2011-11-03 Carlos Garcia Campos [GTK] Show url of history items in a status bar in MiniBrowser https://bugs.webkit.org/show_bug.cgi?id=71474 Reviewed by Martin Robinson. * MiniBrowser/gtk/BrowserWindow.c: (browserWindowSetStatusText): Set status text and show/hide the status label. (resetStatusText): Reset status text when history menu is hidden. (browserWindowHistoryItemSelected): Show url of currently selected history item. (browserWindowCreateBackForwardMenu): Connect to hide signal of menu to reset the status text. (browserWindowConstructed): Use GtkOverlay if available to show status text. 2011-11-03 Carlos Garcia Campos [GTK] Add back/forward menus to MiniBrowser using WebKit2 GTK+ API https://bugs.webkit.org/show_bug.cgi?id=71466 Reviewed by Martin Robinson. * MiniBrowser/gtk/BrowserWindow.c: (browserWindowHistoryItemActivated): Go to selected back formard list item using webkit_web_view_go_to_back_forward_list_item() (browserWindowCreateBackForwardMenu): Create a GtkMenu for the given GList of WebKitBackForwardListItems. (browserWindowUpdateNavigationActions): Enable/disable back and forward buttons depending on whether it's possible to go back/forward. Create a menu for back and forward toolbar buttons. (backForwadlistChanged): Call browserWindowUpdateNavigationActions(). (browserWindowConstructed): Connect to WebKitBackForwardList changed signal. 2011-11-03 Simon Hausmann [Qt] Remove Maemo specific code paths https://bugs.webkit.org/show_bug.cgi?id=71476 Reviewed by Kenneth Rohde Christiansen. * QtTestBrowser/launcherwindow.cpp: (LauncherWindow::~LauncherWindow): (LauncherWindow::init): (LauncherWindow::showLinkHover): (LauncherWindow::selectElements): (LauncherWindow::showFPS): (LauncherWindow::updateFPS): * QtTestBrowser/launcherwindow.h: (WindowOptions::WindowOptions): * QtTestBrowser/locationedit.cpp: * QtTestBrowser/locationedit.h: 2011-11-03 Carlos Garcia Campos [GTK] Show load progress information in MiniBrowser using WebKit2 GTK+ API https://bugs.webkit.org/show_bug.cgi?id=71461 Reviewed by Martin Robinson. * MiniBrowser/gtk/BrowserWindow.c: (resetEntryProgress): Reset the entry progress after a while when load has been completed. (webViewLoadProgressChanged): Update location entry progress. (browserWindowConstructed): Connect to WebView notify::estimated-load-progress signal. 2011-11-03 Carlos Garcia Campos [GTK] Remove WebKit2 C API from MiniBrowser https://bugs.webkit.org/show_bug.cgi?id=71459 Reviewed by Martin Robinson. Use the GTK+ API instead to add minimum functionality. The other features will be ported to GTK+ API in following patches. * MiniBrowser/gtk/BrowserWindow.c: (activateUriEntryCallback): Use webkit_web_view_load_uri(). (goBackCallback): Use webkit_web_view_go_back(). (goForwardCallback): Use webkit_web_view_go_forward(). (webViewURIChanged): Update location entry with current uri using webkit_web_view_get_uri(). (browserWindowFinalize): (browserWindowGetProperty): (browserWindowSetProperty): (browser_window_init): (browserWindowConstructed): Connect to notify::uri signal of WebView to be notified when the URI changes. (browser_window_class_init): (browser_window_new): Use WebKitWebView. (browser_window_get_view): Use WebKitWebView * MiniBrowser/gtk/BrowserWindow.h: * MiniBrowser/gtk/GNUmakefile.am: * MiniBrowser/gtk/WebBundle/WebBundleMain.c: Removed. * MiniBrowser/gtk/main.c: (loadURI): Use webkit_web_view_new(). (main): 2011-11-03 Simon Hausmann [Qt] Unable to start MiniBrowser after run-webkit-tests https://bugs.webkit.org/show_bug.cgi?id=71469 Reviewed by Csaba Osztrogonác. * Scripts/run-launcher: Pass @ARGV instead of @args, which is stripped from things like -2. Neither Gtk or Qt MiniBrowser like unknown command line arguments. 2011-11-03 Philippe Normand [GTK] [WK2] ttf-liberation fonts moved to a new location (in Debian) https://bugs.webkit.org/show_bug.cgi?id=71445 Reviewed by Martin Robinson. * DumpRenderTree/gtk/DumpRenderTree.cpp: (initializeFonts): set directoriesDescription only when needed. * WebKitTestRunner/InjectedBundle/gtk/ActivateFontsGtk.cpp: (WTR::inititializeFontConfigSetting): Add the new font path for Debian fonts and refactored the font files loading code to avoid copy/pastes, ease maintenance and future font directories additions. 2011-11-03 Carlos Garcia Campos [GTK] Remove GtkLauncher2 https://bugs.webkit.org/show_bug.cgi?id=71449 Reviewed by Martin Robinson. GtkLauncher2 is the GtkLauncher code compiled with webkit2. This made sense when we wanted to implement the exactly same wk1 API for wk2. Now the code would require an ifdef for most the api, so it's better to remove GtkLauncher2 and port MiniBrowser to the new GTK+ API. * GNUmakefile.am: * GtkLauncher/main.c: (notifyProgressCb): (closeWebViewCb): (createBrowser): (addWebSettingsGroupToContext): (main): 2011-11-02 Philippe Normand [GTK] [DRT] ttf-liberation fonts moved to a new location (in Debian) https://bugs.webkit.org/show_bug.cgi?id=71359 Reviewed by Martin Robinson. * DumpRenderTree/gtk/DumpRenderTree.cpp: (initializeFonts): Add the new font path for Debian fonts and refactored the font files loading code to avoid copy/pastes, ease maintenance and future font directories additions. 