WebKit-https.git
7 years ago[CMake] Unify coding style for CMake files
commit-queue@webkit.org [Thu, 6 Dec 2012 03:10:13 +0000 (03:10 +0000)]
[CMake] Unify coding style for CMake files
https://bugs.webkit.org/show_bug.cgi?id=103605

Patch by Halton Huo <halton.huo@intel.com> on 2012-12-05
Reviewed by Laszlo Gombos.

Update cmake files(.cmake, CMakeLists.txt) with following style rules:
1. Indentation
1.1 Use spaces, not tabs.
1.2 Four spaces as indent.
2. Spacing
2.1 Place one space between control statements and their parentheses.
    For eg, if (), else (), elseif (), endif (), foreach (),
    endforeach (), while (), endwhile (), break ().
2.2 Do not place spaces between function and macro statements and
    their parentheses. For eg, macro(), endmacro(), function(),
    endfunction().
2.3 Do not place spaces between a command or function or macro and its
    parentheses, or between a parenthesis and its content. For eg,
    message("testing") not message( "testing") or message ("testing" )
2.4 No space at line ending.
3. Lowercase when call commands macros and functions. For eg,
   add_executable() not ADD_EXECUTABLE(), set() not SET().

.:

* CMakeLists.txt:
* Source/CMakeLists.txt:
* Source/PlatformEfl.cmake:
* Source/cmake/EFLHelpers.cmake:
* Source/cmake/FindATK.cmake:
* Source/cmake/FindCFLite.cmake:
* Source/cmake/FindCairo.cmake:
* Source/cmake/FindDBus.cmake:
* Source/cmake/FindDirectX.cmake:
* Source/cmake/FindE_DBus.cmake:
* Source/cmake/FindEcore.cmake:
* Source/cmake/FindEdje.cmake:
* Source/cmake/FindEet.cmake:
* Source/cmake/FindEeze.cmake:
* Source/cmake/FindEfreet.cmake:
* Source/cmake/FindEina.cmake:
* Source/cmake/FindElementary.cmake:
* Source/cmake/FindEnchant.cmake:
* Source/cmake/FindEvas.cmake:
* Source/cmake/FindFontconfig.cmake:
* Source/cmake/FindGLIB.cmake:
* Source/cmake/FindGStreamer.cmake:
* Source/cmake/FindGperf.cmake:
* Source/cmake/FindHarfBuzz.cmake:
* Source/cmake/FindICU.cmake:
* Source/cmake/FindLibSoup.cmake:
* Source/cmake/FindQuickTimeSDK.cmake:
* Source/cmake/FindSqlite.cmake:
* Source/cmake/OptionsBlackBerry.cmake:
* Source/cmake/OptionsCommon.cmake:
* Source/cmake/OptionsEfl.cmake:
* Source/cmake/OptionsWinCE.cmake:
* Source/cmake/OptionsWindows.cmake:
* Source/cmake/WebKitFS.cmake:
* Source/cmake/WebKitFeatures.cmake:
* Source/cmake/WebKitHelpers.cmake:
* Source/cmake/WebKitMacros.cmake:
* Source/cmake/WebKitPackaging.cmake:
* Source/cmake/gtest/CMakeLists.txt:

Source/JavaScriptCore:

* CMakeLists.txt:
* PlatformBlackBerry.cmake:
* PlatformEfl.cmake:
* PlatformWinCE.cmake:
* shell/CMakeLists.txt:
* shell/PlatformBlackBerry.cmake:
* shell/PlatformEfl.cmake:
* shell/PlatformWinCE.cmake:

Source/WebCore:

* CMakeLists.txt:
* PlatformBlackBerry.cmake:
* PlatformEfl.cmake:
* PlatformWinCE.cmake:
* UseJSC.cmake:
* UseV8.cmake:

Source/WebKit:

* CMakeLists.txt:
* PlatformBlackBerry.cmake:
* PlatformEfl.cmake:
* PlatformWinCE.cmake:

Source/WebKit/efl:

* DefaultTheme/CMakeLists.txt:

Source/WebKit2:

* CMakeLists.txt:
* PlatformEfl.cmake:
* win/WebKit2ExportGenerator.vcproj:
* win/WebKit2ExportGeneratorCommon.vsprops:

Source/WTF:

* CMakeLists.txt:
* wtf/CMakeLists.txt:
* wtf/PlatformBlackBerry.cmake:
* wtf/PlatformEfl.cmake:
* wtf/PlatformWinCE.cmake:

Tools:

* CMakeLists.txt:
* DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt:
* DumpRenderTree/efl/CMakeLists.txt:
* EWebLauncher/CMakeLists.txt:
* EWebLauncher/ControlTheme/CMakeLists.txt:
* MiniBrowser/efl/CMakeLists.txt:
* TestWebKitAPI/CMakeLists.txt:
* TestWebKitAPI/PlatformEfl.cmake:
* WebKitTestRunner/CMakeLists.txt:
* WebKitTestRunner/PlatformEfl.cmake:
* WinCELauncher/CMakeLists.txt:
* clang/ReportMemoryUsagePlugin/CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136790 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[V8] Remove unused property names from V8HiddenPropertyName
haraken@chromium.org [Thu, 6 Dec 2012 02:58:22 +0000 (02:58 +0000)]
[V8] Remove unused property names from V8HiddenPropertyName
https://bugs.webkit.org/show_bug.cgi?id=104194

Reviewed by Adam Barth.

We can remove unused property names from V8HiddenPropertyName.

No tests. No change in behavior.

* bindings/v8/V8HiddenPropertyName.h:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCoordinated Graphics: Reorder messages to LayerTreeCoordinatorProxy
commit-queue@webkit.org [Thu, 6 Dec 2012 02:50:02 +0000 (02:50 +0000)]
Coordinated Graphics: Reorder messages to LayerTreeCoordinatorProxy
https://bugs.webkit.org/show_bug.cgi?id=103843

Patch by Huang Dongsung <luxtella@company100.net> on 2012-12-05
Reviewed by Noam Rosenthal.

Send SetRootCompositingLayer message to the UI process in the
constructor instead of sending it on the first flush.

This is in preparation for refactoring TextureMapper to work in an actor
model (http://webkit.org/b/103854).

* UIProcess/CoordinatedGraphics/LayerTreeRenderer.cpp:
(WebKit::LayerTreeRenderer::setLayerState):
(WebKit::LayerTreeRenderer::setRootLayerID):
* WebProcess/WebPage/CoordinatedGraphics/LayerTreeCoordinator.cpp:
(WebKit::LayerTreeCoordinator::LayerTreeCoordinator):
(WebKit::LayerTreeCoordinator::initializeRootCompositingLayer):
(WebKit):
(WebKit::LayerTreeCoordinator::flushPendingLayerChanges):
* WebProcess/WebPage/CoordinatedGraphics/LayerTreeCoordinator.h:
(LayerTreeCoordinator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136788 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Don't use the C API internally in ewk_cookie_manager
commit-queue@webkit.org [Thu, 6 Dec 2012 02:38:22 +0000 (02:38 +0000)]
[EFL][WK2] Don't use the C API internally in ewk_cookie_manager
https://bugs.webkit.org/show_bug.cgi?id=103243

Patch by Jinwoo Song <jinwoo7.song@samsung.com> on 2012-12-05
Reviewed by Gyuyoung Kim.

Used the C++ classes directly instead of the C API wrappers
to avoid a lot of toImpl/toAPI casts.

* UIProcess/API/efl/ewk_context.cpp:
(EwkContext::cookieManager):
* UIProcess/API/efl/ewk_cookie_manager.cpp:
(EwkCookieManager::EwkCookieManager):
(EwkCookieManager::~EwkCookieManager):
(EwkCookieManager::setPersistentStorage):
(EwkCookieManager::setHTTPAcceptPolicy):
(EwkCookieManager::clearHostnameCookies):
(EwkCookieManager::clearAllCookies):
(EwkCookieManager::watchChanges):
(EwkCookieManager::getHostNamesWithCookies):
(EwkCookieManager::getHTTPAcceptPolicy):
* UIProcess/API/efl/ewk_cookie_manager_private.h:
(EwkCookieManager::create):
(EwkCookieManager):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2][EFL] Gardening. Rebase fast/block/basic/020.html
commit-queue@webkit.org [Thu, 6 Dec 2012 02:36:22 +0000 (02:36 +0000)]
[WK2][EFL] Gardening. Rebase fast/block/basic/020.html
https://bugs.webkit.org/show_bug.cgi?id=104195

Unreviewed Gardening.

The layer for "visibility:hidden;position:absolute;" object should be in RenderTree when USE(ACCELERATED_COMPOSITING),
because the HiddenLayers are included in zOrderList when CompositingMode is on.

Patch by KyungTae Kim <ktf.kim@samsung.com> on 2012-12-05

* platform/efl-wk2/fast/block/basic/020-expected.txt: Added.
* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd runtime enable for web intents.
commit-queue@webkit.org [Thu, 6 Dec 2012 02:06:55 +0000 (02:06 +0000)]
Add runtime enable for web intents.
https://bugs.webkit.org/show_bug.cgi?id=103669

Patch by Greg Billock <gbillock@google.com> on 2012-12-05
Reviewed by Adam Barth.

Source/WebCore:

Make web intents Javascript API enabled by a runtime setting.

* Modules/intents/DOMWindowIntents.idl:
* Modules/intents/NavigatorIntents.idl:
* bindings/generic/RuntimeEnabledFeatures.cpp:
(WebCore):
* bindings/generic/RuntimeEnabledFeatures.h:
(RuntimeEnabledFeatures):
(WebCore::RuntimeEnabledFeatures::webkitStartActivityEnabled):
(WebCore::RuntimeEnabledFeatures::webkitIntentEnabled):
(WebCore::RuntimeEnabledFeatures::webKitIntentEnabled):
(WebCore::RuntimeEnabledFeatures::setWebIntentsEnabled):

Source/WebKit/chromium:

Propagate runtime setting to enable/disable web intents Javascript API.

* public/WebRuntimeFeatures.h:
(WebRuntimeFeatures):
* src/WebRuntimeFeatures.cpp:
(WebKit::WebRuntimeFeatures::enableWebIntents):
(WebKit):
(WebKit::WebRuntimeFeatures::isWebIntentsEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Remove linux theme related files and switch to default
commit-queue@webkit.org [Thu, 6 Dec 2012 01:38:29 +0000 (01:38 +0000)]
[chromium] Remove linux theme related files and switch to default
https://bugs.webkit.org/show_bug.cgi?id=103897

Patch by Scott Violet <sky@chromium.org> on 2012-12-05
Reviewed by Tony Chang.

Source/Platform:

* Platform.gypi: Remove WebThemeEngine.h
* chromium/public/linux/WebThemeEngine.h: Removed.

Source/WebCore:

Linux related theme files are now named Default.

No new tests, code cleanup.

* WebCore.gyp/WebCore.gyp: Update files
* WebCore.gypi: Update files
* platform/chromium/PlatformThemeChromiumLinux.cpp: Removed.
* platform/chromium/PlatformThemeChromiumLinux.h: Removed.
* platform/chromium/ScrollbarThemeChromiumLinux.cpp: Removed.
* platform/chromium/ScrollbarThemeChromiumLinux.h: Removed.
* rendering/RenderThemeChromiumAndroid.cpp:
(WebCore::RenderThemeChromiumAndroid::extraDefaultStyleSheet): Linux->Default
* rendering/RenderThemeChromiumAndroid.h: Change superclass
* rendering/RenderThemeChromiumLinux.cpp: Removed.
* rendering/RenderThemeChromiumLinux.h: Removed.

Source/WebKit/chromium:

Linux theme related files are now named Default.

* WebKit.gyp:
* public/linux/WebRenderTheme.h: Removed.
* public/platform/linux/WebThemeEngine.h: Removed.
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::setScrollbarColors): Cleanup ifdefs.
(WebKit::WebViewImpl::setSelectionColors): Cleanup ifdefs.
* src/linux/WebRenderTheme.cpp: Removed.

Tools:

* Scripts/webkitpy/common/config/build_unittest.py:
(ShouldBuildTest): linux->default

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMake the strip line numbers link into a checkbox
ojan@chromium.org [Thu, 6 Dec 2012 01:35:36 +0000 (01:35 +0000)]
Make the strip line numbers link into a checkbox
https://bugs.webkit.org/show_bug.cgi?id=104187

Reviewed by Tony Chang.

The link text was confusing.

* PrettyPatch/PrettyPatch.rb:
* code-review.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIndexedDB: Allow multiple transactions to interleave request execution
jsbell@chromium.org [Thu, 6 Dec 2012 01:30:35 +0000 (01:30 +0000)]
IndexedDB: Allow multiple transactions to interleave request execution
https://bugs.webkit.org/show_bug.cgi?id=97570

Reviewed by Tony Chang.

Source/WebCore:

Implement spec logic for allowing read-only transactions, and read-write transactions
with non-overlapping scopes, to run concurrently. Transactions all still run in the
same thread with tasks triggered via timers, so tasks and the underlying database
operations are interleaved rather than truly parallelized.

Within IDBTransactionCoordinator, rename started->queued, running->started to match
spec terminology and clear up confusion.

Test: storage/indexeddb/transaction-coordination-across-databases.html
      storage/indexeddb/transaction-coordination-within-database.html
      storage/indexeddb/transaction-readwrite-exclusive.html
      storage/indexeddb/transaction-scope-sequencing.html
      storage/indexeddb/transaction-starvation.html

* Modules/indexeddb/IDBDatabaseBackendImpl.cpp: Use IDBTransaction::Mode enum
(WebCore::IDBDatabaseBackendImpl::createTransaction):
* Modules/indexeddb/IDBDatabaseBackendImpl.h: Ditto.
(IDBDatabaseBackendImpl):
* Modules/indexeddb/IDBDatabaseBackendInterface.h: Ditto.
(IDBDatabaseBackendInterface):
* Modules/indexeddb/IDBTransactionBackendImpl.cpp: Convert scope as a HashSet for
fast intersecting.
(WebCore::IDBTransactionBackendImpl::create):
(WebCore::IDBTransactionBackendImpl::IDBTransactionBackendImpl):
* Modules/indexeddb/IDBTransactionBackendImpl.h:
(IDBTransactionBackendImpl):
(WebCore::IDBTransactionBackendImpl::mode):
(WebCore::IDBTransactionBackendImpl::scope):
* Modules/indexeddb/IDBTransactionCoordinator.cpp: Spec logic goes here.
(WebCore::IDBTransactionCoordinator::processStartedTransactions): Extend this
method to test all plausibly runnable transactions.
(WebCore):
(WebCore::IDBTransactionCoordinator::canRunTransaction): Test to see if one
particular transaction can be run.
(WebCore::IDBTransactionCoordinator::doScopesOverlap): Do a quick intersection
test between transaction scopes.
* Modules/indexeddb/IDBTransactionCoordinator.h:
(IDBTransactionCoordinator):

Source/WebKit/chromium:

Map to IDBTransaction::Mode enum as appropriate.

* src/IDBDatabaseBackendProxy.cpp:
(WebKit::IDBDatabaseBackendProxy::createTransaction):
* src/IDBDatabaseBackendProxy.h:
(IDBDatabaseBackendProxy):
* src/WebIDBDatabaseImpl.cpp:
(WebKit::WebIDBDatabaseImpl::createTransaction):
 * tests/IDBDatabaseBackendTest.cpp:

LayoutTests:

New tests for parallel transactions, update existing tests that relied on sequential execution.

* storage/indexeddb/database-close-expected.txt:
* storage/indexeddb/mozilla/clear-expected.txt:
* storage/indexeddb/mozilla/resources/clear.js: Split up transaction steps.
* storage/indexeddb/resources/database-close.js: Don't log on individual transaction completion.
* storage/indexeddb/resources/transaction-coordination-within-database.js: Added.
* storage/indexeddb/resources/transaction-scope-sequencing.js: Added.
* storage/indexeddb/resources/transaction-starvation.js: Added.
* storage/indexeddb/transaction-coordination-within-database-expected.txt: Added.
* storage/indexeddb/transaction-coordination-within-database.html: Added.
* storage/indexeddb/transaction-scope-sequencing-expected.txt: Added.
* storage/indexeddb/transaction-scope-sequencing.html: Added.
* storage/indexeddb/transaction-starvation-expected.txt: Added.
* storage/indexeddb/transaction-starvation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCoordinated Graphics: Enable support for setContentsToBackgroundColor
commit-queue@webkit.org [Thu, 6 Dec 2012 01:23:48 +0000 (01:23 +0000)]
Coordinated Graphics: Enable support for setContentsToBackgroundColor
https://bugs.webkit.org/show_bug.cgi?id=104128

Patch by No'am Rosenthal <noam@webkit.org> on 2012-12-05
Reviewed by Kenneth Rohde Christiansen.

Source/WebCore:

Enable background color in GraphicsLayerTextureMapper, and paint it using TextureMapper::drawSolidColor.
Once https://bugs.webkit.org/show_bug.cgi?id=103786 is enabled, this would enable Qt/GTK/EFL to directly
composite colors that only have a background color, without having to create a backing store for them.
This would eliminate the memory usage for those layers, reduce the time needed for them to draw into the
backing store and upload the texture, and in the future allow direct compositing of things other than
background colors.

