32406

        * Scripts/modules/buildsteps.py:
        * Scripts/modules/commands/download.py:
        * Scripts/modules/commands/download_unittest.py:
        * Scripts/modules/mock_bugzillatool.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52130 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Dec 2009 05:21:33 +0000 (05:21 +0000)]
2009-12-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [bzt] Kill LandingSequence
        https://bugs.webkit.org/show_bug.cgi?id=32464

        Removes LandingSequence in favor of StepSequence.  This required
        changing the Step API slightly to carry a general notion of state
        instead of carrying patches specifically.

        * Scripts/modules/buildsteps.py:
        * Scripts/modules/commands/download.py:
        * Scripts/modules/commands/queues.py:
        * Scripts/modules/landingsequence.py: Removed.
        * Scripts/modules/stepsequence.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52129 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Tue, 15 Dec 2009 03:27:19 +0000 (03:27 +0000)]
2009-12-14  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Sam Weinig.

        in-browser SunSpider suffers excessive penalty under power management
        https://bugs.webkit.org/show_bug.cgi?id=32505

        I have made a few changes to address this:

        1) Change how browser-hosted SunSpider loads tests - preload the content and write it in with
        document.write to reduce triggering of progress UI.
        2) Reduce time between tests to 10ms from 500ms, so that power management doesn't keep the CPU
        stepped all the way down the whole time.
        3) Run the test cycle an extra time for warmup before the runs that count.

        * hosted/sunspider-record-result.js: Removed.
        * make-hosted:
        * resources/TEMPLATE.html:
        * resources/driver-TEMPLATE.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52128 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/7460655> REGRESSION (r47255): Content shifted down at software.opensu...
mitz@apple.com [Tue, 15 Dec 2009 03:05:50 +0000 (03:05 +0000)]
<rdar://problem/7460655> REGRESSION (r47255): Content shifted down at software.opensuse.org
https://bugs.webkit.org/show_bug.cgi?id=32382

Reviewed by Darin Adler.

WebCore:

Test: fast/block/float/in-margin.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::getClearDelta): Floats that do not intrude into the line (because
they are in the margin) do not need to be cleared.

LayoutTests:

* fast/block/float/in-margin.html: Added.
* platform/mac/fast/block/float/in-margin-expected.checksum: Added.
* platform/mac/fast/block/float/in-margin-expected.png: Added.
* platform/mac/fast/block/float/in-margin-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52127 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoPressing backspace inside a table cell erases all empty rows below it.
enrica@apple.com [Tue, 15 Dec 2009 01:14:35 +0000 (01:14 +0000)]
Pressing backspace inside a table cell erases all empty rows below it.
<rdar://problem/5565461>
https://bugs.webkit.org/show_bug.cgi?id=32526

Reviewed by John Sullivan.

WebCore:

No deletion is performed when the caret selection is on an empty table cell.

Test: editing/deleting/delete-empty-table.html

* editing/TypingCommand.cpp:
(WebCore::TypingCommand::deleteKeyPressed): Added check for empty table cells in
case of caret selection.

LayoutTests:

* editing/deleting/delete-empty-table-expected.txt: Added.
* editing/deleting/delete-empty-table.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52126 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRubber-stamped by Brady Eidson.
bweinstein@apple.com [Tue, 15 Dec 2009 00:19:51 +0000 (00:19 +0000)]
Rubber-stamped by Brady Eidson.

Add two http iFrame dynamic form tests to the Windows Skipped
list.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52125 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Evan Stade <estade@chromium.org>
eric@webkit.org [Mon, 14 Dec 2009 22:45:35 +0000 (22:45 +0000)]
2009-12-14  Evan Stade  <estade@chromium.org>

        Reviewed by Darin Fisher.

        https://bugs.webkit.org/show_bug.cgi?id=32347
        Chromium: Context menu launched by menu key has funky behavior

        Make the context menu's position depend on the selection only when an
        editable field has focus. Also, don't send multiple context menu events
        for a single keypress.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::keyEvent):
        (WebKit::WebViewImpl::sendContextMenuEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52124 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Andrei Popescu <andreip@google.com>
eric@webkit.org [Mon, 14 Dec 2009 22:38:09 +0000 (22:38 +0000)]
2009-12-14  Andrei Popescu  <andreip@google.com>

        Reviewed by Darin Adler.

        [Android] Android needs ResourceHandle implementation
        https://bugs.webkit.org/show_bug.cgi?id=32515

        Adds implementation of ResourceHandle on Android.

        No new tests needed, this is platform-specific code, so existing
        tests are sufficient.

        * platform/network/android/ResourceHandleAndroid.cpp: Added.
        (WebCore::ResourceHandleInternal::~ResourceHandleInternal):
        (WebCore::ResourceHandle::~ResourceHandle):
        (WebCore::ResourceHandle::start):
        (WebCore::ResourceHandle::cancel):
        (WebCore::ResourceHandle::bufferedData):
        (WebCore::ResourceHandle::supportsBufferedData):
        (WebCore::ResourceHandle::setDefersLoading):
        (WebCore::ResourceHandle::willLoadFromCache):
        (WebCore::ResourceHandle::loadsBlocked):
        (WebCore::SyncLoader::SyncLoader):
        (WebCore::SyncLoader::~SyncLoader):
        (WebCore::SyncLoader::didReceiveResponse):
        (WebCore::SyncLoader::didReceiveData):
        (WebCore::SyncLoader::didFail):
        (WebCore::ResourceHandle::loadResourceSynchronously):
        * platform/network/android/ResourceLoaderAndroid.h: Added.
        (WebCore::ResourceLoaderAndroid::~ResourceLoaderAndroid):
        (WebCore::ResourceLoaderAndroid::ResourceLoaderAndroid):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52123 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Marco Martin <notmart@gmail.com>
eric@webkit.org [Mon, 14 Dec 2009 22:30:57 +0000 (22:30 +0000)]
2009-12-14  Marco Martin  <notmart@gmail.com>

        Reviewed by Kenneth Rohde Christiansen.

        In the Scrollbar painting of the Qt theme integration,
        intersects any previous clip rects with the one needed to paint the scrollbar.
        This fixes the painting in QGraphicsview, when the parent of the widget that
        paints the QWebPage has the QGraphicsItem::ItemClipsChildrenToShape set and
        a piece of the scrollbar should be cipped away
        https://bugs.webkit.org/show_bug.cgi?id=30366

        No new tests.

        * platform/qt/ScrollbarThemeQt.cpp:
        (WebCore::ScrollbarThemeQt::paint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52122 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoProvide working printing support for WinCairo port.
bfulgham@webkit.org [Mon, 14 Dec 2009 22:23:43 +0000 (22:23 +0000)]
Provide working printing support for WinCairo port.

Reviewed by Adam Roben.

* WebFrame.cpp:
(WebFrame::spoolPage): Conditionalize initialization of
  PlatformGraphicsContext handling for CG vs. Cairo.
(scaleFactor): Add helper function.
(hdcFromContext): Add helper function.
(WebFrame::drawHeader): Correct Cairo variation.
(WebFrame::drawFooter): Correct Cairo variation.
(WebFrame::spoolPages): Correct Cairo variation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52121 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 14 Dec 2009 22:17:19 +0000 (22:17 +0000)]
2009-12-14  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Eric Seidel.

        [V8] Generate bindings for trivial indexers.
        https://bugs.webkit.org/show_bug.cgi?id=32455

        Covered by existing tests.

        * bindings/scripts/CodeGeneratorV8.pm:
            Added detecting indexers and generating code for them.
        * bindings/v8/V8Collection.h: Removed code that's now generated.
        * bindings/v8/V8DOMWrapper.cpp:
        (WebCore::V8DOMWrapper::getTemplate): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52120 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Kinuko Yasuda <kinuko@chromium.com>
eric@webkit.org [Mon, 14 Dec 2009 22:16:33 +0000 (22:16 +0000)]
2009-12-14  Kinuko Yasuda  <kinuko@chromium.com>

        Reviewed by Darin Fisher.

        Make changing-frame-hierarchy-in-onload.html less flaky and
        platform-independent.
        https://bugs.webkit.org/show_bug.cgi?id=32319

        * http/tests/navigation/changing-frame-hierarchy-in-onload.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52119 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Robert Hogan <robert@roberthogan.net>
eric@webkit.org [Mon, 14 Dec 2009 22:09:13 +0000 (22:09 +0000)]
2009-12-14  Robert Hogan  <robert@roberthogan.net>

        Reviewed by Eric Seidel.

        [Qt] Fix https://bugs.webkit.org/show_bug.cgi?id=32437

        Amend the behaviour of DRT::notifyDone to dump only when
        the page has finished loading.

        This fix permits removal of the following tests from the Qt skipped list:
        fast/forms/textarea-linewrap-dynamic.html
        fast/forms/textarea-setvalue-submit.html
        fast/forms/textarea-hard-linewrap-empty.html
        fast/forms/submit-to-url-fragment.html
        http/tests/misc/percent-sign-in-form-field-name.html
        http/tests/security/escape-form-data-field-names.html

        However it also requires the following two to be added to the skipped list:

        http/tests/xmlhttprequest/access-control-basic-denied-preflight-cache.html
        svg/custom/use-instanceRoot-as-event-target.xhtml

        It appears the behaviour of notifyDone was masking problems with these two
        tests.

        * platform/qt/Skipped:
2009-12-14  Robert Hogan  <robert@roberthogan.net>

        Reviewed by Eric Seidel.

        [Qt] Fix https://bugs.webkit.org/show_bug.cgi?id=32437

        Amend the behaviour of DRT::notifyDone to dump only when
        the page has finished loading.

        This fix permits removal of the following tests from the Qt skipped list:
        fast/forms/textarea-linewrap-dynamic.html
        fast/forms/textarea-setvalue-submit.html
        fast/forms/textarea-hard-linewrap-empty.html
        fast/forms/submit-to-url-fragment.html
        http/tests/misc/percent-sign-in-form-field-name.html
        http/tests/security/escape-form-data-field-names.html

        However it also requires the following two to be added to the skipped list:

        http/tests/xmlhttprequest/access-control-basic-denied-preflight-cache.html
        svg/custom/use-instanceRoot-as-event-target.xhtml

        It appears the behaviour of notifyDone was masking problems with these two
        tests.

        * DumpRenderTree/qt/DumpRenderTree.cpp:
        (WebCore::DumpRenderTree::DumpRenderTree):
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::reset):
        (LayoutTestController::processWork):
        (LayoutTestController::maybeDump):
        (LayoutTestController::notifyDone):
        (LayoutTestController::provisionalLoad):
        * DumpRenderTree/qt/LayoutTestControllerQt.h:
        (LayoutTestController::resetLoadFinished):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52118 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Robert Hogan <robert@roberthogan.net>
eric@webkit.org [Mon, 14 Dec 2009 21:57:02 +0000 (21:57 +0000)]
2009-12-14  Robert Hogan  <robert@roberthogan.net>

        Reviewed by Eric Seidel.

