WebKit-https.git
9 years agoDFG 64-bit Branch implementation should not be creating a JSValueOperand that
fpizlo@apple.com [Sun, 25 Mar 2012 02:29:25 +0000 (02:29 +0000)]
DFG 64-bit Branch implementation should not be creating a JSValueOperand that
it isn't going to use
https://bugs.webkit.org/show_bug.cgi?id=82136

Reviewed by Geoff Garen.

* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::emitBranch):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112018 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[wx] Unreviewed. Fix the build after WTF move.
kevino@webkit.org [Sun, 25 Mar 2012 01:41:33 +0000 (01:41 +0000)]
[wx] Unreviewed. Fix the build after WTF move.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[wx] Unreviewed. Add header needed for compilation.
kevino@webkit.org [Sun, 25 Mar 2012 01:32:48 +0000 (01:32 +0000)]
[wx] Unreviewed. Add header needed for compilation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG double voting may be overzealous in the case of variables that end up
fpizlo@apple.com [Sun, 25 Mar 2012 00:46:21 +0000 (00:46 +0000)]
DFG double voting may be overzealous in the case of variables that end up
being used as integers
https://bugs.webkit.org/show_bug.cgi?id=82008

Reviewed by Oliver Hunt.

Cleaned up propagation, making the intent more explicit in most places.
Back-propagate NodeUsedAsInt for cases where a node was used in a context
that is known to strongly prefer integers.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::handleCall):
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::dumpCodeOrigin):
(JSC::DFG::Graph::dump):
* dfg/DFGGraph.h:
(Graph):
* dfg/DFGNodeFlags.cpp:
(JSC::DFG::nodeFlagsAsString):
* dfg/DFGNodeFlags.h:
(DFG):
* dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::run):
(JSC::DFG::PredictionPropagationPhase::propagate):
(PredictionPropagationPhase):
(JSC::DFG::PredictionPropagationPhase::mergeDefaultFlags):
(JSC::DFG::PredictionPropagationPhase::vote):
(JSC::DFG::PredictionPropagationPhase::doRoundOfDoubleVoting):
(JSC::DFG::PredictionPropagationPhase::fixupNode):
* dfg/DFGVariableAccessData.h:
(JSC::DFG::VariableAccessData::shouldUseDoubleFormatAccordingToVote):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoclean up NRWT logging/metering, be less verbose
dpranke@chromium.org [Sat, 24 Mar 2012 21:22:41 +0000 (21:22 +0000)]
clean up NRWT logging/metering, be less verbose
https://bugs.webkit.org/show_bug.cgi?id=73846

Reviewed by Ojan Vafai.

This patch totally rewrites the metered_stream module and
changes the way it and the printing module now interact with the
logging subsystem.

The printing module becomes a dumb layout_tests-specific wrapper around
the metered stream. All of the intelligence of how and when to
actually write (or erase) lines and how things interact with
logging moves into the metered stream.

As of now, anything written to the metered stream *bypasses the
logging system*. If a logger is passed to the meter constructor,
the metered stream will also attach a log handler so that all
messages fed to the logger will also be written to the metered
stream. This detangles the logic that used to exist where
sometimes the meter would write to the logger and sometimes the
logger would write to the meter.

All of this makes the logic generic and reusable for
test-webkitpy (coming in another patch). It also helps clarify
what the mysterious "configure_logging" parameter to the printer
constructor used to do (that parameter itself is gone now, and
by default the printer and metered_stream will not ever touch
the logging system; this makes a bunch of tests less fragile).

All of the above should produce no user-visible changes.

However, to also fix the bug in question, we change the message
format in --verbose mode to %H:%M:%S.%frac %pid %message,
eliminating the date, file:lineno, and level parameters. I am
making this change at the same time because it would be
difficult to write things consistently between the logs and the
metered stream if we were still logging the file:lineno.

* Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
  Remove the configure_logging parameter
* Scripts/webkitpy/layout_tests/controllers/worker.py:
  Use metered stream to configure the logging (we don't need a printer)
  and be careful about when and how we manipulate the log level
  and handlers on the root logger.
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:
  Pass in the root logger so that log messages are formatted
  properly, and set the log level outside of any routines called
  by unit tests so as to not interfere.
* Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
  remove a duplicate log message.
* Scripts/webkitpy/layout_tests/views/metered_stream.py:
  complete rewrite :).
* Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
  complete rewrite :).
* Scripts/webkitpy/layout_tests/views/printing.py:
  move the logging configuration into metered_stream, and also
  move the "when should I log progress messages" into metered stream.
* Scripts/webkitpy/layout_tests/views/printing_unittest.py:
  remove tests that are no longer necessary.
* Scripts/webkitpy/performance_tests/perftestsrunner.py:
  remove the configure_logging parameter.
* Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
  update the test; since the printer no longer writes to the log,
  we need to check what is written to the regular_output instead.
  (test_run_test_pause_before_testing):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG::Node::shouldNotSpeculateInteger() should be eliminated
fpizlo@apple.com [Sat, 24 Mar 2012 20:15:57 +0000 (20:15 +0000)]
DFG::Node::shouldNotSpeculateInteger() should be eliminated
https://bugs.webkit.org/show_bug.cgi?id=82123

Reviewed by Geoff Garen.

* dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::execute):
* dfg/DFGNode.h:
(Node):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compilePutByValForByteArray):
(JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCrash in ApplyStyleCommand::applyInlineStyleToNodeRange.
inferno@chromium.org [Sat, 24 Mar 2012 20:03:28 +0000 (20:03 +0000)]
Crash in ApplyStyleCommand::applyInlineStyleToNodeRange.
https://bugs.webkit.org/show_bug.cgi?id=81959

Reviewed by Ryosuke Niwa.

Source/WebCore:

Test: editing/execCommand/apply-style-command-crash.html

* editing/ApplyStyleCommand.cpp:
(WebCore::ApplyStyleCommand::applyInlineStyleToNodeRange): RefPtr the weak
node iterator |node|.
* editing/ApplyStyleCommand.h:
(ApplyStyleCommand): convert |startNode| and |pastEndNode| into PassRefPtr.

LayoutTests:

* editing/execCommand/apply-style-command-crash-expected.txt: Added.
* editing/execCommand/apply-style-command-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoIncrease getByIdSlowCase ConstantSpace/InstructionSpace for CPU(ARM_TRADITIONAL)
commit-queue@webkit.org [Sat, 24 Mar 2012 19:13:27 +0000 (19:13 +0000)]
Increase getByIdSlowCase ConstantSpace/InstructionSpace for CPU(ARM_TRADITIONAL)
https://bugs.webkit.org/show_bug.cgi?id=81521

Increase sequenceGetByIdSlowCaseConstantSpace and sequenceGetByIdSlowCaseInstructionSpace
for CPU(ARM_TRADITIONAL) to fit actual need.

Patch by Yong Li <yoli@rim.com> on 2012-03-24
Reviewed by Oliver Hunt.

* jit/JIT.h:
(JIT):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Replace occurrences of "uri" in the documentation with "URI"
commit-queue@webkit.org [Sat, 24 Mar 2012 17:01:46 +0000 (17:01 +0000)]
[GTK] Replace occurrences of "uri" in the documentation with "URI"
https://bugs.webkit.org/show_bug.cgi?id=82025

Patch by Martin Robinson <mrobinson@igalia.com> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

In documentation "uri" should really be written as "URI" so
we should fix all existing strings.

* webkit/webkitdownload.cpp: Replace "uri" with "URI."
* webkit/webkitnetworkrequest.cpp: Ditto.
* webkit/webkitnetworkresponse.cpp: Ditto.
* webkit/webkitwebhistoryitem.cpp: Ditto.
* webkit/webkitwebresource.cpp: Ditto.
(webkit_web_resource_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdated German translation
kov@webkit.org [Sat, 24 Mar 2012 16:51:22 +0000 (16:51 +0000)]
Updated German translation
https://bugs.webkit.org/show_bug.cgi?id=77834

Patch by Christian Kirbach <Christian.Kirbach@googlemail.com> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

* de.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdated Norwegian bokmål translation for webkitgtk
kov@webkit.org [Sat, 24 Mar 2012 16:46:46 +0000 (16:46 +0000)]
Updated Norwegian bokmål translation for webkitgtk
https://bugs.webkit.org/show_bug.cgi?id=73891

Patch by Kjartan Maraas <kmaraas@gnome.org> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

* nb.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r90755.
kov@webkit.org [Sat, 24 Mar 2012 16:37:08 +0000 (16:37 +0000)]
Unreviewed, rolling out r90755.
http://trac.webkit.org/changeset/90755
https://bugs.webkit.org/show_bug.cgi?id=44410

This is actually less up-to-date then the one we committed
before

* id.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[l10n] [mr] WebKitGTK+ Marathi Translations
kov@webkit.org [Sat, 24 Mar 2012 16:32:09 +0000 (16:32 +0000)]
[l10n] [mr] WebKitGTK+ Marathi Translations
https://bugs.webkit.org/show_bug.cgi?id=82014

Patch by Sandeep Shedmake <sshedmak@redhat.com> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

* mr.po: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdated en_GB translation of WebKitGTK+
kov@webkit.org [Sat, 24 Mar 2012 16:26:39 +0000 (16:26 +0000)]
Updated en_GB translation of WebKitGTK+
https://bugs.webkit.org/show_bug.cgi?id=82093

Patch by Chris Leonard <cjl@laptop.org> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

* en_GB.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd Galician translations for webkitgtk
kov@webkit.org [Sat, 24 Mar 2012 16:21:48 +0000 (16:21 +0000)]
Add Galician translations for webkitgtk
https://bugs.webkit.org/show_bug.cgi?id=67591

Patch by Fran Dieguez <frandieguez@gnome.org> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

* gl.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111997 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSlovenian translation update
kov@webkit.org [Sat, 24 Mar 2012 16:11:20 +0000 (16:11 +0000)]
Slovenian translation update
https://bugs.webkit.org/show_bug.cgi?id=68504

Patch by Matej Urbančič <mateju@svn.gnome.org> on 2012-03-24
Reviewed by Gustavo Noronha Silva.

* sl.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd myself and Martin Robinson to: * GStreamerGraphics * GtkWebKit2PublicAPI * SoupNe...
kov@webkit.org [Sat, 24 Mar 2012 16:08:06 +0000 (16:08 +0000)]
Add myself and Martin Robinson to: * GStreamerGraphics * GtkWebKit2PublicAPI * SoupNetwork and to the newly created WebKitGTKTranslations.

Rubber-stamped by Martin Robinson.

* Scripts/webkitpy/common/config/watchlist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2012-03-24 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sat, 24 Mar 2012 15:36:59 +0000 (15:36 +0000)]
2012-03-24  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: restoring chromium's expectations for effect-custom-parameters-expected.png.

        * platform/chromium-linux/css3/filters/custom/effect-custom-parameters-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt][WK2] Implement PageClient::isViewWindowActive()
commit-queue@webkit.org [Sat, 24 Mar 2012 15:23:56 +0000 (15:23 +0000)]
[Qt][WK2] Implement PageClient::isViewWindowActive()
https://bugs.webkit.org/show_bug.cgi?id=81143

Patch by Jesus Sanchez-Palencia <jesus.palencia@openbossa.org> on 2012-03-24
Reviewed by Kenneth Rohde Christiansen.

Source/WebKit2:

This patch implements PageClient::isViewWindowActive()
now that QQuickCanvas::isActive() is available (from QWindow).

* UIProcess/qt/QtPageClient.cpp:
(QtPageClient::isViewWindowActive):

Tools:

Fix the WrapperWindow from our PlatformWebView
so it correctly creates the platform related
stuff for QWindow, which is not created unless
QWindow::setVisible() or QWindow::show() are called.

* WebKitTestRunner/qt/PlatformWebViewQt.cpp:
(WTR::PlatformWebView::PlatformWebView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, build fix since we use "-Werror=unused-but-set-variable".
jesus@webkit.org [Sat, 24 Mar 2012 13:34:03 +0000 (13:34 +0000)]
Unreviewed, build fix since we use "-Werror=unused-but-set-variable".

The variable 'it' was set at TextureMapperShaderManager.h:110 but never used.