Compositing tests cover this, but are still skipped for ports using coordinated graphics.
See https://bugs.webkit.org/show_bug.cgi?id=104129.

* platform/graphics/texmap/GraphicsLayerTextureMapper.cpp:
(WebCore::GraphicsLayerTextureMapper::setContentsToBackgroundColor):
(WebCore::GraphicsLayerTextureMapper::backgroundColor):
    Maintain a backgroundColor member variable in GraphicsLayerTextureMapper

* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::TextureMapperGL::drawSolidColor):
    TextureMapperGL::drawSolidColor should allow blending when the solid color has alpha.

* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::TextureMapperLayer::paintSelf):
(WebCore::TextureMapperLayer::flushCompositingStateSelf):
* platform/graphics/texmap/TextureMapperLayer.h:
(State):

Source/WebKit2:

Enable setContentsToBackgroundColor in CoordinatedGraphicsLayer, and pass it through to the UI process.

* Shared/WebLayerTreeInfo.h:
(WebLayerInfo):
* UIProcess/CoordinatedGraphics/LayerTreeRenderer.cpp:
(WebKit::LayerTreeRenderer::setLayerState):
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedGraphicsLayer.cpp:
(WebCore::CoordinatedGraphicsLayer::setContentsToBackgroundColor):
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedGraphicsLayer.h:
(CoordinatedGraphicsLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136781 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: run_webkit_tests_integrationtest.py depends on being able to share a port w...
dpranke@chromium.org [Thu, 6 Dec 2012 00:53:46 +0000 (00:53 +0000)]
nrwt: run_webkit_tests_integrationtest.py depends on being able to share a port w/ between master and worker
https://bugs.webkit.org/show_bug.cgi?id=78168

Reviewed by Ojan Vafai.

This patch adds a single ordered list to ResultSummary that contains
the results of every test run in order of result received, and
modifies the TestResults class to contain the list of references used
for each test as well as the pid of the driver; this allows us to stop
having to override the test drivers to record the tests, and makes the
integration tests that test rdering, retries, restarts, etc. black-box
tests.

* Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
(SingleTestRunner._run_rebaseline):
(SingleTestRunner._compare_output):
(SingleTestRunner._run_reftest):
(SingleTestRunner._compare_output_with_reference):
* Scripts/webkitpy/layout_tests/models/result_summary.py:
(ResultSummary.__init__):
(ResultSummary.add):
* Scripts/webkitpy/layout_tests/models/test_results.py:
(TestResult.__init__):
* Scripts/webkitpy/layout_tests/port/driver.py:
(DriverOutput.__init__):
(Driver.run_test):
* Scripts/webkitpy/layout_tests/port/test.py:
(TestDriver):
(TestDriver.__init__):
(TestDriver.run_test):
(TestDriver.stop):
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
(get_tests_run):
(get_test_batches):
(get_test_results):
(MainTest.test_batch_size):
(MainTest.test_dryrun):
(MainTest.test_natural_order):
(MainTest.test_natural_order_test_specified_multiple_times):
(MainTest.test_random_order):
(MainTest.test_random_order_test_specified_multiple_times):
(MainTest.test_no_order):
(MainTest.test_no_order_test_specified_multiple_times):
(MainTest.test_no_order_with_directory_entries_in_natural_order):
(MainTest.test_repeat_each):
(MainTest.test_ignore_flag):
(MainTest.test_skipped_flag):
(MainTest.test_iterations):
(MainTest.test_run_chunk):
(MainTest.test_run_part):
(MainTest.test_run_singly):
(MainTest.test_skip_failing_tests):
(MainTest.test_single_file):
(MainTest.test_single_file_with_prefix):
(MainTest.test_single_skipped_file):
(MainTest.test_test_list):
(MainTest.test_test_list_with_prefix):
(MainTest.test_exit_after_n_failures):
(MainTest.test_exit_after_n_crashes):
(MainTest.test_run_order__inline):
(MainTest.test_reftest_run):
(MainTest.test_reftest_run_reftests_if_pixel_tests_are_disabled):
(MainTest.test_reftest_skip_reftests_if_no_ref_tests):
(MainTest.test_reftest_expected_html_should_be_ignored):
(MainTest.test_reftest_driver_should_run_expected_html):
(MainTest.test_reftest_driver_should_run_expected_mismatch_html):
(MainTest.test_no_http_tests):
(MainTest.test_platform_tests_are_found):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Make WebCompositorOutputSurface an empty class when USE_CC_OUTPUT_SURFACE...
danakj@chromium.org [Thu, 6 Dec 2012 00:40:25 +0000 (00:40 +0000)]
[chromium] Make WebCompositorOutputSurface an empty class when USE_CC_OUTPUT_SURFACE is defined.
https://bugs.webkit.org/show_bug.cgi?id=103967

Reviewed by James Robinson.

Source/Platform:

When the USE_CC_OUTPUT_SURFACE flag is set:
- Make the WebCompositorOutputSurface class empty.
- Don't define the OutputSurfaceClient.
- Don't define the SoftwareOutputDevice.

WebCompositorSupport has methods to create a WebCompositorOutputSurface. They
will return NULL until the chromium side lands, and then they will be used
in place of the existing WebViewHost methods.

* chromium/public/WebCompositorOutputSurface.h:
(WebKit):
* chromium/public/WebCompositorOutputSurfaceClient.h:
(WebKit):
* chromium/public/WebCompositorSoftwareOutputDevice.h:
(WebKit):
* chromium/public/WebCompositorSupport.h:
(WebKit):
(WebCompositorSupport):
(WebKit::WebCompositorSupport::createOutputSurfaceFor3D):
(WebKit::WebCompositorSupport::createOutputSurfaceForSoftware):

Tools:

Use the WebCompositorSupport methods to create an OutputSurface. If they
fail, then fall back to the WebViewHost methods.

Once the chromium side lands, the WebCompositorSupport methods will stop
failing, and will replace the WebViewHost ones.

* DumpRenderTree/chromium/WebViewHost.cpp:
(WebViewHost::createOutputSurface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r136481.
haraken@chromium.org [Thu, 6 Dec 2012 00:31:31 +0000 (00:31 +0000)]
Unreviewed, rolling out r136481.
http://trac.webkit.org/changeset/136481
https://bugs.webkit.org/show_bug.cgi?id=103868

it might have regressed dom_perf/CloneNodes (See bug 104177)

* dom/ContainerNodeAlgorithms.h:
(ChildNodeInsertionNotifier):
(WebCore::ChildNodeInsertionNotifier::notifyNodeInsertedIntoDocument):
(WebCore::ChildNodeInsertionNotifier::notifyNodeInsertedIntoTree):
(WebCore::ChildNodeInsertionNotifier::notify):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago <rdar://problem/12809890> Should allow sandbox lookup of com.apple.tccd
ap@apple.com [Thu, 6 Dec 2012 00:22:45 +0000 (00:22 +0000)]
    <rdar://problem/12809890> Should allow sandbox lookup of com.apple.tccd

        Reviewed by Beth Dakin and Dan Bernstein.

        * WebProcess/com.apple.WebProcess.sb.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIndexedDB: Abort transactions because of leveldb errors part 3
dgrogan@chromium.org [Thu, 6 Dec 2012 00:08:55 +0000 (00:08 +0000)]
IndexedDB: Abort transactions because of leveldb errors part 3
https://bugs.webkit.org/show_bug.cgi?id=103960

Reviewed by Tony Chang.

Transactions are aborted when there are leveldb problems creating
indexes on an object store with existing data or when adding data to an
objectstore that has indexes.

* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore):
(WebCore::makeIndexWriters):
(WebCore::IDBObjectStoreBackendImpl::setIndexKeys):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreStorageOperation::perform):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix some repaint/paintCounter confusion, and reset it when getting layers out of...
simon.fraser@apple.com [Thu, 6 Dec 2012 00:07:02 +0000 (00:07 +0000)]
Fix some repaint/paintCounter confusion, and reset it when getting layers out of the layer pool
https://bugs.webkit.org/show_bug.cgi?id=104180

Reviewed by Tim Horton.

Layers retrieved from the LayerPool by TileCache need to have their
repaint counters reset, otherwise the scroll performance logging gets
confused.

Also, the counter counts paints, not repaints (invalidations), so
rename it accordingly.

* page/scrolling/mac/ScrollingTreeScrollingNodeMac.mm:
(WebCore::ScrollingTreeScrollingNodeMac::logExposedUnfilledArea): Whitespace fix.
* platform/graphics/GraphicsLayer.h:
* platform/graphics/ca/mac/TileCache.mm:
(WebCore::TileCache::blankPixelCountForTiles):
(WebCore::TileCache::createTileLayer):
(WebCore::TileCache::drawRepaintCounter):
* platform/graphics/ca/mac/WebTileLayer.h:
* platform/graphics/ca/mac/WebTileLayer.mm:
(-[WebTileLayer resetPaintCount]):
(-[WebTileLayer incrementPaintCount]):
(-[WebTileLayer paintCount]):
(-[WebTileLayer logFilledFreshTile]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEmpty parse cache when receiving a low memory warning
oliver@apple.com [Wed, 5 Dec 2012 23:43:06 +0000 (23:43 +0000)]
Empty parse cache when receiving a low memory warning
https://bugs.webkit.org/show_bug.cgi?id=104161

Reviewed by Filip Pizlo.

Source/JavaScriptCore:

This adds a function to the globaldata to empty all code related data
structures (code in the heap and the code cache).
It also adds a function to allow the CodeCache to actually be cleared
at all.

* runtime/CodeCache.h:
(CacheMap):
(JSC::CacheMap::clear):
(JSC::CodeCache::clear):
(CodeCache):
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::discardAllCode):
(JSC):
* runtime/JSGlobalData.h:
(JSGlobalData):

Source/WebCore:

Use new discardAllCode() function on the global data, rather than
directly interacting with the heap.

* bindings/js/GCController.cpp:
(WebCore::GCController::discardAllCompiledCode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSanitize content on copy in the code review tool
ojan@chromium.org [Wed, 5 Dec 2012 23:39:09 +0000 (23:39 +0000)]
Sanitize content on copy in the code review tool
https://bugs.webkit.org/show_bug.cgi?id=104155

Reviewed by Tony Chang.

Always remove expand/header/annotate links. Provide an option
to remove line numbers as well. Store the option in localStorage so
people can always get whichever behavior they want.

A better solution would be to restructure the DOM, but that would require gutting
the whole code review tool and would make it difficult to include line numbers if
you wanted them.

* PrettyPatch/PrettyPatch.rb:
* code-review-test.html:
* code-review.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: split all of the buildbot-specific output into its own file
dpranke@chromium.org [Wed, 5 Dec 2012 23:28:42 +0000 (23:28 +0000)]
nrwt: split all of the buildbot-specific output into its own file
https://bugs.webkit.org/show_bug.cgi?id=104165

Reviewed by Ojan Vafai.

Now that we have the RunDetails object we can move all of the
buildbot-specific knowledge out of the core test-running code
and into the outermost run_webkit_tests layer.

* Scripts/webkitpy/layout_tests/run_webkit_tests.py:
(run):
(main):
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
(passing_run):
(logging_run):
(run_and_capture):
(MainTest.test_all):
(MainTest.test_max_locked_shards):
(MainTest.test_child_processes_2):
(MainTest.test_child_processes_min):
(MainTest.test_full_results_html):
(MainTest.test_hung_thread):
(MainTest.test_keyboard_interrupt):
(MainTest.test_no_tests_found):
(MainTest.test_no_tests_found_2):
(MainTest.test_repeat_each_iterations_num_tests):
(MainTest.test_run_singly_actually_runs_tests):
(MainTest.test_test_list):
(MainTest.test_unexpected_failures):
(MainTest.test_missing_and_unexpected_results):
(MainTest.test_pixel_test_directories):
(MainTest.test_missing_and_unexpected_results_with_custom_exit_code):
(MainTest.test_crash_with_stderr):
(MainTest.test_no_image_failure_with_image_diff):
(MainTest.test_crash_log):
(MainTest.test_web_process_crash_log):
(MainTest.test_exit_after_n_failures_upload):
(MainTest.test_results_directory_absolute):
(MainTest.test_results_directory_default):
(MainTest.test_results_directory_relative):
(MainTest.test_retrying_and_flaky_tests):
(MainTest.test_reftest_should_not_use_naming_convention_if_not_listed_in_reftestlist):
(MainTest.test_output_diffs):
(MainTest.test_unsupported_platform):
(MainTest.test_verbose_in_child_processes):
(EndToEndTest.test_end_to_end):
(EndToEndTest.test_reftest_with_two_notrefs):
(RebaselineTest.test_reset_results):
(RebaselineTest.test_missing_results):
(RebaselineTest.test_new_baseline):
* Scripts/webkitpy/layout_tests/views/buildbot_results.py: Added.
(BuildBotPrinter):
(BuildBotPrinter.__init__):
(BuildBotPrinter.print_results):
(BuildBotPrinter._print):
(BuildBotPrinter.print_result_summary):
(BuildBotPrinter._print_result_summary_entry):
(BuildBotPrinter.print_unexpected_results):
(BuildBotPrinter.print_unexpected_results.add_to_dict_of_lists):
(BuildBotPrinter.print_unexpected_results.add_result):
* Scripts/webkitpy/layout_tests/views/buildbot_results_unittest.py: Added.
(BuildBotPrinterTests):
(BuildBotPrinterTests.assertEmpty):
(BuildBotPrinterTests.assertNotEmpty):
(BuildBotPrinterTests.get_printer):
(BuildBotPrinterTests.get_result):
(BuildBotPrinterTests.get_result_summary):
(BuildBotPrinterTests.test_print_unexpected_results):
(BuildBotPrinterTests.test_print_unexpected_results.get_unexpected_results):
* Scripts/webkitpy/layout_tests/views/printing.py:
(Printer):
(Printer.__init__):
(Printer.print_results):
(Printer._print_statistics_for_test_timings):
(Printer._print_baseline):
(Printer._print_debug):
* Scripts/webkitpy/layout_tests/views/printing_unittest.py:
(Testprinter.get_printer):
(Testprinter.get_result):
(Testprinter.test_configure_and_cleanup):
(Testprinter.test_print_config):
(Testprinter.test_print_one_line_summary):
(Testprinter.test_test_status_line):
(Testprinter.test_details):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago [WK2] Would like to queue messages while NetworkProcess is launching
ap@apple.com [Wed, 5 Dec 2012 23:28:31 +0000 (23:28 +0000)]
    [WK2] Would like to queue messages while NetworkProcess is launching
        https://bugs.webkit.org/show_bug.cgi?id=104143

        Reviewed by Anders Carlsson.

        Factored out process launching and message sending code to a ChildProcessProxy class.
        Process closing and crashing code should be moved once we better understand common
        requirements for the processes.

        * CMakeLists.txt:
        * GNUmakefile.list.am:
        * Target.pri:
        * WebKit2.xcodeproj/project.pbxproj:
        * win/WebKit2.vcproj:
        Added ChildProcessProxy files.

        * Shared/ChildProcessProxy.cpp: Added.
        (WebKit::ChildProcessProxy::ChildProcessProxy):
        (WebKit::ChildProcessProxy::~ChildProcessProxy):
        (WebKit::ChildProcessProxy::fromConnection):
        (WebKit::ChildProcessProxy::connect):
        (WebKit::ChildProcessProxy::terminate):
        (WebKit::ChildProcessProxy::sendMessage):
        (WebKit::ChildProcessProxy::isLaunching):
        (WebKit::ChildProcessProxy::didFinishLaunching):
        (WebKit::ChildProcessProxy::clearConnection):
        * Shared/ChildProcessProxy.h: Added.
        (WebKit::ChildProcessProxy::connection):
        (WebKit::ChildProcessProxy::isValid):
        (WebKit::ChildProcessProxy::canSendMessage):
        (WebKit::ChildProcessProxy::processIdentifier):
        (WebKit::ChildProcessProxy::send):
        (WebKit::ChildProcessProxy::sendSync):
        Moved code from WebProcessProxy. Every ChildProcessProxy is a connection client,
        but every one is a connection queue client, so that latter is passed as an argument.

        * UIProcess/Network/NetworkProcessProxy.h:
        * UIProcess/Network/NetworkProcessProxy.cpp:
        Removed unused syncMessageSendTimedOut(), thank you OVERRIDE.

        * UIProcess/Network/NetworkProcessProxy.cpp:
        (WebKit::NetworkProcessProxy::NetworkProcessProxy):
        (WebKit::NetworkProcessProxy::getLaunchOptions):
        (WebKit::NetworkProcessProxy::~NetworkProcessProxy):
        (WebKit::NetworkProcessProxy::getNetworkProcessConnection):
        (WebKit::NetworkProcessProxy::didFinishLaunching):
        * UIProcess/Network/NetworkProcessProxy.h:
        * UIProcess/Network/mac/NetworkProcessProxyMac.mm:
        (WebKit::NetworkProcessProxy::setApplicationIsOccluded):
        * UIProcess/WebProcessProxy.cpp:
        (WebKit::WebProcessProxy::WebProcessProxy):
        (WebKit::WebProcessProxy::~WebProcessProxy):
        (WebKit::WebProcessProxy::getLaunchOptions):
        (WebKit::WebProcessProxy::disconnect):
        (WebKit::WebProcessProxy::didFinishLaunching):
        Use shared code.