        Bug 32519 -  mouseEvent fires mutiple times in
        LayoutTests/svg/custom/use-instanceRoot-as-event-target.xhtml

        Add test to Qt skipped list (was passing due to bug in DRT, see
        https://bugs.webkit.org/show_bug.cgi?id=32437). Updated detail on bug
        in gtk and win skipped lists.

        https://bugs.webkit.org/show_bug.cgi?id=32519

        * platform/gtk/Skipped:
        * platform/qt/Skipped:
        * platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52117 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Dave Hyatt.
ap@apple.com [Mon, 14 Dec 2009 21:47:04 +0000 (21:47 +0000)]
    Reviewed by Dave Hyatt.

        https://bugs.webkit.org/show_bug.cgi?id=31660
        REGRESSION (r49394): Image load event fires before the document fragment is attached.

        Test: fast/images/image-load-event-in-fragment.html

        We cannot fire load events immediately after parsing a fragment - an element inserted via
        innerHTML or equivalent should be reachable via getElementById() by the time its load event
        fires.

        * dom/Document.cpp: (WebCore::Document::implicitClose): Call dispatchPendingBeforeLoadEvents()
        and dispatchPendingLoadEvents() individually. I'm not sure why these calls are here, just
        preserving existing behavior.

        * dom/XMLTokenizer.cpp: (WebCore::XMLTokenizer::write):
        * html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::write):
        Load events for elements that have just been parsed cannot be dispatched yet, we should give
        the elements a chance to be inserted into a document first. There is no compatibility baggage
        for beforeload, so it seems OK to dispatch it right away.

        * loader/ImageLoader.cpp:
        (WebCore::ImageLoader::dispatchPendingBeforeLoadEvents):
        (WebCore::ImageLoader::dispatchPendingLoadEvents):
        * loader/ImageLoader.h:
        Separated dispatchPendingEvents() into individual functions for each event kind.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52116 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Jakob Petsovits <jpetsovits@rim.com>
eric@webkit.org [Mon, 14 Dec 2009 21:27:11 +0000 (21:27 +0000)]
2009-12-14  Jakob Petsovits  <jpetsovits@rim.com>

        Reviewed by Simon Fraser.

        Add FloatQuad::isRectilinear() to check whether it can be represented as FloatRect.
        https://bugs.webkit.org/show_bug.cgi?id=30442

        This enables optimizations for distinguishing between
        arbitrary FloatQuads and rectilinear ones.

        * platform/graphics/FloatQuad.cpp:
        (WebCore::FloatQuad::isRectilinear):
        * platform/graphics/FloatQuad.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52115 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRubber-stamped by Adam Roben.
bweinstein@apple.com [Mon, 14 Dec 2009 21:08:55 +0000 (21:08 +0000)]
Rubber-stamped by Adam Roben.

Add two XSS tests that intermittently time out to the Windows
Skipped list.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52114 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Simon Hausmann <hausmann@webkit.org>
eric@webkit.org [Mon, 14 Dec 2009 20:57:32 +0000 (20:57 +0000)]
2009-12-14  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        [Qt] Extend basic touch event test with a simple keyboard modifier test.

        https://bugs.webkit.org/show_bug.cgi?id=32482

        * fast/events/basic-touch-events-expected.txt:
        * fast/events/script-tests/basic-touch-events.js:
        (singleTouchSequence):
2009-12-14  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        [Qt] Add support for keyboard modifiers to TouchEvent

        https://bugs.webkit.org/show_bug.cgi?id=32482

        Similar to other ui events with keyboard state, get the alt, shift,
        meta and ctrl modifiers straight from the platform event.

        * dom/TouchEvent.cpp:
        (WebCore::TouchEvent::TouchEvent):
        (WebCore::TouchEvent::initTouchEvent):
        * dom/TouchEvent.h:
        (WebCore::TouchEvent::create):
        * dom/TouchEvent.idl:
        * page/EventHandler.cpp:
        (WebCore::EventHandler::handleTouchEvent):
        * platform/PlatformTouchEvent.h:
        (WebCore::PlatformTouchEvent::PlatformTouchEvent):
        (WebCore::PlatformTouchEvent::shiftKey):
        (WebCore::PlatformTouchEvent::ctrlKey):
        (WebCore::PlatformTouchEvent::altKey):
        (WebCore::PlatformTouchEvent::metaKey):
        * platform/qt/PlatformTouchEventQt.cpp:
        (WebCore::PlatformTouchEvent::PlatformTouchEvent):
2009-12-14  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        [Qt] Add support for keyboard modifiers to Qt DRT's EventSender for touch events

        https://bugs.webkit.org/show_bug.cgi?id=32482

        * DumpRenderTree/qt/EventSenderQt.cpp:
        (EventSender::setTouchModifier):
        (EventSender::clearTouchPoints):
        (EventSender::sendTouchEvent):
        * DumpRenderTree/qt/EventSenderQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52113 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Benjamin Poulain <benjamin.poulain@nokia.com>
eric@webkit.org [Mon, 14 Dec 2009 20:49:51 +0000 (20:49 +0000)]
2009-12-14  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] crash when clicking "Sign out" in GMail
        https://bugs.webkit.org/show_bug.cgi?id=32300

        When the ResourceLoader cancels the handle we have to make sure not to access
        the resource handle afterwards again.
        This complete r29515

        * platform/network/qt/QNetworkReplyHandler.cpp:
        (WebCore::QNetworkReplyHandler::sendResponseIfNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52112 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Benjamin Poulain <benjamin.poulain@nokia.com>
eric@webkit.org [Mon, 14 Dec 2009 20:42:41 +0000 (20:42 +0000)]
2009-12-14  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Improve the autotests of QtWebkit
        https://bugs.webkit.org/show_bug.cgi?id=32216

        Refactor tst_qwebelement to remove the qWait()

        * tests/qwebelement/tst_qwebelement.cpp:
        (tst_QWebElement::style):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52111 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Andreas Kling <andreas.kling@nokia.com>
eric@webkit.org [Mon, 14 Dec 2009 20:35:31 +0000 (20:35 +0000)]
2009-12-14  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Simon Hausmann.

        Fix the QWebPage inputMethods() autotest after r51758
        to compare the Qt::ImFont property's family against an explicitly
        previously configured family.

        https://bugs.webkit.org/show_bug.cgi?id=32491

        * tests/qwebpage/tst_qwebpage.cpp:
        (tst_QWebPage::inputMethods):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52110 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Robert Hogan <robert@roberthogan.net>
eric@webkit.org [Mon, 14 Dec 2009 20:28:24 +0000 (20:28 +0000)]
2009-12-14  Robert Hogan  <robert@roberthogan.net>

        Reviewed by Eric Seidel.

        access-control-basic-denied-preflight-cache.html fails manual testing.

        LayoutTests/http/tests/xmlhttprequest/access-control-basic-denied-preflight-cache.html
        fails manual testing. It passes the automatic layout tests in Qt due to a bug with
        notifyDone (see https://bugs.webkit.org/show_bug.cgi?id=32437).

        Performing the test manually in QtLauncher shows that an exception is received.

        Add access-control-basic-denied-preflight-cache.html to the Qt skipped list.

        https://bugs.webkit.org/show_bug.cgi?id=32521

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52109 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Alexander Pavlov <apavlov@chromium.org>
eric@webkit.org [Mon, 14 Dec 2009 20:21:13 +0000 (20:21 +0000)]
2009-12-14  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Implement InjectedScript.evaluateOnSelf() that evaluates functions in the InjectedScript context.
        https://bugs.webkit.org/show_bug.cgi?id=32517

        * inspector/front-end/InjectedScript.js:
        (InjectedScript.evaluateOnSelf):
        * inspector/front-end/InjectedScriptAccess.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52108 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Nayan Kumar K <nayankk@gmail.com>
eric@webkit.org [Mon, 14 Dec 2009 20:13:45 +0000 (20:13 +0000)]
2009-12-14  Nayan Kumar K  <nayankk@gmail.com>

        Reviewed by Darin Adler.

        Fix for compilation errors in WML enabled builds.

        https://bugs.webkit.org/show_bug.cgi?id=32481

        * wml/WMLInputElement.cpp:
        (WebCore::WMLInputElement::setValue):
        (WebCore::WMLInputElement::setValueForUser):
        * wml/WMLInputElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52107 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Mon, 14 Dec 2009 20:06:47 +0000 (20:06 +0000)]
2009-12-14  Eric Carlson  <eric.carlson@apple.com>

        Not reviewed, build fix.

        Add missing newline at end of test result.

        * platform/mac-tiger/media/video-no-audio-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52106 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Android] Android needs ResourceError, ResourceResponse and
benm@google.com [Mon, 14 Dec 2009 20:01:22 +0000 (20:01 +0000)]
[Android] Android needs ResourceError, ResourceResponse and
ResourceRequest implementation.
https://bugs.webkit.org/show_bug.cgi?id=32522

Patch by Andrei Popescu <andreip@google.com> on 2009-12-14
Reviewed by Darin Adler.

Add implementation of ResourceError, ResourceResponse and ResourceRequest.
No new tests needed, the existing tests should be sufficient.

* platform/network/android/ResourceError.h: Added.
(WebCore::ResourceError::ResourceError):
* platform/network/android/ResourceRequest.h: Added.
(WebCore::ResourceRequest::ResourceRequest):
(WebCore::ResourceRequest::doUpdatePlatformRequest):
(WebCore::ResourceRequest::doUpdateResourceRequest):
(WebCore::ResourceRequest::setUserGesture):
(WebCore::ResourceRequest::getUserGesture):
* platform/network/android/ResourceResponse.h: Added.
(WebCore::ResourceResponse::ResourceResponse):
(WebCore::ResourceResponse::doUpdateResourceResponse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52105 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix the build.
weinig@apple.com [Mon, 14 Dec 2009 19:38:37 +0000 (19:38 +0000)]
Fix the build.

* page/Geolocation.cpp:
* page/Geolocation.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52104 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix for https://bugs.webkit.org/show_bug.cgi?id=32499
weinig@apple.com [Mon, 14 Dec 2009 19:26:50 +0000 (19:26 +0000)]
Fix for https://bugs.webkit.org/show_bug.cgi?id=32499
Add client based Geolocation provider

Reviewed by Dan Bernstein.

Add first cut of a client based Geolocation provider. This is guarded by
ENABLE(CLIENT_BASED_GEOLOCATION) and is off by default for now. This adds a
GeolocationControllerClient interface that no-one currently implements,
but will in a subsequent patch.

