- 19 Nov, 2011 4 commits
-
-
ojan@chromium.org authored
Patch by Alice Boxhall <aboxhall@chromium.org> on 2011-11-19 Reviewed by Ojan Vafai. * TestResultServer/static-dashboards/dashboard_base.js: (ADD_RESULTS): (flattenTrie): * TestResultServer/static-dashboards/flakiness_dashboard_tests.js: (testFlattenTrie): git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100864 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
ojan@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72814 Reviewed by Adam Barth. * TestResultServer/model/jsonresults.py: * TestResultServer/model/jsonresults_unittest.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100863 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
ojan@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72813 Reviewed by Adam Barth. Once this lands, we can start storing pure json in the test results server and then we can delete the code with all the FIXMEs added here. * Scripts/webkitpy/layout_tests/controllers/manager.py: * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py: Only add jsonp for full_results.json. * TestResultServer/model/jsonresults.py: * TestResultServer/model/jsonresults_unittest.py: Accept pure json uploads. * TestResultServer/static-dashboards/dashboard_base.js: (appendJSONScriptElementFor): Use the callback parameter so that the server can start returning pure json if it's left out. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100862 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
ojan@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72809 Reviewed by Adam Barth. I'm moving the server to storing and serving up raw json instead of jsonp. You can still get the jsonp by passing a "callback" parameter, but there's no need for run-webkit-tests to get jsonp when all it wants is the raw json. * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py: Make the wrapper stripping only happen if the json is actually wrapped. * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100861 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
- 18 Nov, 2011 12 commits
-
-
dbates@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72768 .: Reviewed by Adam Barth and Antonio Gomes. * Source/CMakeLists.txt: Add the BlackBerry port and support for building on QNX. * Source/cmake/OptionsBlackBerry.cmake: Added. * Source/cmake/OptionsCommon.cmake: Add QNX QCC-variant of CODE_GENERATOR_PREPROCESSOR. We may be able to fold this into the non-MSVC CODE_GENERATOR_PREPROCESSOR. Source/JavaScriptCore: Reviewed by Antonio Gomes. * PlatformBlackBerry.cmake: Added. * shell/PlatformBlackBerry.cmake: Added. * wtf/PlatformBlackBerry.cmake: Added. Source/WebCore: Reviewed by Antonio Gomes. * CMakeLists.txt: At this time the BlackBerry port doesn't support generating DOM bindings from the SVG IDLs. See WebKit bug #72773. * PlatformBlackBerry.cmake: Added. Source/WebKit: Reviewed by Antonio Gomes. * blackberry: Added. * blackberry/CMakeListsBlackBerry.txt: Added. Tools: Reviewed by Antonio Gomes. Add optional --blackberry command line argument to build-webkit to build the BlackBerry port. * Scripts/build-webkit: * Scripts/webkitdirs.pm: (determineBaseProductDir): (argumentsForConfiguration): (builtDylibPathForName): (determineIsQt): (isBlackBerry): Added. (determineIsBlackBerry): Added. (blackberryTargetArchitecture): Added. (isAppleWebKit): git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100844 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
jamesr@google.com authored
http://trac.webkit.org/changeset/100835 https://bugs.webkit.org/show_bug.cgi?id=72778 Broke some page scale tests * DumpRenderTree/chromium/EventSender.cpp: (EventSender::scalePageBy): git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100837 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
abarth@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72782 Reviewed by Eric Seidel. .: Update these files to refer to the new location of ManualTests. * .gitattributes: * Source/cmake/WebKitPackaging.cmake: Tools: Update this file to refer to the new location of ManualTests. * Scripts/webkitpy/common/config/build.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100836 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
jamesr@google.com authored
https://bugs.webkit.org/show_bug.cgi?id=72778 Reviewed by Kenneth Russell. Set the min/max page scale clamps when a test calls eventSender.scalePageBy() so that the scale isn't clamped. * DumpRenderTree/chromium/EventSender.cpp: (EventSender::scalePageBy): git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100835 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