        * UIProcess/WebProcessProxy.h: (WebKit::WebProcessProxy::fromConnection):
        NetworkProcess inherits from conneciton client privately, so we need to do most
        of the work in ClientProcessProxy, and further upcast here.

        * UIProcess/efl/WebProcessProxyEfl.cpp:
        (WebKit::WebProcessProxy::platformGetLaunchOptions):
        * UIProcess/gtk/WebProcessProxyGtk.cpp:
        (WebKit::WebProcessProxy::platformGetLaunchOptions):
        * UIProcess/mac/WebProcessProxyMac.mm:
        (WebKit::WebProcessProxy::platformGetLaunchOptions):
        * UIProcess/qt/WebProcessProxyQt.cpp:
        (WebKit::WebProcessProxy::platformGetLaunchOptions):
        * UIProcess/win/WebProcessProxyWin.cpp:
        (WebKit::WebProcessProxy::platformGetLaunchOptions):
        Updated for renaming, platformConnect -> platformGetLaunchOptions().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: collect all of the information about a run into a new RunDetails class
dpranke@chromium.org [Wed, 5 Dec 2012 23:26:24 +0000 (23:26 +0000)]
nrwt: collect all of the information about a run into a new RunDetails class
https://bugs.webkit.org/show_bug.cgi?id=104158

Reviewed by Ojan Vafai.

This patch collects the information from both test passes (the
main one and the retries) as well as the summarized results and
the exit code into a RunDetails class, so that the integration
tests can more easily tell what happened.

Also, change the way the --lint-test-files works slightly so that
we don't need to create a printer object (and hence we need to
rework the tests that are testing logging).

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(RunDetails):
(RunDetails.__init__):
(Manager.run):
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:
(run):
(main):
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
(passing_run):
(run_and_capture):
(LintTest.test_lint_test_files):
(LintTest.test_lint_test_files__errors):
(MainTest.test_verbose_in_child_processes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: remove --no-record-results
dpranke@chromium.org [Wed, 5 Dec 2012 23:25:07 +0000 (23:25 +0000)]
nrwt: remove --no-record-results
https://bugs.webkit.org/show_bug.cgi?id=104072

Reviewed by Ojan Vafai.

The command line flag was only being used in the unit tests and
wasn't really needed once we started using a mock filesystem.

Also, reformat a bunch of long lines ...

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager.run):
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:
(parse_args):
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
(parse_args):
(passing_run):
(logging_run):
(MainTest.test_repeat_each_iterations_num_tests):
(MainTest.test_test_list):
(MainTest.test_missing_and_unexpected_results):
(MainTest.test_pixel_test_directories):
(MainTest.test_missing_and_unexpected_results_with_custom_exit_code):
(MainTest.test_crash_with_stderr):
(MainTest.test_no_image_failure_with_image_diff):
(MainTest.test_crash_log):
(MainTest.test_web_process_crash_log):
(MainTest.test_exit_after_n_failures_upload):
(MainTest.test_results_directory_absolute):
(MainTest.test_results_directory_relative):
(MainTest.test_reftest_should_not_use_naming_convention_if_not_listed_in_reftestlist):
(MainTest.test_additional_expectations):
(MainTest.test_output_diffs):
(EndToEndTest.test_end_to_end):
(EndToEndTest.test_reftest_with_two_notrefs):
(RebaselineTest.test_reset_results):
(RebaselineTest.test_missing_results):
(RebaselineTest.test_new_baseline):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: only summarize the results one way
dpranke@chromium.org [Wed, 5 Dec 2012 23:14:42 +0000 (23:14 +0000)]
nrwt: only summarize the results one way
https://bugs.webkit.org/show_bug.cgi?id=104064

Reviewed by Ojan Vafai.

Change manager.summarize_results so that we always (and only)
include both expected and unexpected results. Previously we
would summarize the results twice, once with expected results
and once without. This made the code much more confusing than
it needed to be.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(summarize_results):
(Manager.run):
* Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
(ResultSummaryTest.summarized_results):
(ResultSummaryTest.test_no_svn_revision):
(ResultSummaryTest.test_svn_revision):
(ResultSummaryTest.test_summarized_results_wontfix):
* Scripts/webkitpy/layout_tests/views/printing.py:
(Printer.print_results):
(Printer._print_unexpected_results):
(Printer._print_unexpected_results.add_result):
* Scripts/webkitpy/layout_tests/views/printing_unittest.py:
(Testprinter.test_print_unexpected_results.get_unexpected_results):
(Testprinter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWebKit2 child processes need to initialize timer coalescing policy on Mac.
commit-queue@webkit.org [Wed, 5 Dec 2012 23:14:14 +0000 (23:14 +0000)]
WebKit2 child processes need to initialize timer coalescing policy on Mac.
https://bugs.webkit.org/show_bug.cgi?id=103613

Patch by Kiran Muppala <cmuppala@apple.com> on 2012-12-05
Reviewed by Mark Rowe.

Set timer coalescing policy of WebKit2 child processes to that of visible applications
until they can manage the policy based on UI process visibility.

* Shared/mac/ChildProcessMac.mm:
(WebKit::initializeTimerCoalescingPolicy): Set task_latency and task_throughput QOS tiers as appropriate
for visible applications.
(WebKit::ChildProcess::platformInitialize): Add call to initializeTimerCoalescingPolicy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoText decorations are rotated when text-combine takes effect
mitz@apple.com [Wed, 5 Dec 2012 23:07:31 +0000 (23:07 +0000)]
Text decorations are rotated when text-combine takes effect
https://bugs.webkit.org/show_bug.cgi?id=104172

Reviewed by Dave Hyatt.

Source/WebCore:

Test: fast/text/decorations-with-text-combine.html

* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::paint): Applied a rotation to the graphics context when painting
the text decorations for combined text.

LayoutTests:

* fast/text/decorations-with-text-combine.html: Added.
* platform/chromium/TestExpectations: Added to tests missing platform-specific results.
* platform/mac/fast/text/decorations-with-text-combine-expected.png: Added.
* platform/mac/fast/text/decorations-with-text-combine-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Geolocation prompts won't respond to the correct geolocation object...
commit-queue@webkit.org [Wed, 5 Dec 2012 22:46:56 +0000 (22:46 +0000)]
[BlackBerry] Geolocation prompts won't respond to the correct geolocation object if the user opens two pages and respond in the reversed order
https://bugs.webkit.org/show_bug.cgi?id=104162

PR 257505
Patch by Otto Derek Cheung <otcheung@rim.com> on 2012-12-05
Reviewed by Yong Li.

GeolocationClientBlackBerry will register itself to GeolocationHandler when it has
a pending request. When browser responds with WebPageClientImpl::setAllowGeolocation,
we go through the list and let all the pending GeolocationListeners know that a certain
origin has been aceepted/denied.

* WebCoreSupport/GeolocationClientBlackBerry.cpp:
(GeolocationClientBlackBerry::geolocationDestroyed):
(GeolocationClientBlackBerry::requestPermission):
(GeolocationClientBlackBerry::cancelPermissionRequest):
(GeolocationClientBlackBerry::onPermission):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Remove WebLayerTreeView API to set font atlas
commit-queue@webkit.org [Wed, 5 Dec 2012 22:38:48 +0000 (22:38 +0000)]
[chromium] Remove WebLayerTreeView API to set font atlas
https://bugs.webkit.org/show_bug.cgi?id=102960

Patch by Eberhard Graether <egraether@google.com> on 2012-12-05
Reviewed by James Robinson.

Removes old font atlas creation from WebViewImpl.

Source/Platform:

* chromium/public/WebLayerTreeView.h:

Source/WebKit/chromium:

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::WebViewImpl):
(WebKit::WebViewImpl::setShowFPSCounter):
(WebKit::WebViewImpl::setIsAcceleratedCompositingActive):
(WebKit):
* src/WebViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoConnection::waitForMessage shouldn't use the message ID
andersca@apple.com [Wed, 5 Dec 2012 22:16:44 +0000 (22:16 +0000)]
Connection::waitForMessage shouldn't use the message ID
https://bugs.webkit.org/show_bug.cgi?id=104157

Reviewed by Andreas Kling.

Pass the message receiver name and message name to waitForMessage and use them for lookups instead of
the message ID.

* Platform/CoreIPC/Connection.cpp:
(CoreIPC::Connection::createSyncMessageEncoder):
(CoreIPC::Connection::waitForMessage):
(CoreIPC::Connection::processIncomingMessage):
* Platform/CoreIPC/Connection.h:
(CoreIPC::Connection::waitForAndDispatchImmediately):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Better crash fix for resetting tiles on a backingstore takeover.
jpetsovits@rim.com [Wed, 5 Dec 2012 22:12:55 +0000 (22:12 +0000)]
[BlackBerry] Better crash fix for resetting tiles on a backingstore takeover.
https://bugs.webkit.org/show_bug.cgi?id=103969
RIM PR 258415

Reviewed by Adam Treat.

The previous attempt for releasing tile buffers from the
previously active backingstore did not take into account
that resetTiles() would sometimes return early without
having swapped the tiles and thus reclaiming them.

This patch removes the early returns, because it's safe
to swap in an empty geometry at any time and we should
always take the chance when we get it.

Also put the code that calls resetTiles() on an owner
switch right into setCurrentBackingStoreOwner(), so that
it will also release the buffers when the BackingStore
is destroyed (the function is then called with 0).

* Api/BackingStore.cpp:
(BlackBerry::WebKit::BackingStorePrivate::resetTiles):
(WebKit):
(BlackBerry::WebKit::BackingStorePrivate::setCurrentBackingStoreOwner):
* Api/BackingStore.h: Had to add the friend here to make the moved code work from the static function.
* Api/BackingStore_p.h:
(BackingStorePrivate):
* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::resumeBackingStore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCoordinated Graphics: Move AreaAllocator and UpdateAtlas to CoordinatedGraphics
commit-queue@webkit.org [Wed, 5 Dec 2012 21:57:23 +0000 (21:57 +0000)]
Coordinated Graphics: Move AreaAllocator and UpdateAtlas to CoordinatedGraphics
https://bugs.webkit.org/show_bug.cgi?id=103864

Patch by Jae Hyun Park <jae.park@company100.net> on 2012-12-05
Reviewed by Noam Rosenthal.

AreaAllocator and UpdateAtlas are only used for Coordinated Graphics.
So, these should be moved to CoordinatedGraphics to clarify the code.

No new test, because no change of behavior.

* CMakeLists.txt:
* Target.pri:
* WebProcess/WebPage/CoordinatedGraphics/AreaAllocator.cpp: Renamed from Source/WebKit2/WebProcess/WebPage/AreaAllocator.cpp.
(WebKit):
(WebKit::AreaAllocator::AreaAllocator):
(WebKit::AreaAllocator::~AreaAllocator):
(WebKit::AreaAllocator::expand):
(WebKit::AreaAllocator::expandBy):
(WebKit::AreaAllocator::release):
(WebKit::AreaAllocator::overhead):
(WebKit::AreaAllocator::roundAllocation):
(WebKit::GeneralAreaAllocator::GeneralAreaAllocator):
(WebKit::GeneralAreaAllocator::~GeneralAreaAllocator):
(WebKit::GeneralAreaAllocator::freeNode):
(WebKit::GeneralAreaAllocator::expand):
(WebKit::fitsWithin):
(WebKit::GeneralAreaAllocator::allocate):
(WebKit::GeneralAreaAllocator::allocateFromNode):
(WebKit::GeneralAreaAllocator::splitNode):
(WebKit::GeneralAreaAllocator::updateLargestFree):
(WebKit::GeneralAreaAllocator::release):
(WebKit::GeneralAreaAllocator::overhead):
* WebProcess/WebPage/CoordinatedGraphics/AreaAllocator.h: Renamed from Source/WebKit2/WebProcess/WebPage/AreaAllocator.h.
(WebCore::nextPowerOfTwo):
(WebCore):
(WebKit):
(AreaAllocator):
(WebKit::AreaAllocator::size):
(WebKit::AreaAllocator::minimumAllocation):
(WebKit::AreaAllocator::setMinimumAllocation):
(WebKit::AreaAllocator::margin):
(WebKit::AreaAllocator::setMargin):
(GeneralAreaAllocator):
(Node):
* WebProcess/WebPage/CoordinatedGraphics/UpdateAtlas.cpp: Renamed from Source/WebKit2/WebProcess/WebPage/UpdateAtlas.cpp.
(WebKit):
(WebKit::UpdateAtlas::UpdateAtlas):
(WebKit::UpdateAtlas::~UpdateAtlas):
(WebKit::UpdateAtlas::buildLayoutIfNeeded):
(WebKit::UpdateAtlas::didSwapBuffers):
(WebKit::UpdateAtlas::beginPaintingOnAvailableBuffer):
* WebProcess/WebPage/CoordinatedGraphics/UpdateAtlas.h: Renamed from Source/WebKit2/WebProcess/WebPage/UpdateAtlas.h.
(WebCore):
(WebKit):
(UpdateAtlasClient):
(UpdateAtlas):
(WebKit::UpdateAtlas::size):
(WebKit::UpdateAtlas::flags):
(WebKit::UpdateAtlas::addTimeInactive):
(WebKit::UpdateAtlas::isInactive):
(WebKit::UpdateAtlas::isInUse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, add Mac baselines after r136409 (mathml sup/sub box changes).
tony@chromium.org [Wed, 5 Dec 2012 21:47:30 +0000 (21:47 +0000)]
Unreviewed, add Mac baselines after r136409 (mathml sup/sub box changes).

* platform/mac/TestExpectations:
* platform/mac/mathml/presentation/roots-expected.txt:
* platform/mac/mathml/presentation/row-alignment-expected.txt:
* platform/mac/mathml/presentation/sub-expected.txt:
* platform/mac/mathml/presentation/subsup-expected.txt:
* platform/mac/mathml/presentation/sup-expected.txt:
* platform/mac/mathml/presentation/tables-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoExtra space in fast/css/image-resolution/image-resolution.html
leoyang@rim.com [Wed, 5 Dec 2012 21:41:36 +0000 (21:41 +0000)]
Extra space in fast/css/image-resolution/image-resolution.html
https://bugs.webkit.org/show_bug.cgi?id=104130

Reviewed by Emil A Eklund.

String returned by cssText doesn't include an extra tailing space.
This patch is removing an expectation of the extra tailing space in the test.

* fast/css/image-resolution/image-resolution-expected.txt:
* fast/css/image-resolution/image-resolution.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDFG profiler should display more information
fpizlo@apple.com [Wed, 5 Dec 2012 21:40:32 +0000 (21:40 +0000)]
DFG profiler should display more information
https://bugs.webkit.org/show_bug.cgi?id=104163

Reviewed by Gavin Barraclough.

Added the following:

- Distinguish between source counts and bytecode counts.

- Add a "full" summary view that shows both kinds of counts, plus inline counts.

- Add a "source" command to see the source.

- Add a "bytecode" command to see the bytecode and counts.

* Scripts/display-profiler-output:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[skia] Improve performance of GraphicsContext::createCompatibleBuffer by using SkDevi...
junov@google.com [Wed, 5 Dec 2012 21:40:26 +0000 (21:40 +0000)]
[skia] Improve performance of GraphicsContext::createCompatibleBuffer by using SkDevice:createCompatibleDevice
https://bugs.webkit.org/show_bug.cgi?id=103896

Reviewed by Stephen White.

Source/WebCore:

Refactored GraphicsContext::createCompatibleBuffer (platform common
code) to use platform-specific implementations provided by ImageBuffer.
There is no change in behavior for non-skia ports. The skia
implementation uses skia'a own createCompatibleDevice implementation,
which offers several performance benefits:
1. For accelerated contexts, the backing store may be allocated from
the scratch texture pool, which minimizes texture allocation and
deallocation overhead.
2. The backing store will not be initially cleared if it is known
in advance that fully opaque contents will be drawn into the buffer.
3. For non-accelerated contexts, if the backing store is flagged as
opaque, faster blitter loop implementations will be used for drawing
the buffer contents into other buffers. (e.g. generated opaque gradient
fills)

No new tests: covered by existing layout tests

* platform/graphics/Generator.h:
(Generator):
Added hasAlpha method so that it can be used by GeneratorGeneratedImage
* platform/graphics/GeneratorGeneratedImage.cpp:
(WebCore::GeneratorGeneratedImage::drawPattern):
Passing m_generator->hasAlpha() to createCompatibleBuffer in order to
take advantage of optimizations that apply to opaque buffers.
* platform/graphics/Gradient.h:
(Gradient):
Made hasAlpha virtual so that it now overrides Generator::hasAlpha
* platform/graphics/GraphicsContext.cpp:
(WebCore::GraphicsContext::createCompatibleBuffer):
Refactored to use platform-specific implementation
* platform/graphics/GraphicsContext.h:
* platform/graphics/ImageBuffer.cpp:
(WebCore::createCompatibleBuffer):
Non-skia implementation. Mimics old
GraphicsContext::createCompatibleBuffer
* platform/graphics/ImageBuffer.h:
(ImageBuffer):
New skia-specific constructor
* platform/graphics/skia/ImageBufferSkia.cpp:
(WebCore::ImageBuffer::createCompatibleBuffer):
(WebCore::ImageBuffer::ImageBuffer):
* platform/graphics/skia/PlatformContextSkia.cpp:
(WebCore::PlatformContextSkia::createCompatibleDevice):
(WebCore):
* platform/graphics/skia/PlatformContextSkia.h:
(PlatformContextSkia):

LayoutTests:

Added image failure expectation for test needing new baselines

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION (r136683): css3/calc/background-position-parsing.html failing on EFL Linux...
alexis@webkit.org [Wed, 5 Dec 2012 21:38:52 +0000 (21:38 +0000)]
REGRESSION (r136683): css3/calc/background-position-parsing.html failing on EFL Linux 64-bit Debug WK2
https://bugs.webkit.org/show_bug.cgi?id=104131

Reviewed by Antti Koivisto.

css3/calc/background-position-parsing.html assert in debug because we
call CSSParser::validUnit multiple times in a row. The problem was with
validUnit which check calc() values and save the result inside
m_parsedCalculation for later usage. validUnit expects you to
use m_parsedCalculation therefore calling validUnit again with
m_parsedCalculation being set asserts. As parseFillBackgroundPosition
just want to check wether the current value is maybe valid for
background-position we can just relax the check to allow either the
valid keywords or any other units (we will anyway filter the incorrect
values later in the parsing). The most important check at this point
for the shorthand is the validity of the keyword.