WebCore:

* GNUmakefile.am:
* WebCore.base.exp:
* WebCore.gypi:
* WebCore.pro:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* WebCoreSources.bkl:
* page/Geolocation.cpp:
(WebCore::createGeopositionFromGeolocationPosition):
(WebCore::createPositionErrorFromGeolocationError):
(WebCore::Geolocation::Geolocation):
(WebCore::Geolocation::~Geolocation):
(WebCore::Geolocation::disconnectFrame):
(WebCore::Geolocation::lastPosition):
(WebCore::Geolocation::startRequest):
(WebCore::Geolocation::fatalErrorOccurred):
(WebCore::Geolocation::requestTimedOut):
(WebCore::Geolocation::clearWatch):
(WebCore::Geolocation::handleError):
(WebCore::Geolocation::positionChanged):
(WebCore::Geolocation::makeSuccessCallbacks):
(WebCore::Geolocation::errorOccurred):
(WebCore::Geolocation::geolocationServicePositionChanged):
(WebCore::Geolocation::geolocationServiceErrorOccurred):
(WebCore::Geolocation::startUpdating):
(WebCore::Geolocation::stopUpdating):
* page/Geolocation.h:
* page/GeolocationController.cpp: Added.
(WebCore::GeolocationController::GeolocationController):
(WebCore::GeolocationController::~GeolocationController):
(WebCore::GeolocationController::addObserver):
(WebCore::GeolocationController::removeObserver):
(WebCore::GeolocationController::positionChanged):
(WebCore::GeolocationController::errorOccurred):
(WebCore::GeolocationController::lastPosition):
* page/GeolocationController.h: Added.
* page/GeolocationControllerClient.h: Added.
(WebCore::GeolocationControllerClient::~GeolocationControllerClient):
* page/GeolocationError.h: Added.
(WebCore::GeolocationError::):
(WebCore::GeolocationError::create):
(WebCore::GeolocationError::code):
(WebCore::GeolocationError::message):
* page/GeolocationPosition.h: Added.
(WebCore::GeolocationPosition::create):
(WebCore::GeolocationPosition::timestamp):
(WebCore::GeolocationPosition::latitude):
(WebCore::GeolocationPosition::longitude):
(WebCore::GeolocationPosition::accuracy):
(WebCore::GeolocationPosition::altitude):
(WebCore::GeolocationPosition::altitudeAccuracy):
(WebCore::GeolocationPosition::heading):
(WebCore::GeolocationPosition::speed):
(WebCore::GeolocationPosition::canProvideAltitude):
(WebCore::GeolocationPosition::canProvideAltitudeAccuracy):
(WebCore::GeolocationPosition::canProvideHeading):
(WebCore::GeolocationPosition::canProvideSpeed):
(WebCore::GeolocationPosition::GeolocationPosition):
* page/Geoposition.h:
(WebCore::Geoposition::create):
* page/Page.cpp:
(WebCore::Page::Page):
* page/Page.h:
(WebCore::Page::geolocationController):
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::dataChanged):

WebKit/chromium:

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::WebViewImpl):

WebKit/gtk:

* webkit/webkitwebview.cpp:
(webkit_web_view_init):

WebKit/mac:

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

WebKit/qt:

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

WebKit/win:

* WebView.cpp:
(WebView::initWithFrame):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52103 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Adam Langley <agl@google.com>
agl@chromium.org [Mon, 14 Dec 2009 18:45:10 +0000 (18:45 +0000)]
2009-12-14  Adam Langley  <agl@google.com>

        Reviewed by Darin Adler.

        Test that an image's size is valid before reading it.

        Previously, an image that failed to load (m_failed == true) would
        trigger an assertion failure when WebKit tried to get its size.

        https://bugs.webkit.org/show_bug.cgi?id=32176
        http://code.google.com/p/chromium/issues/detail?id=28566

        Test: fast/images/size-failure.html

        * platform/graphics/ImageSource.cpp:
        (WebCore::ImageSource::createFrameAtIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52102 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=32524
ggaren@apple.com [Mon, 14 Dec 2009 18:40:59 +0000 (18:40 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=32524
REGRESSION(52084): fast/dom/prototypes.html failing two CSS tests

Reviewed by Simon Fraser.

* wtf/StdLibExtras.h:
(WTF::bitCount): The original patch put the parentheses in the wrong
place, completely changing the calculation and making it almost always
wrong. Moved the parentheses around the '+' operation, like the original
compiler warning suggested.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52101 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/7458272> REGRESSION (r50470) Reproducible ASSERT in
mitz@apple.com [Mon, 14 Dec 2009 18:04:37 +0000 (18:04 +0000)]
<rdar://problem/7458272> REGRESSION (r50470) Reproducible ASSERT in
RenderWidget::setWidgetGeometry()

Reviewed by Darin Adler.

WebCore:

Test: fast/forms/defer-updateFromElement.html

* html/HTMLFormControlElement.cpp:
(WebCore::updateFromElementCallback): Added this post-attach callback
that invokes updateFromElement().
(WebCore::HTMLFormControlElement::recalcStyle): Instead of calling
updateFromElement(), enqueue a post-attach (in this case, post-style
recalc) callback to do it. updateFromElement() can cause the selection
to change, and in turn trigger synchronous layout, so it must not be
called during style recalc.

LayoutTests:

* fast/forms/defer-updateFromElement-expected.txt: Added.
* fast/forms/defer-updateFromElement.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52100 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 14 Dec 2009 18:01:08 +0000 (18:01 +0000)]
2009-12-14  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: [regression] inspector deletes nodes when editing css.

        Now all the treeoutlines are focusable, sections and their titles
        are also focusable. Change contains couple of drive-by fixes such as
        canceling of the attribute edit nuking attr info and maintaining proper
        selection while deleting nodes.

        https://bugs.webkit.org/show_bug.cgi?id=32460

        * inspector/front-end/CallStackSidebarPane.js:
        (WebInspector.CallStackSidebarPane.prototype.handleShortcut):
        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleView.prototype._promptKeyDown):
        * inspector/front-end/DataGrid.js:
        (WebInspector.DataGrid.prototype._keyDown):
        * inspector/front-end/DatabaseQueryView.js:
        (WebInspector.DatabaseQueryView):
        (WebInspector.DatabaseQueryView.prototype._promptKeyDown):
        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel.this.treeOutline.focusedNodeChanged):
        (WebInspector.ElementsPanel.prototype.get defaultFocusedElement):
        (WebInspector.ElementsPanel.prototype.handleShortcut):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeOutline):
        (WebInspector.ElementsTreeOutline.prototype._keyDown):
        (WebInspector.ElementsTreeElement.prototype._textNodeEditingCommitted):
        (WebInspector.ElementsTreeElement.prototype._editingCancelled):
        * inspector/front-end/Panel.js:
        (WebInspector.Panel.prototype.show):
        (WebInspector.Panel.prototype.get defaultFocusedElement):
        * inspector/front-end/PropertiesSection.js:
        (WebInspector.PropertiesSection):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel):
        (WebInspector.ScriptsPanel.prototype.get defaultFocusedElement):
        (WebInspector.ScriptsPanel.prototype.handleShortcut):
        * inspector/front-end/SidebarPane.js:
        (WebInspector.SidebarPane):
        (WebInspector.SidebarPane.prototype.toggleExpanded):
        (WebInspector.SidebarPane.prototype._onTitleKeyDown):
        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype._loaded):
        * inspector/front-end/TextPrompt.js:
        (WebInspector.TextPrompt):
        * inspector/front-end/inspector.css:
        * inspector/front-end/inspector.html:
        * inspector/front-end/inspector.js:
        (WebInspector.loaded):
        (WebInspector.documentKeyDown):
        (WebInspector.documentCanCopy):
        (WebInspector.documentCopy):
        (WebInspector._searchFieldManualFocus):
        (WebInspector._searchKeyDown):
        (WebInspector.startEditing.cleanUpAfterEditing):
        (WebInspector.startEditing.keyDownEventListener):
        (WebInspector.startEditing):
        * inspector/front-end/treeoutline.js:
        (TreeOutline):
        (TreeOutline._removeChildAtIndex):
        (TreeOutline.prototype._treeKeyDown):
        (TreeElement.prototype.deselect):
        * inspector/front-end/utilities.js:
        ():

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52099 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoChange IWebFramePrivate's vtable to be compatible with Safari 4.0.4
aroben@apple.com [Mon, 14 Dec 2009 17:46:11 +0000 (17:46 +0000)]
Change IWebFramePrivate's vtable to be compatible with Safari 4.0.4

Reviewed by Steve Falkenburg.

Fixes <http://webkit.org/b/32433> REGRESSION (r51567): Right click on
a link element crashes WebKit nightly

* Interfaces/IWebFramePrivate.idl: Moved pauseSVGAnimation to the end
of the interface so the vtable that Safari 4.0.4 sees is unchanged.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52098 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Mon, 14 Dec 2009 17:43:03 +0000 (17:43 +0000)]
2009-12-14  Eric Carlson  <eric.carlson@apple.com>

        Not reviewed, build fix.

        Update Tiger media test results after r52094.

        * platform/mac-tiger/accessibility: Added.
        * platform/mac-tiger/accessibility/media-element-expected.txt: Added.
        * platform/mac-tiger/media/audio-controls-rendering-expected.txt: Added.
        * platform/mac-tiger/media/controls-after-reload-expected.txt: Added.
        * platform/mac-tiger/media/controls-drag-timebar-expected.txt: Added.
        * platform/mac-tiger/media/controls-strict-expected.txt: Added.
        * platform/mac-tiger/media/controls-styling-expected.txt: Added.
        * platform/mac-tiger/media/video-controls-rendering-expected.txt: Added.
        * platform/mac-tiger/media/video-display-toggle-expected.txt: Added.
        * platform/mac-tiger/media/video-no-audio-expected.txt: Added.
        * platform/mac-tiger/media/video-zoom-controls-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52097 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Mon, 14 Dec 2009 17:20:36 +0000 (17:20 +0000)]
2009-12-14  Dimitri Glazkov  <dglazkov@chromium.org>

        Unreviewed, build fix.

        [Chromium] Guard all of JSC's wrapper cache-related members on Document.

        * dom/Document.cpp:
        (WebCore::Document::createWrapperCache): Widened the guards to the whole method.
        * dom/Document.h: Added JSC guards to wrapper cache-related members.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52096 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAX: there's an infinite loop going on for some webpages - facebook.com
cfleizach@apple.com [Mon, 14 Dec 2009 16:25:00 +0000 (16:25 +0000)]
AX: there's an infinite loop going on for some webpages - facebook.com
https://bugs.webkit.org/show_bug.cgi?id=32501

Reviewed by Adele Peterson.

WebCore:

The problem is that parentObjectUnignored() cannot be called
while in accessibilityIsIgnored, because it leads to
recursiveness. Specifically, there were checks to determine
if an object was within a ARIA tree. That check should
1) be moved to all platforms, and 2) not call parentObjectUnignored().

Test: accessibility/hang-in-isignored.html

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::isAllowedChildOfTree):
(WebCore::AccessibilityRenderObject::accessibilityIsIgnored):
* accessibility/AccessibilityRenderObject.h:
* accessibility/mac/AccessibilityObjectMac.mm:
(WebCore::AccessibilityObject::accessibilityPlatformIncludesObject):

LayoutTests:

* accessibility/hang-in-isignored-expected.txt: Added.
* accessibility/hang-in-isignored.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52095 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Mon, 14 Dec 2009 16:16:57 +0000 (16:16 +0000)]
2009-12-14  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

        Use MediaControllerThemeQuickTime on Leopard.

        * rendering/RenderThemeMac.mm:
        (WebCore::mediaControllerTheme):

2009-12-14  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

        Regenerate layout and pixel test results for the new UI.