* platform/graphics/texmap/TextureMapperShaderManager.h:
(WebCore::TextureMapperShaderManager::getShaderProgram):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdated chromium baselines wrt button selection changes in r111946.
pfeldman@chromium.org [Sat, 24 Mar 2012 11:51:01 +0000 (11:51 +0000)]
Updated chromium baselines wrt button selection changes in r111946.
Not reviewed.

* platform/chromium-linux/editing/selection/3690703-2-expected.png:
* platform/chromium-linux/editing/selection/3690703-expected.png:
* platform/chromium-linux/editing/selection/3690719-expected.png:
* platform/chromium-mac-leopard/editing/selection/3690703-2-expected.png:
* platform/chromium-mac-leopard/editing/selection/3690703-expected.png:
* platform/chromium-mac-leopard/editing/selection/3690719-expected.png:
* platform/chromium-mac-snowleopard/editing/selection/3690703-2-expected.png:
* platform/chromium-mac-snowleopard/editing/selection/3690703-expected.png:
* platform/chromium-mac-snowleopard/editing/selection/3690719-expected.png:
* platform/chromium-mac/editing/selection/3690703-2-expected.png:
* platform/chromium-mac/editing/selection/3690703-expected.png:
* platform/chromium-mac/editing/selection/3690719-expected.png:
* platform/chromium-win-vista/editing/selection/3690703-2-expected.png: Added.
* platform/chromium-win-vista/editing/selection/3690703-expected.png: Added.
* platform/chromium-win-vista/editing/selection/3690719-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: updating chromium test expectations.
pfeldman@chromium.org [Sat, 24 Mar 2012 10:45:42 +0000 (10:45 +0000)]
Not reviewed: updating chromium test expectations.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded Mac expected results for these two tests. The results were accidentally removed in
mitz@apple.com [Sat, 24 Mar 2012 09:07:44 +0000 (09:07 +0000)]
Added Mac expected results for these two tests. The results were accidentally removed in
r111545, even though these tests were not converted into reftests.

* platform/mac/fast/regions/position-writing-modes-in-variable-width-regions-expected.png: Copied from LayoutTests/platform/mac/fast/regions/position-writing-modes-in-variable-width-regions-expected.png.
* platform/mac/fast/regions/position-writing-modes-in-variable-width-regions-expected.txt: Copied from LayoutTests/platform/mac/fast/regions/position-writing-modes-in-variable-width-regions-expected.txt.
* platform/mac/fast/regions/positioned-objects-inline-static-spanning-regions-rtl-expected.png: Copied from LayoutTests/platform/mac/fast/regions/positioned-objects-inline-static-spanning-regions-rtl-expected.png.
* platform/mac/fast/regions/positioned-objects-inline-static-spanning-regions-rtl-expected.txt: Copied from LayoutTests/platform/mac/fast/regions/positioned-objects-inline-static-spanning-regions-rtl-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded Mac expected results for this test that was added in r111139.
mitz@apple.com [Sat, 24 Mar 2012 08:54:54 +0000 (08:54 +0000)]
Added Mac expected results for this test that was added in r111139.

* platform/mac/compositing/layer-creation/fixed-position-scroll-expected.png: Added.
* platform/mac/compositing/layer-creation/fixed-position-scroll-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded Mac expected results for this test that was added in r111946.
mitz@apple.com [Sat, 24 Mar 2012 08:49:39 +0000 (08:49 +0000)]
Added Mac expected results for this test that was added in r111946.

* platform/mac/editing/selection/selection-button-text-expected.png: Added.
* platform/mac/editing/selection/selection-button-text-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt][WK2] Make TextureMapperShaderManager::getShaderProgram() not be a template.
zeno.albisser@nokia.com [Sat, 24 Mar 2012 08:37:51 +0000 (08:37 +0000)]
[Qt][WK2] Make TextureMapperShaderManager::getShaderProgram() not be a template.
https://bugs.webkit.org/show_bug.cgi?id=82049

Change the getShaderProgram() function to not be a template.
This is a workaround for a compiler bug that leads to an assert
when compiling in debug mode on mac.

Reviewed by Noam Rosenthal.

* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::TextureMapperGL::drawTexture):
(WebCore::TextureMapperGL::beginClip):
* platform/graphics/texmap/TextureMapperShaderManager.h:
(TextureMapperShaderProgram):
(WebCore::TextureMapperShaderManager::getShaderProgram):
(TextureMapperShaderManager):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: chromium expectations updated.
pfeldman@chromium.org [Sat, 24 Mar 2012 08:36:12 +0000 (08:36 +0000)]
Not reviewed: chromium expectations updated.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: chromium expectations updated.
pfeldman@chromium.org [Sat, 24 Mar 2012 08:10:41 +0000 (08:10 +0000)]
Not reviewed: chromium expectations updated.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: chromium expectations updated.
pfeldman@chromium.org [Sat, 24 Mar 2012 07:01:28 +0000 (07:01 +0000)]
Not reviewed: chromium expectations updated.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Incorrect replica originTransform used in CCDamageTracker
shawnsingh@chromium.org [Sat, 24 Mar 2012 06:51:34 +0000 (06:51 +0000)]
[chromium] Incorrect replica originTransform used in CCDamageTracker
https://bugs.webkit.org/show_bug.cgi?id=82118

Reviewed by Adrienne Walker.

Source/WebCore:

Unit test added to CCDamageTrackerTest.cpp

* platform/graphics/chromium/cc/CCDamageTracker.cpp:
(WebCore::CCDamageTracker::extendDamageForRenderSurface):

Source/WebKit/chromium:

* tests/CCDamageTrackerTest.cpp:
(WebKitTests::TEST_F):
(WebKitTests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111982 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: chromium expectations updated.
pfeldman@chromium.org [Sat, 24 Mar 2012 06:44:52 +0000 (06:44 +0000)]
Not reviewed: chromium expectations updated.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: adding chromium baselines.
pfeldman@chromium.org [Sat, 24 Mar 2012 06:27:12 +0000 (06:27 +0000)]
Not reviewed: adding chromium baselines.

* platform/chromium-linux/editing/selection/selection-button-text-expected.png: Added.
* platform/chromium-linux/editing/selection/selection-button-text-expected.txt: Added.
* platform/chromium-mac-leopard/editing/selection/selection-button-text-expected.png: Added.
* platform/chromium-mac-snowleopard/editing/selection/selection-button-text-expected.png: Added.
* platform/chromium-mac/editing/selection/selection-button-text-expected.png: Added.
* platform/chromium-mac/editing/selection/selection-button-text-expected.txt: Added.
* platform/chromium-win-vista/editing/selection/selection-button-text-expected.png: Added.
* platform/chromium-win-vista/editing/selection/selection-button-text-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNot reviewed: chromium expectations updated.
pfeldman@chromium.org [Sat, 24 Mar 2012 06:22:02 +0000 (06:22 +0000)]
Not reviewed: chromium expectations updated.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] When prepainting fails, tiles dirty rects may be cleared
commit-queue@webkit.org [Sat, 24 Mar 2012 05:54:38 +0000 (05:54 +0000)]
[chromium] When prepainting fails, tiles dirty rects may be cleared
https://bugs.webkit.org/show_bug.cgi?id=82107

Patch by Dana Jansens <danakj@chromium.org> on 2012-03-23
Reviewed by Adrienne Walker.

Source/WebCore:

When prepainting, if a tile is unable to be reserved due to memory
limits, we bail out of prepareToUpdateTiles. But we would have
cleared the dirty rect of any previous tiles. This leaves them
in a bad state where their textures are reserved, but their textureIds
are set to 0, and they are not marked dirty. This means that they will
not be updated and displayed if they become visible, since it is
assumed that valid textures with zero textureId must have a dirty
region.

We fix this by not clearing the dirty rects until we know we are
going to update the layer.

Unit test: TiledLayerChromiumTest.pushTilesAfterIdlePaintFailed

* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::prepareToUpdateTiles):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::hasTextureIdForTileAt):
(WebCore):
* platform/graphics/chromium/cc/CCTiledLayerImpl.h:
(CCTiledLayerImpl):

Source/WebKit/chromium:

* tests/TiledLayerChromiumTest.cpp:
(WTF::FakeTextureAllocator::createTexture):
(WTF::FakeLayerTextureUpdater::Texture::updateRect):
(FakeCCTiledLayerImpl):
(WTF::FakeCCTiledLayerImpl::hasTextureIdForTileAt):
(WTF::TEST):
(WTF):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111978 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=81963 WebProcess can get stuck in GC during...
slewis@apple.com [Sat, 24 Mar 2012 05:21:26 +0000 (05:21 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=81963 WebProcess can get stuck in GC during many low memory signals.
<rdar://problem/11094830> WebProcess appears to get stuck in its GC handler (81963).
Remove the call to garbage collect  in low memory signal handler.  Did some testing with hitting the low memory handler
during Membuster and we would get back at most 100k - 200k.  That isn't enough to help the system, and in
that state the GC collection can take a substantial amount of time.

Reviewed by Geoff Garen.

Performance Change, no change in behavior.

* platform/mac/MemoryPressureHandlerMac.mm:
(WebCore::MemoryPressureHandler::releaseMemory):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111977 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdated the Mac expected results for this test after CSS shaders were disabled in...
mitz@apple.com [Sat, 24 Mar 2012 03:30:35 +0000 (03:30 +0000)]
Updated the Mac expected results for this test after CSS shaders were disabled in r111893.
WebKitCSSFilterValue.CSS_FILTER_CUSTOM is no longer defined.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111975 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG Fixup should be able to short-circuit trivial ValueToInt32's
fpizlo@apple.com [Sat, 24 Mar 2012 03:23:02 +0000 (03:23 +0000)]
DFG Fixup should be able to short-circuit trivial ValueToInt32's
https://bugs.webkit.org/show_bug.cgi?id=82030

Reviewed by Michael Saboff.

Takes the fixup() method of the prediction propagation phase and makes it
into its own phase. Adds the ability to short-circuit trivial ValueToInt32
nodes, and mark pure ValueToInt32's as such.

* CMakeLists.txt:
* GNUmakefile.list.am:
* JavaScriptCore.xcodeproj/project.pbxproj:
* Target.pri:
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::makeSafe):
(JSC::DFG::ByteCodeParser::handleCall):
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCommon.h:
* dfg/DFGDriver.cpp:
(JSC::DFG::compile):
* dfg/DFGFixupPhase.cpp: Added.
(DFG):
(FixupPhase):
(JSC::DFG::FixupPhase::FixupPhase):
(JSC::DFG::FixupPhase::run):
(JSC::DFG::FixupPhase::fixupNode):
(JSC::DFG::FixupPhase::fixIntEdge):
(JSC::DFG::performFixup):
* dfg/DFGFixupPhase.h: Added.
(DFG):
* dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::run):
(PredictionPropagationPhase):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111974 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agotryReallocate could break the zero-ed memory invariant of CopiedBlocks
mhahnenberg@apple.com [Sat, 24 Mar 2012 03:11:45 +0000 (03:11 +0000)]
tryReallocate could break the zero-ed memory invariant of CopiedBlocks
https://bugs.webkit.org/show_bug.cgi?id=82087

Reviewed by Filip Pizlo.

Removing this optimization turned out to be ~1% regression on kraken, so I simply
undid the modification to the current block if we fail.

* heap/CopiedSpace.cpp:
(JSC::CopiedSpace::tryReallocate): Undid the reset in the CopiedAllocator if we fail
to reallocate from the current block.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix more Chromium 10.6/10.5 tests that broke with r111917. There's a bug
ojan@chromium.org [Sat, 24 Mar 2012 02:27:51 +0000 (02:27 +0000)]
Fix more Chromium 10.6/10.5 tests that broke with r111917. There's a bug
with garden-o-matic when rebaselining all the Chromium Mac ports.
Other ports seem fine.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] CCLayerTreeHostImpl::scrollBegin() should return ScrollFailed for CCInputH...
wjmaclean@chromium.org [Sat, 24 Mar 2012 02:23:46 +0000 (02:23 +0000)]
[chromium] CCLayerTreeHostImpl::scrollBegin() should return ScrollFailed for CCInputHandlerClient::Gesture type when wheel handlers found.
https://bugs.webkit.org/show_bug.cgi?id=82106

Reviewed by Adrienne Walker.

Source/WebCore:

Existing unit tests updated.