No new tests : the assert was covered by css3/calc/background-position-parsing.html.

* css/CSSParser.cpp:
(WebCore::CSSParser::isPotentialPositionValue):
(WebCore::CSSParser::parseFillBackgroundPosition):
* css/CSSParser.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Keep screen updates suspended until after a viewport resize.
jpetsovits@rim.com [Wed, 5 Dec 2012 21:37:34 +0000 (21:37 +0000)]
[BlackBerry] Keep screen updates suspended until after a viewport resize.
https://bugs.webkit.org/show_bug.cgi?id=104041

Reviewed by Adam Treat.

Now that screen and backing store updates can be suspended
and resumed separately from each other, prevent blits for
the whole duration of setViewportSize() and always issue
a blit at the end.

Makes WebPage::setViewportSize() more robust in threaded usage,
easier to read/understand, and gets rid of a couple of FIXMEs.

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::setViewportSize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Unreviewed gardening.
senorblanco@chromium.org [Wed, 5 Dec 2012 21:21:09 +0000 (21:21 +0000)]
[Chromium] Unreviewed gardening.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r136735.
senorblanco@chromium.org [Wed, 5 Dec 2012 21:17:45 +0000 (21:17 +0000)]
Unreviewed, rolling out r136735.
http://trac.webkit.org/changeset/136735
https://bugs.webkit.org/show_bug.cgi?id=102699

Broke Chrome/Win build.

Source/WebCore:

* platform/chromium/PlatformSupport.h:
(PlatformSupport):
* platform/chromium/ScrollbarThemeChromiumWin.cpp:
(WebCore::ScrollbarThemeChromiumWin::paintTrackPiece):
(WebCore::ScrollbarThemeChromiumWin::paintButton):
(WebCore::ScrollbarThemeChromiumWin::paintThumb):
* rendering/RenderThemeChromiumWin.cpp:
(WebCore):
(WebCore::RenderThemeChromiumWin::paintButton):
(WebCore::RenderThemeChromiumWin::paintSliderTrack):
(WebCore::RenderThemeChromiumWin::paintSliderThumb):
(WebCore::RenderThemeChromiumWin::paintMenuList):
(WebCore::RenderThemeChromiumWin::paintTextFieldInternal):
(WebCore::RenderThemeChromiumWin::paintInnerSpinButton):
(WebCore::RenderThemeChromiumWin::paintProgressBar):

Source/WebKit/chromium:

* src/PlatformSupport.cpp:
(WebCore):
(WebCore::PlatformSupport::paintButton):
(WebCore::PlatformSupport::paintMenuList):
(WebCore::PlatformSupport::paintScrollbarArrow):
(WebCore::PlatformSupport::paintScrollbarThumb):
(WebCore::PlatformSupport::paintScrollbarTrack):
(WebCore::PlatformSupport::paintSpinButton):
(WebCore::PlatformSupport::paintTextField):
(WebCore::PlatformSupport::paintTrackbar):
(WebCore::PlatformSupport::paintProgressBar):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r136741.
senorblanco@chromium.org [Wed, 5 Dec 2012 21:16:43 +0000 (21:16 +0000)]
Unreviewed, rolling out r136741.
http://trac.webkit.org/changeset/136741
https://bugs.webkit.org/show_bug.cgi?id=104151

Broke Chrome/Win build.

* WebKit.gyp:
* src/PlatformSupport.cpp:
(WebCore):
(WebCore::WorkerContextProxy::create):
* src/WorkerContextProxy.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Scroll position shifts while FCC & selection handle are dragged
commit-queue@webkit.org [Wed, 5 Dec 2012 21:14:17 +0000 (21:14 +0000)]
[BlackBerry] Scroll position shifts while FCC & selection handle are dragged
https://bugs.webkit.org/show_bug.cgi?id=104156

Patch by Andrew Lo <anlo@rim.com> on 2012-12-05
Reviewed by Rob Buis.
Internally reviewed by Mike Fenton.

Only ensureFocusTextElementVisible on selectionChanged when
an element is being re-focused.
Internal PR256329

* WebKitSupport/InputHandler.cpp:
(BlackBerry::WebKit::InputHandler::InputHandler):
(BlackBerry::WebKit::InputHandler::setElementFocused):
(BlackBerry::WebKit::InputHandler::ensureFocusTextElementVisible):
(BlackBerry::WebKit::InputHandler::selectionChanged):
* WebKitSupport/InputHandler.h:
(InputHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Build with libjpegturbo
yoli@rim.com [Wed, 5 Dec 2012 21:10:18 +0000 (21:10 +0000)]
[BlackBerry] Build with libjpegturbo
https://bugs.webkit.org/show_bug.cgi?id=104152

Reviewed by Rob Buis.

.:

RIM PR# 196975.
This change is made by Ming Xie to link with libjpegturbo.

* Source/cmake/OptionsBlackBerry.cmake:

Tools:

RIM PR# 196975.
Make sure the right include path for libjpeg is included.

* Scripts/webkitdirs.pm:
(blackberryCMakeArguments):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWebGL: use pointer to the source data for Image directly in texImage2D and texSubImag...
commit-queue@webkit.org [Wed, 5 Dec 2012 21:03:01 +0000 (21:03 +0000)]
WebGL: use pointer to the source data for Image directly in texImage2D and texSubImage2D to avoid memory copy if no conversion is needed
https://bugs.webkit.org/show_bug.cgi?id=103885

Patch by Jun Jiang <jun.a.jiang@intel.com> on 2012-12-05
Reviewed by Kenneth Russell.

In texImage2D() and texSubImage2D(), there is a redundant memory copy from Image data to the intermediate vector when no conversion is needed. If the Image resource can be locked properly, it is better to use the pointer to the image source directly and hence avoid the memory copy and improve the performance.

Already covered by current tests.

* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::texImage2DBase):
(WebCore::WebGLRenderingContext::texImage2DImpl):
(WebCore::WebGLRenderingContext::texSubImage2DBase):
(WebCore::WebGLRenderingContext::texSubImage2DImpl):
* html/canvas/WebGLRenderingContext.h:
(WebGLRenderingContext):
* platform/graphics/GraphicsContext3D.h:
(ImageExtractor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd infrastructure for :before and :after in DOM
commit-queue@webkit.org [Wed, 5 Dec 2012 21:00:54 +0000 (21:00 +0000)]
Add infrastructure for :before and :after in DOM
https://bugs.webkit.org/show_bug.cgi?id=103705

Patch by Elliott Sprehn <esprehn@chromium.org> on 2012-12-05
Reviewed by Eric Seidel.

Add all infrastructure for reimplementing generated content :before and
:after as DOM Elements. Now ElementRareData has two pointers to PseudoElements
for the generated content and Node has methods for traversing the tree
including generated content.

This will allow the generated content to be treated as real nodes instead
of anonymous and take part in the usual recalcStyle and attach flow which
fixes many bugs and vastly simplifies the ifecycle of generated content.

Instead of attempting to land both the infrastructure and enable it at
the same time which has proven problematic due to how drastic this change
is, this patch contains only the support code so a much smaller future
patch can be used to switch it on.

No new behavior, this is just the infrastructure.

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* dom/DOMAllInOne.cpp:
* dom/Element.cpp:
(WebCore::Element::~Element):
(WebCore::Element::detach):
(WebCore::Element::updatePseudoElement):
(WebCore::Element::createPseudoElementIfNeeded):
(WebCore::Element::beforePseudoElement):
(WebCore::Element::afterPseudoElement):
* dom/Element.h:
(Element):
* dom/ElementRareData.h:
(ElementRareData):
(WebCore::ElementRareData::ElementRareData):
(WebCore::ElementRareData::~ElementRareData):
(WebCore::ElementRareData::setPseudoElement):
(WebCore::ElementRareData::pseudoElement):
(WebCore::ElementRareData::releasePseudoElement):
* dom/Node.cpp:
(WebCore::Node::pseudoAwarePreviousSibling):
(WebCore::Node::pseudoAwareNextSibling):
(WebCore::Node::rendererIsEditable):
(WebCore::checkAcceptChild):
* dom/Node.h:
(Node):
(WebCore::Node::isPseudoElement):
(WebCore::Node::isBeforePseudoElement):
(WebCore::Node::isAfterPseudoElement):
(WebCore::Node::pseudoId):
(WebCore::Node::customPseudoId):
* dom/NodeRenderingContext.cpp:
(WebCore::NodeRenderingContext::nextRenderer):
(WebCore::NodeRenderingContext::previousRenderer):
* dom/Position.cpp:
(WebCore::Position::hasRenderedNonAnonymousDescendantsWithHeight):
* dom/PseudoElement.cpp: Added.
(WebCore::pseudoElementTagName):
(WebCore::PseudoElement::PseudoElement):
(WebCore::PseudoElement::customStyleForRenderer):
(WebCore::PseudoElement::attach):
(WebCore::PseudoElement::rendererIsNeeded):
(WebCore::PseudoElement::updateChildStyle):
(WebCore::PseudoElement::didRecalcStyle):
* dom/PseudoElement.h: Added.
(PseudoElement):
(WebCore::PseudoElement::create):
(WebCore::pseudoElementIsNeeded):
* editing/visible_units.cpp:
(WebCore::logicallyPreviousBox):
(WebCore::logicallyNextBox):
(WebCore::startPositionForLine):
(WebCore::endPositionForLine):
* page/animation/AnimationController.cpp:
(WebCore::AnimationController::updateAnimations):
* rendering/HitTestResult.cpp:
(WebCore::HitTestResult::setInnerNode):
(WebCore::HitTestResult::setInnerNonSharedNode):
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::isSelectionRoot):
(WebCore::RenderBlock::renderName):
* rendering/RenderCounter.cpp:
(WebCore::RenderCounter::originalText):
* rendering/RenderDeprecatedFlexibleBox.cpp:
(WebCore::RenderDeprecatedFlexibleBox::renderName):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::renderName):
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::renderName):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::createObject):
(WebCore::RenderObject::setPseudoStyle):
(WebCore::RenderObject::createVisiblePosition):
* rendering/RenderObject.h:
(WebCore::RenderObject::isPseudoElement):
(RenderObject):
(WebCore::RenderObject::generatingNode):
* rendering/RenderTableCell.h:
(WebCore::RenderTableCell::renderName):
* rendering/RenderTableRow.h:
(WebCore::RenderTableRow::renderName):
* rendering/RenderTableSection.h:
(WebCore::RenderTableSection::renderName):
* rendering/RenderTreeAsText.cpp:
(WebCore::RenderTreeAsText::writeRenderObject):
* rendering/style/ContentData.cpp:
(WebCore::ImageContentData::createRenderer):
(WebCore::TextContentData::createRenderer):
(WebCore::CounterContentData::createRenderer):
(WebCore::QuoteContentData::createRenderer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove stale check in check-for-webkit-framework-include-consistency
commit-queue@webkit.org [Wed, 5 Dec 2012 20:55:53 +0000 (20:55 +0000)]
Remove stale check in check-for-webkit-framework-include-consistency
https://bugs.webkit.org/show_bug.cgi?id=103775

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2012-12-05
Reviewed by David Kilzer.

The underlying issue was fixed in r81173.

* Scripts/check-for-webkit-framework-include-consistency:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUse of uninitialized variable in WebCore::RenderBox::paintFillLayers
junov@google.com [Wed, 5 Dec 2012 20:53:24 +0000 (20:53 +0000)]
Use of uninitialized variable in WebCore::RenderBox::paintFillLayers
https://bugs.webkit.org/show_bug.cgi?id=104154

Reviewed by Stephen White.

Method FillLayer::clipOccludesNextLayers performs an internal
initialization when called on the first layer of a list of layers.
Without this initialization, calls to clipOccludesNextLayers on
subsequent layers will use uninitialized data.  In some cases, the
call to clipOccludesNextLayers was being short-circuited in
RenderBox::paintFillLayers.
Fix: Predicate was permuted to ensure that clipOccludesNextLayers
is never short-circuited.

Test: fast/backgrounds/size/contain-and-cover-zoomed.html
Running with valgrind reveals the error.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Move WorkerContextProxy::create out of PlatformSupport
pilgrim@chromium.org [Wed, 5 Dec 2012 20:49:21 +0000 (20:49 +0000)]
[Chromium] Move WorkerContextProxy::create out of PlatformSupport
https://bugs.webkit.org/show_bug.cgi?id=104151

Reviewed by Adam Barth.

Doesn't really solve the layering violation as much as kick it
down the road a bit. Part of a refactoring series; see tracking
bug 82948.

* WebKit.gyp:
* src/PlatformSupport.cpp:
(WebCore):
* src/WorkerContextProxy.cpp: Added.
(WebCore):
(WebCore::WebCore::WorkerContextProxy::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Re-enabling encrypted-media tests now that the issues are fixed.
commit-queue@webkit.org [Wed, 5 Dec 2012 20:47:34 +0000 (20:47 +0000)]
[Chromium] Re-enabling encrypted-media tests now that the issues are fixed.
https://bugs.webkit.org/show_bug.cgi?id=104137

Unreviewed revert of r136312.
The issues were fixed in Chromium r171039 & r171081.

Patch by David Dorwin <ddorwin@chromium.org> on 2012-12-05

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEncapsulate ElementRareData for possible future sharing
commit-queue@webkit.org [Wed, 5 Dec 2012 20:43:40 +0000 (20:43 +0000)]
Encapsulate ElementRareData for possible future sharing
https://bugs.webkit.org/show_bug.cgi?id=103948

Patch by Elliott Sprehn <esprehn@gmail.com> on 2012-12-05
Reviewed by Eric Seidel.

Encapsulate ElementRareData and make all access go through getters and
setters. This is already true for NodeRareData, but ElementRareData
had many places where people directly access it.

This the first step in possible future sharing of ElementRareData
instances across nodes. All nodes in a shadow tree have rare data to
store the TreeScope, but in most cases all they have is the TreeScope!
Instead of making every node in a shadow subtree have rare data
and use lots of memory we could just share a single instance of rare data
and only allocate a unique one on modification. Encapsulation lets us
assert about mutations on shared instances.