        * platform/mac-leopard/Skipped:
        * platform/mac-leopard/accessibility/media-element-expected.txt:
        * platform/mac-leopard/media/audio-controls-rendering-expected.checksum:
        * platform/mac-leopard/media/audio-controls-rendering-expected.txt:
        * platform/mac-leopard/media/controls-after-reload-expected.checksum: Added.
        * platform/mac-leopard/media/controls-after-reload-expected.png: Added.
        * platform/mac-leopard/media/controls-after-reload-expected.txt:
        * platform/mac-leopard/media/controls-drag-timebar-expected.txt:
        * platform/mac-leopard/media/controls-strict-expected.checksum: Added.
        * platform/mac-leopard/media/controls-strict-expected.png: Added.
        * platform/mac-leopard/media/controls-strict-expected.txt:
        * platform/mac-leopard/media/controls-strict-mode-expected.txt: Added.
        * platform/mac-leopard/media/controls-styling-expected.checksum: Added.
        * platform/mac-leopard/media/controls-styling-expected.png: Added.
        * platform/mac-leopard/media/controls-styling-expected.txt:
        * platform/mac-leopard/media/video-controls-rendering-expected.checksum:
        * platform/mac-leopard/media/video-controls-rendering-expected.txt:
        * platform/mac-leopard/media/video-display-toggle-expected.checksum:
        * platform/mac-leopard/media/video-display-toggle-expected.txt:
        * platform/mac-leopard/media/video-no-audio-expected.checksum: Added.
        * platform/mac-leopard/media/video-no-audio-expected.png: Added.
        * platform/mac-leopard/media/video-no-audio-expected.txt:
        * platform/mac-leopard/media/video-zoom-controls-expected.checksum:
        * platform/mac-leopard/media/video-zoom-controls-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52094 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoBuild fix
aroben@apple.com [Mon, 14 Dec 2009 15:44:00 +0000 (15:44 +0000)]
Build fix

* WebKit.vcproj/WebKit.vcproj: Added $(DXSDK_DIR)\Lib\x86 to the
libpath for all non-Cairo configurations (it was missing from most of
them).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52093 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMove QuartzCore-related build settings into a vsprops file
aroben@apple.com [Mon, 14 Dec 2009 15:38:01 +0000 (15:38 +0000)]
Move QuartzCore-related build settings into a vsprops file

Fixes <http://webkit.org/b/32518>

Reviewed by Dan Bernstein.

* WebCore.vcproj/WebCore.vcproj: Let VS reformat the files list, added
WebCoreQuartzCore.vsprops to the non-Cairo configurations, moved all
other QuartzCore-related build settings from here...
* WebCore.vcproj/WebCoreQuartzCore.vsprops: ...to here.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52092 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Yael Aharon <yael.aharon@nokia.com>
eric@webkit.org [Mon, 14 Dec 2009 15:21:41 +0000 (15:21 +0000)]
2009-12-14  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Antti Koivisto.

        handling scripts can block UI
        https://bugs.webkit.org/show_bug.cgi?id=27612

        Break execution of external scrips to smaller chunks.
        No new tests since no new functionality was introduced.

        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::HTMLTokenizer):
        (WebCore::HTMLTokenizer::reset):
        (WebCore::HTMLTokenizer::notifyFinished):
        (WebCore::HTMLTokenizer::executeExternalScriptsIfReady):
        (WebCore::HTMLTokenizer::executeExternalScriptsTimerFired):
        (WebCore::HTMLTokenizer::continueExecutingExternalScripts):
        * html/HTMLTokenizer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52091 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed trivial buildfix.
ossy@webkit.org [Mon, 14 Dec 2009 14:13:31 +0000 (14:13 +0000)]
Unreviewed trivial buildfix.

Patch by Gabor Loki <loki@inf.u-szeged.hu> on 2009-12-14
Fix crosses initialization of usedPrimaryBlocks for JSValue32

* runtime/Collector.cpp:
(JSC::Heap::markConservatively):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52090 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Connect QNetworkReply via direct connections.
zecke@webkit.org [Mon, 14 Dec 2009 11:09:13 +0000 (11:09 +0000)]
[Qt] Connect QNetworkReply via direct connections.

https://bugs.webkit.org/show_bug.cgi?id=32113

In the past we have switched from auto connections to queued
connections to get some more stability. The problem with the
Queued connections is latency. Even when running simple loading
tests the time it takes to runs highly varies and by using
direct connections these tests produce a stable result.

We will have to look out for crashes coming from within the
QNetworkAccessManager due this patch and address them then.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52089 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-14 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Mon, 14 Dec 2009 11:03:09 +0000 (11:03 +0000)]
2009-12-14  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Gustavo Noronha Silva.

        media/video-played-collapse.html is failing
        https://bugs.webkit.org/show_bug.cgi?id=31434

        * platform/gtk/Skipped: Give this test another chance, we believe
        the bug above has been fixed in a previous commit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52088 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Xan Lopez.
kov@webkit.org [Mon, 14 Dec 2009 10:51:29 +0000 (10:51 +0000)]
    Reviewed by Xan Lopez.

        New setting to allow applications to completely supress the
        default context menu that is generated by WebKit. This allows them
        to still pass the vent to the default handler without needing to
        use ugly hacks such as handling populate-popup, and removing all
        the items from the default context menu.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52087 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] QtWebKit should use QImageReader::setQuality(49)
zecke@webkit.org [Mon, 14 Dec 2009 10:48:14 +0000 (10:48 +0000)]
[Qt] QtWebKit should use QImageReader::setQuality(49)

https://bugs.webkit.org/show_bug.cgi?id=30740

Using QImageReader::setQuality(49) will influence the JPEG
decoder of Qt to use the JDCT_IFAST routines for decompressing. This
is showing a 5% improvement in a reduction that decodes all images
used during the page loading test. The libjpeg documentation mentions
that the IFAST routine is not as accurate as other routines but in
our tests we couldn't see a difference.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52086 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoemit volumeChanged and durationChanged upon notification from GStreamer.
philn@webkit.org [Mon, 14 Dec 2009 10:01:33 +0000 (10:01 +0000)]
emit volumeChanged and durationChanged upon notification from GStreamer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52085 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoGCC 4.3.x warning fixed. Suggested parantheses added.
ossy@webkit.org [Mon, 14 Dec 2009 09:56:51 +0000 (09:56 +0000)]
GCC 4.3.x warning fixed. Suggested parantheses added.
warning: ../../../JavaScriptCore/wtf/StdLibExtras.h:77: warning: suggest parentheses around + or - in operand of &

Reviewed by Simon Hausmann.

* wtf/StdLibExtras.h:
(WTF::bitCount):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52084 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRS by Maciej Stachowiak.
adele@apple.com [Mon, 14 Dec 2009 08:39:54 +0000 (08:39 +0000)]
RS by Maciej Stachowiak.

Adding platform specific results for these test as part of reverting changes from http://trac.webkit.org/changeset/49585 to fix:
<rdar://problem/7382815> REGRESSION (4.0.4-TOT): Home photo slider is too narrow at http://www.ziprealty.com/
https://bugs.webkit.org/show_bug.cgi?id=32078

* platform/win/fast/css/large-number-round-trip-expected.txt: Added.
* platform/win/fast/css/opacity-float-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52083 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: Changed GC from mark-sweep to mark-allocate.
ggaren@apple.com [Mon, 14 Dec 2009 08:13:24 +0000 (08:13 +0000)]
JavaScriptCore: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

Added WeakGCMap to keep WebCore blissfully ignorant about objects that
have become garbage but haven't run their destructors yet.

1% SunSpider speedup.
7.6% v8 speedup (37% splay speedup).
17% speedup on bench-alloc-nonretained.js.
18% speedup on bench-alloc-retained.js.

* API/JSBase.cpp:
(JSGarbageCollect):
* API/JSContextRef.cpp:
* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj: Updated for renames and new
files.

* debugger/Debugger.cpp:
(JSC::Debugger::recompileAllJSFunctions): Updated to use the Collector
iterator abstraction.

* jsc.cpp:
(functionGC): Updated for rename.

* runtime/Collector.cpp: Slightly reduced the number of allocations per
collection, so that small workloads only allocate on collector block,
rather than two.

(JSC::Heap::Heap): Updated to use the new allocateBlock function.

(JSC::Heap::destroy): Updated to use the new freeBlocks function.

(JSC::Heap::allocateBlock): New function to initialize a block when
allocating it.

(JSC::Heap::freeBlock): Consolidated the responsibility for running
destructors into this function.

(JSC::Heap::freeBlocks): Updated to use freeBlock.

(JSC::Heap::recordExtraCost): Sweep the heap in this reporting function,
so that allocation, which is more common, doesn't have to check extraCost.

(JSC::Heap::heapAllocate): Run destructors right before recycling a
garbage cell. This has better cache utilization than a separate sweep phase.

(JSC::Heap::resizeBlocks):
(JSC::Heap::growBlocks):
(JSC::Heap::shrinkBlocks): New set of functions for managing the size of
the heap, now that the heap doesn't maintain any information about its
size.

(JSC::isPointerAligned):
(JSC::isHalfCellAligned):
(JSC::isPossibleCell):
(JSC::isCellAligned):
(JSC::Heap::markConservatively): Cleaned up this code a bit.

(JSC::Heap::clearMarkBits):
(JSC::Heap::markedCells): Some helper functions for examining the the mark
bitmap.

(JSC::Heap::sweep): Simplified this function by using a DeadObjectIterator.

(JSC::Heap::markRoots): Reordered some operations for clarity.

(JSC::Heap::objectCount):
(JSC::Heap::addToStatistics):
(JSC::Heap::statistics): Rewrote these functions to calculate an object
count on demand, since the heap doesn't maintain this information by
itself.

(JSC::Heap::reset): New function for resetting the heap once we've
exhausted heap space.

(JSC::Heap::collectAllGarbage): This function matches the old collect()
behavior, but it's now an uncommon function used only by API.

* runtime/Collector.h:
(JSC::CollectorBitmap::count):
(JSC::CollectorBitmap::isEmpty): Added some helper functions for managing
the collector mark bitmap.

(JSC::Heap::reportExtraMemoryCost): Changed reporting from cell equivalents
to bytes, so it's easier to understand.

* runtime/CollectorHeapIterator.h:
(JSC::CollectorHeapIterator::CollectorHeapIterator):
(JSC::CollectorHeapIterator::operator!=):
(JSC::CollectorHeapIterator::operator*):
(JSC::CollectorHeapIterator::advance):
(JSC::::LiveObjectIterator):
(JSC::::operator):
(JSC::::DeadObjectIterator):
(JSC::::ObjectIterator): New iterators for encapsulating details about
heap layout, and what's live and dead on the heap.

* runtime/JSArray.cpp:
(JSC::JSArray::putSlowCase):
(JSC::JSArray::increaseVectorLength): Delay reporting extra cost until
we're fully constructed, so the heap mark phase won't visit us in an
invalid state.

* runtime/JSCell.h:
(JSC::JSCell::):
(JSC::JSCell::createDummyStructure):
(JSC::JSCell::JSCell):
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::JSGlobalData):
* runtime/JSGlobalData.h: Added a dummy cell to simplify allocation logic.