commit-queue@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=71753 Patch by Scott Graham <scottmg@chromium.org> on 2011-11-18 Reviewed by Adam Barth. Source/WebCore: IDL changes and associated plumbing to expose list of gamepad objects on navigator object (per current spec). Full patch is https://bugs.webkit.org/show_bug.cgi?id=69451. Only basic existence test until more plumbing in future patches. Test: gamepad/gamepad-api.html * WebCore.gypi: * bindings/generic/RuntimeEnabledFeatures.h: (WebCore::RuntimeEnabledFeatures::setWebkitGamepadsEnabled): (WebCore::RuntimeEnabledFeatures::webkitGamepadsEnabled): * page/Gamepad.cpp: Added. (WebCore::Gamepad::Gamepad): (WebCore::Gamepad::axes): (WebCore::Gamepad::buttons): (WebCore::Gamepad::~Gamepad): * page/Gamepad.h: Added. * page/Gamepad.idl: Added. * page/GamepadList.cpp: Added. (WebCore::GamepadList::~GamepadList): (WebCore::GamepadList::set): (WebCore::GamepadList::length): (WebCore::GamepadList::item): * page/GamepadList.h: Added. (WebCore::GamepadList::create): (WebCore::GamepadList::GamepadList): * page/GamepadList.idl: Added. * page/Navigator.cpp: (WebCore::Navigator::webkitGamepads): * page/Navigator.h: * page/Navigator.idl: Source/WebKit/chromium: Update to use vendor-prefixed enable. * src/WebRuntimeFeatures.cpp: (WebKit::WebRuntimeFeatures::enableGamepad): (WebKit::WebRuntimeFeatures::isGamepadEnabled): Tools: Runtime enable gamepad in chromium test shell. * DumpRenderTree/chromium/TestShell.cpp: (TestShell::TestShell): LayoutTests: Add basic api existence test. * gamepad/gamepad-api-expected.txt: Added. * gamepad/gamepad-api.html: Added. * gamepad/gamepad-test.js: Added. (logConsole): (testExpected): (reportExpected): (waitForEventAndEnd): (waitForEvent._eventCallback): (waitForEvent): (waitForEventAndTest._eventCallback): (waitForEventAndTest): (waitForEventTestAndEnd): (endTest): (logResult): (consoleWrite): * platform/chromium/fast/dom/navigator-detached-no-crash-expected.txt: * platform/efl/Skipped: * platform/gtk/Skipped: * platform/mac/Skipped: * platform/qt/Skipped: * platform/win/Skipped: * platform/wincairo/Skipped: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100833 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
darin@chromium.org authored
[chromium] Improve fullscreen API https://bugs.webkit.org/show_bug.cgi?id=70477 Reviewed by Adam Barth. Move fullscreen entry points from WebView to WebWidget since they are just about widget geometry. Delay webkitfullscreenchange event until we know that we have entered fullscreen mode. This is accomplished by expanding the WebKit APIs to differentiate will{Enter,Exit} from did{Enter,Exit} state changes. WebViewImpl is forced to keep a reference to the requested fullscreen element while we are requesting to enter fullscreen mode. Once in fullscreen mode, we switch to keeping only a reference to the Frame containing the fullscreen element, which we drop upon exiting fullscreen mode. * public/WebView.h: * public/WebViewClient.h: * public/WebWidget.h: (WebKit::WebWidget::willEnterFullScreen): (WebKit::WebWidget::didEnterFullScreen): (WebKit::WebWidget::willExitFullScreen): (WebKit::WebWidget::didExitFullScreen): * public/WebWidgetClient.h: (WebKit::WebWidgetClient::enterFullScreen): (WebKit::WebWidgetClient::exitFullScreen): * src/ChromeClientImpl.cpp: (WebKit::ChromeClientImpl::supportsFullscreenForNode): (WebKit::ChromeClientImpl::enterFullscreenForNode): (WebKit::ChromeClientImpl::exitFullscreenForNode): (WebKit::ChromeClientImpl::supportsFullScreenForElement): (WebKit::ChromeClientImpl::enterFullScreenForElement): (WebKit::ChromeClientImpl::exitFullScreenForElement): * src/WebViewImpl.cpp: (WebKit::WebViewImpl::willEnterFullScreen): (WebKit::WebViewImpl::didEnterFullScreen): (WebKit::WebViewImpl::willExitFullScreen): (WebKit::WebViewImpl::didExitFullScreen): (WebKit::WebViewImpl::enterFullScreenForElement): (WebKit::WebViewImpl::exitFullScreenForElement): * src/WebViewImpl.h: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100821 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
* Scripts/old-run-webkit-tests: (countAndPrintLeaks): * Scripts/webkitpy/layout_tests/port/leakdetector.py: (LeakDetector._callstacks_to_exclude_from_leaks): Added ScanFromString to the call stacks to exclude on all OS versions. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100799 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