* dom/Element.cpp:
(WebCore::Element::~Element):
(WebCore::Element::attributes):
(WebCore::Element::shadow):
(WebCore::Element::ensureShadow):
(WebCore::Element::minimumSizeForResizing):
(WebCore::Element::setMinimumSizeForResizing):
(WebCore::Element::computedStyle):
(WebCore::Element::classList):
(WebCore::Element::optionalClassList):
(WebCore::Element::dataset):
(WebCore::Element::hasNamedNodeMap):
(WebCore::Element::savedLayerScrollOffset):
(WebCore::Element::setSavedLayerScrollOffset):
* dom/ElementRareData.cpp:
(WebCore::ElementRareData::reportMemoryUsage):
* dom/ElementRareData.h:
(WebCore::ElementRareData::needsFocusAppearanceUpdateSoonAfterAttach):
(WebCore::ElementRareData::setNeedsFocusAppearanceUpdateSoonAfterAttach):
(ElementRareData):
(WebCore::ElementRareData::styleAffectedByEmpty):
(WebCore::ElementRareData::setStyleAffectedByEmpty):
(WebCore::ElementRareData::isInCanvasSubtree):
(WebCore::ElementRareData::setIsInCanvasSubtree):
(WebCore::ElementRareData::containsFullScreenElement):
(WebCore::ElementRareData::setContainsFullScreenElement):
(WebCore::ElementRareData::isInTopLayer):
(WebCore::ElementRareData::setIsInTopLayer):
(WebCore::ElementRareData::childrenAffectedByHover):
(WebCore::ElementRareData::setChildrenAffectedByHover):
(WebCore::ElementRareData::childrenAffectedByActive):
(WebCore::ElementRareData::setChildrenAffectedByActive):
(WebCore::ElementRareData::childrenAffectedByDrag):
(WebCore::ElementRareData::setChildrenAffectedByDrag):
(WebCore::ElementRareData::childrenAffectedByFirstChildRules):
(WebCore::ElementRareData::setChildrenAffectedByFirstChildRules):
(WebCore::ElementRareData::childrenAffectedByLastChildRules):
(WebCore::ElementRareData::setChildrenAffectedByLastChildRules):
(WebCore::ElementRareData::childrenAffectedByDirectAdjacentRules):
(WebCore::ElementRareData::setChildrenAffectedByDirectAdjacentRules):
(WebCore::ElementRareData::childrenAffectedByForwardPositionalRules):
(WebCore::ElementRareData::setChildrenAffectedByForwardPositionalRules):
(WebCore::ElementRareData::childrenAffectedByBackwardPositionalRules):
(WebCore::ElementRareData::setChildrenAffectedByBackwardPositionalRules):
(WebCore::ElementRareData::childIndex):
(WebCore::ElementRareData::setChildIndex):
(WebCore::ElementRareData::shadow):
(WebCore::ElementRareData::setShadow):
(WebCore::ElementRareData::attributeMap):
(WebCore::ElementRareData::setAttributeMap):
(WebCore::ElementRareData::computedStyle):
(WebCore::ElementRareData::setComputedStyle):
(WebCore::ElementRareData::classList):
(WebCore::ElementRareData::setClassList):
(WebCore::ElementRareData::dataset):
(WebCore::ElementRareData::setDataset):
(WebCore::ElementRareData::minimumSizeForResizing):
(WebCore::ElementRareData::setMinimumSizeForResizing):
(WebCore::ElementRareData::savedLayerScrollOffset):
(WebCore::ElementRareData::setSavedLayerScrollOffset):
(WebCore::ElementRareData::resetComputedStyle):
* dom/NodeRareData.h:
(NodeRareData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agohttp/tests/loading/authentication-after-redirect-stores-wrong-credentials/resources...
commit-queue@webkit.org [Wed, 5 Dec 2012 20:39:33 +0000 (20:39 +0000)]
http/tests/loading/authentication-after-redirect-stores-wrong-credentials/resources/wrong-credential-2-auth-then-redirect-to-finish.php returns a malformed authentication header
https://bugs.webkit.org/show_bug.cgi?id=104052

Patch by Martin Robinson <mrobinson@igalia.com> on 2012-12-05
Reviewed by Alexey Proskuryakov.

Add an authentication realm to the WWW-Authenticate header of the PHP script. Some networking
libraries do not cope well with WWW-Authenticate headers that do not include realms. Additionally
RFC2617 specifies that this part of the header is not-optional.

This test will be unskipped once the rest of the authentication support is complete in the
libsoup backend and GTK+ DRT/WTR.

* http/tests/loading/authentication-after-redirect-stores-wrong-credentials/resources/wrong-credential-2-auth-then-redirect-to-finish.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed Windows re-rebaseline.
roger_fong@apple.com [Wed, 5 Dec 2012 20:35:28 +0000 (20:35 +0000)]
Unreviewed Windows re-rebaseline.

* platform/win/fast/text/orientation-sideways-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium][Win] Remove theme-related functions from PlatformSupport
pilgrim@chromium.org [Wed, 5 Dec 2012 20:30:16 +0000 (20:30 +0000)]
[Chromium][Win] Remove theme-related functions from PlatformSupport
https://bugs.webkit.org/show_bug.cgi?id=102699

Reviewed by Adam Barth.

Call theme engine directly instead of going through
PlatformSupport. Part of a refactoring series. See tracking bug 82948.

Source/WebCore:

* platform/chromium/PlatformSupport.h:
(PlatformSupport):
* platform/chromium/ScrollbarThemeChromiumWin.cpp:
(WebCore::ScrollbarThemeChromiumWin::paintTrackPiece):
(WebCore::ScrollbarThemeChromiumWin::paintButton):
* rendering/RenderThemeChromiumWin.cpp:
(WebCore):
(WebCore::RenderThemeChromiumWin::paintButton):
(WebCore::RenderThemeChromiumWin::paintSliderTrack):
(WebCore::RenderThemeChromiumWin::paintSliderThumb):
(WebCore::RenderThemeChromiumWin::paintMenuList):
(WebCore::RenderThemeChromiumWin::paintTextFieldInternal):
(WebCore::RenderThemeChromiumWin::paintInnerSpinButton):
(WebCore::RenderThemeChromiumWin::paintProgressBar):

Source/WebKit/chromium:

* src/PlatformSupport.cpp:
(WebCore::PlatformSupport::pluginScriptableObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r136609.
senorblanco@chromium.org [Wed, 5 Dec 2012 20:28:07 +0000 (20:28 +0000)]
Unreviewed, rolling out r136609.
http://trac.webkit.org/changeset/136609
https://bugs.webkit.org/show_bug.cgi?id=103793

Breaking Chrome/Mac Debug builds.

Source/Platform:

* Platform.gyp/Platform.gyp:
* chromium/public/WebGraphicsContext3D.h:
(Attributes):

Source/WebCore:

* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::create):
* platform/chromium/support/GraphicsContext3DChromium.cpp:
(WebCore::GraphicsContext3D::create):
* platform/graphics/GraphicsContext3D.h:
(Attributes):

Source/WebKit/chromium:

* WebKit.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: simplify the finding of tests to retry
dpranke@chromium.org [Wed, 5 Dec 2012 20:24:58 +0000 (20:24 +0000)]
nrwt: simplify the finding of tests to retry
https://bugs.webkit.org/show_bug.cgi?id=104067

Reviewed by Ryosuke Niwa.

I think this makes the code a little clearer; note that
we were no longer using the include_missing flag.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager.run):
(Manager._failed_test_names):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: remove two unused parameters from a function
dpranke@chromium.org [Wed, 5 Dec 2012 20:22:23 +0000 (20:22 +0000)]
nrwt: remove two unused parameters from a function
https://bugs.webkit.org/show_bug.cgi?id=104059

Reviewed by Ryosuke Niwa.

Also remove an old, stale FIXME.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(interpret_test_failures):
(summarize_results):
* Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
(ResultSummaryTest.test_interpret_test_failures):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: return result_summaries from LayoutTestRunner.run_tests()
dpranke@chromium.org [Wed, 5 Dec 2012 20:19:09 +0000 (20:19 +0000)]
nrwt: return result_summaries from LayoutTestRunner.run_tests()
https://bugs.webkit.org/show_bug.cgi?id=104051

Reviewed by Ojan Vafai.

This patch completes the refactoring that makes the result summary
an object returned from the layout test runner, rather than having
it passed in and modified. This makes the split of duties between
the manager and runner clearer and makes the data flow more
functionally.

Also clean up some minor lint issues.

* Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
(LayoutTestRunner.__init__):
(LayoutTestRunner.run_tests):
(LayoutTestRunner._mark_interrupted_tests_as_skipped):
* Scripts/webkitpy/layout_tests/controllers/layout_test_runner_unittest.py:
(FakePrinter.print_expected):
(LockCheckingRunner.__init__):
(LayoutTestRunnerTests._runner):
(LayoutTestRunnerTests._run_tests):
(LayoutTestRunnerTests.test_interrupt_if_at_failure_limits):
* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager.__init__):
(Manager.run):
(Manager._run_tests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agonrwt: move creation of the resultsummary objects into Manager._run_tests()
dpranke@chromium.org [Wed, 5 Dec 2012 20:18:02 +0000 (20:18 +0000)]
nrwt: move creation of the resultsummary objects into Manager._run_tests()
https://bugs.webkit.org/show_bug.cgi?id=104048

Reviewed by Ojan Vafai.

This is a preparatory step to making the resultsummary just get
returned from the layout_test_runner; this does most of manager-side
changes.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager._prepare_lists):
(Manager.run):
(Manager._run_tests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSS Exclusions] Enable shape-inside for multiple-segment polygons
commit-queue@webkit.org [Wed, 5 Dec 2012 20:17:30 +0000 (20:17 +0000)]
[CSS Exclusions] Enable shape-inside for multiple-segment polygons
https://bugs.webkit.org/show_bug.cgi?id=91878

Patch by Bear Travis <betravis@adobe.com> on 2012-12-05
Reviewed by David Hyatt.

Source/WebCore:

Patch adding support for multiple-segment polygons for shape-insides from
the CSS Exclusions specification. The layout code has been modified to first
support dividing text into multiple segments per line, and then to lay out
multiple segments per line box.

Tests:  fast/exclusions/shape-inside/shape-inside-multiple-segments-001.html
        fast/exclusions/shape-inside/shape-inside-multiple-segments-002.html
        fast/exclusions/shape-inside/shape-inside-multiple-segments-003.html
        fast/exclusions/shape-inside/shape-inside-multiple-segments-004.html

* platform/text/BidiResolver.h:
(WebCore::BidiCharacterRun::BidiCharacterRun):
(BidiCharacterRun): Added a 'startsSegment' bitfield to track whether a
run begins a new segment. The field is here, rather than in BidiRun, in
order to save space. See Bug 100173.
* rendering/BidiRun.h:
(WebCore::BidiRun::BidiRun):
(BidiRun): Set the initial value for 'startsSegment'.
* rendering/ExclusionShapeInsideInfo.cpp:
(WebCore::ExclusionShapeInsideInfo::computeSegmentsForLine): Clear the
segment ranges when computing segments for a new line.
* rendering/ExclusionShapeInsideInfo.h:
(LineSegmentRange): Added a new type to track segment offsets.
(WebCore::ExclusionShapeInsideInfo::segmentRanges): Added a getter for
the line segment ranges vector.
* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::placeBoxesInInlineDirection): Modified to call
a refactored box placement algorithm that takes a beginning and end position.
(WebCore::InlineFlowBox::placeBoxRangeInInlineDirection): Similar to the
placeBoxes method, but takes an end position that may occur before the end
of the line.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::exclusionShapeInsideInfo): Factored out calls
to ExclusionShapeInsideInfo to the .cpp file, as ExclusionShapeInsideInfo
requires InlineIterator.h, which would create a circular dependency in
ExclusionShapeInsideInfo.h.
* rendering/RenderBlock.h:
(WebCore::RenderBlock::exclusionShapeInsideInfo): Modified to call
exclusionShapeInsideInfo to be non-inline.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::LineWidth::LineWidth): LineWidth now uses the number of computed
segmentRanges to determine which LineSegment to use when calculating width.
(WebCore::RenderBlock::createLineBoxes): Now takes a boolean to determine
whether or not it can share the same parentBox as its siblings (segments
must be able to position themselves separately in the RootInlineBox).
(WebCore::RenderBlock::constructLine):
(WebCore::computeExpansionForJustifiedText): Do include runs that start new
segments when calculating justification for the current segment.
(WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Modified to
use the computeInlineDirectionPositionsForSegment, which can position
multiple segments of text per line.
(WebCore::RenderBlock::computeInlineDirectionPositionsForSegment): Positions
the given segment of text and calculates its layout offsets.
(WebCore::constructBidiRuns): Modified some variables from 'endOfLine' to 'endOfSegment'.
(WebCore::constructBidiRunsForLine): Construct the bidi runs for each segment in
the given line.
(WebCore::RenderBlock::layoutRunsAndFloatsInRange): Call constructBidiRunsForLine.
(WebCore::RenderBlock::LineBreaker::nextLineBreak): Moved the line breaking code
to break segments, with nextLineBreak iterating over segment breaks for the entire line.
(WebCore::RenderBlock::LineBreaker::nextSegmentBreak): Calculates how much text a
segment can hold. This code was originally in nextLineBreak.

LayoutTests:

Adding tests for multiple segment polygons. These tests make sure that text can be
properly broken up into multiple segments per line to fit into a shape-inside.

* fast/exclusions/shape-inside/shape-inside-multiple-segments-001-expected.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-001.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-002-expected.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-002.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-003-expected.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-003.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-004-expected.html: Added.
* fast/exclusions/shape-inside/shape-inside-multiple-segments-004.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r136630.
senorblanco@chromium.org [Wed, 5 Dec 2012 20:14:46 +0000 (20:14 +0000)]
Unreviewed, rolling out r136630.
http://trac.webkit.org/changeset/136630
https://bugs.webkit.org/show_bug.cgi?id=103555

It's breaking Chrome/Win Debug builds.

* platform/graphics/chromium/LazyDecodingPixelRef.cpp:
(WebCore::LazyDecodingPixelRef::LazyDecodingPixelRef):
* platform/graphics/chromium/LazyDecodingPixelRef.h:
(LazyDecodingPixelRef):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Enable CSS_IMAGE_RESOLUTION
leoyang@rim.com [Wed, 5 Dec 2012 20:05:25 +0000 (20:05 +0000)]
[BlackBerry] Enable CSS_IMAGE_RESOLUTION
https://bugs.webkit.org/show_bug.cgi?id=104132

Reviewed by Yong Li.

.:

* Source/cmake/OptionsBlackBerry.cmake:

Source/WebKit/blackberry:

* WebCoreSupport/AboutDataEnableFeatures.in:

Tools:

* Scripts/webkitperl/FeatureList.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Re-add Windows specific results removed in r136699 that were actually...
roger_fong@apple.com [Wed, 5 Dec 2012 20:04:34 +0000 (20:04 +0000)]
Unreviewed. Re-add Windows specific results removed in r136699 that were actually correct.

* platform/win/plugins/npp-set-window-called-during-destruction-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed chromium rebaselines.
eae@chromium.org [Wed, 5 Dec 2012 19:55:57 +0000 (19:55 +0000)]
Unreviewed chromium rebaselines.

* platform/chromium-linux/css3/filters/effect-reference-hw-expected.png: Added.
* platform/chromium-linux/plugins/npp-set-window-called-during-destruction-expected.txt: Added.
* platform/chromium-mac-lion/fast/harness: Added.
* platform/chromium-mac-lion/fast/harness/perftests: Added.
* platform/chromium-mac-lion/fast/harness/perftests/runs-per-second-log-expected.png: Added.
* platform/chromium-mac-lion/fast/harness/perftests/runs-per-second-log-expected.txt: Added.
* platform/chromium-mac-snowleopard/fast/harness: Added.
* platform/chromium-mac-snowleopard/fast/harness/perftests: Added.
* platform/chromium-mac-snowleopard/fast/harness/perftests/runs-per-second-log-expected.png: Added.
* platform/chromium-win-xp/plugins/npp-set-window-called-during-destruction-expected.txt: Added.
* platform/chromium-win/fast/harness/perftests: Added.
* platform/chromium-win/fast/harness/perftests/runs-per-second-log-expected.png: Added.
* platform/chromium-win/fast/harness/perftests/runs-per-second-log-expected.txt: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed gardening.
zandobersek@gmail.com [Wed, 5 Dec 2012 19:51:57 +0000 (19:51 +0000)]
Unreviewed gardening.

Manage failing fast/media reftests for EFL and Qt.

* platform/efl-wk2/TestExpectations:
* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Wed, 5 Dec 2012 19:33:36 +0000 (19:33 +0000)]
Unreviewed GTK gardening.

Adding GTK-specific reference files for two fast/media tests after r136708.
The GTK port isn't yet supporting 3D transforms so these two reference files
should reflect that.

* platform/gtk/fast/media/mq-transform-02-expected.html: Added.
* platform/gtk/fast/media/mq-transform-03-expected.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r136615): Incorrect style sharing in view-source documents.
akling@apple.com [Wed, 5 Dec 2012 19:31:20 +0000 (19:31 +0000)]
REGRESSION(r136615): Incorrect style sharing in view-source documents.
<http://webkit.org/b/104089>

Reviewed by Antti Koivisto.

Source/WebCore:

Make sure we collect feature data about the default view-source style sheet before evaluating
style sharing candidates, otherwise it won't know which class selectors to care about.

* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectFeatures):

LayoutTests:

Unskip: fast/frames/viewsource-attribute.html
        fast/frames/viewsource-on-image-file.html

* platform/chromium/TestExpectations:
* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:
* platform/mac/TestExpectations:
* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed rebaseline on Windows after r136640.
roger_fong@apple.com [Wed, 5 Dec 2012 19:23:26 +0000 (19:23 +0000)]
Unreviewed rebaseline on Windows after r136640.

* platform/win/fast/text/orientation-sideways-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoJSC profiler should not count executions of op_call_put_result because doing so chang...
fpizlo@apple.com [Wed, 5 Dec 2012 19:18:52 +0000 (19:18 +0000)]
JSC profiler should not count executions of op_call_put_result because doing so changes DFG codegen
https://bugs.webkit.org/show_bug.cgi?id=104102

Reviewed by Oliver Hunt.

Source/JavaScriptCore:

This removes op_call_put_result from profiling, since profiling it has an effect on
codegen. This fix enables all of SunSpider, V8, and Kraken to be profiled with the
new profiler.