* runtime/JSString.h:
(JSC::jsSubstring): Don't report extra cost for substrings, since they
share a buffer that's already reported extra cost.

* runtime/Tracing.d:
* runtime/Tracing.h: Changed these dtrace hooks not to report object
counts, since they're no longer cheap to compute.

* runtime/UString.h: Updated for renames.

* runtime/WeakGCMap.h: Added.
(JSC::WeakGCMap::isEmpty):
(JSC::WeakGCMap::uncheckedGet):
(JSC::WeakGCMap::uncheckedBegin):
(JSC::WeakGCMap::uncheckedEnd):
(JSC::::get):
(JSC::::take):
(JSC::::set):
(JSC::::uncheckedRemove): Mentioned above.

* wtf/StdLibExtras.h:
(WTF::bitCount): Added a bit population count function, so the heap can
count live objects to fulfill statistics questions.

JavaScriptGlue: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

* JavaScriptGlue.cpp:
(JSCollect): Updated for rename. Fixed a bug where JSGlue would not check
to avoid nested GC calls.

WebCore: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

* ForwardingHeaders/runtime/WeakGCMap.h: Added.
* bindings/js/GCController.cpp:
(WebCore::collect):
(WebCore::GCController::gcTimerFired):
(WebCore::GCController::garbageCollectNow): Updated for rename.

* bindings/js/JSDOMBinding.cpp:
(WebCore::removeWrappers):
(WebCore::hasCachedDOMObjectWrapperUnchecked):
(WebCore::hasCachedDOMObjectWrapper):
(WebCore::hasCachedDOMNodeWrapperUnchecked):
(WebCore::forgetDOMObject):
(WebCore::forgetDOMNode):
(WebCore::isObservableThroughDOM):
(WebCore::markDOMNodesForDocument):
(WebCore::markDOMObjectWrapper):
(WebCore::markDOMNodeWrapper):
* bindings/js/JSDOMBinding.h: Changed DOM wrapper maps to be WeakGCMaps.
Don't ASSERT that an item must be in the WeakGCMap when its destructor
runs, since it might have been overwritten in the map first.

* bindings/js/JSDocumentCustom.cpp:
(WebCore::toJS): Changed Document from a DOM object wrapper to a DOM node
wrapper, to simplify some code.

* bindings/js/JSInspectedObjectWrapper.cpp:
(WebCore::JSInspectedObjectWrapper::JSInspectedObjectWrapper):
(WebCore::JSInspectedObjectWrapper::~JSInspectedObjectWrapper):
* bindings/js/JSInspectorCallbackWrapper.cpp: Use a WeakGCMap for these
wrappers.

* bindings/js/JSNodeCustom.cpp:
(WebCore::JSNode::markChildren): Updated for WeakGCMap and Document using
a DOM node wrapper instead of a DOM object wrapper.

* bindings/js/JSSVGPODTypeWrapper.h:
(WebCore::JSSVGDynamicPODTypeWrapperCache::wrapperMap):
(WebCore::JSSVGDynamicPODTypeWrapperCache::lookupOrCreateWrapper):
(WebCore::JSSVGDynamicPODTypeWrapperCache::forgetWrapper):
(WebCore::::~JSSVGDynamicPODTypeWrapper): Shined a small beam of sanity light
on this code. Use hashtable-based lookup in JSSVGPODTypeWrapper.h instead
of linear lookup through iteration, since that's what hashtables were
invented for. Make JSSVGPODTypeWrapper.h responsible for reomving itself
from the table, instead of its JS wrapper, to decouple these objects from
GC, and because these objects are refCounted, not solely owned by their
JS wrappers.

* bindings/scripts/CodeGeneratorJS.pm:
* dom/Document.h: Adopted changes above.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52082 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRS by Maciej Stachowiak.
adele@apple.com [Mon, 14 Dec 2009 07:01:26 +0000 (07:01 +0000)]
RS by Maciej Stachowiak.

Readding platform specific results for this test as part of reverting changes from http://trac.webkit.org/changeset/49585 to fix:
<rdar://problem/7382815> REGRESSION (4.0.4-TOT): Home photo slider is too narrow at http://www.ziprealty.com/
https://bugs.webkit.org/show_bug.cgi?id=32078

* platform/mac/fast/css/opacity-float-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52081 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Charles Reis <creis@chromium.org>
abarth@webkit.org [Mon, 14 Dec 2009 05:01:01 +0000 (05:01 +0000)]
2009-12-13  Charles Reis  <creis@chromium.org>

        Reviewed by Adam Barth.

        Refactor some security code out of V8 bindings
        https://bugs.webkit.org/show_bug.cgi?id=32326

        No new tests. There should be no functionality changes in this patch,
        since it is only refactoring code.

        * WebCore.gyp/WebCore.gyp:
        * WebCore.gypi:
        * bindings/BindingSecurity.h: Added.
        (WebCore::BindingSecurity::BindingSecurity):
        (WebCore::::canAccessWindow):
        (WebCore::::canAccessFrame):
        (WebCore::::checkNodeSecurity):
        * bindings/BindingSecurityBase.cpp: Added.
        (WebCore::BindingSecurityBase::getDOMWindow):
        (WebCore::BindingSecurityBase::getFrame):
        (WebCore::BindingSecurityBase::canAccessWindow):
        * bindings/BindingSecurityBase.h: Added.
        * bindings/GenericBinding.h: Added.
        (WebCore::):
        * bindings/scripts/CodeGeneratorV8.pm:
        * bindings/v8/ScriptController.cpp:
        (WebCore::ScriptController::isSafeScript):
        * bindings/v8/V8Binding.h:
        * bindings/v8/V8BindingState.cpp: Added.
        (WebCore::::Only):
        (WebCore::::getActiveWindow):
        (WebCore::::immediatelyReportUnsafeAccessTo):
        * bindings/v8/V8BindingState.h: Added.
        (WebCore::):
        * bindings/v8/V8Proxy.cpp:
        (WebCore::V8Proxy::reportUnsafeAccessTo):
        (WebCore::reportUnsafeJavaScriptAccess):
        * bindings/v8/V8Proxy.h:
        (WebCore::V8Proxy::):
        * bindings/v8/custom/V8CustomBinding.cpp:
        (WebCore::allowSettingFrameSrcToJavascriptUrl):
        (WebCore::INDEXED_ACCESS_CHECK):
        (WebCore::NAMED_ACCESS_CHECK):
        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::V8Custom::WindowSetTimeoutImpl):
        (WebCore::ACCESSOR_GETTER):
        (WebCore::ACCESSOR_SETTER):
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::V8Custom::ClearTimeoutImpl):
        (WebCore::NAMED_ACCESS_CHECK):
        (WebCore::INDEXED_ACCESS_CHECK):
        * bindings/v8/custom/V8LocationCustom.cpp:
        (WebCore::ACCESSOR_GETTER):
        (WebCore::CALLBACK_FUNC_DECL):
        (WebCore::INDEXED_ACCESS_CHECK):
        (WebCore::NAMED_ACCESS_CHECK):
2009-12-13  Charles Reis  <creis@chromium.org>

        Reviewed by Adam Barth.

        Refactor some security code out of V8 bindings
        https://bugs.webkit.org/show_bug.cgi?id=32326

        * src/WebBindings.cpp:
        (WebKit::getDragDataImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52080 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Implement initializeMaximumHTTPConnectionCountPerHost.
zecke@webkit.org [Mon, 14 Dec 2009 03:10:56 +0000 (03:10 +0000)]
[Qt] Implement initializeMaximumHTTPConnectionCountPerHost.

Qt allows to have three connections per host on Symbian, for
every other platform the limit is set to six connections per
host but the default implementation will only schedule four
connections per host.
In a manual test we seem to get the best loading speed by
going with Qt's limit and schedule two more jobs to allow Qt
to prepare them.

* platform/network/ResourceRequestBase.cpp:
* platform/network/qt/ResourceRequestQt.cpp:
(WebCore::initializeMaximumHTTPConnectionCountPerHost):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52079 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Eric Seidel <eric@webkit.org>
mjs@apple.com [Mon, 14 Dec 2009 02:38:22 +0000 (02:38 +0000)]
2009-12-13  Eric Seidel  <eric@webkit.org>

        Reviewed by Gavin Barraclough.

        string-base64 test does not compute a valid base64 string
        http://bugs.webkit.org/show_bug.cgi?id=16806

        * tests/string-base64.js: change str[i] to str.charCodeAt(i)
2009-12-13  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Gavin Barraclough.

        SunSpider/tests/string-base64.js does not compute a valid base64 encoded string
        https://bugs.webkit.org/show_bug.cgi?id=16806

        Based on a patch by Eric Seidel.

        Fix the base64 computation to actually compute correct results. The impact on runtime of
        the test is pretty small, but noticeable for some browsers. But at least it's not
        doing a wrong and meaningless computation any more.

        * tests/sunspider-0.9.1/string-base64.js:
        ():
        (base64ToString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52078 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix minor problem in sunspider-compare-results which left it broken.
mjs@apple.com [Mon, 14 Dec 2009 02:26:17 +0000 (02:26 +0000)]
Fix minor problem in sunspider-compare-results which left it broken.

Reviewed by Gavin Barraclaugh.

* Scripts/sunspider-compare-results: Declare $parseonly

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52077 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago* runtime/JSGlobalObjectFunctions.cpp:
barraclough@apple.com [Mon, 14 Dec 2009 01:32:48 +0000 (01:32 +0000)]
* runtime/JSGlobalObjectFunctions.cpp:

Reviewed by NOBODY (speculative Windows build fix).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52076 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=32496
barraclough@apple.com [Mon, 14 Dec 2009 00:27:07 +0000 (00:27 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=32496
Switch remaining cases of string construction to use StringBuilder.
Builds strings using a vector rather than using string append / addition.

Reviewed by Sam Weinig.

* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj:
* runtime/Executable.cpp:
(JSC::FunctionExecutable::paramString):
* runtime/FunctionConstructor.cpp:
(JSC::constructFunction):
* runtime/JSGlobalObjectFunctions.cpp:
(JSC::encode):
(JSC::decode):
(JSC::globalFuncEscape):
(JSC::globalFuncUnescape):
* runtime/JSONObject.cpp:
(JSC::Stringifier::stringify):
(JSC::Stringifier::indent):
* runtime/JSString.h:
* runtime/LiteralParser.cpp:
(JSC::LiteralParser::Lexer::lexString):
* runtime/NumberPrototype.cpp:
(JSC::integerPartNoExp):
(JSC::numberProtoFuncToFixed):
(JSC::numberProtoFuncToPrecision):
* runtime/Operations.h:
(JSC::jsString):
* runtime/StringPrototype.cpp:
(JSC::substituteBackreferencesSlow):
(JSC::substituteBackreferences):
(JSC::stringProtoFuncConcat):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52075 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd a style guideline concerning floating point literals
mitz@apple.com [Mon, 14 Dec 2009 00:14:13 +0000 (00:14 +0000)]
Add a style guideline concerning floating point literals
https://bugs.webkit.org/show_bug.cgi?id=32497

Reviewed by Sam Weinig.