With gesture scroll events now being handled as mouse wheel events, we should not start a gesture scroll when
wheel handlers are present.

* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::scrollBegin):

Source/WebKit/chromium:

* tests/CCLayerTreeHostImplTest.cpp:
(WebKit::TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMove Notifications APIs from WorkerContext.idl to WorkerContextNotifications.idl
abarth@webkit.org [Sat, 24 Mar 2012 02:20:42 +0000 (02:20 +0000)]
Move Notifications APIs from WorkerContext.idl to WorkerContextNotifications.idl
https://bugs.webkit.org/show_bug.cgi?id=79635

Reviewed by Kentaro Hara.

This patch moves the NotificationCenter from WorkerContext to
WorkerContextNotificiations, reducing clutter in WorkerContext.  After
this patch, Notifications are almost entirely contained in the
notifications directory.

* CMakeLists.txt:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* workers/WorkerContext.cpp:
(WebCore::WorkerContext::~WorkerContext):
* workers/WorkerContext.h:
(WebCore):
(WorkerContext):
* workers/WorkerContext.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] RootLayer was not being checked for null causing segfaults very rarely.
commit-queue@webkit.org [Sat, 24 Mar 2012 02:15:59 +0000 (02:15 +0000)]
[chromium] RootLayer was not being checked for null causing segfaults very rarely.
https://bugs.webkit.org/show_bug.cgi?id=82111

Patch by Michal Mocny <mmocny@google.com> on 2012-03-23
Reviewed by Adrienne Walker.

* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::beginDrawingFrame):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::drawLayers):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Fix race bug that clobbers CCLayerImpl updateRect
shawnsingh@chromium.org [Sat, 24 Mar 2012 01:28:12 +0000 (01:28 +0000)]
[chromium] Fix race bug that clobbers CCLayerImpl updateRect
https://bugs.webkit.org/show_bug.cgi?id=82109

Reviewed by Dirk Pranke.

Source/WebCore:

If the main thread commits twice before the impl thread actually
draws, then the updateRect of the first frame gets lost forever,
and not propagated to the damage tracker.

The solution is to accumulate the updateRect. The CCLayerImpl
updateRect is already being correctly cleared at the appropriate
time after drawing.

Unit test added to LayerChromiumTest.cpp.

* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::pushPropertiesTo):

Source/WebKit/chromium:

* tests/LayerChromiumTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Chromium 10.6/10.5 tests that broke with r111917. There's a bug
ojan@chromium.org [Sat, 24 Mar 2012 01:13:46 +0000 (01:13 +0000)]
Fix Chromium 10.6/10.5 tests that broke with r111917. There's a bug
with garden-o-matic when rebaselining all the Chromium Mac ports.
Other ports seem fine.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] Add WebPageCompositor class to BlackBerry WebKit API
commit-queue@webkit.org [Sat, 24 Mar 2012 01:12:23 +0000 (01:12 +0000)]
[BlackBerry] Add WebPageCompositor class to BlackBerry WebKit API
https://bugs.webkit.org/show_bug.cgi?id=81121
RIM PR: 136687

Patch by Arvid Nilsson <anilsson@rim.com> on 2012-03-23
Reviewed by Rob Buis.

The idea is for the API client to use a WebPageCompositor object to
render the WebPage when there's no screen window available.

The WebPageCompositorPrivate becomes reference counted in order to be
kept alive either because it's used for accelerated compositing layers
or used for the WebPageCompositor API. It can still work standalone.

Clean up some code - no need to keep track of a separate boolean
WebPagePrivate::m_isAcceleratedCompositingActive when the state of
m_compositor can tell you.

Also remove duplicated code from WebPagePrivate - go directly to the
compositor object instead, if you need to draw the AC layers.

Reviewed internally by Robin Cao and the hasBlitJobs/blitOnIdle change
by Filip Spacek.

* Api/BackingStore.cpp:
(BlackBerry::WebKit::BackingStore::hasBlitJobs):
(BlackBerry::WebKit::BackingStore::blitOnIdle):
(BlackBerry::WebKit::BackingStorePrivate::blitContents):
(BlackBerry::WebKit::BackingStorePrivate::drawSubLayers):
(BlackBerry::WebKit::BackingStorePrivate::drawAndBlendLayersForDirectRendering):
(BlackBerry::WebKit::BackingStorePrivate::surfaceSize):
(BlackBerry::WebKit::BackingStorePrivate::buffer):
* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::WebPagePrivate):
(BlackBerry::WebKit::WebPagePrivate::drawLayersOnCommit):
(BlackBerry::WebKit::WebPagePrivate::setCompositor):
(BlackBerry::WebKit::WebPagePrivate::setRootLayerCompositingThread):
(BlackBerry::WebKit::WebPagePrivate::createCompositor):
(BlackBerry::WebKit::WebPagePrivate::destroyCompositor):
* Api/WebPage.h:
* Api/WebPageCompositor.cpp:
(BlackBerry::WebKit::WebPageCompositorPrivate::WebPageCompositorPrivate):
(BlackBerry::WebKit::WebPageCompositorPrivate::setContext):
(BlackBerry::WebKit::WebPageCompositorPrivate::hardwareCompositing):
(BlackBerry::WebKit::WebPageCompositorPrivate::setRootLayer):
(BlackBerry::WebKit::WebPageCompositorPrivate::render):
(BlackBerry::WebKit::WebPageCompositorPrivate::drawLayers):
(BlackBerry::WebKit::WebPageCompositorPrivate::animationTimerFired):
(BlackBerry::WebKit::WebPageCompositorPrivate::compositorDestroyed):
(BlackBerry::WebKit::WebPageCompositor::WebPageCompositor):
(BlackBerry::WebKit::WebPageCompositor::~WebPageCompositor):
(BlackBerry::WebKit::WebPageCompositor::client):
(BlackBerry::WebKit::WebPageCompositor::prepareFrame):
(BlackBerry::WebKit::WebPageCompositor::render):
(BlackBerry::WebKit::WebPageCompositor::cleanup):
(BlackBerry::WebKit::WebPageCompositor::contextLost):
* Api/WebPageCompositor.h: Added.
* Api/WebPageCompositorClient.h: Added.
* Api/WebPageCompositor_p.h:
(BlackBerry::WebKit::WebPageCompositorPrivate::create):
(BlackBerry::WebKit::WebPageCompositorPrivate::context):
(BlackBerry::WebKit::WebPageCompositorPrivate::animationFrameTimestamp):
(BlackBerry::WebKit::WebPageCompositorPrivate::page):
(BlackBerry::WebKit::WebPageCompositorPrivate::client):
* Api/WebPage_p.h:
(BlackBerry::WebKit::WebPagePrivate::client):
(BlackBerry::WebKit::WebPagePrivate::isAcceleratedCompositingActive):
(BlackBerry::WebKit::WebPagePrivate::compositor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[MutationObservers] attributeFilter should be case sensitive at all times
rafaelw@chromium.org [Sat, 24 Mar 2012 00:58:06 +0000 (00:58 +0000)]
[MutationObservers] attributeFilter should be case sensitive at all times
https://bugs.webkit.org/show_bug.cgi?id=81822

Reviewed by Ryosuke Niwa.

Source/WebCore:

This removes the behavior in MutationObserverRegistration which treats
attributeFilter as case insensitive for HTML elements in HTML documents.

Relevant tests have been updated.

* dom/MutationObserverRegistration.cpp:
(WebCore::MutationObserverRegistration::resetObservation):
(WebCore::MutationObserverRegistration::shouldReceiveMutationFrom):
* dom/MutationObserverRegistration.h:
(MutationObserverRegistration):

LayoutTests:

* fast/mutation/observe-attributes-expected.txt:
* fast/mutation/observe-attributes.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMove WebSocket APIs from WorkerContext.idl to WorkerContextWebSocket.idl
haraken@chromium.org [Sat, 24 Mar 2012 00:55:59 +0000 (00:55 +0000)]
Move WebSocket APIs from WorkerContext.idl to WorkerContextWebSocket.idl
https://bugs.webkit.org/show_bug.cgi?id=79663

Reviewed by Adam Barth.

For WebKit modularization, this patch moves WebSocket APIs from WorkerContext.idl
to WorkerContextWebSocket.idl.

Tests: http/tests/websocket/tests/hybi/workers/* (No change in test results)

* CMakeLists.txt:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Modules/websockets/WorkerContextWebSocket.idl: Added.
* WebCore.gypi:
* WebCore.xcodeproj/project.xcodeproj:
* workers/WorkerContext.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION: can select text of an input button
rniwa@webkit.org [Sat, 24 Mar 2012 00:54:04 +0000 (00:54 +0000)]
REGRESSION: can select text of an input button
https://bugs.webkit.org/show_bug.cgi?id=13624

Patch by Parag Radke <parag@motorola.com> on 2012-03-23
Reviewed by Ryosuke Niwa.

Source/WebCore:

Text in a button should never have a selection background of its own painted
but if the button is inside an editable area, then the whole button should
have it's selection background painted.

Test: editing/selection/selection-button-text.html

* rendering/RenderButton.h:
Added implementation for the virtual function canBeSelectionLeaf()
which returns true if the button renderer is editable, false other wise.
* rendering/RenderTextFragment.h:
Added implementation for the virtual function canBeSelectionLeaf()
which returns true if the text fragment renderer is editable, false other wise.

LayoutTests:

Added a pixel test case to test the selection background for button label text.

* editing/selection/selection-button-text.html: Added.
* platform/gtk/editing/selection/selection-button-text-expected.png: Added.
* platform/gtk/editing/selection/selection-button-text-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agonrwt: don't wait for safari to exit before exiting after showing the results file
dpranke@chromium.org [Sat, 24 Mar 2012 00:53:34 +0000 (00:53 +0000)]
nrwt: don't wait for safari to exit before exiting after showing the results file
https://bugs.webkit.org/show_bug.cgi?id=81845

Reviewed by Ryosuke Niwa.

Prior to this change, if you ran new-run-webkit-tests and
displayed the HTML results file at the end, we would block
waiting for the user to quit the browser. There doesn't seem to
be a need for that, and the Chromium ports don't do this.

Also, update the mac tests to capture the output and be quiet :).

* Scripts/webkitpy/common/system/executive_mock.py:
  Make popen() testable, implement should_log for it.
* Scripts/webkitpy/layout_tests/port/mac.py:
(MacPort.show_results_html_file):
* Scripts/webkitpy/layout_tests/port/mac_unittest.py:
  Silence some of the tests via outputcapture.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRemove unnecessary ASSERT from LoadableTextTrack::trackElementIndex
eric.carlson@apple.com [Sat, 24 Mar 2012 00:34:32 +0000 (00:34 +0000)]
Remove unnecessary ASSERT from LoadableTextTrack::trackElementIndex
https://bugs.webkit.org/show_bug.cgi?id=82095

Reviewed by Dan Bernstein.

No new tests, removed a bogus ASSERT that was sometimes triggered by tests.

* html/track/LoadableTextTrack.cpp:
(WebCore::LoadableTextTrack::trackElementIndex): Remove ASSERT.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoLayout Test media/video-source-load.html is failing
eric.carlson@apple.com [Sat, 24 Mar 2012 00:29:48 +0000 (00:29 +0000)]
Layout Test media/video-source-load.html is failing
https://bugs.webkit.org/show_bug.cgi?id=82094

Reviewed by Alexey Proskuryakov.

No new tests, tested by media/video-source-load.html.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::sourceWasAdded): Set m_nextChildNodeToConsider when called
    with networkState == NETWORK_EMPTY because the resource selection algorithm has
    not started yet.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded Mac WebKit2 expected results for this test. The fact that it is failing is...
mitz@apple.com [Sat, 24 Mar 2012 00:22:42 +0000 (00:22 +0000)]
Added Mac WebKit2 expected results for this test. The fact that it is failing is tracked by
<http://webkit.org/b/82020>.

* platform/mac-wk2/plugins/netscape-dom-access-and-reload-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago [Mac] No need for platform-specific ENABLE_BLOB values
ap@apple.com [Fri, 23 Mar 2012 23:55:45 +0000 (23:55 +0000)]
    [Mac] No need for platform-specific ENABLE_BLOB values
        https://bugs.webkit.org/show_bug.cgi?id=82102

        Reviewed by David Kilzer.

        * Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd TextFieldDecorator and TextFieldDecorationElement
tkent@chromium.org [Fri, 23 Mar 2012 23:41:22 +0000 (23:41 +0000)]
Add TextFieldDecorator and TextFieldDecorationElement
https://bugs.webkit.org/show_bug.cgi?id=80580

Reviewed by Dimitri Glazkov.

These classes will be used to add a decoration icon to a text field
<input> from outside of WebCore. We have a plan to use them in Chromium
port. So this patch doesn't add TextFieldDecorationElement.{cpp,h} to
non-Chromium build files.

How it works:
When a shadow subtree for a text field is created, we ask ChromeClient
to decorate it. ChromeClient appends a TextFieldDecorationElement to the
existing shadow subtree or another shadow tree. A TextFieldDecorator is
associated to a TextFieldDecorationElement, and it provides decoration
icon images and a click handler.

No new tests. This patch won't make any behavior change for now.

* WebCore.gypi: Add TextFieldDecorationElement.cpp and TextFieldDecorationElement.h
* dom/Element.h:
(WebCore::Element::isTextFieldDecoration):
* html/TextFieldInputType.cpp:
(WebCore::TextFieldInputType::createShadowSubtree):
Calls ChromeClient::willAddTextFieldDecorationsTo() and addTextFieldDecorationsTo().
* html/shadow/TextFieldDecorationElement.cpp: Added.
(WebCore::TextFieldDecorator::~TextFieldDecorator):
(WebCore::TextFieldDecorationElement::TextFieldDecorationElement):
(WebCore::TextFieldDecorationElement::create):
(WebCore::TextFieldDecorationElement::isTextFieldDecoration):
(WebCore::TextFieldDecorationElement::updateImage):
(WebCore::TextFieldDecorationElement::customStyleForRenderer):
Sets the size to a square of the font-size.
(WebCore::TextFieldDecorationElement::createRenderer):
Always creates a RenderImage.
(WebCore::TextFieldDecorationElement::attach):
Sets an image for appropriate state.
(WebCore::TextFieldDecorationElement::isMouseFocusable):
(WebCore::TextFieldDecorationElement::defaultEventHandler):
Delegate a click event to the corresponding function of TextFieldDecorator.
* html/shadow/TextFieldDecorationElement.h: Added.
(TextFieldDecorator): Added.
(TextFieldDecorationElement): Added.
(WebCore::toTextFieldDecorationElement): Added.
* page/ChromeClient.h:
(WebCore::ChromeClient::willAddTextFieldDecorationsTo): Add an empty implementation.
(WebCore::ChromeClient::addTextFieldDecorationsTo): ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] libgcrypt and p11-kit should not be in jhbuild modules
kov@webkit.org [Fri, 23 Mar 2012 23:31:49 +0000 (23:31 +0000)]
[GTK] libgcrypt and p11-kit should not be in jhbuild modules
https://bugs.webkit.org/show_bug.cgi?id=82073

Reviewed by Martin Robinson.

* Scripts/webkitdirs.pm:
(jhbuildConfigurationChanged): new function that isolates the
jhbuild configuration change checking from the autogen logic.
(mustReRunAutogen): code for the new function lived here.
(buildAutotoolsProject): now calls jhbuildConfigurationChanged
itself, and cleans up the jhbuild root if configuration has
changed; it also forces autogen to be run in that case, to
make sure the new libraries are used.
* gtk/jhbuild.modules: removed libgcrypt and p11-kit.
* jhbuild/jhbuild-wrapper:
(ensure_jhbuild): this might be run inside a jhbuild environment,
in which case aclocal fails when trying to use the now deleted
aclocal directory in the jhbuild prefix, so work around that.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r108851.
commit-queue@webkit.org [Fri, 23 Mar 2012 23:00:52 +0000 (23:00 +0000)]
Unreviewed, rolling out r108851.
http://trac.webkit.org/changeset/108851
https://bugs.webkit.org/show_bug.cgi?id=82098

"Revert the addition of unnecessary, deprecated functions in
WKPreferences.h" (Requested by estes on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-03-23

* UIProcess/API/C/WKPreferences.cpp:
* UIProcess/API/C/WKPreferences.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed test_expectations update for chromium.
eae@chromium.org [Fri, 23 Mar 2012 22:54:19 +0000 (22:54 +0000)]
Unreviewed test_expectations update for chromium.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMatch DOM4 spec with respect to DocumentFragment insertion
adamk@chromium.org [Fri, 23 Mar 2012 22:51:25 +0000 (22:51 +0000)]
Match DOM4 spec with respect to DocumentFragment insertion
https://bugs.webkit.org/show_bug.cgi?id=82089

Reviewed by Ryosuke Niwa.

Source/WebCore:

DOM4 specifies the behavior of appendChild, insertBefore, and replaceChild
in terms of "mutation algorithms":

http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#mutation-algorithms

This change updates WebKit to match, in particular with regard to DocumentFragments.
Previously, ContainerNode would remove nodes one at a time, then add them to the new parent.
When combined with MutationObservers, this results in overly-verbose mutation records.
Now we create as few records as possible, matching the spec as well as Gecko's implementation
of MutationObservers.

Note that we still need to check validity each time through the loop,
since inserting a node may dispatch events. In a future change, I hope
to move these events so that they fire only after all nodes are inserted,
but that's too much to tackle all in one.

Tests: fast/mutation/document-fragment-insertion.html

* dom/ContainerNode.cpp:
(WebCore::collectChildrenAndRemoveFromOldParent): New helper method
combining collectTargetNodes() with the removal of the collected nodes from
their old parent, if any.
(WebCore::ContainerNode::insertBefore): Use new helper method instead
of removing nodes one at a time from the fragment.
(WebCore::ContainerNode::replaceChild): ditto. Also removed some redundant asserts
and moved the "do nothing" check out of the loop.
(WebCore::ContainerNode::appendChild): Use new helper method.

LayoutTests:

* fast/dom/Node/fragment-mutation-expected.txt:
* fast/dom/Node/fragment-mutation.html: Removed tests that no longer make sense
since the DocumentFragment has no children by the time DOMSubtreeModified is called.
* fast/mutation/document-fragment-insertion-expected.txt: Added.
* fast/mutation/document-fragment-insertion.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed rebaseline of css3 filter test for chromium linux.
eae@chromium.org [Fri, 23 Mar 2012 22:43:05 +0000 (22:43 +0000)]
Unreviewed rebaseline of css3 filter test for chromium linux.

* platform/chromium-linux/css3/filters/custom/effect-custom-parameters-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[skia] Switch to Skia's implementation of the feMorphology filter.
senorblanco@chromium.org [Fri, 23 Mar 2012 22:41:34 +0000 (22:41 +0000)]
[skia] Switch to Skia's implementation of the feMorphology filter.
https://bugs.webkit.org/show_bug.cgi?id=82085

Reviewed by Dirk Schulze.

Covered by existing tests in svg/filters/.

* WebCore.gypi:
* platform/graphics/filters/FEMorphology.h:
(FEMorphology):
* platform/graphics/filters/skia/FEMorphologySkia.cpp: Added.
(WebCore):
(WebCore::FEMorphology::platformApplySkia):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoPutting the y in 'already'
mitz@apple.com [Fri, 23 Mar 2012 22:37:40 +0000 (22:37 +0000)]
Putting the y in 'already'

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMove Notifications APIs from DOMWindow.idl to DOMWindowNotifications.idl (Part 3)
abarth@webkit.org [Fri, 23 Mar 2012 22:35:53 +0000 (22:35 +0000)]
Move Notifications APIs from DOMWindow.idl to DOMWindowNotifications.idl (Part 3)
https://bugs.webkit.org/show_bug.cgi?id=82027

Reviewed by Kentaro Hara.

Rather than watching for willDetachPage, NotificationCenter can just
override ActiveDOMObject::stop directly to be notified when it needs to
stop.

Switching NotificationCenter over to using ActiveDOMObject callbacks
frees DOMWindowNotification to implement reconnectFrame.  However, this
is only really a victory in princple because NotificationCenter always
returns false for canSuspend(), which means we'll never actually use
the reconnectFrame functionality.  However, this patch does move us
closer to the time when NotificationCenter can implement
ActiveDOMObject::suspend/resume to play nice with the PageCache.

* notifications/DOMWindowNotifications.cpp:
(WebCore::DOMWindowNotifications::disconnectFrame):
(WebCore::DOMWindowNotifications::reconnectFrame):
* notifications/DOMWindowNotifications.h:
(DOMWindowNotifications):
* notifications/NotificationCenter.cpp:
(WebCore::NotificationCenter::NotificationCenter):
(WebCore::NotificationCenter::stop):
* notifications/NotificationCenter.h:
(NotificationCenter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Support file drag out using DataTransferItemList::add(File)
dcheng@chromium.org [Fri, 23 Mar 2012 22:25:16 +0000 (22:25 +0000)]
[chromium] Support file drag out using DataTransferItemList::add(File)
https://bugs.webkit.org/show_bug.cgi?id=81261

Reviewed by Tony Chang.

Source/WebCore:

Original patch by Varun Jain <varunjain@chromium.org>.
I've updated the original patch to remove 'downloadurl' handling since it's not tested, nor
is it necessary for File objects, since File objects correspond to a native filesystem path.

Test: fast/events/drag-dataTransferItemList-file-handling.html

* platform/chromium/DataTransferItemListChromium.cpp:
(WebCore::DataTransferItemListChromium::add):

Source/WebKit/chromium:

Original patch by Varun Jain <varunjain@chromium.org>.
I've updated the original patch to add some asserts for conditions we shouldn't encounter as
well as removing some redundant null checks.

* src/WebDragData.cpp:
(WebKit::WebDragData::items):

LayoutTests:

* fast/events/drag-dataTransferItemList-file-handling-expected.txt: Added.
* fast/events/drag-dataTransferItemList-file-handling.html: Added.
* platform/gtk/Skipped:
* platform/mac/Skipped:
* platform/qt/Skipped:
* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded all tests that crashed more than once on the Lion WebKit2 bots between r111867
mitz@apple.com [Fri, 23 Mar 2012 22:22:08 +0000 (22:22 +0000)]
Added all tests that crashed more than once on the Lion WebKit2 bots between r111867
and r111906 to the Mac WebKit2 skip list (that weren’t alread on the list).

Rubber-stamped by Geoff Garen.

* platform/mac-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agogarden-o-matic scrolls to the top of the page when you click an accordion item
ojan@chromium.org [Fri, 23 Mar 2012 22:19:46 +0000 (22:19 +0000)]
garden-o-matic scrolls to the top of the page when you click an accordion item
https://bugs.webkit.org/show_bug.cgi?id=82088

Reviewed by Adam Barth.

Check the scroll position and make sure that the clicked accordion item is visible.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMark fast/dom/nodesFromRect-basic.html as passing; it seems to have been fixed by...
adamk@chromium.org [Fri, 23 Mar 2012 22:17:30 +0000 (22:17 +0000)]
Mark fast/dom/nodesFromRect-basic.html as passing; it seems to have been fixed by r110594.

Unreviewed test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Gtk] Webkit fails to build with --disable-geolocation
mrobinson@webkit.org [Fri, 23 Mar 2012 22:14:00 +0000 (22:14 +0000)]
[Gtk] Webkit fails to build with --disable-geolocation
https://bugs.webkit.org/show_bug.cgi?id=81451

Patch by Zan Dobersek <zandobersek@gmail.com> on 2012-03-23
Reviewed by Benjamin Poulain.

Wrap uses of WebCore::Geolocation in WebKitGeolocationPolicyDecision
and WebKitWebView in ifdefs, removing compilation errors when compiling
with geolocation support disabled. Affected public API methods are then
stubs, throwing a warning that feature was not enabled at compile-time.

* webkit/webkitgeolocationpolicydecision.cpp:
(_WebKitGeolocationPolicyDecisionPrivate):
(webkit_geolocation_policy_decision_new):
(webkit_geolocation_policy_allow):
(webkit_geolocation_policy_deny):
* webkit/webkitgeolocationpolicydecisionprivate.h:
* webkit/webkitwebview.cpp:
(webkit_web_view_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GObject bindings] Supplemental interfaces are not disabled with the "Conditional...
mrobinson@webkit.org [Fri, 23 Mar 2012 22:10:35 +0000 (22:10 +0000)]
[GObject bindings] Supplemental interfaces are not disabled with the "Conditional" attribute
https://bugs.webkit.org/show_bug.cgi?id=80030

Patch by Zan Dobersek <zandobersek@gmail.com> on 2012-03-23
Reviewed by Martin Robinson.