* Scripts/old-run-webkit-tests: (countAndPrintLeaks): * Scripts/webkitpy/layout_tests/port/leakdetector.py: (LeakDetector._callstacks_to_exclude_from_leaks): Added SecTransformExecute to the call stacks to exclude. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100790 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
yutak@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72687 Reviewed by Simon Hausmann. Source/WebKit/qt: * WebCoreSupport/DumpRenderTreeSupportQt.cpp: (DumpRenderTreeSupportQt::defaultHixie76WebSocketProtocolEnabled): (DumpRenderTreeSupportQt::setHixie76WebSocketProtocolEnabled): Call WebCore::Settings::setUseHixie76WebSocketProtocol() to switch WebSocket protocols. * WebCoreSupport/DumpRenderTreeSupportQt.h: Tools: Let LayoutTestControllerQt be able to switch WebSocket protocols via overridePreference(). * DumpRenderTree/qt/DumpRenderTreeQt.cpp: (WebCore::WebPage::resetSettings): * DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::overridePreference): LayoutTests: * platform/qt/Skipped: Unskip WebSocket hybi tests. Skip some of them which are known to fail probably due to a bug in SocketStreamHandleQt implementation. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100763 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
sergio@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72707 Reviewed by Philippe Normand. Tools: Make new-run-webkit-tests use the right path to locate the test plugin used by WebKitTestRunner. * Scripts/webkitpy/layout_tests/port/gtk.py: LayoutTests: Unskipped a test that is passing now. * platform/gtk-wk2/Skipped: unskipped plugins/get-url-notify-with-url-that-fails-to-load.html git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100755 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
hausmann@webkit.org authored
[Qt] WTR and InjectedBundle should not link statically against JSC/WTF https://bugs.webkit.org/show_bug.cgi?id=72697 Reviewed by Kenneth Rohde Christiansen. .: * Source/api.pri: On Linux, use a GNU LD version script to manage the symbol exports, which includes the public WK1/WK2 API as well as a bunch of private WTF symbols needed for WTR and the Injected Bundle. * Source/qtwebkit-export.map: Added. Tools: * Tools.pro: Make WTR linux only for the moment. * WebKitTestRunner/InjectedBundle/Target.pri: Remove static jsc/wtf linkage. * WebKitTestRunner/Target.pri: Ditto. * qmake/mkspecs/features/default_post.prf: Don't build the main webkit sources with ELF visibility on Linux. Instead we're temporarily using an LD version script. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100746 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
kenneth@webkit.org authored
Reviewed by Simon Hausmann. As we lay out desktop pages using the width of 980, change our default size to 980x735 so that we are not scaling down by default. * MiniBrowser/qt/MiniBrowserApplication.h: (WindowOptions::WindowOptions): git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100743 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
- 17 Nov, 2011 24 commits
-
-
commit-queue@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72681 Patch by Raphael Kubo da Costa <kubo@profusion.mobi> on 2011-11-17 Reviewed by Daniel Bates. .: * Source/cmake/OptionsEfl.cmake: Rename DATA_DIR to DATA_INSTALL_DIR and add it to the cache; add a variable with the path of the generated theme and remove the -DDATA_DIR definition, it was moved to PlatformEFL.cmake in WebCore. Source/WebCore: Add the -DDATA_DIR definition here instead of defining it globally in OptionsEfl.cmake, as WebCore is the only place which needs it. No new tests, this is a buildsystem change. * PlatformEfl.cmake: Source/WebKit/efl: Instead of messing with BUILD_DATA_DIR and PARENT_SCOPE, just use THEME_BINARY_DIR as defined in OptionsEfl.cmake. * CMakeListsEfl.txt: Tools: Instead of adding -DDATA_DIR in all scopes, only define it in WebCore/PlatformEFL.cmake, as WebCore is the only place that uses it. We can then define DATA_DIR to the generated theme directory for EWebLauncher, since it's what we want here. * CMakeListsEfl.txt: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100722 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
commit-queue@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72684 Patch by Raphael Kubo da Costa <kubo@profusion.mobi> on 2011-11-17 Reviewed by Daniel Bates. This is the first step in fixing the build when a platform include (in this case, Valgrind 3.7.0 ships its config.h) and the wrong config.h is included instead of WebKit's. * DumpRenderTree/efl/CMakeLists.txt: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100720 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