* coding/coding-style.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52074 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/7341364> Crash at HTMLParser::popOneBlockCommon() after
mitz@apple.com [Mon, 14 Dec 2009 00:12:14 +0000 (00:12 +0000)]
<rdar://problem/7341364> Crash at HTMLParser::popOneBlockCommon() after
handling misnested residual style tags

Reviewed by Simon Fraser.

WebCore:

Test: fast/parser/residual-style-close-ref-clone.html

* html/HTMLParser.cpp:
(WebCore::HTMLParser::handleResidualStyleCloseTagAcrossBlocks): Gave the
block stack a strong reference to the cloned residual style element.

LayoutTests:

* fast/parser/residual-style-close-ref-clone-expected.txt: Added.
* fast/parser/residual-style-close-ref-clone.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52073 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFixing commit error...
mjs@apple.com [Sun, 13 Dec 2009 23:53:34 +0000 (23:53 +0000)]
Fixing commit error...

I accidentally committed my last patch in a form that broke Web-hosted SunSpider. Fixing.

* resources/driver-TEMPLATE.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52072 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoReverting changes from http://trac.webkit.org/changeset/49585 to fix:
adele@apple.com [Sun, 13 Dec 2009 23:00:09 +0000 (23:00 +0000)]
Reverting changes from trac.webkit.org/changeset/49585 to fix:
<rdar://problem/7382815> REGRESSION (4.0.4-TOT): Home photo slider is too narrow at http://www.ziprealty.com/
https://bugs.webkit.org/show_bug.cgi?id=32078

Reviewed by Sam Weinig.

WebCore:

* css/CSSPrimitiveValue.cpp: (WebCore::CSSPrimitiveValue::cssText):

LayoutTests:

* fast/css/large-number-round-trip-expected.txt:
* fast/css/large-number-round-trip.html: Updated to run in Firefox.
* fast/css/opacity-float-expected.txt:
* fast/css/opacity-float.html:
* transforms/2d/compound-2d-transforms-expected.txt:
* transforms/2d/compound-2d-transforms.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52071 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Xan Lopez.
kov@webkit.org [Sun, 13 Dec 2009 22:48:23 +0000 (22:48 +0000)]
    Reviewed by Xan Lopez.

        Also give the right click event to the frame's event handler, if
        there WebCore decides to not generate the default context
        menu. This fixes some pages that handle right-button clicks for
        things other than context menu creation.

        * webkit/webkitwebview.cpp:
        (webkit_web_view_forward_context_menu_event):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52070 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Benjamin Poulain <benjamin.poulain@nokia.com>
eric@webkit.org [Sun, 13 Dec 2009 22:34:54 +0000 (22:34 +0000)]
2009-12-13  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Simon Hausmann.

        Add a test in Qt for https://bugs.webkit.org/show_bug.cgi?id=29005
        https://bugs.webkit.org/show_bug.cgi?id=29008

        * tests/qwebframe/tst_qwebframe.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52069 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sun, 13 Dec 2009 20:31:02 +0000 (20:31 +0000)]
2009-12-13  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Sam Weinig.

        Use JSON.parse instead of eval for Web-hosted SunSpider results processing
        https://bugs.webkit.org/show_bug.cgi?id=32490

        * hosted/json2.js: Added. Incorporated from json.org
        * resources/driver-TEMPLATE.html: Fix an HTML validation bug I noticed.
        * resources/results-TEMPLATE.html: Use JSON.parse instead of eval to process
        results.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52068 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoReviewed by Eric Seidel.
jeremy@chromium.org [Sun, 13 Dec 2009 13:33:40 +0000 (13:33 +0000)]
Reviewed by Eric Seidel.

Add code to allow toggling ATSUI/Core Text rendering at runtime in ComplexTextController.
https://bugs.webkit.org/show_bug.cgi?id=31802

The goal here is to allow for a zero runtime hit for ports that decide to select
the API at compile time.
When both USE(ATSUI) and USE(CORE_TEXT) are true, the API is toggled
at runtime.  Core Text is used for OS Versions >= 10.6.

No test since this is already covered by existing pixel tests.

* platform/graphics/SimpleFontData.h:  Change #ifdef to define getNSFont() on Chrome/Mac .
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::shouldUseATSUIAPI): Cache decision about which API to use.
(WebCore::ComplexTextController::ComplexTextRun::indexAt):
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Stub function to call through
to ATSUI/Core text variants.
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): The actual work for object construction
is delegate to createTextRunFromFontDataATSUI/createTextRunFromFontDataCoreText.
* platform/graphics/mac/ComplexTextController.h:
* platform/graphics/mac/ComplexTextControllerATSUI.cpp:
(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation): Small style fix.
(WebCore::ComplexTextController::ComplexTextRun::createTextRunFromFontDataATSUI):
(WebCore::ComplexTextController::collectComplexTextRunsForCharactersATSUI):
* platform/graphics/mac/ComplexTextControllerCoreText.cpp:
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun):
(WebCore::ComplexTextController::ComplexTextRun::createTextRunFromFontDataCoreText):
(WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52067 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Christian Dywan <christian@twotoasts.de>
christian@webkit.org [Sun, 13 Dec 2009 11:32:15 +0000 (11:32 +0000)]
2009-12-13  Christian Dywan  <christian@twotoasts.de>

        Reviewed by Holger Freyther.

        http://bugs.webkit.org/show_bug.cgi?id=31014
        [Gtk] Build fails with gtk+-2.10 (no tested with older releases)

        Guard GTK+ 2.12 specific tooltip code with #ifdef.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52066 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Simon Hausmann <hausmann@webkit.org>
eric@webkit.org [Sun, 13 Dec 2009 09:31:41 +0000 (09:31 +0000)]
2009-12-13  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        [Qt] Re-enable QWebView::renderHints property for Qt for Symbian

        https://bugs.webkit.org/show_bug.cgi?id=28273

        The bug in Qt's moc that triggered a linking error when declaring this
        property has been fixed and we can remove the workaround.

        * Api/qwebview.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52065 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-13 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Sun, 13 Dec 2009 08:44:00 +0000 (08:44 +0000)]
2009-12-13  Eric Carlson  <eric.carlson@apple.com>

        No review, rolling out r52059 until I can generate new Tiger
        test results.

        * rendering/RenderThemeMac.mm:
        (WebCore::mediaControllerTheme):

2009-12-13  Eric Carlson  <eric.carlson@apple.com>

        No review, rolling out r52059 until I can generate new Tiger test
        results.

        * platform/mac-leopard/Skipped:
        * platform/mac-leopard/media/controls-after-reload-expected.txt:
        * platform/mac-leopard/media/controls-drag-timebar-expected.txt:
        * platform/mac-leopard/media/controls-strict-expected.txt:
        * platform/mac-leopard/media/controls-styling-expected.txt:
        * platform/mac-leopard/media/video-controls-rendering-expected.checksum:
        * platform/mac-leopard/media/video-controls-rendering-expected.txt:
        * platform/mac-leopard/media/video-display-toggle-expected.checksum:
        * platform/mac-leopard/media/video-display-toggle-expected.txt:
        * platform/mac-leopard/media/video-no-audio-expected.txt:
        * platform/mac-leopard/media/video-zoom-controls-expected.checksum:
        * platform/mac-leopard/media/video-zoom-controls-expected.txt:
        * platform/mac/media/audio-controls-rendering-expected.checksum:
        * platform/mac/media/video-zoom-expected.checksum:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52064 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRubber-stamped by Dan Bernstein.
weinig@apple.com [Sun, 13 Dec 2009 04:45:31 +0000 (04:45 +0000)]
Rubber-stamped by Dan Bernstein.

Fix the EXPERIMENTAL_SINGLE_VIEW_MODE build.

* Configurations/WebCore.xcconfig:
* page/EventHandler.cpp:
(WebCore::EventHandler::EventHandler):
(WebCore::EventHandler::clearDragState):
* page/mac/DragControllerMac.mm:
(WebCore::DragController::isCopyKeyDown):
* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::eventActivatedView):
* platform/PlatformKeyboardEvent.h:
* platform/PlatformMouseEvent.h:
* platform/PlatformWheelEvent.h:
(WebCore::PlatformWheelEvent::PlatformWheelEvent):
* platform/ScrollView.cpp:
(WebCore::ScrollView::wheelEvent):
* platform/mac/KeyEventMac.mm:
(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
* platform/mac/PlatformMouseEventMac.mm:
(WebCore::PlatformMouseEvent::PlatformMouseEvent):
* platform/mac/PopupMenuMac.mm:
(WebCore::PopupMenu::show):
* platform/mac/WidgetMac.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52063 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sun, 13 Dec 2009 04:26:23 +0000 (04:26 +0000)]
2009-12-12  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        Enable Web-hosted version of SunSpider to handle multiple versions
        https://bugs.webkit.org/show_bug.cgi?id=32478

        * make-hosted: Now generate the hosted version to be able to run both the
        0.9 and the 0.9.1 test suites.
        * hosted: Modified property svn:ignore.
        * hosted/sunspider-driver.html: Removed.
        * hosted/sunspider-results.html: Removed.
        * hosted/sunspider.html:
        * hosted/versions.html: Added.
        * resources/TEMPLATE.html:
        * resources/driver-TEMPLATE.html: Copied from hosted/sunspider-driver.html.
        * resources/driver-TEMPLATE.html: Copied from hosted/sunspider-results.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52062 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRubber-stamped by Dan Bernstein.
weinig@apple.com [Sun, 13 Dec 2009 03:59:30 +0000 (03:59 +0000)]
Rubber-stamped by Dan Bernstein.

Fix the !ENABLE(VIDEO) build.

* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::requiresCompositingForVideo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52061 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRubber-stamped by Dan Bernstein.
weinig@apple.com [Sun, 13 Dec 2009 03:56:00 +0000 (03:56 +0000)]
Rubber-stamped by Dan Bernstein.

Fix the !ENABLE(DATABASE) build.

* storage/SQLTransactionClient.cpp:
* storage/SQLTransactionClient.h:
* storage/SQLTransactionCoordinator.cpp:
* storage/SQLTransactionCoordinator.h:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::stopLoading):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52060 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Sun, 13 Dec 2009 03:28:42 +0000 (03:28 +0000)]
2009-12-12  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

        Use MediaControllerThemeQuickTime on Leopard.

        * rendering/RenderThemeMac.mm:
        (WebCore::mediaControllerTheme):

2009-12-12  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

        Regenerate layout and pixel test results for the new UI.