Source/WebCore:

Changes in CodeGeneratorGObject.pm:
- group implementation of private helpers and guard them inside
an ifdef if a root conditional attribute is present.
- always define the type of a GObject binding, even if the root
conditional feature is not enabled
- the methods are generated as well, but are simply stubs if the
feature is not enabled, throwing a warning upon interaction.
- if the method itself is guarded by a conditional attribute,
the stub implementation gains one more level of depth, throwing
a warning if the method's conditional feature is not enabled
but the root conditional feature is.
- small style fixes in random places.

Also changed are GObject baselines for binding tests.

No new tests - no new testable functionality.

* bindings/gobject/GNUmakefile.am:
* bindings/scripts/CodeGeneratorGObject.pm:
(HumanReadableConditional):
(GenerateConditionalWarn):
(GenerateProperty):
(GenerateProperties):
(GenerateFunction):
(GenerateFunctions):
(GenerateCFile):
(GenerateEventTargetIface):
(Generate):
(WriteData):
* bindings/scripts/test/GObject/WebKitDOMFloat64Array.cpp:
(WebKit):
(WebKit::kit):
(WebKit::wrapFloat64Array):
(webkit_dom_float64array_finalize):
(webkit_dom_float64array_foo):
* bindings/scripts/test/GObject/WebKitDOMFloat64ArrayPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestActiveDOMObject.cpp:
(WebKit):
(WebKit::kit):
(WebKit::wrapTestActiveDOMObject):
(webkit_dom_test_active_dom_object_finalize):
(webkit_dom_test_active_dom_object_exciting_function):
(webkit_dom_test_active_dom_object_post_message):
(webkit_dom_test_active_dom_object_get_exciting_attr):
* bindings/scripts/test/GObject/WebKitDOMTestActiveDOMObjectPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestCallback.cpp:
(WebKit):
(WebKit::kit):
(WebKit::core):
(WebKit::wrapTestCallback):
(webkit_dom_test_callback_finalize):
(webkit_dom_test_callback_callback_with_no_param):
(webkit_dom_test_callback_callback_with_class1param):
(webkit_dom_test_callback_callback_with_class2param):
(webkit_dom_test_callback_callback_with_non_bool_return_type):
(webkit_dom_test_callback_callback_with_string_list):
* bindings/scripts/test/GObject/WebKitDOMTestCallbackPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestCustomNamedGetter.cpp:
(WebKit):
(WebKit::kit):
(WebKit::wrapTestCustomNamedGetter):
(webkit_dom_test_custom_named_getter_finalize):
(webkit_dom_test_custom_named_getter_another_function):
* bindings/scripts/test/GObject/WebKitDOMTestCustomNamedGetterPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestEventConstructor.cpp:
(WebKit):
(WebKit::kit):
(WebKit::wrapTestEventConstructor):
(webkit_dom_test_event_constructor_finalize):
(webkit_dom_test_event_constructor_get_attr1):
(webkit_dom_test_event_constructor_get_attr2):
* bindings/scripts/test/GObject/WebKitDOMTestEventConstructorPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestEventTarget.cpp:
(WebKit):
(WebKit::core):
(WebKit::wrapTestEventTarget):
(webkit_dom_test_event_target_finalize):
(webkit_dom_test_event_target_item):
(webkit_dom_test_event_target_dispatch_event):
* bindings/scripts/test/GObject/WebKitDOMTestEventTargetPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestInterface.cpp:
(WebKit):
(WebKit::kit):
(WebKit::core):
(WebKit::wrapTestInterface):
(webkit_dom_test_interface_finalize):
(webkit_dom_test_interface_supplemental_method1):
(webkit_dom_test_interface_supplemental_method2):
(webkit_dom_test_interface_supplemental_method4):
(webkit_dom_test_interface_get_supplemental_str1):
(webkit_dom_test_interface_get_supplemental_str2):
(webkit_dom_test_interface_set_supplemental_str2):
(webkit_dom_test_interface_get_supplemental_node):
(webkit_dom_test_interface_set_supplemental_node):
* bindings/scripts/test/GObject/WebKitDOMTestInterfacePrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestMediaQueryListListener.cpp:
(WebKit):
(WebKit::kit):
(WebKit::wrapTestMediaQueryListListener):
(webkit_dom_test_media_query_list_listener_finalize):
* bindings/scripts/test/GObject/WebKitDOMTestMediaQueryListListenerPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestNamedConstructor.cpp:
(WebKit):
(WebKit::kit):
(WebKit::wrapTestNamedConstructor):
(webkit_dom_test_named_constructor_finalize):
* bindings/scripts/test/GObject/WebKitDOMTestNamedConstructorPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
(WebKit):
(WebKit::kit):
(WebKit::core):
(WebKit::wrapTestObj):
(webkit_dom_test_obj_finalize):
(webkit_dom_test_obj_set_property):
(webkit_dom_test_obj_get_property):
(webkit_dom_test_obj_constructed):
(webkit_dom_test_obj_class_init):
(webkit_dom_test_obj_init):
(webkit_dom_test_obj_void_method):
(webkit_dom_test_obj_void_method_with_args):
(webkit_dom_test_obj_int_method):
(webkit_dom_test_obj_int_method_with_args):
(webkit_dom_test_obj_obj_method):
(webkit_dom_test_obj_obj_method_with_args):
(webkit_dom_test_obj_method_with_sequence_arg):
(webkit_dom_test_obj_method_that_requires_all_args_and_throws):
(webkit_dom_test_obj_serialized_value):
(webkit_dom_test_obj_idb_key):
(webkit_dom_test_obj_options_object):
(webkit_dom_test_obj_method_with_exception):
(webkit_dom_test_obj_with_script_state_void):
(webkit_dom_test_obj_with_script_state_obj):
(webkit_dom_test_obj_with_script_state_void_exception):
(webkit_dom_test_obj_with_script_state_obj_exception):
(webkit_dom_test_obj_with_script_execution_context):
(webkit_dom_test_obj_with_script_execution_context_and_script_state):
(webkit_dom_test_obj_with_script_execution_context_and_script_state_obj_exception):
(webkit_dom_test_obj_with_script_execution_context_and_script_state_with_spaces):
(webkit_dom_test_obj_method_with_optional_arg):
(webkit_dom_test_obj_method_with_non_optional_arg_and_optional_arg):
(webkit_dom_test_obj_method_with_non_optional_arg_and_two_optional_args):
(webkit_dom_test_obj_method_with_optional_string):
(webkit_dom_test_obj_method_with_optional_string_is_undefined):
(webkit_dom_test_obj_method_with_optional_string_is_null_string):
(webkit_dom_test_obj_conditional_method1):
(webkit_dom_test_obj_conditional_method2):
(webkit_dom_test_obj_conditional_method3):
(webkit_dom_test_obj_class_method):
(webkit_dom_test_obj_class_method_with_optional):
(webkit_dom_test_obj_overloaded_method1):
(webkit_dom_test_obj_convert1):
(webkit_dom_test_obj_convert2):
(webkit_dom_test_obj_convert3):
(webkit_dom_test_obj_convert4):
(webkit_dom_test_obj_convert5):
(webkit_dom_test_obj_mutable_point_function):
(webkit_dom_test_obj_immutable_point_function):
(webkit_dom_test_obj_orange):
(webkit_dom_test_obj_strict_function):
(webkit_dom_test_obj_get_read_only_int_attr):
(webkit_dom_test_obj_get_read_only_string_attr):
(webkit_dom_test_obj_get_read_only_test_obj_attr):
(webkit_dom_test_obj_get_short_attr):
(webkit_dom_test_obj_set_short_attr):
(webkit_dom_test_obj_get_unsigned_short_attr):
(webkit_dom_test_obj_set_unsigned_short_attr):
(webkit_dom_test_obj_get_int_attr):
(webkit_dom_test_obj_set_int_attr):
(webkit_dom_test_obj_get_long_long_attr):
(webkit_dom_test_obj_set_long_long_attr):
(webkit_dom_test_obj_get_unsigned_long_long_attr):
(webkit_dom_test_obj_set_unsigned_long_long_attr):
(webkit_dom_test_obj_get_string_attr):
(webkit_dom_test_obj_set_string_attr):
(webkit_dom_test_obj_get_test_obj_attr):
(webkit_dom_test_obj_set_test_obj_attr):
(webkit_dom_test_obj_get_xml_obj_attr):
(webkit_dom_test_obj_set_xml_obj_attr):
(webkit_dom_test_obj_get_create):
(webkit_dom_test_obj_set_create):
(webkit_dom_test_obj_get_reflected_string_attr):
(webkit_dom_test_obj_set_reflected_string_attr):
(webkit_dom_test_obj_get_reflected_integral_attr):
(webkit_dom_test_obj_set_reflected_integral_attr):
(webkit_dom_test_obj_get_reflected_unsigned_integral_attr):
(webkit_dom_test_obj_set_reflected_unsigned_integral_attr):
(webkit_dom_test_obj_get_reflected_boolean_attr):
(webkit_dom_test_obj_set_reflected_boolean_attr):
(webkit_dom_test_obj_get_reflected_url_attr):
(webkit_dom_test_obj_set_reflected_url_attr):
(webkit_dom_test_obj_get_reflected_custom_integral_attr):
(webkit_dom_test_obj_set_reflected_custom_integral_attr):
(webkit_dom_test_obj_get_reflected_custom_boolean_attr):
(webkit_dom_test_obj_set_reflected_custom_boolean_attr):
(webkit_dom_test_obj_get_reflected_custom_url_attr):
(webkit_dom_test_obj_set_reflected_custom_url_attr):
(webkit_dom_test_obj_get_attr_with_getter_exception):
(webkit_dom_test_obj_set_attr_with_getter_exception):
(webkit_dom_test_obj_get_attr_with_setter_exception):
(webkit_dom_test_obj_set_attr_with_setter_exception):
(webkit_dom_test_obj_get_string_attr_with_getter_exception):
(webkit_dom_test_obj_set_string_attr_with_getter_exception):
(webkit_dom_test_obj_get_string_attr_with_setter_exception):
(webkit_dom_test_obj_set_string_attr_with_setter_exception):
(webkit_dom_test_obj_get_with_script_state_attribute):
(webkit_dom_test_obj_set_with_script_state_attribute):
(webkit_dom_test_obj_get_with_script_execution_context_attribute):
(webkit_dom_test_obj_set_with_script_execution_context_attribute):
(webkit_dom_test_obj_get_with_script_state_attribute_raises):
(webkit_dom_test_obj_set_with_script_state_attribute_raises):
(webkit_dom_test_obj_get_with_script_execution_context_attribute_raises):
(webkit_dom_test_obj_set_with_script_execution_context_attribute_raises):
(webkit_dom_test_obj_get_with_script_execution_context_and_script_state_attribute):
(webkit_dom_test_obj_set_with_script_execution_context_and_script_state_attribute):
(webkit_dom_test_obj_get_with_script_execution_context_and_script_state_attribute_raises):
(webkit_dom_test_obj_set_with_script_execution_context_and_script_state_attribute_raises):
(webkit_dom_test_obj_get_with_script_execution_context_and_script_state_with_spaces_attribute):
(webkit_dom_test_obj_set_with_script_execution_context_and_script_state_with_spaces_attribute):
(webkit_dom_test_obj_get_conditional_attr1):
(webkit_dom_test_obj_set_conditional_attr1):
(webkit_dom_test_obj_get_conditional_attr2):
(webkit_dom_test_obj_set_conditional_attr2):
(webkit_dom_test_obj_get_conditional_attr3):
(webkit_dom_test_obj_set_conditional_attr3):
(webkit_dom_test_obj_get_content_document):
(webkit_dom_test_obj_get_mutable_point):
(webkit_dom_test_obj_set_mutable_point):
(webkit_dom_test_obj_get_immutable_point):
(webkit_dom_test_obj_set_immutable_point):
(webkit_dom_test_obj_get_strict_float):
(webkit_dom_test_obj_set_strict_float):
(webkit_dom_test_obj_get_description):
(webkit_dom_test_obj_get_id):
(webkit_dom_test_obj_set_id):
(webkit_dom_test_obj_get_hash):
* bindings/scripts/test/GObject/WebKitDOMTestObjPrivate.h:
* bindings/scripts/test/GObject/WebKitDOMTestSerializedScriptValueInterface.cpp:
(WebKit):
(WebKit::kit):
(WebKit::core):
(WebKit::wrapTestSerializedScriptValueInterface):
(webkit_dom_test_serialized_script_value_interface_finalize):
(webkit_dom_test_serialized_script_value_interface_accept_transfer_list):
(webkit_dom_test_serialized_script_value_interface_multi_transfer_list):
(webkit_dom_test_serialized_script_value_interface_get_value):
(webkit_dom_test_serialized_script_value_interface_set_value):
(webkit_dom_test_serialized_script_value_interface_get_readonly_value):
(webkit_dom_test_serialized_script_value_interface_get_cached_value):
(webkit_dom_test_serialized_script_value_interface_set_cached_value):
(webkit_dom_test_serialized_script_value_interface_get_cached_readonly_value):
* bindings/scripts/test/GObject/WebKitDOMTestSerializedScriptValueInterfacePrivate.h:

Source/WebKit/gtk:

Add a private macro that can be used when necessary to throw a warning
about a future not enabled at compile-time.