rniwa@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72636 Reviewed by Eric Seidel. Make contributors_by_fuzz_match search string in shorthand names of contributors such as "Ryosuke N" and user names of email addreses such as rniwa in rniwa@webkit.org. Also import legacy contributor names and mispell lists from Eric's script on the bug 26533 as a test. * Scripts/webkitpy/common/config/committers.py: * Scripts/webkitpy/common/config/committers_unittest.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100700 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72678 Reviewed by Adam Barth. After talking to Dave Levin we've decided to remove this code. I know of no users of check-webkit-style w/o a webkit checkout and this needlessly adds complexity. * Scripts/webkitpy/style/main.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100699 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72677 Reviewed by Adam Barth. Remove more uses of os.path, including refactoring CommitterValidator to use a Host object instead of manual hacks to emulate what SCM and FileSystem provide. * Scripts/webkitpy/common/checkout/changelog.py: * Scripts/webkitpy/common/config/committervalidator.py: * Scripts/webkitpy/common/config/committervalidator_unittest.py: * Scripts/webkitpy/common/system/filesystem_mock.py: - Turns out none of our unittests actually looked at this path. now I've made it be a valid path. * Scripts/webkitpy/tool/bot/feeders.py: * Scripts/webkitpy/tool/commands/queues.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100697 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72673 Reviewed by Adam Barth. Unfortunately TextFileReader doesn't use FileSystem everywhere yet, so we can't move the unittests to using MockFileSystem, but we're close. * Scripts/webkitpy/style/filereader.py: * Scripts/webkitpy/style/filereader_unittest.py: * Scripts/webkitpy/style/main.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100694 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72670 Reviewed by Adam Barth. * Scripts/webkitpy/common/checkout/scm/detection.py: * Scripts/webkitpy/common/checkout/scm/scm_unittest.py: - We don't run these unittests, but this one was looking for the wrong exception. * Scripts/webkitpy/style/main.py: * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py: - This was create a new scm object even though it already had one? git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100689 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72664 Reviewed by Adam Barth. * Scripts/check-webkit-style: * Scripts/webkitpy/style/main.py: * Scripts/webkitpy/style/main_unittest.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100684 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
commit-queue@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72601 Re-enable ref tests. Run the driver under evenly numbered display if performing pixel tests or under oddly numbered display otherwise. This prevents the ref tests being run in an already used Xvfb instance if pixel tests are disabled. Patch by Zan Dobersek <zandobersek@gmail.com> on 2011-11-17 Reviewed by Tony Chang. * Scripts/webkitpy/layout_tests/port/gtk.py: Also set Xvfb variable to None after the process is killed git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100683 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
caio.oliveira@openbossa.org authored
https://bugs.webkit.org/show_bug.cgi?id=72620 Reviewed by David Levin. Qt's convention of keeping private (but exposed) API suffixed by "_p.h" but not adding the suffix the cpp file. One example of false positive https://bugs.webkit.org/show_bug.cgi?id=72319#c12. * Scripts/webkitpy/style/checkers/cpp.py: accept headers ending with _p as primary if its basename contains the cpp's basename. * Scripts/webkitpy/style/checkers/cpp_unittest.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100678 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72657 Reviewed by Adam Barth. * Scripts/check-webkit-style: * Scripts/webkitpy/style/main.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100675 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