To make this all fit together, the profiler now also reports in its output the exact
bytecode opcode name for each instruction (in addition to the stringified dump of that
bytecode), so that tools that grok the output can take note of op_call_put_result and
work around the fact that it has no counts.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
(JSC::DFG::ByteCodeParser::parseCodeBlock):
* dfg/DFGDriver.cpp:
(JSC::DFG::compile):
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
* profiler/ProfilerBytecode.cpp:
(JSC::Profiler::Bytecode::toJS):
* profiler/ProfilerBytecode.h:
(JSC::Profiler::Bytecode::Bytecode):
(JSC::Profiler::Bytecode::opcodeID):
(Bytecode):
* profiler/ProfilerDatabase.cpp:
(JSC::Profiler::Database::ensureBytecodesFor):
* runtime/CommonIdentifiers.h:

Tools:

Modify the profiler to not output counts for op_call_put_result, since there
won't be any. Also fix a few weird bugs, like providing better error reporting
when you type something incorrectly and not reporting counts for slow paths
in the old JIT since those counts are actually not what you think they are
(we don't actually count slow path executions separately).

* Scripts/display-profiler-output:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoStop generating call stacks inside ContentSecurityPolicy
mkwst@chromium.org [Wed, 5 Dec 2012 19:18:13 +0000 (19:18 +0000)]
Stop generating call stacks inside ContentSecurityPolicy
https://bugs.webkit.org/show_bug.cgi?id=104126

Reviewed by Yury Semikhatsky.

Now that bug 100650 has made it possible for ConsoleMessage to generate
call stacks, we can remove the stack being generated in
ContentSecurityPolicy::logToConsole. We need the ScriptState*, however,
so this patch makes it possible to pipe it through addConsoleMessage
down into the guts of the inspector.

This patch also removes an unused addConsoleMessage variant from
ScriptExecutionContext, which should have been removed from the patch to
100650 before landing it. My bad.

* dom/Document.cpp:
(WebCore::Document::addMessage):
* dom/Document.h:
(Document):
    Accept a ScriptState* parameter in Document::addMessage, and pass it
    through to Console::addMessage.
* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::addConsoleMessage):
* dom/ScriptExecutionContext.h:
(ScriptExecutionContext):
    Accept a ScriptState* parameter in Document::addMessage, and pass it
    through to ScriptExecutionContext::addMessage. Also, remove an unused
    variant of ScriptExecutionContext::addConsoleMessage that accepted
    only a call stack in addition to the basics.
* inspector/ConsoleMessage.cpp:
(WebCore::ConsoleMessage::ConsoleMessage):
* inspector/ConsoleMessage.h:
(ConsoleMessage):
    Accept a ScriptState* and pass it into autogenerateMetadata for
    call stack generational goodness.
* inspector/InspectorConsoleAgent.cpp:
(WebCore::InspectorConsoleAgent::addMessageToConsole):
    Accept a ScriptState* and pass it into addConsoleMessage.
(WebCore::InspectorConsoleAgent::didFinishXHRLoading):
(WebCore::InspectorConsoleAgent::didReceiveResponse):
(WebCore::InspectorConsoleAgent::didFailLoading):
    Add an empty ScriptState parameter to the addConsoleMessage call.
* inspector/InspectorConsoleAgent.h:
(InspectorConsoleAgent):
* inspector/InspectorConsoleInstrumentation.h:
(WebCore::InspectorInstrumentation::addMessageToConsole):
* inspector/InspectorInstrumentation.cpp:
(WebCore):
(WebCore::InspectorInstrumentation::addMessageToConsoleImpl):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
* page/Console.cpp:
(WebCore::Console::addMessage):
* page/Console.h:
(Console):
    Pipe ScriptState through the intermediate classes.
* page/ContentSecurityPolicy.cpp:
(WebCore::ContentSecurityPolicy::logToConsole):
    Drop call stack generation, and pass ScriptState* into addConsoleMessage.
* workers/WorkerContext.cpp:
(WebCore::WorkerContext::addConsoleMessage):
(WebCore::WorkerContext::addMessage):
(WebCore::WorkerContext::addMessageToWorkerConsole):
* workers/WorkerContext.h:
(WorkerContext):
    Accept ScriptState* and pass it into addMessageToConsole.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed gardening.
zandobersek@gmail.com [Wed, 5 Dec 2012 19:10:48 +0000 (19:10 +0000)]
Unreviewed gardening.

Removing redundant pixel and render tree baselines for fast/media tests.
This is the second and last patch covering non-Chromium ports.

Too many files to list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDOMImplementation.createDocument should call appendChild rather than parserAppendChil...
rafaelw@chromium.org [Wed, 5 Dec 2012 19:02:06 +0000 (19:02 +0000)]
DOMImplementation.createDocument should call appendChild rather than parserAppendChild to add docType and documentElement
https://bugs.webkit.org/show_bug.cgi?id=104040

Reviewed by Ojan Vafai.

createDocument now calls appendChild.

No new tests (no observable changes).

* dom/ContainerNode.cpp:
(WebCore::ContainerNode::parserAppendChild):
* dom/DOMImplementation.cpp:
(WebCore::DOMImplementation::createDocument):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed rebaseline.
roger_fong@apple.com [Wed, 5 Dec 2012 19:00:14 +0000 (19:00 +0000)]
Unreviewed rebaseline.
https://bugs.webkit.org/show_bug.cgi?id=98209.

* platform/win/fast/text/international/float-as-only-child-of-isolate-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CMake] Enable to build WebKit sources without Tools
commit-queue@webkit.org [Wed, 5 Dec 2012 18:59:16 +0000 (18:59 +0000)]
[CMake] Enable to build WebKit sources without Tools
https://bugs.webkit.org/show_bug.cgi?id=103918

Patch by Laszlo Gombos <l.gombos@samsung.com> on 2012-12-05
Reviewed by Gyuyoung Kim.

Turn on building Tools by default only if ENABLE_TOOLS is not explicitly set
(enabled or disabled) and the Tools directory exists for all CMake based ports.

This change enables the possibility to build WebKit even if the Tools directory
does not exists.

* CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIndexedDB: Stub out transaction-backend methods
alecflett@chromium.org [Wed, 5 Dec 2012 18:52:18 +0000 (18:52 +0000)]
IndexedDB: Stub out transaction-backend methods
https://bugs.webkit.org/show_bug.cgi?id=103921

Reviewed by Dimitri Glazkov.

Source/WebCore:

In preparation for a refactoring. Legacy methods
will be removed in https://bugs.webkit.org/show_bug.cgi?id=103923.

No new tests, just adding unused methods for refactoring.

* Modules/indexeddb/IDBCallbacks.h:
(WebCore::IDBCallbacks::onUpgradeNeeded):
* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::onAbort):
(WebCore):
(WebCore::IDBDatabase::onComplete):
* Modules/indexeddb/IDBDatabase.h:
(IDBDatabase):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::createTransaction):
(WebCore):
* Modules/indexeddb/IDBDatabaseBackendImpl.h:
(IDBDatabaseBackendImpl):
(WebCore::IDBDatabaseBackendImpl::commit):
(WebCore::IDBDatabaseBackendImpl::abort):
* Modules/indexeddb/IDBDatabaseBackendInterface.h:
(IDBDatabaseBackendInterface):
* Modules/indexeddb/IDBDatabaseCallbacks.h:
(IDBDatabaseCallbacks):
* Modules/indexeddb/IDBDatabaseCallbacksImpl.cpp:
(WebCore::IDBDatabaseCallbacksImpl::onAbort):
(WebCore):
(WebCore::IDBDatabaseCallbacksImpl::onComplete):
* Modules/indexeddb/IDBDatabaseCallbacksImpl.h:
(IDBDatabaseCallbacksImpl):
* Modules/indexeddb/IDBFactoryBackendImpl.cpp:
(WebCore::IDBFactoryBackendImpl::open):
* Modules/indexeddb/IDBFactoryBackendImpl.h:
(WebCore::IDBFactoryBackendImpl::open):
(IDBFactoryBackendImpl):
* Modules/indexeddb/IDBFactoryBackendInterface.h:
(IDBFactoryBackendInterface):
* Modules/indexeddb/IDBOpenDBRequest.cpp:
(WebCore::IDBOpenDBRequest::onUpgradeNeeded):
(WebCore):
* Modules/indexeddb/IDBOpenDBRequest.h:
(IDBOpenDBRequest):
* Modules/indexeddb/IDBTransactionBackendInterface.h:
(WebCore):
* Modules/indexeddb/IDBTransactionCoordinator.h:

Source/WebKit/chromium:

Add new signatures for refactoring, to allow
transaction_ids to make is through the open/onUpgradeNeeded
loop.

* public/WebIDBCallbacks.h:
(WebKit::WebIDBCallbacks::onUpgradeNeeded):
* public/WebIDBDatabase.h:
(WebKit::WebIDBDatabase::createTransaction):
(WebKit::WebIDBDatabase::abort):
(WebKit::WebIDBDatabase::commit):
(WebIDBDatabase):
* public/WebIDBDatabaseCallbacks.h:
(WebIDBDatabaseCallbacks):
(WebKit::WebIDBDatabaseCallbacks::onAbort):
(WebKit::WebIDBDatabaseCallbacks::onComplete):
* public/WebIDBFactory.h:
(WebKit::WebIDBFactory::open):
* src/IDBCallbacksProxy.cpp:
(WebKit::IDBCallbacksProxy::onUpgradeNeeded):
(WebKit):
* src/IDBCallbacksProxy.h:
(IDBCallbacksProxy):
* src/IDBDatabaseBackendProxy.cpp:
(WebKit::IDBDatabaseBackendProxy::createTransaction):
(WebKit):
(WebKit::IDBDatabaseBackendProxy::commit):
(WebKit::IDBDatabaseBackendProxy::abort):
* src/IDBDatabaseBackendProxy.h:
(IDBDatabaseBackendProxy):
* src/IDBDatabaseCallbacksProxy.cpp:
(WebKit::IDBDatabaseCallbacksProxy::onAbort):
(WebKit):
(WebKit::IDBDatabaseCallbacksProxy::onComplete):
* src/IDBDatabaseCallbacksProxy.h:
(IDBDatabaseCallbacksProxy):
* src/IDBFactoryBackendProxy.cpp:
(WebKit::IDBFactoryBackendProxy::open):
(WebKit):
* src/IDBFactoryBackendProxy.h:
(IDBFactoryBackendProxy):
* src/WebIDBCallbacksImpl.cpp:
(WebKit::WebIDBCallbacksImpl::onUpgradeNeeded):
(WebKit):
* src/WebIDBCallbacksImpl.h:
(WebIDBCallbacksImpl):
* src/WebIDBDatabaseCallbacksImpl.cpp:
(WebKit::WebIDBDatabaseCallbacksImpl::onAbort):
(WebKit):
(WebKit::WebIDBDatabaseCallbacksImpl::onComplete):
* src/WebIDBDatabaseCallbacksImpl.h:
(WebIDBDatabaseCallbacksImpl):
* src/WebIDBDatabaseImpl.cpp:
(WebKit::WebIDBDatabaseImpl::createTransaction):
(WebKit):
(WebKit::WebIDBDatabaseImpl::abort):
(WebKit::WebIDBDatabaseImpl::commit):
* src/WebIDBDatabaseImpl.h:
(WebIDBDatabaseImpl):
* src/WebIDBFactoryImpl.cpp:
(WebKit::WebIDBFactoryImpl::open):
(WebKit):
* src/WebIDBFactoryImpl.h:
(WebIDBFactoryImpl):
* tests/IDBAbortOnCorruptTest.cpp:
* tests/IDBDatabaseBackendTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] REGRESSION: 2 tests started to fail with newer Qt5
ossy@webkit.org [Wed, 5 Dec 2012 18:48:12 +0000 (18:48 +0000)]
[Qt] REGRESSION: 2 tests started to fail with newer Qt5
https://bugs.webkit.org/show_bug.cgi?id=104140

Unreviewed gardening, skip new failing tests to paint the bots green.

* platform/qt-5.0/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r126683): Table cell are getting borders when the style doesn't mention any
robert@webkit.org [Wed, 5 Dec 2012 18:22:34 +0000 (18:22 +0000)]
REGRESSION(r126683): Table cell are getting borders when the style doesn't mention any
https://bugs.webkit.org/show_bug.cgi?id=101677

Reviewed by Julien Chaffraix.

Source/WebCore:

The bordercolor attribute on a cell or row has no effect on FF and Opera, even when the parent
table has a border attribute. IE displays the bordercolor of the cell and row in all cases except when
the parent has no border attribute. WebKit's old behaviour (before r126683) was to display the
bordercolor of the row but not the cells (I don't think this behaviour was particularly deliberate). No
browser displays the bordercolor for a col or tbody element.

Bring WebKit into line with Opera and FF by ignoring the bordercolor attribute on cells, rows, cols, rowgroups, and
colgroups in all cases, regardless of whether the table has a border or not.

Test: fast/table/td-bordercolor-attribute.html

* html/HTMLTablePartElement.cpp:
(WebCore::HTMLTablePartElement::isPresentationAttribute):
(WebCore::HTMLTablePartElement::collectStyleForPresentationAttribute):

LayoutTests:

* fast/table/td-bordercolor-attribute-expected.txt: Added.
* fast/table/td-bordercolor-attribute.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed gardening.
zandobersek@gmail.com [Wed, 5 Dec 2012 18:07:32 +0000 (18:07 +0000)]
Unreviewed gardening.

Removing redundant pixel and render tree baselines for fast/media tests.
This first patch of a two-part cleanup covers Chromium ports.

Too many files to list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Introduce WebKit-EFL to Assistive Technologies
commit-queue@webkit.org [Wed, 5 Dec 2012 17:56:12 +0000 (17:56 +0000)]
[EFL][WK2] Introduce WebKit-EFL to Assistive Technologies
https://bugs.webkit.org/show_bug.cgi?id=104000

Patch by Krzysztof Czech <k.czech@samsung.com> on 2012-12-05
Reviewed by Chris Fleizach.

Introduce Webkit-EFL so that Assistive Technologies can distinguish it.

* accessibility/atk/WebKitAccessibleWrapperAtk.cpp:
(webkitAccessibleGetAttributes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove unneeded WTF prefix from WTF types in RenderFlexibleBox
commit-queue@webkit.org [Wed, 5 Dec 2012 17:52:04 +0000 (17:52 +0000)]
Remove unneeded WTF prefix from WTF types in RenderFlexibleBox
https://bugs.webkit.org/show_bug.cgi?id=104109

Patch by Carlos Garcia Campos <cgarcia@igalia.com> on 2012-12-05
Reviewed by Eric Seidel.

WTF prefix is not needed because WTF headers have a using
statement at the end of the file.

* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::layoutBlock):
(WebCore::RenderFlexibleBox::repositionLogicalHeightDependentFlexItems):
(WebCore::RenderFlexibleBox::layoutFlexItems):
(WebCore::RenderFlexibleBox::freezeViolations):
(WebCore::RenderFlexibleBox::resolveFlexibleLengths):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren):
(WebCore::RenderFlexibleBox::alignFlexLines):
(WebCore::RenderFlexibleBox::alignChildren):
(WebCore::RenderFlexibleBox::flipForWrapReverse):
* rendering/RenderFlexibleBox.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTurn pixel tests in fast/media into reftests
zandobersek@gmail.com [Wed, 5 Dec 2012 17:51:34 +0000 (17:51 +0000)]
Turn pixel tests in fast/media into reftests
https://bugs.webkit.org/show_bug.cgi?id=103632

Reviewed by Darin Adler.

Turn the pixel tests under fast/media into reftests.
The reference files are tailored so they present the same output
that the pixel baselines. The latter will be removed in two subsequent commits.