        * platform/mac-leopard/Skipped:
        * platform/mac-leopard/media/controls-after-reload-expected.txt:
        * platform/mac-leopard/media/controls-drag-timebar-expected.txt:
        * platform/mac-leopard/media/controls-strict-expected.txt:
        * platform/mac-leopard/media/controls-styling-expected.txt:
        * platform/mac-leopard/media/video-controls-rendering-expected.checksum:
        * platform/mac-leopard/media/video-controls-rendering-expected.png:
        * platform/mac-leopard/media/video-controls-rendering-expected.txt:
        * platform/mac-leopard/media/video-display-toggle-expected.checksum:
        * platform/mac-leopard/media/video-display-toggle-expected.png:
        * platform/mac-leopard/media/video-display-toggle-expected.txt:
        * platform/mac-leopard/media/video-no-audio-expected.txt:
        * platform/mac-leopard/media/video-zoom-controls-expected.checksum:
        * platform/mac-leopard/media/video-zoom-controls-expected.png:
        * platform/mac-leopard/media/video-zoom-controls-expected.txt:
        * platform/mac/media/audio-controls-rendering-expected.checksum:
        * platform/mac/media/audio-controls-rendering-expected.png:
        * platform/mac/media/video-zoom-expected.checksum:
        * platform/mac/media/video-zoom-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52059 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Kenneth Russell <kbr@google.com>
eric@webkit.org [Sun, 13 Dec 2009 03:02:14 +0000 (03:02 +0000)]
2009-12-12  Kenneth Russell  <kbr@google.com>

        Reviewed by Oliver Hunt.

        Errors in WebGL[T]Array set() and get() methods
        https://bugs.webkit.org/show_bug.cgi?id=32456

        Test: fast/canvas/webgl/bug-32456.html

        * fast/canvas/webgl/array-unit-tests.html:
        * fast/canvas/webgl/bug-32456-expected.txt: Added.
        * fast/canvas/webgl/bug-32456.html: Added.
2009-12-12  Kenneth Russell  <kbr@google.com>

        Reviewed by Oliver Hunt.

        Errors in WebGL[T]Array set() and get() methods
        https://bugs.webkit.org/show_bug.cgi?id=32456

        Test: fast/canvas/webgl/bug-32456.html

        * bindings/js/JSWebGLFloatArrayCustom.cpp:
        (WebCore::JSWebGLFloatArray::set):
        * bindings/js/JSWebGLIntArrayCustom.cpp:
        (WebCore::JSWebGLIntArray::set):
        * bindings/js/JSWebGLShortArrayCustom.cpp:
        (WebCore::JSWebGLShortArray::set):
        * bindings/js/JSWebGLUnsignedByteArrayCustom.cpp:
        (WebCore::JSWebGLUnsignedByteArray::set):
        * bindings/js/JSWebGLUnsignedIntArrayCustom.cpp:
        (WebCore::JSWebGLUnsignedIntArray::set):
        * bindings/js/JSWebGLUnsignedShortArrayCustom.cpp:
        (WebCore::JSWebGLUnsignedShortArray::set):
        * bindings/v8/custom/V8WebGLArrayCustom.h:
        (WebCore::setWebGLArray):
        * html/canvas/WebGLFloatArray.idl:
        * html/canvas/WebGLUnsignedIntArray.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52058 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sun, 13 Dec 2009 02:18:53 +0000 (02:18 +0000)]
2009-12-12  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        Give command-line sunspider the ability to handle multiple suites and versions
        https://bugs.webkit.org/show_bug.cgi?id=32477

        * Scripts/run-sunspider: Updated for changes to command-line parameters.
        * Scripts/sunspider-compare-results: ditto
2009-12-12  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        Give command-line sunspider the ability to handle multiple suites and versions
        https://bugs.webkit.org/show_bug.cgi?id=32477

        Each suite/version's set of tests are now in a separate subdirectory of tests/, with a file named LIST
        enumerating the tests. Current suites are:

        sunspider-0.9 - riginal SunSpider content
        sunspider-0.9.1 - SunSpider suite with a few tests fixed (and a few more to come)
        ubench - "ubench" microbenchmark suite
        v8-v4 - version 4 of the v8 benchmark
        parse-only - some parser tests

        Sunspider can now be run with the --suite= parameter to select a
        suite. The default is --suite=sunspider-0.9.

        * .: Modified property svn:ignore to ignore new results directories.
        * make-hosted: Generate based on sunspider-0.9 suite.
        * resources/TEMPLATE.html: Fix an HTML compliance problem.
        * resources/sunspider-standalone-driver.js: Become aware of suite names.
        * sunspider: Modified to handle multiple suites as described above.
        * sunspider-compare-results: ditto
        * tests/LIST: Moved to tests/sunspider-0.9/ and tests/sunspider-0.9.1/
        * tests/LIST-PARSE-ONLY: Moved to parse-only/LIST
        * tests/LIST-UBENCH: Removed.
        * tests/LIST-V8: Removed.
        * tests/3d-cube.js: Moved to tests/sunspider-0.9 and tests/sunspider-0.9.1
        * tests/3d-morph.js: ditto
        * tests/3d-raytrace.js: ditto
        * tests/access-binary-trees.js: ditto
        * tests/access-fannkuch.js: ditto
        * tests/access-nbody.js: ditto
        * tests/access-nsieve.js: ditto
        * tests/bitops-3bit-bits-in-byte.js: ditto
        * tests/bitops-bits-in-byte.js: ditto
        * tests/bitops-bitwise-and.js: ditto
        * tests/bitops-nsieve-bits.js: ditto
        * tests/controlflow-recursive.js: ditto
        * tests/crypto-aes.js: ditto
        * tests/crypto-md5.js: ditto
        * tests/crypto-sha1.js: ditto
        * tests/date-format-tofte.js: ditto
        * tests/date-format-xparb.js: ditto
        * tests/math-cordic.js: ditto
        * tests/math-partial-sums.js: ditto
        * tests/math-spectral-norm.js: ditto
        * tests/regexp-dna.js: ditto
        * tests/string-base64.js: ditto
        * tests/string-fasta.js: ditto
        * tests/string-tagcloud.js: ditto
        * tests/string-unpack-code.js: ditto
        * tests/string-validate-input.js: ditto
        * tests/sunspider-0.9/: Added
        * tests/sunspider-0.9/*.js: Moved from one level up
        * tests/sunspider-0.9.1: Added.
        * tests/sunspider-0.9.1/*.js: Moved from one level up.
        * tests/v8-crypto.js: Moved to v8-v4/ directory.
        * tests/v8-deltablue.js: ditto
        * tests/v8-earley-boyer.js: ditto
        * tests/v8-raytrace.js: ditto
        * tests/v8-regexp.js: ditto
        * tests/v8-richards.js: ditto
        * tests/v8-splay.js: ditto
        * tests/v8-v4: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52057 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Kenneth Russell <kbr@google.com>
eric@webkit.org [Sat, 12 Dec 2009 23:12:31 +0000 (23:12 +0000)]
2009-12-12  Kenneth Russell  <kbr@google.com>

        Reviewed by Oliver Hunt.

        Update WebGL tests to released versions
        https://bugs.webkit.org/show_bug.cgi?id=32457

        * fast/canvas/webgl/resources/utils3d.js:
        (initWebGL.gl.console.window.console.log):
        (initWebGL):
        (loadShader):
        (makeBox):
        (makeSphere):
        (processLoadObj):
        (doLoadObj):
        (doLoadImageTexture):
2009-12-12  Kenneth Russell  <kbr@google.com>

        Reviewed by Oliver Hunt.

        Update WebGL tests to released versions
        https://bugs.webkit.org/show_bug.cgi?id=32457

        * manual-tests/webgl/Earth.html:
        * manual-tests/webgl/ManyPlanetsDeep.html:
        * manual-tests/webgl/SpinningBox.html:
        * manual-tests/webgl/TeapotPerPixel.html:
        * manual-tests/webgl/TeapotPerVertex.html:
        * manual-tests/webgl/resources/CanvasMatrix.js:
        (CanvasMatrix4.prototype.getAsWebGLFloatArray):
        * manual-tests/webgl/resources/utils3d.js:
        (initWebGL.gl.console.window.console.log):
        (initWebGL):
        (loadShader):
        (makeBox):
        (makeSphere):
        (processLoadObj):
        (doLoadObj):
        (doLoadImageTexture):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52056 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Sat, 12 Dec 2009 21:29:36 +0000 (21:29 +0000)]
2009-12-12  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Darin Adler.

       <rdar://problem/7453726> Pull shared UI code into WebKit

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

2009-12-12  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Darin Adler.

       <rdar://problem/7453726> Pull shared UI code into WebKit

       Update media tests for small change in timeline slider thumb position

        * platform/mac/media/audio-controls-rendering-expected.txt:
        * platform/mac/media/controls-after-reload-expected.txt:
        * platform/mac/media/controls-styling-expected.txt:
        * platform/mac/media/video-no-audio-expected.txt:
        * platform/mac/media/video-zoom-controls-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52055 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Christian Dywan <christian@twotoasts.de>
christian@webkit.org [Sat, 12 Dec 2009 19:10:38 +0000 (19:10 +0000)]
2009-12-12  Christian Dywan  <christian@twotoasts.de>

        Reviewed by Dirk Schulze.

        Remove the include to SVGFEDisplacementMap.h, it's not needed at all.

        * platform/graphics/filters/FEComponentTransfer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52054 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNot reviewed, fixing previous comment.
mjs@apple.com [Sat, 12 Dec 2009 18:32:14 +0000 (18:32 +0000)]
Not reviewed, fixing previous comment.

Remove accidental change to sunspider-compare-results

* Scripts/sunspider-compare-results:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52053 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Christian Dywan <christian@twotoasts.de>
christian@webkit.org [Sat, 12 Dec 2009 18:24:42 +0000 (18:24 +0000)]
2009-12-12  Christian Dywan  <christian@twotoasts.de>

        Rubber-stamped by Maciej Stachowiak.

        Only include SVGFEDisplacementMap.h if SVG is enabled.

        * platform/graphics/filters/FEComponentTransfer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52052 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-11 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sat, 12 Dec 2009 17:52:21 +0000 (17:52 +0000)]
2009-12-11  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Oliver Hunt.

        Unify codegen for forward and backward variants of branches
        https://bugs.webkit.org/show_bug.cgi?id=32463

        * jit/JIT.h:
        (JSC::JIT::emit_op_loop): Implemented in terms of forward variant.
        (JSC::JIT::emit_op_loop_if_true): ditto
        (JSC::JIT::emitSlow_op_loop_if_true): ditto
        (JSC::JIT::emit_op_loop_if_false): ditto
        (JSC::JIT::emitSlow_op_loop_if_false): ditto
        (JSC::JIT::emit_op_loop_if_less): ditto
        (JSC::JIT::emitSlow_op_loop_if_less): ditto
        * jit/JITOpcodes.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52051 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-11 Christian Dywan <christian@twotoasts.de>
christian@webkit.org [Sat, 12 Dec 2009 17:14:39 +0000 (17:14 +0000)]
2009-12-11  Christian Dywan  <christian@twotoasts.de>

        Reviewed by Eric Seidel.

        [GTK] Build with --disable-svg is broken
        http://bugs.webkit.org/show_bug.cgi?id=32361

        * webkit/webkitwebframe.cpp:
        (webkit_web_frame_pause_svg_animation): Fix ENABLE(SVG) guards.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52050 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2009-12-12 Dirk Schulze <krit@webkit.org>
eric@webkit.org [Sat, 12 Dec 2009 10:20:57 +0000 (10:20 +0000)]
2009-12-12  Dirk Schulze  <krit@webkit.org>

        Reviewed by Eric Seidel.