eric@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=68691 Reviewed by Adam Barth. Make NRWT use a clean environment, just like ORWT did. * Scripts/webkitpy/layout_tests/port/base.py: * Scripts/webkitpy/layout_tests/port/chromium.py: * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: * Scripts/webkitpy/layout_tests/port/chromium_win.py: * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py: * Scripts/webkitpy/layout_tests/port/efl.py: * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py: * Scripts/webkitpy/layout_tests/port/mac.py: * Scripts/webkitpy/layout_tests/port/qt.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/win.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100674 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
dpranke@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72615 Reviewed by James Robinson. * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100673 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
hausmann@webkit.org authored
[Qt] Layer violation: qt_runtime.cpp accesses QWebElement and QTDRTNode https://bugs.webkit.org/show_bug.cgi?id=72595 Reviewed by Noam Rosenthal. Source/WebCore: Removed QWebElement and QtDRTNode usage that reached from WebCore into WebKit/qt and replaced it with the ability to register custom JSValue conversion functions. The old code has been moved to WebKit/qt. * bridge/qt/qt_instance.cpp: (JSC::Bindings::QtInstance::QtInstance): Remove unnecessary meta type registration (now done in QtWebElementRuntime::initialize in WebKit/qt). * bridge/qt/qt_runtime.cpp: (JSC::Bindings::registerCustomType): (JSC::Bindings::convertValueToQVariant): (JSC::Bindings::convertQVariantToValue): * bridge/qt/qt_runtime.h: Source/WebKit: * WebKit.pri: Add qwebelement_p.h to HEADERS. Source/WebKit/qt: Move QWebElement and QtDRTNode JS bindings from WebCore to WebKit/qt. * Api/qwebelement.cpp: Add QWebElement to JSValue and JSValue to QWebElement conversion through intermediate QVariant. (QtWebElementRuntime::create): (QtWebElementRuntime::get): (convertJSValueToWebElementVariant): (convertWebElementVariantToJSValue): (QtWebElementRuntime::initialize): Register conversion functions for QWebElement. * Api/qwebelement.h: * Api/qwebelement_p.h: Added. * Api/qwebpage.cpp: (QWebPagePrivate::QWebPagePrivate): Initialize QWebElement conversion functions for the JS runtime. * WebCoreSupport/DumpRenderTreeSupportQt.cpp: Add QtDRTNode to JSValue and JSValue to QtDRTNode conversion through intermediate QVariant. (QtDRTNodeRuntime::create): (QtDRTNodeRuntime::get): (convertJSValueToNodeVariant): (convertNodeVariantToJSValue): (QtDRTNodeRuntime::initialize): (DumpRenderTreeSupportQt::initialize): Register conversion functions for QtDRTNode. * WebCoreSupport/DumpRenderTreeSupportQt.h: Tools: * DumpRenderTree/qt/DumpRenderTreeQt.cpp: (WebCore::DumpRenderTree::DumpRenderTree): Use new initialize() function to ensure the registration of the DRTNode JS bindings. * DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::LayoutTestController): Remove unnecessary explicit qt meta type registration of QWebElement (now done implicitly through QWebPage -> QtWebElementRunTime::initialize code path) git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100658 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
Fixes <http://webkit.org/b/72642> REGRESSION (r100648): NRWT crashes on Chromium when handling a crashing test Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/chromium.py: (ChromiumDriver.run_test): Get the driver name from the port. * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: (ChromiumDriverTest.test_crashed_process_name): Simulate a crash and check that we got the right crashed process name. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100657 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
Fixes <http://webkit.org/b/72635> NRWT crashes when parsing leaks files from which certain leaks were excluded Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/leakdetector.py: (LeakDetector._parse_leaks_output): Extract just the number of excluded leaks from the regular expression match, not the entire matched string. * Scripts/webkitpy/layout_tests/port/leakdetector_unittest.py: (LeakDetectorTest): Added some example output that shows how things look when leaks are excluded. (LeakDetectorTest.test_parse_leaks_output): Test that we parse the number of exclusions correctly. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100655 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