* fast/media/implicit-media-all-expected.html: Added.
* fast/media/media-descriptor-syntax-01-expected.html: Added.
* fast/media/media-descriptor-syntax-02-expected.html: Added.
* fast/media/media-descriptor-syntax-03-expected.html: Added.
* fast/media/media-descriptor-syntax-04-expected.html: Added.
* fast/media/media-descriptor-syntax-05-expected.html: Added.
* fast/media/media-descriptor-syntax-06-expected.html: Added.
* fast/media/media-query-invalid-value-expected.html: Added.
* fast/media/media-query-invalid-value-expected.txt: Removed.
* fast/media/media-type-syntax-01-expected.html: Added.
* fast/media/media-type-syntax-02-expected.html: Added.
* fast/media/monochrome-expected.html: Added.
* fast/media/mq-animation-expected.html: Added.
* fast/media/mq-aspect-ratio-expected.html: Added.
* fast/media/mq-compound-query-01-expected.html: Added.
* fast/media/mq-compound-query-02-expected.html: Added.
* fast/media/mq-compound-query-03-expected.html: Added.
* fast/media/mq-compound-query-04-expected.html: Added.
* fast/media/mq-compound-query-05-expected.html: Added.
* fast/media/mq-grid-01-expected.html: Added.
* fast/media/mq-grid-02-expected.html: Added.
* fast/media/mq-invalid-media-feature-01-expected.html: Added.
* fast/media/mq-invalid-media-feature-02-expected.html: Added.
* fast/media/mq-invalid-media-feature-03-expected.html: Added.
* fast/media/mq-invalid-media-feature-04-expected.html: Added.
* fast/media/mq-invalid-syntax-01-expected.html: Added.
* fast/media/mq-invalid-syntax-02-expected.html: Added.
* fast/media/mq-invalid-syntax-03-expected.html: Added.
* fast/media/mq-invalid-syntax-04-expected.html: Added.
* fast/media/mq-invalid-syntax-05-expected.html: Added.
* fast/media/mq-js-media-except-01-expected.html: Added.
* fast/media/mq-js-media-except-01-expected.txt: Removed.
* fast/media/mq-js-media-except-02-expected.html: Added.
* fast/media/mq-js-media-except-02-expected.txt: Removed.
* fast/media/mq-js-media-except-03-expected.html: Added.
* fast/media/mq-js-media-except-03-expected.txt: Removed.
* fast/media/mq-js-media-forward-syntax-expected.html: Added.
* fast/media/mq-js-stylesheet-media-01-expected.html: Added.
* fast/media/mq-js-stylesheet-media-02-expected.html: Added.
* fast/media/mq-js-stylesheet-media-03-expected.html: Added.
* fast/media/mq-js-stylesheet-media-04-expected.html: Added.
* fast/media/mq-max-pixel-ratio-expected.html: Added.
* fast/media/mq-min-constraint-expected.html: Added.
* fast/media/mq-min-pixel-ratio-expected.html: Added.
* fast/media/mq-min-pixel-ratio-expected.txt: Removed.
* fast/media/mq-orientation-expected.html: Added.
* fast/media/mq-pixel-ratio-expected.html: Added.
* fast/media/mq-relative-constraints-02-expected.html: Added.
* fast/media/mq-relative-constraints-03-expected.html: Added.
* fast/media/mq-relative-constraints-04-expected.html: Added.
* fast/media/mq-relative-constraints-05-expected.html: Added.
* fast/media/mq-relative-constraints-06-expected.html: Added.
* fast/media/mq-relative-constraints-07-expected.html: Added.
* fast/media/mq-relative-constraints-08-expected.html: Added.
* fast/media/mq-relative-constraints-08-expected.txt: Removed.
* fast/media/mq-relative-constraints-09-expected.html: Added.
* fast/media/mq-simple-neg-query-01-expected.html: Added.
* fast/media/mq-simple-neg-query-02-expected.html: Added.
* fast/media/mq-simple-neg-query-03-expected.html: Added.
* fast/media/mq-simple-neg-query-04-expected.html: Added.
* fast/media/mq-simple-neg-query-05-expected.html: Added.
* fast/media/mq-simple-query-01-expected.html: Added.
* fast/media/mq-simple-query-02-expected.html: Added.
* fast/media/mq-simple-query-03-expected.html: Added.
* fast/media/mq-simple-query-04-expected.html: Added.
* fast/media/mq-simple-query-05-expected.html: Added.
* fast/media/mq-transform-01-expected.html: Added.
* fast/media/mq-transform-02-expected.html: Added.
* fast/media/mq-transform-03-expected.html: Added.
* fast/media/mq-transform-04-expected.html: Added.
* fast/media/mq-transition-expected.html: Added.
* fast/media/mq-valueless-expected.html: Added.
* fast/media/mq-width-absolute-01-expected.html: Added.
* fast/media/mq-width-absolute-02-expected.html: Added.
* fast/media/mq-width-absolute-03-expected.html: Added.
* fast/media/mq-width-absolute-04-expected.html: Added.
* fast/media/view-mode-media-feature-expected.html: Added.
* fast/media/view-mode-media-feature-expected.txt: Removed.
* fast/media/viewport-media-query-expected.html: Added.
* platform/gtk/fast/media/mq-transform-02-expected.html: Added as 3D transforms support
isn't yet enabled on the GTK port.
* platform/gtk/fast/media/mq-transform-03-expected.html: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix compilation for Qt5.0.0 stable branch.
ossy@webkit.org [Wed, 5 Dec 2012 17:48:15 +0000 (17:48 +0000)]
Fix compilation for Qt5.0.0 stable branch.
https://bugs.webkit.org/show_bug.cgi?id=103870

Source/WebCore:

Patch by Andras Becsi <andras.becsi@digia.com> on 2012-12-05
Reviewed by Simon Hausmann.

This is a squashed patch consisting of patches by multiple authors.

Fixed use of to-be-removed compatibility functions in QWindow
Patch by Samuel Rødal <samuel.rodal@digia.com>

Build fix. No new tests needed.

* platform/graphics/gstreamer/PlatformVideoWindowQt.cpp:
(FullScreenVideoWindow::FullScreenVideoWindow):
(PlatformVideoWindow::PlatformVideoWindow):

Source/WebKit2:

Patch by Michael Brüning <michael.bruning@digia.com> on 2012-12-05
Reviewed by Simon Hausmann.

QWindow::pos() and QQuickItem::pos() have been renamed to position().
Patch by Lars Knoll <lars.knoll@digia.com>

* UIProcess/qt/PageViewportControllerClientQt.cpp:
(WebKit::PageViewportControllerClientQt::setContentRectVisiblePositionAtScale):
(WebKit::PageViewportControllerClientQt::setViewportPosition):

Tools:

Patch by Andras Becsi <andras.becsi@digia.com> on 2012-12-05
Reviewed by Simon Hausmann.

Fix the use of renamed QWindow functions.
Patch by Andras Becsi <andras.becsi@digia.com>

* MiniBrowser/qt/BrowserWindow.cpp:
(BrowserWindow::BrowserWindow):
(BrowserWindow::onTitleChanged):
* MiniBrowser/qt/raw/View.cpp:
(View::View):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: more robust treeoutline.findTreeElement
commit-queue@webkit.org [Wed, 5 Dec 2012 17:45:56 +0000 (17:45 +0000)]
Web Inspector: more robust treeoutline.findTreeElement
https://bugs.webkit.org/show_bug.cgi?id=103911

Patch by John J. Barton <johnjbarton@chromium.org> on 2012-12-05
Reviewed by Vsevolod Vlasov.

Non-recursive algorithm to populate the treeoutline given an DOMNode
as representedObject. Walk up the parents from the desired object
until we find one that is currently represented in the treeoutline
cache. Then walk down, putting children in the cache until we get
to the desired object again.
With Pavel Feldeman <pfeldman@chromium.org>

No new tests, refactoring, no new function

       * inspector/front-end/treeoutline.js:
(TreeOutline.prototype.findTreeElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][GTK] List value response headers are not handled in RespourceResponse (libsoup...
commit-queue@webkit.org [Wed, 5 Dec 2012 17:44:40 +0000 (17:44 +0000)]
[EFL][GTK] List value response headers are not handled in RespourceResponse (libsoup specific)
https://bugs.webkit.org/show_bug.cgi?id=95181

Patch by Alexander Shalamov <alexander.shalamov@intel.com> on 2012-12-05
Reviewed by Martin Robinson.

Source/WebCore:

When list-value response headers are received, they are handled as single
value headers. This patch fixes incorrect behavior. New test was added to
verify list-value header handling.

Test: http/tests/xmlhttprequest/xmlhttprequest-test-custom-headers.html

* platform/network/soup/ResourceResponseSoup.cpp:
(WebCore::ResourceResponse::updateFromSoupMessage):

LayoutTests:

Added test that checks if list-value response headers handled correctly.

* http/tests/xmlhttprequest/resources/custom-headers.php: Added.
* http/tests/xmlhttprequest/xmlhttprequest-test-custom-headers-expected.txt: Added.
* http/tests/xmlhttprequest/xmlhttprequest-test-custom-headers.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Introduce WebKit-EFL to Assistive Technologies - new baselines
commit-queue@webkit.org [Wed, 5 Dec 2012 17:42:27 +0000 (17:42 +0000)]
[EFL][WK2] Introduce WebKit-EFL to Assistive Technologies - new baselines
https://bugs.webkit.org/show_bug.cgi?id=104000

Patch by Krzysztof Czech <k.czech@samsung.com> on 2012-12-05
Reviewed by Chris Fleizach.

Added EFL baselines for accessibility/table-cells.html and accessibility/table-cell-spans.html.

* platform/efl-wk2/TestExpectations:
* platform/efl-wk2/accessibility/table-cell-spans-expected.txt: Added.
* platform/efl-wk2/accessibility/table-cells-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Don't include error message text in the editor buffer.
pfeldman@chromium.org [Wed, 5 Dec 2012 17:41:01 +0000 (17:41 +0000)]
Web Inspector: Don't include error message text in the editor buffer.
https://bugs.webkit.org/show_bug.cgi?id=103932

Reviewed by Alexander Pavlov.

Source/WebCore:

Skip entire decoration content when collecting dirty regions.

* inspector/front-end/DefaultTextEditor.js:
(WebInspector.TextEditorMainPanel.prototype._collectLinesFromDOM):

LayoutTests:

* inspector/editor/editor-test.js:
(initialize_EditorTests.InspectorTest.getLineElement):
(initialize_EditorTests):
* inspector/editor/text-editor-type-over-decoration-expected.txt: Added.
* inspector/editor/text-editor-type-over-decoration.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, adding my new email address to watchlist for blackberry.
yoli@rim.com [Wed, 5 Dec 2012 17:34:04 +0000 (17:34 +0000)]
Unreviewed, adding my new email address to watchlist for blackberry.

 * Scripts/webkitpy/common/config/watchlist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Wed, 5 Dec 2012 17:26:43 +0000 (17:26 +0000)]
Unreviewed GTK gardening.

Skipping back the Media Stream layout tests, they're not behaving well.
Removing a few failure expectations for tests that are passing.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] meta viewport pages are rendered at a bad initial scale
zhajiang@rim.com [Wed, 5 Dec 2012 17:25:15 +0000 (17:25 +0000)]
[BlackBerry] meta viewport pages are rendered at a bad initial scale
https://bugs.webkit.org/show_bug.cgi?id=104122

Patch by Jacky Jiang <zhajiang@rim.com>.
Reviewed by Rob Buis.
Internally reviewed by Arvid Nilsson.

PR: 240378
Many websites are rendered at a very bad initial scale because they
both specify meta viewport tag and have large width elements. In such
a case we will think the meta viewport tag is broken if the width of
any element is larger than the calculated layout viewport width.
The fix to this is straightforward that we will zoom the page to fit
the screen to have better user experience instead of respecting the
meta viewport. This can fix the issue on the websites such as
www.travelpod.com, www.forbes.com, interfacelift.com, www.sky.com,
raphaeljs.com and so on.
If non-empty m_userViewportArguments is applied, then we will force to
respect the viewport arguments. In this way we will restrict the fix to
browser only.

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::WebPagePrivate):
(BlackBerry::WebKit::WebPagePrivate::setLoadState):
(BlackBerry::WebKit::WebPagePrivate::respectViewport):
(WebKit):
(BlackBerry::WebKit::WebPagePrivate::initialScale):
(BlackBerry::WebKit::WebPagePrivate::maximumScale):
(BlackBerry::WebKit::WebPagePrivate::dispatchViewportPropertiesDidChange):
* Api/WebPage_p.h:
(BlackBerry::WebKit::WebPagePrivate::isUserScalable):
(WebPagePrivate):
(BlackBerry::WebKit::WebPagePrivate::minimumScale):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed chromium rebaselines.
eae@chromium.org [Wed, 5 Dec 2012 17:22:23 +0000 (17:22 +0000)]
Unreviewed chromium rebaselines.

* platform/chromium-mac/plugins/npp-set-window-called-during-destruction-expected.txt: Added.
* platform/chromium-win-xp/css3/flexbox: Added.
* platform/chromium-win-xp/css3/flexbox/repaint-column-reverse-expected.png: Added.
* platform/chromium-win/css3/filters/effect-reference-hw-expected.png:
* platform/chromium-win/plugins/npp-set-window-called-during-destruction-expected.txt:
* platform/chromium/plugins/npp-set-window-called-during-destruction-expected.txt: Removed.
* platform/win-7sp0/plugins: Added.
* platform/win-7sp0/plugins/npp-set-window-called-during-destruction-expected.txt: Added.
* platform/win/plugins/npp-set-window-called-during-destruction-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Unreviewed gardening.
thiago.santos@intel.com [Wed, 5 Dec 2012 17:20:54 +0000 (17:20 +0000)]
[EFL] Unreviewed gardening.

* platform/efl-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Gtk] navigator.plugins contains too many plugin entries. First one are garbages
commit-queue@webkit.org [Wed, 5 Dec 2012 17:09:34 +0000 (17:09 +0000)]
[Gtk] navigator.plugins contains too many plugin entries. First one are garbages
https://bugs.webkit.org/show_bug.cgi?id=102438

Patch by Arnaud Renevier <a.renevier@sisa.samsung.com> on 2012-12-05
Reviewed by Xan Lopez.

Source/WebKit/gtk:

In getPluginInfo, outPlugins is resized to plugins.size and then,
plugins are appended to it. So at the end, outPlugins will be twice
too large, and first half will contain null objects. As outPlugins
size is 0 when calling getPluginInfo, we don't need to resize it.

* WebCoreSupport/PlatformStrategiesGtk.cpp:
(PlatformStrategiesGtk::getPluginInfo):

LayoutTests:

Add a test to check that navigator.plugins only contain valid plugin
objects.

* plugins/navigator-plugins-expected.txt: Added.
* plugins/navigator-plugins.html: Added.
* plugins/script-tests/navigator-plugins.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIndexedDB: Replace use of ScriptExecutionContext::Task (Part 2)
jsbell@chromium.org [Wed, 5 Dec 2012 17:06:27 +0000 (17:06 +0000)]
IndexedDB: Replace use of ScriptExecutionContext::Task (Part 2)
https://bugs.webkit.org/show_bug.cgi?id=103931

Reviewed by Tony Chang.

Source/WebCore:

Actually drop use of ScriptExecutionContext::Task and remove incorrect usage of
ThreadSafeRefCounted<>. Define a new IDBTransactionBackendImpl::Operation base class
for operations; storage of per-operation data becomes explicit.

No new tests - just a refactor.