        [Cairo][Qt] Alpha blending in SVG filters is wrong
        https://bugs.webkit.org/show_bug.cgi?id=32427

        This patch is a follow-up to:
        http://trac.webkit.org/changeset/51950
        It fixes the alpha blending for Cairo and Qt on SVG Filters.

        Covered by: * LayoutTests/svg/filters/feGaussianBlur.svg
                    * LayoutTests/svg/filters/feComposite.svg
                     and others.

        * platform/graphics/cairo/ImageBufferCairo.cpp:
        (WebCore::ImageBuffer::platformTransformColorSpace):
        * platform/graphics/qt/ImageBufferQt.cpp:
        (WebCore::ImageBuffer::platformTransformColorSpace):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52049 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Missing forward entries when dumping back/forward list
ossy@webkit.org [Sat, 12 Dec 2009 10:13:04 +0000 (10:13 +0000)]
[Qt] Missing forward entries when dumping back/forward list
https://bugs.webkit.org/show_bug.cgi?id=32470

Failing tests added to skiplist until bug fixed.

* platform/qt/Skipped:
- http/tests/navigation/dynamic-iframe-dynamic-form-back-entry.html added
- http/tests/navigation/parsed-iframe-dynamic-form-back-entry.html added

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52048 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRolled out my last patch because the bots were crashing
ggaren@apple.com [Sat, 12 Dec 2009 07:20:27 +0000 (07:20 +0000)]
Rolled out my last patch because the bots were crashing

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52047 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAllow WTFs concept of the main thread to differ from pthreads when necessary.
weinig@apple.com [Sat, 12 Dec 2009 06:30:47 +0000 (06:30 +0000)]
Allow WTFs concept of the main thread to differ from pthreads when necessary.

Reviewed by Anders Carlsson.

* wtf/ThreadingPthreads.cpp:
(WTF::initializeThreading):
(WTF::isMainThread):
* wtf/mac/MainThreadMac.mm:
(WTF::initializeMainThreadPlatform):
(WTF::scheduleDispatchFunctionsOnMainThread):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52046 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: Export some new symbols.
ggaren@apple.com [Sat, 12 Dec 2009 06:22:35 +0000 (06:22 +0000)]
Windows build fix: Export some new symbols.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52045 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: Removed an incorrect #ifdef.
ggaren@apple.com [Sat, 12 Dec 2009 06:02:24 +0000 (06:02 +0000)]
Windows build fix: Removed an incorrect #ifdef.

* bindings/js/GCController.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52044 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed build fix.
eroman@chromium.org [Sat, 12 Dec 2009 05:50:17 +0000 (05:50 +0000)]
Unreviewed build fix.

        [V8] Add a method on ScriptController to get it compiling after http://trac.webkit.org/changeset/52033.

        * bindings/v8/ScriptController.cpp:
        (WebCore::ScriptController::anyPageIsProcessingUserGesture):
        * bindings/v8/ScriptController.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52043 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: Removed some old exports.
ggaren@apple.com [Sat, 12 Dec 2009 05:49:16 +0000 (05:49 +0000)]
Windows build fix: Removed some old exports.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52042 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWindows build fix: Use unsigned instead of uint32_t to avoid dependencies.
ggaren@apple.com [Sat, 12 Dec 2009 05:39:33 +0000 (05:39 +0000)]
Windows build fix: Use unsigned instead of uint32_t to avoid dependencies.

* wtf/StdLibExtras.h:
(WTF::bitCount):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52041 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: Changed GC from mark-sweep to mark-allocate.
ggaren@apple.com [Sat, 12 Dec 2009 05:29:56 +0000 (05:29 +0000)]
JavaScriptCore: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

Added WeakGCMap to keep WebCore blissfully ignorant about objects that
have become garbage but haven't run their destructors yet.

1% SunSpider speedup.
7.6% v8 speedup (37% splay speedup).
17% speedup on bench-alloc-nonretained.js.
18% speedup on bench-alloc-retained.js.

* API/JSBase.cpp:
(JSGarbageCollect):
* API/JSContextRef.cpp:
* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj: Updated for renames and new
files.

* debugger/Debugger.cpp:
(JSC::Debugger::recompileAllJSFunctions): Updated to use the Collector
iterator abstraction.

* jsc.cpp:
(functionGC): Updated for rename.

* runtime/Collector.cpp: Slightly reduced the number of allocations per
collection, so that small workloads only allocate on collector block,
rather than two.

(JSC::Heap::Heap): Updated to use the new allocateBlock function.

(JSC::Heap::destroy): Updated to use the new freeBlocks function.

(JSC::Heap::allocateBlock): New function to initialize a block when
allocating it.

(JSC::Heap::freeBlock): Consolidated the responsibility for running
destructors into this function.

(JSC::Heap::freeBlocks): Updated to use freeBlock.

(JSC::Heap::recordExtraCost): Sweep the heap in this reporting function,
so that allocation, which is more common, doesn't have to check extraCost.

(JSC::Heap::heapAllocate): Run destructors right before recycling a
garbage cell. This has better cache utilization than a separate sweep phase.

(JSC::Heap::resizeBlocks):
(JSC::Heap::growBlocks):
(JSC::Heap::shrinkBlocks): New set of functions for managing the size of
the heap, now that the heap doesn't maintain any information about its
size.

(JSC::isPointerAligned):
(JSC::isHalfCellAligned):
(JSC::isPossibleCell):
(JSC::isCellAligned):
(JSC::Heap::markConservatively): Cleaned up this code a bit.

(JSC::Heap::clearMarkBits):
(JSC::Heap::markedCells): Some helper functions for examining the the mark
bitmap.

(JSC::Heap::sweep): Simplified this function by using a DeadObjectIterator.

(JSC::Heap::markRoots): Reordered some operations for clarity.

(JSC::Heap::objectCount):
(JSC::Heap::addToStatistics):
(JSC::Heap::statistics): Rewrote these functions to calculate an object
count on demand, since the heap doesn't maintain this information by
itself.

(JSC::Heap::reset): New function for resetting the heap once we've
exhausted heap space.

(JSC::Heap::collectAllGarbage): This function matches the old collect()
behavior, but it's now an uncommon function used only by API.

* runtime/Collector.h:
(JSC::CollectorBitmap::count):
(JSC::CollectorBitmap::isEmpty): Added some helper functions for managing
the collector mark bitmap.

(JSC::Heap::reportExtraMemoryCost): Changed reporting from cell equivalents
to bytes, so it's easier to understand.

* runtime/CollectorHeapIterator.h:
(JSC::CollectorHeapIterator::CollectorHeapIterator):
(JSC::CollectorHeapIterator::operator!=):
(JSC::CollectorHeapIterator::operator*):
(JSC::CollectorHeapIterator::advance):
(JSC::::LiveObjectIterator):
(JSC::::operator):
(JSC::::DeadObjectIterator):
(JSC::::ObjectIterator): New iterators for encapsulating details about
heap layout, and what's live and dead on the heap.

* runtime/JSArray.cpp:
(JSC::JSArray::putSlowCase):
(JSC::JSArray::increaseVectorLength): Delay reporting extra cost until
we're fully constructed, so the heap mark phase won't visit us in an
invalid state.

* runtime/JSCell.h:
(JSC::JSCell::):
(JSC::JSCell::createDummyStructure):
(JSC::JSCell::JSCell):
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::JSGlobalData):
* runtime/JSGlobalData.h: Added a dummy cell to simplify allocation logic.

* runtime/JSString.h:
(JSC::jsSubstring): Don't report extra cost for substrings, since they
share a buffer that's already reported extra cost.

* runtime/Tracing.d:
* runtime/Tracing.h: Changed these dtrace hooks not to report object
counts, since they're no longer cheap to compute.

* runtime/UString.h: Updated for renames.

* runtime/WeakGCMap.h: Added.
(JSC::WeakGCMap::isEmpty):
(JSC::WeakGCMap::uncheckedGet):
(JSC::WeakGCMap::uncheckedBegin):
(JSC::WeakGCMap::uncheckedEnd):
(JSC::::get):
(JSC::::take):
(JSC::::set):
(JSC::::uncheckedRemove): Mentioned above.

* wtf/StdLibExtras.h:
(WTF::bitCount): Added a bit population count function, so the heap can
count live objects to fulfill statistics questions.

JavaScriptGlue: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

* JavaScriptGlue.cpp:
(JSCollect): Updated for rename. Fixed a bug where JSGlue would not check
to avoid nested GC calls.

WebCore: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

* ForwardingHeaders/runtime/WeakGCMap.h: Added.
* bindings/js/GCController.cpp:
(WebCore::collect):
(WebCore::GCController::gcTimerFired):
(WebCore::GCController::garbageCollectNow): Updated for rename.

* bindings/js/JSDOMBinding.cpp:
(WebCore::removeWrappers):
(WebCore::hasCachedDOMObjectWrapperUnchecked):
(WebCore::hasCachedDOMObjectWrapper):
(WebCore::hasCachedDOMNodeWrapperUnchecked):
(WebCore::forgetDOMObject):
(WebCore::forgetDOMNode):
(WebCore::isObservableThroughDOM):
(WebCore::markDOMNodesForDocument):
(WebCore::markDOMObjectWrapper):
(WebCore::markDOMNodeWrapper):
* bindings/js/JSDOMBinding.h: Changed DOM wrapper maps to be WeakGCMaps.
Don't ASSERT that an item must be in the WeakGCMap when its destructor
runs, since it might have been overwritten in the map first.

* bindings/js/JSDocumentCustom.cpp:
(WebCore::toJS): Changed Document from a DOM object wrapper to a DOM node
wrapper, to simplify some code.

* bindings/js/JSInspectedObjectWrapper.cpp:
(WebCore::JSInspectedObjectWrapper::JSInspectedObjectWrapper):
(WebCore::JSInspectedObjectWrapper::~JSInspectedObjectWrapper):
* bindings/js/JSInspectorCallbackWrapper.cpp: Use a WeakGCMap for these
wrappers.

* bindings/js/JSNodeCustom.cpp:
(WebCore::JSNode::markChildren): Updated for WeakGCMap and Document using
a DOM node wrapper instead of a DOM object wrapper.

* bindings/js/JSSVGPODTypeWrapper.h:
(WebCore::JSSVGDynamicPODTypeWrapperCache::wrapperMap):
(WebCore::JSSVGDynamicPODTypeWrapperCache::lookupOrCreateWrapper):
(WebCore::JSSVGDynamicPODTypeWrapperCache::forgetWrapper):
(WebCore::::~JSSVGDynamicPODTypeWrapper): Shined a small beam of sanity
on this code. Use hashtable-based lookup in JSSVGPODTypeWrapper.h instead
of linear lookup through iteration, since that's what hashtables were
invented for. Make JSSVGPODTypeWrapper.h responsible for reomving itself
from the table, instead of its JS wrapper, to decouple these objects from
GC, and because these objects are refCounted, not solely owned by their
JS wrappers.

* bindings/scripts/CodeGeneratorJS.pm:
* dom/Document.h: Adopted changes above.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@52040 268f45cc-cd09-0410-ab3c-d52691b4dbfc