Fixes <http://webkit.org/b/72526> REGRESSION (NRWT): WebKitTestRunner crash log gets saved when web process crashes, but WebProcess crash log should get saved instead Reviewed by Eric Seidel. * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py: (write_test_result): Simplified a little to share more code between the case where we crashed when running the test and where we crashed when checking the expected rendering for a reftest. Changed to pass the crashed process name down to write_crash_report. (TestResultWriter.write_crash_report): Added crashed_process_name parameter, which we pass along to CrashLogs.find_newest_log rather than always using the driver name. * Scripts/webkitpy/layout_tests/port/chromium.py: (ChromiumDriver.run_test): Pass the driver name as the name of the crashed process if we crashed. * Scripts/webkitpy/layout_tests/port/test.py: (TestInstance.__init__): Initialize new web_process_crash attribute. (unit_test_list): Added new web-process-crash-with-stderr test, which is used by the new test this patch adds. (TestDriver.run_test): Pass a crashed process name to DriverOutput based on whether the test specifies that the driver crashed or the web process crashed. * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: (MainTest.test_web_process_crash_log): Added. Similar to test_crash_log, but uses a test for which we pretend the web process crashed rather than the driver process. (MainTest.test_run_singly_actually_runs_tests): (MainTest.test_unexpected_failures): (EndToEndTest.test_end_to_end): Updated magic numbers to account for new unexpected failing test added to unit_test_list. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100648 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
tony@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72629 Reviewed by Ojan Vafai. We toggle $blobSupport on if $mediaStreamSupport is on. Since $mediaStreamSupport is on by default for chromium, make the default for $blobSupport also on. * Scripts/build-webkit: git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100646 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
Fixes <http://webkit.org/b/72623> REGRESSION (NRWT): run-webkit-tests --leaks reports leaks that are supposed to be ignored Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/leakdetector.py: (LeakDetector._leaks_args): Removed extra quoting of call stacks and types. Quoting is only required when passing arguments via the shell (which we aren't doing here). It is not required by Perl's option-parsing code (as the comment I removed mistakenly claimed). * Scripts/webkitpy/layout_tests/port/leakdetector_unittest.py: (LeakDetectorTest.test_leaks_args): Updated expectations. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100644 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
philn@webkit.org authored
* Scripts/run-bindings-tests: Use scm.detection module. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100636 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
jochen@chromium.org authored
https://bugs.webkit.org/show_bug.cgi?id=72616 Reviewed by Tony Gentilcore. * DumpRenderTree/mac/AccessibilityUIElementMac.mm: (AccessibilityUIElement::verticalScrollbar): git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100625 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
aroben@apple.com authored
Fixes <http://webkit.org/b/72609> Lion Leaks bot complains about a bunch of leaks that aren't WebKit's fault Reviewed by Antti Koivisto. * Scripts/old-run-webkit-tests: (countAndPrintLeaks): * Scripts/webkitpy/layout_tests/port/leakdetector.py: (LeakDetector._callstacks_to_exclude_from_leaks): Added some call stacks to exclude on Lion that represent leaks in lower-level frameworks. * Scripts/webkitpy/layout_tests/port/mac.py: (MacPort.is_lion): Added. * Scripts/webkitpy/layout_tests/port/mac_unittest.py: (MacPortTest.test_is_version_methods): Added tests for the is_leopard/is_snowleopard/is_lion methods. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100618 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
philn@webkit.org authored
https://bugs.webkit.org/show_bug.cgi?id=72599 Reviewed by Xan Lopez. * Scripts/webkitpy/layout_tests/port/gtk.py: Enforce disabled ref tests to avoid massive flakiness. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100605 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-
vestbo@webkit.org authored
A new extension object has been added to QQuickWebView (the same approach should be used for other API classes that need experimental APIs). The QML extension mechanism is then built on top of the experimental object. https://bugs.webkit.org/show_bug.cgi?id=72522 Reviewed by Simon Hausmann. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@100602 268f45cc-cd09-0410-ab3c-d52691b4dbfc
-