* webkit/webkitglobalsprivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Record painted pixel counts for measuring effectiveness of per-tile painting
commit-queue@webkit.org [Fri, 23 Mar 2012 22:09:01 +0000 (22:09 +0000)]
[chromium] Record painted pixel counts for measuring effectiveness of per-tile painting
https://bugs.webkit.org/show_bug.cgi?id=82057

Patch by Dana Jansens <danakj@chromium.org> on 2012-03-23
Reviewed by Adrienne Walker.

Source/WebCore:

Adds painted pixel metrics to CCOverdrawMetrics class, and calls in
to report painted areas when updating tiles in the TiledLayerChromium
class.

Unit test: TiledLayerChromiumTest.opaqueContentsRegion
           TiledLayerChromiumTest.pixelsPaintedMetrics

* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::prepareToUpdateTiles):
* platform/graphics/chromium/cc/CCOverdrawMetrics.cpp:
(WebCore::CCOverdrawMetrics::CCOverdrawMetrics):
(WebCore::CCOverdrawMetrics::didPaint):
(WebCore):
(WebCore::CCOverdrawMetrics::recordMetricsInternal):
* platform/graphics/chromium/cc/CCOverdrawMetrics.h:
(CCOverdrawMetrics):
(WebCore::CCOverdrawMetrics::pixelsPainted):

Source/WebKit/chromium:

* tests/TiledLayerChromiumTest.cpp:
(WTF::TEST):
(WTF):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111913 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCrash in RenderBlock::splitBlocks.
inferno@chromium.org [Fri, 23 Mar 2012 22:05:21 +0000 (22:05 +0000)]
Crash in RenderBlock::splitBlocks.
https://bugs.webkit.org/show_bug.cgi?id=81926

Reviewed by Julien Chaffraix.

Source/WebCore:

We are updating the :after content before calling splitFlow. The :after content
gets blown away since it will go to the continuation. beforeChild was earlier
set to the first child. Being the last anonymous block, its children gets pulled
up in collapseAnonymousBoxChild and it gets destroyed. So, we need to update
beforeChild value.

Test: fast/multicol/span/update-after-content-before-child-crash.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks):

LayoutTests:

* fast/multicol/span/update-after-content-before-child-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[WinCairo] Unreviewed build fix.
bfulgham@webkit.org [Fri, 23 Mar 2012 22:03:47 +0000 (22:03 +0000)]
[WinCairo] Unreviewed build fix.

* win/WebKit2CFLite.def: Add missing export symbols so that other
tools can link properly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoIntegrate IETC CSS : textshadow tests
tomz@codeaurora.org [Fri, 23 Mar 2012 21:47:24 +0000 (21:47 +0000)]
Integrate IETC CSS : textshadow tests
https://bugs.webkit.org/show_bug.cgi?id=81936

Patch by Dave Tharp <dtharp@codeaurora.org> on 2012-03-23
Reviewed by Adam Barth.

* ietestcenter/TestSupport/FeatureDetection.js: Added.
* ietestcenter/css3/support/TahomaNarrow.ttf: Added.
* ietestcenter/css3/support/TahomaNormal.ttf: Added.
* ietestcenter/css3/support/TahomaT.ttf: Added.
* ietestcenter/css3/support/ahem.ttf: Added.
* ietestcenter/css3/support/black_color.png: Added.
* ietestcenter/css3/support/black_red.png: Added.
* ietestcenter/css3/support/blue_color.png: Added.
* ietestcenter/css3/support/cat.png: Added.
* ietestcenter/css3/support/flexbox_100r_50g_100r_50g_100g.png: Added.
* ietestcenter/css3/support/flexbox_100x100_green.png: Added.
* ietestcenter/css3/support/flexbox_100x100_red.png: Added.
* ietestcenter/css3/support/flexbox_100x100g_red.png: Added.
* ietestcenter/css3/support/flexbox_200x100_red.png: Added.
* ietestcenter/css3/support/flexbox_200x100g_red.png: Added.
* ietestcenter/css3/support/flexbox_300x100_red.png: Added.
* ietestcenter/css3/support/flexbox_300x150_100red.png: Added.
* ietestcenter/css3/support/flexbox_300x200_red.png: Added.
* ietestcenter/css3/support/flexbox_300x50_100_80_baseline_red.png: Added.
* ietestcenter/css3/support/flexbox_300x50_100_80_red.png: Added.
* ietestcenter/css3/support/green_color.png: Added.
* ietestcenter/css3/support/red_color.png: Added.
* ietestcenter/css3/support/red_space_pb.png: Added.
* ietestcenter/css3/support/scope-003.css: Added.
* ietestcenter/css3/support/space_border_box.png: Added.
* ietestcenter/css3/support/syntax-018.css: Added.
* ietestcenter/css3/support/syntax-019.css: Added.
* ietestcenter/css3/support/white_color.png: Added.
* ietestcenter/css3/text/textshadow-001.htm: Added.
* ietestcenter/css3/text/textshadow-002.htm: Added.
* ietestcenter/css3/text/textshadow-003.htm: Added.
* ietestcenter/css3/text/textshadow-004.htm: Added.
* ietestcenter/css3/text/textshadow-005.htm: Added.
* ietestcenter/css3/text/textshadow-006.htm: Added.
* ietestcenter/css3/text/textshadow-007.htm: Added.
* ietestcenter/css3/text/textshadow-008.htm: Added.
* ietestcenter/css3/text/textshadow-009.htm: Added.
* ietestcenter/css3/text/textshadow-010.htm: Added.
* platform/chromium/test_expectations.txt:
* platform/efl/test_expectations.txt:
* platform/gtk/test_expectations.txt:
* platform/mac/test_expectations.txt:
* platform/qt/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Improved composited debug borders
commit-queue@webkit.org [Fri, 23 Mar 2012 21:45:58 +0000 (21:45 +0000)]
[chromium] Improved composited debug borders
https://bugs.webkit.org/show_bug.cgi?id=82071

Patch by Dana Jansens <danakj@chromium.org> on 2012-03-23
Reviewed by Adrienne Walker.

Fix the problem with missing debug borders. Since we append quads
to the list front to back now, we draw the quads in the list from
back to front of the list. Which means if the debug quad is added
after the layer it is drawing around, the layer can draw over top
of its debug quad. Reverse the ordering when appending to fix it.

Make the border red around tiles that are checkerboarded. Missing
tiles in non-root layers are drawn clear, so this makes them more
visible.

Put a blue border around each render surface, and a purple border
on their replicas.

No new tests.

* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore):
(WebCore::CCRenderPass::appendQuadsForLayer):
(WebCore::CCRenderPass::appendQuadsForRenderSurfaceLayer):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::hasReplica):
(WebCore):
(WebCore::CCRenderSurface::createReplicaSharedQuadState):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(CCRenderSurface):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore):
(WebCore::CCTiledLayerImpl::appendQuads):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMove code that calls setAcceleratesDrawing() on GraphicsLayers into RenderLayerBacking
simon.fraser@apple.com [Fri, 23 Mar 2012 21:40:21 +0000 (21:40 +0000)]
Move code that calls setAcceleratesDrawing() on GraphicsLayers into RenderLayerBacking
https://bugs.webkit.org/show_bug.cgi?id=82092

Reviewed by Dean Jackson.

A RenderLayerBacking manages several GraphicsLayers, but RenderLayerCompositor
was only calling setAcceleratesDrawing() on the one it knows about. Fix by moving
this setting code down into RenderLayerBacking.

RenderLayerCompositor now caches the setAcceleratesDrawing setting, as it does
for other compositing-related settings, so that RenderLayerBacking doesn't have
to grovel for Settings every time.

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::createGraphicsLayer):
(WebCore::RenderLayerBacking::createPrimaryGraphicsLayer):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::RenderLayerCompositor):
(WebCore::RenderLayerCompositor::cacheAcceleratedCompositingFlags):
(WebCore::RenderLayerCompositor::updateBacking):
* rendering/RenderLayerCompositor.h:
(WebCore::RenderLayerCompositor::acceleratedDrawingEnabled):
(RenderLayerCompositor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoReverting unintential change to JavaScriptCore xcode project file
msaboff@apple.com [Fri, 23 Mar 2012 21:38:22 +0000 (21:38 +0000)]
Reverting unintential change to JavaScriptCore xcode project file
checked in with change set r111906.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG::compileValueToInt32 Sometime Generates GPR to FPR reg back to GPR
msaboff@apple.com [Fri, 23 Mar 2012 21:27:20 +0000 (21:27 +0000)]
DFG::compileValueToInt32 Sometime Generates GPR to FPR reg back to GPR
https://bugs.webkit.org/show_bug.cgi?id=81805

Reviewed by Filip Pizlo.

Added SpeculativeJIT::checkGeneratedType() to determine the current format
of an operand.  Used that information in SpeculativeJIT::compileValueToInt32
to generate code that will use integer and JSValue types in integer
format directly without a conversion to double.

* JavaScriptCore.xcodeproj/project.pbxproj:
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::checkGeneratedType):
(DFG):
(JSC::DFG::SpeculativeJIT::compileValueToInt32):
* dfg/DFGSpeculativeJIT.h:
(DFG):
(SpeculativeJIT):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded all tests that crashed more than once on the Lion WebKit2 bots between r111885
mitz@apple.com [Fri, 23 Mar 2012 21:19:27 +0000 (21:19 +0000)]
Added all tests that crashed more than once on the Lion WebKit2 bots between r111885
and r111892 to the Mac WebKit2 skip list (that weren’t alread on the list).

Rubber-stamped by Geoff Garen.

* platform/mac-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Unreviewed, move a comment closer to where it matters and
tony@chromium.org [Fri, 23 Mar 2012 21:16:43 +0000 (21:16 +0000)]
[chromium] Unreviewed, move a comment closer to where it matters and
make it more specific.

* WTF.gyp/WTF.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111904 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Unreviewed, rebaseline platform/chromium/compositing/3d-corners.html
enne@google.com [Fri, 23 Mar 2012 21:10:35 +0000 (21:10 +0000)]
[chromium] Unreviewed, rebaseline platform/chromium/compositing/3d-corners.html
https://bugs.webkit.org/show_bug.cgi?id=69124

Somebody checked in an incorrect windows expectation missing a
newline. Fixing.

* platform/chromium-mac/platform/chromium/compositing/3d-corners-expected.png: Renamed from LayoutTests/platform/chromium/platform/chromium/compositing/3d-corners-expected.png.
* platform/chromium-win/platform/chromium/compositing/3d-corners-expected.txt: Removed.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111903 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoLayoutTestHelper could get torn down earlier (mostly to reset color profile)
dpranke@chromium.org [Fri, 23 Mar 2012 21:09:57 +0000 (21:09 +0000)]
LayoutTestHelper could get torn down earlier (mostly to reset color profile)
https://bugs.webkit.org/show_bug.cgi?id=80567

Reviewed by Ryosuke Niwa.