* Modules/indexeddb/IDBCallbacks.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBCursorBackendImpl.cpp:
(WebCore::IDBCursorBackendImpl::CursorIterationOperation::create):
(IDBCursorBackendImpl::CursorIterationOperation):
(WebCore::IDBCursorBackendImpl::CursorIterationOperation::CursorIterationOperation):
(WebCore::IDBCursorBackendImpl::CursorAdvanceOperation::create):
(IDBCursorBackendImpl::CursorAdvanceOperation):
(WebCore::IDBCursorBackendImpl::CursorAdvanceOperation::CursorAdvanceOperation):
(WebCore::IDBCursorBackendImpl::CursorPrefetchIterationOperation::create):
(IDBCursorBackendImpl::CursorPrefetchIterationOperation):
(WebCore::IDBCursorBackendImpl::CursorPrefetchIterationOperation::CursorPrefetchIterationOperation):
(WebCore::IDBCursorBackendImpl::CursorAdvanceOperation::perform):
(WebCore::IDBCursorBackendImpl::CursorIterationOperation::perform):
(WebCore::IDBCursorBackendImpl::CursorPrefetchIterationOperation::perform):
* Modules/indexeddb/IDBCursorBackendInterface.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBDatabase.cpp:
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::CreateObjectStoreOperation::create):
(IDBDatabaseBackendImpl::CreateObjectStoreOperation):
(WebCore::IDBDatabaseBackendImpl::CreateObjectStoreOperation::CreateObjectStoreOperation):
(WebCore::IDBDatabaseBackendImpl::DeleteObjectStoreOperation::create):
(IDBDatabaseBackendImpl::DeleteObjectStoreOperation):
(WebCore::IDBDatabaseBackendImpl::DeleteObjectStoreOperation::DeleteObjectStoreOperation):
(WebCore::IDBDatabaseBackendImpl::VersionChangeOperation::create):
(IDBDatabaseBackendImpl::VersionChangeOperation):
(WebCore::IDBDatabaseBackendImpl::VersionChangeOperation::VersionChangeOperation):
(WebCore::IDBDatabaseBackendImpl::CreateObjectStoreAbortOperation::create):
(IDBDatabaseBackendImpl::CreateObjectStoreAbortOperation):
(WebCore::IDBDatabaseBackendImpl::CreateObjectStoreAbortOperation::CreateObjectStoreAbortOperation):
(WebCore::IDBDatabaseBackendImpl::DeleteObjectStoreAbortOperation::create):
(IDBDatabaseBackendImpl::DeleteObjectStoreAbortOperation):
(WebCore::IDBDatabaseBackendImpl::DeleteObjectStoreAbortOperation::DeleteObjectStoreAbortOperation):
(WebCore::IDBDatabaseBackendImpl::VersionChangeAbortOperation::create):
(IDBDatabaseBackendImpl::VersionChangeAbortOperation):
(WebCore::IDBDatabaseBackendImpl::VersionChangeAbortOperation::VersionChangeAbortOperation):
(WebCore::IDBDatabaseBackendImpl::createObjectStore):
(WebCore::IDBDatabaseBackendImpl::CreateObjectStoreOperation::perform):
(WebCore::IDBDatabaseBackendImpl::deleteObjectStore):
(WebCore::IDBDatabaseBackendImpl::DeleteObjectStoreOperation::perform):
(WebCore::IDBDatabaseBackendImpl::VersionChangeOperation::perform):
(WebCore::IDBDatabaseBackendImpl::runIntVersionChangeTransaction):
(WebCore::IDBDatabaseBackendImpl::CreateObjectStoreAbortOperation::perform):
(WebCore::IDBDatabaseBackendImpl::DeleteObjectStoreAbortOperation::perform):
(WebCore::IDBDatabaseBackendImpl::VersionChangeAbortOperation::perform):
* Modules/indexeddb/IDBDatabaseBackendInterface.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBDatabaseCallbacks.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBFactoryBackendImpl.cpp:
* Modules/indexeddb/IDBFactoryBackendInterface.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBIndexBackendImpl.cpp:
(WebCore::IDBIndexBackendImpl::OpenIndexCursorOperation::create):
(IDBIndexBackendImpl::OpenIndexCursorOperation):
(WebCore::IDBIndexBackendImpl::OpenIndexCursorOperation::OpenIndexCursorOperation):
(WebCore::IDBIndexBackendImpl::IndexCountOperation::create):
(IDBIndexBackendImpl::IndexCountOperation):
(WebCore::IDBIndexBackendImpl::IndexCountOperation::IndexCountOperation):
(WebCore::IDBIndexBackendImpl::IndexReferencedValueRetrievalOperation::create):
(IDBIndexBackendImpl::IndexReferencedValueRetrievalOperation):
(WebCore::IDBIndexBackendImpl::IndexReferencedValueRetrievalOperation::IndexReferencedValueRetrievalOperation):
(WebCore::IDBIndexBackendImpl::IndexValueRetrievalOperation::create):
(IDBIndexBackendImpl::IndexValueRetrievalOperation):
(WebCore::IDBIndexBackendImpl::IndexValueRetrievalOperation::IndexValueRetrievalOperation):
(WebCore::IDBIndexBackendImpl::OpenIndexCursorOperation::perform):
(WebCore::IDBIndexBackendImpl::openCursor):
(WebCore::IDBIndexBackendImpl::openKeyCursor):
(WebCore::IDBIndexBackendImpl::IndexCountOperation::perform):
(WebCore::IDBIndexBackendImpl::count):
(WebCore::IDBIndexBackendImpl::IndexReferencedValueRetrievalOperation::perform):
(WebCore::IDBIndexBackendImpl::IndexValueRetrievalOperation::perform):
(WebCore::IDBIndexBackendImpl::get):
(WebCore::IDBIndexBackendImpl::getKey):
* Modules/indexeddb/IDBIndexBackendInterface.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBKey.h: Remove bogus "ThreadSafe"
(IDBKey):
* Modules/indexeddb/IDBKeyRange.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreRetrievalOperation::create):
(IDBObjectStoreBackendImpl::ObjectStoreRetrievalOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreRetrievalOperation::ObjectStoreRetrievalOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreStorageOperation::create):
(IDBObjectStoreBackendImpl::ObjectStoreStorageOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreStorageOperation::ObjectStoreStorageOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreIndexesReadyOperation::create):
(IDBObjectStoreBackendImpl::ObjectStoreIndexesReadyOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreIndexesReadyOperation::ObjectStoreIndexesReadyOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreDeletionOperation::create):
(IDBObjectStoreBackendImpl::ObjectStoreDeletionOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreDeletionOperation::ObjectStoreDeletionOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreClearOperation::create):
(IDBObjectStoreBackendImpl::ObjectStoreClearOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreClearOperation::ObjectStoreClearOperation):
(WebCore::IDBObjectStoreBackendImpl::CreateIndexOperation::create):
(IDBObjectStoreBackendImpl::CreateIndexOperation):
(WebCore::IDBObjectStoreBackendImpl::CreateIndexOperation::CreateIndexOperation):
(WebCore::IDBObjectStoreBackendImpl::DeleteIndexOperation::create):
(IDBObjectStoreBackendImpl::DeleteIndexOperation):
(WebCore::IDBObjectStoreBackendImpl::DeleteIndexOperation::DeleteIndexOperation):
(WebCore::IDBObjectStoreBackendImpl::OpenObjectStoreCursorOperation::create):
(IDBObjectStoreBackendImpl::OpenObjectStoreCursorOperation):
(WebCore::IDBObjectStoreBackendImpl::OpenObjectStoreCursorOperation::OpenObjectStoreCursorOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreCountOperation::create):
(IDBObjectStoreBackendImpl::ObjectStoreCountOperation):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreCountOperation::ObjectStoreCountOperation):
(WebCore::IDBObjectStoreBackendImpl::CreateIndexAbortOperation::create):
(IDBObjectStoreBackendImpl::CreateIndexAbortOperation):
(WebCore::IDBObjectStoreBackendImpl::CreateIndexAbortOperation::CreateIndexAbortOperation):
(WebCore::IDBObjectStoreBackendImpl::DeleteIndexAbortOperation::create):
(IDBObjectStoreBackendImpl::DeleteIndexAbortOperation):
(WebCore::IDBObjectStoreBackendImpl::DeleteIndexAbortOperation::DeleteIndexAbortOperation):
(WebCore::IDBObjectStoreBackendImpl::get):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreRetrievalOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::put):
(WebCore):
(WebCore::IDBObjectStoreBackendImpl::setIndexesReady):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreIndexesReadyOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreStorageOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::deleteFunction):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreDeletionOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::clear):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreClearOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::createIndex):
(WebCore::IDBObjectStoreBackendImpl::CreateIndexOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::deleteIndex):
(WebCore::IDBObjectStoreBackendImpl::DeleteIndexOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::openCursor):
(WebCore::IDBObjectStoreBackendImpl::OpenObjectStoreCursorOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::count):
(WebCore::IDBObjectStoreBackendImpl::ObjectStoreCountOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::CreateIndexAbortOperation::perform):
(WebCore::IDBObjectStoreBackendImpl::DeleteIndexAbortOperation::perform):
* Modules/indexeddb/IDBObjectStoreBackendInterface.h: Remove bogus "ThreadSafe"
* Modules/indexeddb/IDBRequest.h: Remove bogus "ThreadSafe"
(IDBRequest):
* Modules/indexeddb/IDBTransactionBackendImpl.cpp:
(WebCore::IDBTransactionBackendImpl::IDBTransactionBackendImpl): Appease RefPtr<> adoption strictness.
(WebCore::IDBTransactionBackendImpl::scheduleTask):
(WebCore::IDBTransactionBackendImpl::abort):
(WebCore::IDBTransactionBackendImpl::taskTimerFired):
* Modules/indexeddb/IDBTransactionBackendImpl.h:
(Operation):
(WebCore::IDBTransactionBackendImpl::Operation::~Operation):
(IDBTransactionBackendImpl):
(WebCore::IDBTransactionBackendImpl::scheduleTask):
* Modules/indexeddb/IDBTransactionBackendInterface.h: Remove bogus "ThreadSafe"

Source/WebKit/chromium:

Now that IDBCallbacks uses real RefPtr<> a mock class needs to be heap allocated.

* tests/IDBAbortOnCorruptTest.cpp:
(WebCore::MockIDBCallbacks::create):
(WebCore::MockIDBCallbacks::MockIDBCallbacks):
(MockIDBCallbacks):
(WebCore::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoExtend StringHasher to take a stream of characters
jonlee@apple.com [Wed, 5 Dec 2012 17:06:26 +0000 (17:06 +0000)]
Extend StringHasher to take a stream of characters
https://bugs.webkit.org/show_bug.cgi?id=104076
<rdar://problem/12811887>

Reviewed by Darin Adler.

We essentially move computeHash() to a member function of StringHasher. This allows clients of the
hasher to append strings to the general hash, without having to first concatenate the strings.

* wtf/StringHasher.h:
(WTF::StringHasher::addCharacters): Same as computeHash(), except also take into account the possibility
that there is a pending character to hash. If this is the case, we align the word boundary by consuming the
first character in data, adjust the length, move the data pointer forward one, and update the hasher's
pending character state. Also add an early return so that we are guaranteed at least one character in case
we run into this situation.
(WTF::StringHasher::computeHash): Refactor into addCharacters().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] GeolocationClientBB only holding the latest geolocation permission request
commit-queue@webkit.org [Wed, 5 Dec 2012 16:48:22 +0000 (16:48 +0000)]
[BlackBerry] GeolocationClientBB only holding the latest geolocation permission request
https://bugs.webkit.org/show_bug.cgi?id=104123

PR 256089

Patch by Otto Derek Cheung <otcheung@rim.com> on 2012-12-05
Reviewed by Yong Li.

Add a mapping between origin URL and it's geolocation object so we could
respond to the correct geolocation when libwebview responds with onPermission
in GeolocationClient.

Using the mapping, we could also prevent requestGeolocation requests from the same origin
to be sent twice. This prevents pages that hold two iframes to the same resource to request
geolocation permission on the same security origin twice.

Tested this by running a page that includes 2 or more iframes that requests geolocation
permission.

* WebCoreSupport/GeolocationClientBlackBerry.cpp:
(GeolocationClientBlackBerry::requestPermission):
(GeolocationClientBlackBerry::cancelPermissionRequest):
(GeolocationClientBlackBerry::onPermission):
* WebCoreSupport/GeolocationClientBlackBerry.h:
(GeolocationClientBlackBerry):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoENABLE(IMAGE_DECODER_DOWN_SAMPLING): Don't swizzle decode down sampled images
noel.gordon@gmail.com [Wed, 5 Dec 2012 16:36:19 +0000 (16:36 +0000)]
ENABLE(IMAGE_DECODER_DOWN_SAMPLING): Don't swizzle decode down sampled images
https://bugs.webkit.org/show_bug.cgi?id=103856

Reviewed by Yong Li.

For ports using ENABLE(IMAGE_DECODER_DOWN_SAMPLING), turbo swizzle decodes provide
no speed-up when the decoded image will be down sampled because the data must pass
through the buffer.setRGBA() slow path. That is not the swizzled path. Instead use
JCS_RGB decodes to clarify the output color space handling in outputScanlines().

No new tests. Covered by existing JPEG decoding tests.

* platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
(WebCore::JPEGImageReader::decode): libjpeg-turbo swizzle decodes provides no real
speed-up if the image will be down sampled. Revert to using JCS_RGA in this case.
(WebCore::setPixel): Remove the libjpeg-turbo JCS_EXT_BGRA and JCS_EXT_RGBA cases.
(WebCore::JPEGImageDecoder::outputScanlines): ditto.
* platform/image-decoders/jpeg/JPEGImageDecoder.h:
(WebCore::JPEGImageDecoder::willDownSample): Return true if image down sampling
will be applied to the decoded image (m_scaled). Note: the willDownSample() return
value is valid only after setSize() has been called: add an ASSERT for that.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Wed, 5 Dec 2012 16:27:07 +0000 (16:27 +0000)]
Unreviewed GTK gardening.

Moving some expectations for expected failures under the appropriate section.

Rather than skipping expected failures, they should be run (as long as they
don't time out in great numbers) just to get the extra test coverage (mostly
to confirm they don't crash unexpectably). Most of such expectations are being
unskipped and changed into a proper failure expectation, but more can folllow.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] EWK2UnitTestBase.ewk_favicon_database_async_icon_get is crashing with...
christophe.dumez@intel.com [Wed, 5 Dec 2012 16:10:37 +0000 (16:10 +0000)]
[EFL][WK2] EWK2UnitTestBase.ewk_favicon_database_async_icon_get is crashing with new Ewk_Context
https://bugs.webkit.org/show_bug.cgi?id=104110

Reviewed by Laszlo Gombos.

Unregister the WKFaviconDatabase client in EwkFaviconDatabase
destructor to avoid crashing if the callbacks get called
after the EwkFaviconDatabase object has been destroyed (i.e.
the parent EwkContext object was destroyed).

* UIProcess/API/efl/ewk_favicon_database.cpp:
(EwkFaviconDatabase::~EwkFaviconDatabase):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Unreviewed gardening.
senorblanco@chromium.org [Wed, 5 Dec 2012 15:57:42 +0000 (15:57 +0000)]
[Chromium] Unreviewed gardening.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Avoid showing FCC on empty text field in a different way
commit-queue@webkit.org [Wed, 5 Dec 2012 15:52:50 +0000 (15:52 +0000)]
[BlackBerry] Avoid showing FCC on empty text field in a different way
https://bugs.webkit.org/show_bug.cgi?id=104019

Patch by Yongxin Dai <yodai@rim.com> on 2012-12-05
Reviewed by Rob Buis.

PR #222796

Add text field empty flag along with the caret change notification so that
FineCursorContnrol is able to avoid showing FCC on empty text filed.
Previous approach, notifying client with empty caret on empty text field,
causes problem. Its code is removed.

Reviewed Internally by Mike Fenton.

* Api/WebPageClient.h:
* WebKitSupport/SelectionHandler.cpp:
(BlackBerry::WebKit::SelectionHandler::caretPositionChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening, updating baselines after r136657 and r136672.
zandobersek@gmail.com [Wed, 5 Dec 2012 15:51:13 +0000 (15:51 +0000)]
Unreviewed GTK gardening, updating baselines after r136657 and r136672.

* platform/gtk/http/tests/inspector/console-websocket-error-expected.txt:
* platform/gtk/http/tests/xmlhttprequest/access-control-preflight-async-header-denied-expected.txt: Removed.
* platform/gtk/http/tests/xmlhttprequest/access-control-preflight-async-method-denied-expected.txt: Removed.
* platform/gtk/http/tests/xmlhttprequest/access-control-preflight-async-not-supported-expected.txt: Removed.
* platform/gtk/http/tests/xmlhttprequest/origin-whitelisting-https-expected.txt: Removed.
* platform/gtk/http/tests/xmlhttprequest/simple-cross-origin-denied-events-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTools: [WK2] InjectedBundle should check Intent url for null
commit-queue@webkit.org [Wed, 5 Dec 2012 15:43:14 +0000 (15:43 +0000)]
Tools: [WK2] InjectedBundle should check Intent url for null
https://bugs.webkit.org/show_bug.cgi?id=104117

Patch by Jussi Kukkonen <jussi.kukkonen@intel.com> on 2012-12-05
Reviewed by Kenneth Rohde Christiansen.

WKIntentServiceInfoCopyHref() can return null, we need to check
for that in InjectedBundlePage::registerIntentServiceForFrame()
to avoid crashing.

* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::InjectedBundlePage::registerIntentServiceForFrame):

LayoutTests: [EFL] InjectedBundle should check Intent url for null
https://bugs.webkit.org/show_bug.cgi?id=104117

Patch by Jussi Kukkonen <jussi.kukkonen@intel.com> on 2012-12-05
Reviewed by Kenneth Rohde Christiansen.

EFL is slightly more verbose than chromium on this one.

* platform/efl/fast/images/crossfade-client-not-removed-crash-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIndexedDB: Implement IndexedDB bindings for JSC
commit-queue@webkit.org [Wed, 5 Dec 2012 15:37:44 +0000 (15:37 +0000)]
IndexedDB: Implement IndexedDB bindings for JSC
https://bugs.webkit.org/show_bug.cgi?id=103484

Patch by Michael Pruett <michael@68k.org> on 2012-12-05
Reviewed by Kentaro Hara.

IndexedDB is currently implemented only for V8. This change adds
the bindings necessary for IndexedDB to work with JSC. With this
patch, IndexedDB for JSC passes 112 out of the 196 layout tests in
storage/indexeddb. IndexedDB is still not enabled for any JSC port.

Tests: storage/indexeddb/*

* CMakeLists.txt:
* GNUmakefile.list.am:
* Modules/indexeddb/IDBOpenDBRequest.idl:
* Modules/indexeddb/IDBVersionChangeRequest.idl:
* UseJSC.cmake:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/DOMRequestState.h: Added to match bindings/v8/DOMRequestState.h.
(WebCore):
(DOMRequestState):
(WebCore::DOMRequestState::DOMRequestState):
(WebCore::DOMRequestState::clear):
(Scope):
(WebCore::DOMRequestState::Scope::Scope):
(WebCore::DOMRequestState::exec):
* bindings/js/IDBBindingUtilities.cpp: Add utility functions matching those in bindings/v8/IDBBindingUtilities.cpp.
(WebCore::get):
(WebCore):
(WebCore::canSet):
(WebCore::set):
(WebCore::createIDBKeyFromValue):
(WebCore::getNthValueOnKeyPath):
(WebCore::createIDBKeyFromScriptValueAndKeyPath):
(WebCore::ensureNthValueOnKeyPath):
(WebCore::canInjectNthValueOnKeyPath):
(WebCore::injectIDBKeyIntoScriptValue):
(WebCore::canInjectIDBKeyIntoScriptValue):
(WebCore::deserializeIDBValue):
(WebCore::idbKeyToScriptValue):
* bindings/js/IDBBindingUtilities.h: Add utility functions matching those in bindings/v8/IDBBindingUtilities.h.
(WebCore):
* bindings/js/JSIDBAnyCustom.cpp:
(WebCore::toJS):
* bindings/js/JSIDBKeyCustom.cpp:
(WebCore::toJS):
* bindings/js/JSIDBVersionChangeRequestCustom.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@136686 268f45cc-cd09-0410-ab3c-d52691b4dbfc