Move clean_up_run to occur right after we finish testing and
before we start processing the results. Also makes set_up_run()
and clean_up_run() private since there's no real need for
callers to be aware of them.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager._set_up_run):
(Manager.run):
(Manager._clean_up_run):
* Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
(ManagerTest.test_http_locking):
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:
(run):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=82083
bdakin@apple.com [Fri, 23 Mar 2012 21:02:51 +0000 (21:02 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=82083

Rubber-stamped by Simon Fraser.

This patch is just a name change. getExistingScrollAnimator() (which was
added earlier today) is now existingScrollAnimator().
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::transitionToCommitted):
* page/EventHandler.cpp:
(WebCore::EventHandler::handleGestureTapDown):
* page/FrameView.cpp:
(WebCore::FrameView::setAnimatorsAreActive):
* platform/ScrollView.cpp:
(WebCore::ScrollView::updateScrollbars):
* platform/ScrollableArea.cpp:
(WebCore::ScrollableArea::willStartLiveResize):
(WebCore::ScrollableArea::willEndLiveResize):
(WebCore::ScrollableArea::contentAreaWillPaint):
(WebCore::ScrollableArea::mouseEnteredContentArea):
(WebCore::ScrollableArea::mouseExitedContentArea):
(WebCore::ScrollableArea::mouseMovedInContentArea):
(WebCore::ScrollableArea::contentAreaDidShow):
(WebCore::ScrollableArea::contentAreaDidHide):
(WebCore::ScrollableArea::contentsResized):
(WebCore::ScrollableArea::serviceScrollAnimations):
* platform/ScrollableArea.h:
(WebCore::ScrollableArea::existingScrollAnimator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRevert r111891. Instead, add editing/inserting/insert-bg-font.html to the list of...
rniwa@webkit.org [Fri, 23 Mar 2012 21:00:25 +0000 (21:00 +0000)]
Revert r111891. Instead, add editing/inserting/insert-bg-font.html to the list of failing tests.
Again, the failure is tracked by the bug https://bugs.webkit.org/show_bug.cgi?id=82079.

* platform/mac-lion/Skipped:
* platform/mac-lion/editing: Removed.
* platform/mac-lion/editing/inserting: Removed.
* platform/mac-lion/editing/inserting/insert-bg-font-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCrash due to accessing removed parent lineboxes when clearing selection.
inferno@chromium.org [Fri, 23 Mar 2012 20:56:10 +0000 (20:56 +0000)]
Crash due to accessing removed parent lineboxes when clearing selection.
https://bugs.webkit.org/show_bug.cgi?id=81359

Reviewed by Eric Seidel.

Source/WebCore:

Similar to r110323, adds the canUpdateSelectionOnRootLineBoxes
check to more places.

Test: editing/selection/clear-selection-crash.html

* rendering/RenderObject.cpp:
(WebCore::RenderObject::canUpdateSelectionOnRootLineBoxes):
* rendering/RenderSelectionInfo.h:
(WebCore::RenderSelectionInfo::RenderSelectionInfo):
(WebCore::RenderBlockSelectionInfo::RenderBlockSelectionInfo):

LayoutTests:

* editing/selection/clear-selection-crash-expected.txt: Added.
* editing/selection/clear-selection-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdate Apple Windows build files for WTF move
sfalken@apple.com [Fri, 23 Mar 2012 20:47:20 +0000 (20:47 +0000)]
Update Apple Windows build files for WTF move
https://bugs.webkit.org/show_bug.cgi?id=82069

Reviewed by Jessie Berlin.

../JavaScriptCore:

* JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: Removed WTF and WTFGenerated.

../WTF:

* WTF.vcproj/WTF.make: Added.
* WTF.vcproj/WTF.sln: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoInitial triage pass of css3/selectors3/xml for the Chromium ports.
ojan@chromium.org [Fri, 23 Mar 2012 20:40:15 +0000 (20:40 +0000)]
Initial triage pass of css3/selectors3/xml for the Chromium ports.
The difference from the Apple Mac results are all text rendering or
cases where the Chromium result is clearly correct and the Apple Mac
result is clearly incorrect.

All changes done to the non-chromium ports were done automatically
by "webkit-patch optimize-baselines".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=82083
bdakin@apple.com [Fri, 23 Mar 2012 20:36:31 +0000 (20:36 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=82083
Too many ScrollAnimators are allocated on pages with frames

Reviewed by Geoffrey Garen.

This patch adds a new getter to ScrollableArea called
getExistingScrollAnimator() unlike scrollAnimator() this new function will
just return null if there is not already a ScrollAnimator. Callers should use
this new function when they have business with ScrollAnimator that does not
need to do anything when there are no scrollbars.
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::transitionToCommitted):
* page/EventHandler.cpp:
(WebCore::EventHandler::handleGestureTapDown):
* page/FrameView.cpp:
(WebCore::FrameView::setAnimatorsAreActive):
* platform/ScrollView.cpp:
(WebCore::ScrollView::updateScrollbars):
* platform/ScrollableArea.cpp:
(WebCore::ScrollableArea::willStartLiveResize):
(WebCore::ScrollableArea::willEndLiveResize):
(WebCore::ScrollableArea::contentAreaWillPaint):
(WebCore::ScrollableArea::mouseEnteredContentArea):
(WebCore::ScrollableArea::mouseExitedContentArea):
(WebCore::ScrollableArea::mouseMovedInContentArea):
(WebCore::ScrollableArea::contentAreaDidShow):
(WebCore::ScrollableArea::contentAreaDidHide):
(WebCore::ScrollableArea::willRemoveHorizontalScrollbar):
(WebCore::ScrollableArea::contentsResized):
(WebCore::ScrollableArea::serviceScrollAnimations):
* platform/ScrollableArea.h:
(ScrollableArea):
(WebCore::ScrollableArea::getExistingScrollAnimator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDeal with DOM modifications when evaluating source elements.
eric.carlson@apple.com [Fri, 23 Mar 2012 20:26:17 +0000 (20:26 +0000)]
Deal with DOM modifications when evaluating source elements.
https://bugs.webkit.org/show_bug.cgi?id=81163

Reviewed by Alexey Proskuryakov.

Source/WebCore:

Test: media/video-beforeload-remove-source.html

* dom/ContainerNode.cpp: Make NodeVector and collectNodes public, renamed as getChildNodes.
(WebCore::ContainerNode::takeAllChildrenFrom): collectNodes -> getChildNodes.
(WebCore::ContainerNode::willRemove): collectNodes -> getChildNodes.
(WebCore::ContainerNode::willRemoveChildren): collectNodes -> getChildNodes.
(WebCore::ContainerNode::insertedIntoDocument): collectNodes -> getChildNodes.
(WebCore::ContainerNode::removedFromDocument): collectNodes -> getChildNodes.
* dom/ContainerNode.h:
(WebCore::getChildNodes):

* editing/ReplaceSelectionCommand.cpp: Remove unused NodeVector declaration.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement): m_nextChildNodeToConsider and m_currentSourceNode
    are now RefPtrs.
(WebCore::HTMLMediaElement::loadTimerFired): Protect HTMLMediaElement from being deleted during
    a DOM modification during an event callback.
(WebCore::HTMLMediaElement::load): Ditto.
(WebCore::HTMLMediaElement::selectMediaResource): Set m_nextChildNodeToConsider to the first
    child node, it will be the first node considered.
(WebCore::HTMLMediaElement::havePotentialSourceChild): m_nextChildNodeToConsider and m_currentSourceNode
    are now RefPtrs.
(WebCore::HTMLMediaElement::selectNextSourceChild): Collect all child nodes in a vector before
    looking for <source> nodes because 'beforeload' event handlers can mutate the DOM. Don't
    use a <source> that is no longer a child node after 'beforeload'. Use 0 to represent the end
    of the child node list because m_nextChildNodeToConsider is now a RefPtr so using the previous
    sentinel, "this", would cause a retain cycle.
(WebCore::HTMLMediaElement::sourceWasAdded):  m_nextChildNodeToConsider and m_currentSourceNode
    are now RefPtrs.
(WebCore::HTMLMediaElement::sourceWillBeRemoved): Ditto.
(WebCore::HTMLMediaElement::getPluginProxyParams): Protect HTMLMediaElement from being deleted during
    a DOM modification during an event callback.
* html/HTMLMediaElement.h:

LayoutTests:

* media/video-beforeload-remove-source-expected.txt: Added.
* media/video-beforeload-remove-source.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdded all tests that crashed more than once on the Lion WebKit2 bots between r111865
mitz@apple.com [Fri, 23 Mar 2012 20:24:29 +0000 (20:24 +0000)]
Added all tests that crashed more than once on the Lion WebKit2 bots between r111865
and r111885 to the Mac WebKit2 skip list (that weren’t alread on the list).

Rubber-stamped by Geoff Garen.

* platform/mac-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDisable CSS_SHADERS in Apple builds https://bugs.webkit.org/show_bug.cgi?id=81996
dino@apple.com [Fri, 23 Mar 2012 20:18:55 +0000 (20:18 +0000)]
Disable CSS_SHADERS in Apple builds https://bugs.webkit.org/show_bug.cgi?id=81996

Source/JavaScriptCore:

Reviewed by Simon Fraser.

Remove ENABLE_CSS_SHADERS from FeatureDefines. It's now in Platform.h.

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Reviewed by Simon Fraser.

Remove ENABLE_CSS_SHADERS from FeatureDefines. It's now in Platform.h.

* Configurations/FeatureDefines.xcconfig:

Source/WebKit/mac:

Reviewed by Simon Fraser.

Remove ENABLE_CSS_SHADERS from FeatureDefines. It's now in Platform.h.

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

Reviewed by Simon Fraser.

Remove ENABLE_CSS_SHADERS from FeatureDefines. It's now in Platform.h.

* Configurations/FeatureDefines.xcconfig:

Source/WTF:

Reviewed by Simon Fraser.

Put ENABLE_CSS_SHADERS into Platform.h, but disable for
Apple builds.

* wtf/Platform.h:

Tools:

Reviewed by Simon Fraser.

* Scripts/build-webkit:

LayoutTests:

Add css3/filters/custom to the skipped list.

Reviewed by Simon Fraser.

* platform/mac/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdated Mac expected results for these tests by reverting r111339, now that bug 81826 was
mitz@apple.com [Fri, 23 Mar 2012 20:07:33 +0000 (20:07 +0000)]
Updated Mac expected results for these tests by reverting r111339, now that bug 81826 was
fixed in r111882.

* platform/mac/tables/mozilla/bugs/bug27038-1-expected.txt:
* platform/mac/tables/mozilla/bugs/bug27038-2-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd a failing expectation for the bug 82079.
rniwa@webkit.org [Fri, 23 Mar 2012 20:03:47 +0000 (20:03 +0000)]
Add a failing expectation for the bug 82079.

* platform/editing: Added.
* platform/editing/inserting: Added.
* platform/mac-lion/editing: Added.
* platform/mac-lion/editing/inserting: Added.
* platform/mac-lion/editing/inserting/insert-bg-font-expected.txt: Copied from LayoutTests/editing/inserting/insert-bg-font-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRelanding(r111754): HTMLPluginElement is not destroyed on reload or navigation if...
commit-queue@webkit.org [Fri, 23 Mar 2012 20:00:10 +0000 (20:00 +0000)]
Relanding(r111754): HTMLPluginElement is not destroyed on reload or navigation if getNPObject is called
https://bugs.webkit.org/show_bug.cgi?id=80428

.:

Patch by Dave Michael <dmichael@chromium.org> on 2012-03-23

Source/WebCore:

Patch by Dave Michael <dmichael@chromium.org> on 2012-03-23
Reviewed by Eric Seidel and Ryosuke Niwa.

Make HTMLPluginElement release its m_NPObject in detach() to break a
reference-counting cycle that happens on reload or navigation. With this
change, HTMLPlugInElement::removedFromDocument is unnecessary, so it
was removed. Note that Releasing m_NPObject does not result in a call to
the plugin; it simply releases a reference count on the wrapper object
for this HTMLPlugInElement. (The plugin's NPP_Deallocate is invoked
when the render tree is destroyed, when PluginView calls
PluginPackage::unload.) Thus, it is safe to release m_NPObject in
detach, because it can not result in layout or style changes.

Also added numberOfLiveNodes() and numberOfLiveDocuments() to
window.internals to enable testing.

Test: plugins/netscape-dom-access-and-reload.html

* WebCore.exp.in:
* html/HTMLPlugInElement.cpp:
(WebCore::HTMLPlugInElement::detach):
* html/HTMLPlugInElement.h:
(HTMLPlugInElement):
* testing/Internals.cpp:
(WebCore):
(WebCore::Internals::numberOfLiveNodes):
(WebCore::Internals::numberOfLiveDocuments):
* testing/Internals.h:
(Internals):
* testing/Internals.idl:

Source/WebKit2:

Patch by Dave Michael <dmichael@chromium.org> on 2012-03-23
Reviewed by Eric Seidel and Ryosuke Niwa.

* win/WebKit2.def: Export a symbol for InspectorCounters::counterValue
* win/WebKit2CFLite.def: Export a symbol for InspectorCounters::counterValue

LayoutTests:

Patch by Dave Michael <dmichael@chromium.org> on 2012-03-23
Reviewed by Eric Seidel and Ryosuke Niwa.

Due to unfortunate copy/paste laziness, the new test was using the same
window.sessionStorage as plugins/reloadplugins-and-pages.html, so that
if the tests were run in the same session, reloadplugins-and-pages.html
would *not* reload as it was supposed to, causing a text mismatch. This
patch uses a more appropriate and unique name so that these two tests
won't affect each other.

* plugins/netscape-dom-access-and-reload-expected.txt: Added.
* plugins/netscape-dom-access-and-reload.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRexExp constructor last match properties should not rely on previous ovector
barraclough@apple.com [Fri, 23 Mar 2012 19:57:28 +0000 (19:57 +0000)]
RexExp constructor last match properties should not rely on previous ovector
https://bugs.webkit.org/show_bug.cgi?id=82077

Reviewed by Oliver Hunt.

This change simplifies matching, and will enable subpattern results to be fully lazily generated in the future.

This patch changes the scheme used to lazily generate the last match properties of the RegExp object.
Instead of relying on the results in the ovector, we can instead lazily generate the subpatters using
a RegExpMatchesArray. To do so we just need to store the input, the regexp matched, and the match
location (the MatchResult). When the match is accessed or the input is set, we reify results. We use
a special value of setting the saved result to MatchResult::failed() to indicated that we're in a
reified state. This means that next time a match is performed, the store of the result will
automatically blow away the reified value.

* JavaScriptCore.xcodeproj/project.pbxproj:
    - Added new files.
* runtime/RegExp.cpp:
(JSC::RegExpFunctionalTestCollector::outputOneTest):
    - changed 'subPattern' -> 'subpattern' (there was a mix in JSC, 'subpattern' was more common).
* runtime/RegExpCachedResult.cpp: Added.
(JSC::RegExpCachedResult::visitChildren):
(JSC::RegExpCachedResult::lastResult):
(JSC::RegExpCachedResult::setInput):
    - New methods, mark GC objects, lazily create the matches array, and record a user provided input (via assignment to RegExp.inupt).
* runtime/RegExpCachedResult.h: Added.
(RegExpCachedResult):
    - Added new class.
(JSC::RegExpCachedResult::RegExpCachedResult):
(JSC::RegExpCachedResult::record):
(JSC::RegExpCachedResult::input):
    - Initialize the object, record the result of a RegExp match, access the stored input property.
* runtime/RegExpConstructor.cpp:
(JSC::RegExpConstructor::RegExpConstructor):
    - Initialize m_result/m_multiline properties.
(JSC::RegExpConstructor::visitChildren):
    - Make sure the cached results (or lazy source for them) are marked.
(JSC::RegExpConstructor::getBackref):
(JSC::RegExpConstructor::getLastParen):
(JSC::RegExpConstructor::getLeftContext):
(JSC::RegExpConstructor::getRightContext):
    - Moved from RegExpConstructor, moved to RegExpCachedResult, and using new caching scheme.
(JSC::regExpConstructorInput):
(JSC::setRegExpConstructorInput):
    - Changed to use RegExpCachedResult.
* runtime/RegExpConstructor.h:
(JSC::RegExpConstructor::create):
(RegExpConstructor):
(JSC::RegExpConstructor::setMultiline):
(JSC::RegExpConstructor::multiline):
    - Move multiline property onto the constructor object; it is not affected by the last match.
(JSC::RegExpConstructor::setInput):
(JSC::RegExpConstructor::input):
    - These defer to RegExpCachedResult.
(JSC::RegExpConstructor::performMatch):
* runtime/RegExpMatchesArray.cpp: Added.
(JSC::RegExpMatchesArray::visitChildren):
    - Eeeep! added missing visitChildren!
(JSC::RegExpMatchesArray::finishCreation):
(JSC::RegExpMatchesArray::reifyAllProperties):
(JSC::RegExpMatchesArray::reifyMatchProperty):
    - Moved from RegExpConstructor.cpp.
(JSC::RegExpMatchesArray::leftContext):
(JSC::RegExpMatchesArray::rightContext):
    - Since the match start/
* runtime/RegExpMatchesArray.h:
(RegExpMatchesArray):
    - Declare new methods & structure flags.
* runtime/RegExpObject.cpp:
(JSC::RegExpObject::match):
    - performMatch now requires the JSString input, to cache.
* runtime/StringPrototype.cpp:
(JSC::removeUsingRegExpSearch):
(JSC::replaceUsingRegExpSearch):
(JSC::stringProtoFuncMatch):
(JSC::stringProtoFuncSearch):
    - performMatch now requires the JSString input, to cache.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCSSParser doesn't set border-*-width/style/color to initial by border shorthand property
rniwa@webkit.org [Fri, 23 Mar 2012 19:37:39 +0000 (19:37 +0000)]
CSSParser doesn't set border-*-width/style/color to initial by border shorthand property
https://bugs.webkit.org/show_bug.cgi?id=82040

Reviewed by Antti Koivisto.

Source/WebCore:

The border shorthand property sets values for border-width, border-style, and border-color shorthand properties.
While CSSParser::parseValue can process these shorthand properties properly and set the longhand properties
such as border-top-width, border-right-width, ... border-left-color, CSSParser::addProperty can't and the
initialization in parseShorthand fails for the border property.

Fixed the bug by explicitly initializing longhand properties.

Changing the behavior here is unlikely to break the Web since our behavior already differs from that of Firefox
and Internet Explorer. Both of those browsers return the actual initial values such as "medium" and "currentColor".
This discrepancy is tracked by https://bugs.webkit.org/show_bug.cgi?id=82078.

Test: fast/css/border-shorthand-initialize-longhands.html

* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue): Uses borderAbridgedLonghand.
(WebCore::CSSParser::parseShorthand): Uses longhand properties for initialization if one is available.
This allows us to initialize multiple properties (e.g. border-*-color) for a single property missing in the set.
* css/CSSPropertyLonghand.cpp:
(WebCore::borderAbridgedLonghand): Added. The longhand here (border-width, border-style, border-color) is
"abridged" in the sense that they're still shorthands.
* css/CSSPropertyLonghand.h:
(WebCore::CSSPropertyLonghand::CSSPropertyLonghand):
(CSSPropertyLonghand): Added the version that takes longhand instances for initialization purposes.
(WebCore::CSSPropertyLonghand::longhandsForInitialization):

LayoutTests:

Added a comprehensive regression test.

* fast/css/border-shorthand-initialize-longhands-expected.txt: Added.
* fast/css/border-shorthand-initialize-longhands.html: Added.
* inspector/styles/styles-new-API-expected.txt: Correctly lists border's longhand properties.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] rename newwtf target back to wtf
tony@chromium.org [Fri, 23 Mar 2012 19:30:21 +0000 (19:30 +0000)]
[chromium] rename newwtf target back to wtf
https://bugs.webkit.org/show_bug.cgi?id=82064

Reviewed by Adam Barth.

Source/JavaScriptCore:

* JavaScriptCore.gyp/JavaScriptCore.gyp:

Source/Platform:

* Platform.gyp/Platform.gyp:

Source/WebCore:

No new tests, only renaming a build target.

* WebCore.gyp/WebCore.gyp:

Source/WTF:

* WTF.gyp/WTF.gyp:

Tools:

* DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Turn off painting of scrollbars on compositor thread for unsupported platforms
enne@google.com [Fri, 23 Mar 2012 19:24:40 +0000 (19:24 +0000)]
[chromium] Turn off painting of scrollbars on compositor thread for unsupported platforms
https://bugs.webkit.org/show_bug.cgi?id=82066

Reviewed by James Robinson.

With threaded compositing, OSX will crash and Windows will have black
scrollbars. Fix that by reverting to the previosu behavior of updating
scrollbars on the main thread instead.

* page/scrolling/chromium/ScrollingCoordinatorChromium.cpp:
(WebCore::scrollbarLayerDidChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoInitial triage pass of css3/selectors3/xhtml for the Chromium ports.
ojan@chromium.org [Fri, 23 Mar 2012 18:40:16 +0000 (18:40 +0000)]
Initial triage pass of css3/selectors3/xhtml for the Chromium ports.
The difference from the Apple Mac results are all text rendering or
cases where the Chromium results is clearly correct and the Apple Mac
result is clearly incorrect.

All changes done to the non-chromium ports were done automatically
by "webkit-patch optimize-baselines".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago<rdar://problem/7883987> ASSERTION FAILED: ASSERT(input == AnimationStateInputEndTime...
mitz@apple.com [Fri, 23 Mar 2012 18:34:31 +0000 (18:34 +0000)]
<rdar://problem/7883987> ASSERTION FAILED: ASSERT(input == AnimationStateInputEndTimerFired || input == AnimationStateInputPlayStatePaused) in AnimationBase::updateStateMachine causing multiple "crashes" on Lion Intel Debug WebKit testers
https://bugs.webkit.org/show_bug.cgi?id=81351

Rubber-stamped by Jessie Berlin.

* page/animation/AnimationBase.cpp:
(WebCore::AnimationBase::updateStateMachine): Replaced the assertion with a LOG_ERROR so that
the issue is still noticeable, but it does not automatically make the debug bots red.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCSS3 calc: image sizing works with mixed percentage/absolute
mikelawther@chromium.org [Fri, 23 Mar 2012 18:30:01 +0000 (18:30 +0000)]
CSS3 calc: image sizing works with mixed percentage/absolute
https://bugs.webkit.org/show_bug.cgi?id=82017

Reviewed by Ojan Vafai.

Source/WebCore:

Test: css3/calc/img-size-expected.html

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computeReplacedLogicalWidthUsing):
(WebCore::RenderBox::computeReplacedLogicalHeightUsing):

LayoutTests:

Test is now expected to match.

* css3/calc/img-size-expected.html: Renamed from LayoutTests/css3/calc/img-size-expected-mismatch.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION(107971): Google Voice contact list is broken in WebKit due to badly alloca...
jchaffraix@webkit.org [Fri, 23 Mar 2012 18:27:30 +0000 (18:27 +0000)]
REGRESSION(107971): Google Voice contact list is broken in WebKit due to badly allocating the extra height
https://bugs.webkit.org/show_bug.cgi?id=81826

Reviewed by Tony Chang.

Source/WebCore:

Covered by tables/mozilla/bugs/bug27038-{1|2}.html.

This partly reverts r107971: the extra logical height distribution change was not needed
to fix the bug (it is needed by the test though). We revert to giving all the extra height
to the first tbody and not the first section.

This is broken but unfortunately some websites are relying on that. Getting a real
distribution algorithm is covered by bug 81824. However this is super tricky to get
right and I did not want to add more compatibility risks until I have something solid.

* rendering/RenderTable.cpp:
(WebCore::RenderTable::distributeExtraLogicalHeight):

LayoutTests:

* fast/table/double-height-table-no-tbody.html-disabled: Renamed from LayoutTests/fast/table/double-height-table-no-tbody.html.
Disabled this test as it relies on our algorithm not to discriminate between first sections.

* platform/chromium/test_expectations.txt:
* platform/qt/Skipped:
Mark those 2 tests as needing a new baseline again.

* platform/efl/test_expectations.txt:
* platform/gtk/Skipped:
* platform/mac/test_expectations.txt:
* platform/win/Skipped:
Those platforms did not rebaseline those 2 tests so they should automatically pass them.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@111882 268f45cc-cd09-0410-ab3c-d52691b4dbfc