2011-11-02 Eric Seidel Move Mocks into _mock files near their implementations https://bugs.webkit.org/show_bug.cgi?id=71425 Reviewed by Adam Barth. This entire change is splitting mocktool.py into separate _mock.py files and placing them next to their real implementations. I also deleted MockPortFactory (since it was wrong) and was just hiding the fact that the "skipped-ports" command was broken (and has been for a long time). So I made MockHost use a real PortFactory (passed a MockHost) and changed the skipped-ports implementation to use modern PortFactory methods. * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py: * Scripts/webkitpy/common/checkout/checkout_mock.py: Added. * Scripts/webkitpy/common/checkout/checkout_unittest.py: * Scripts/webkitpy/common/checkout/deps_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/__init__.py. * Scripts/webkitpy/common/checkout/scm/scm_mock.py: Added. * Scripts/webkitpy/common/checkout/scm/scm_unittest.py: * Scripts/webkitpy/common/config/ports_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/gtk_unittest.py. * Scripts/webkitpy/common/host.py: * Scripts/webkitpy/common/host_mock.py: Added. * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py: Added. * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py: * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py: Added. * Scripts/webkitpy/common/net/credentials_unittest.py: * Scripts/webkitpy/common/net/failuremap_unittest.py: * Scripts/webkitpy/common/net/irc/irc_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/__init__.py. * Scripts/webkitpy/common/net/statusserver_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/gtk_unittest.py. * Scripts/webkitpy/common/net/statusserver_unittest.py: * Scripts/webkitpy/common/net/web_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/gtk_unittest.py. * Scripts/webkitpy/common/system/executive.py: * Scripts/webkitpy/common/system/executive_mock.py: * Scripts/webkitpy/common/system/platforminfo_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/__init__.py. * Scripts/webkitpy/common/system/user_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/gtk_unittest.py. * Scripts/webkitpy/common/system/workspace_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/__init__.py. * Scripts/webkitpy/common/system/workspace_unittest.py: * Scripts/webkitpy/common/watchlist/watchlist_mock.py: Copied from Tools/Scripts/webkitpy/layout_tests/port/__init__.py. * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py: * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py: * Scripts/webkitpy/layout_tests/port/__init__.py: * Scripts/webkitpy/layout_tests/port/base_unittest.py: * Scripts/webkitpy/layout_tests/port/efl_unittest.py: * Scripts/webkitpy/layout_tests/port/factory.py: * Scripts/webkitpy/layout_tests/port/factory_unittest.py: * Scripts/webkitpy/layout_tests/port/gtk_unittest.py: * Scripts/webkitpy/layout_tests/port/http_lock_unittest.py: * Scripts/webkitpy/layout_tests/port/leakdetector_unittest.py: * Scripts/webkitpy/layout_tests/port/mac_unittest.py: * Scripts/webkitpy/layout_tests/port/port_testcase.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: * Scripts/webkitpy/layout_tests/port/win_unittest.py: * Scripts/webkitpy/test/main.py: * Scripts/webkitpy/tool/bot/botinfo_unittest.py: * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: * Scripts/webkitpy/tool/commands/queries_unittest.py: * Scripts/webkitpy/tool/commands/rebaseline_unittest.py: * Scripts/webkitpy/tool/mocktool.py: * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: 2011-11-02 Dan Bernstein Fixed the build and removed another copy to ColorBits.ttf. * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: * WebKitTestRunner/fonts/ColorBits.ttf: Removed. 2011-11-02 Dan Bernstein REGRESSION (r98759-r98773): platform/mac/fonts/color-bitmap.html failing on Lion Intel Debug (Tests) https://bugs.webkit.org/show_bug.cgi?id=71424 Reviewed by Sam Weinig. Removed the ColorBits font and the code to register it in the process running the tests. The font was only used by the aforementioned test, which is now being removed. * DumpRenderTree/fonts/ColorBits-A.png: Removed. * DumpRenderTree/fonts/ColorBits.ttf: Removed. * DumpRenderTree/mac/DumpRenderTree.mm: (allowedFontFamilySet): (activateTestingFonts): * WebKitTestRunner/InjectedBundle/gtk/ActivateFontsGtk.cpp: (WTR::inititializeFontConfigSetting): * WebKitTestRunner/InjectedBundle/mac/ActivateFonts.mm: (WTR::allowedFontFamilySet): (WTR::activateFonts): 2011-11-02 Dan Bernstein Fix an intermittent hang in RenderedImageFromDOMRange. Reviewed by Anders Carlsson. * TestWebKitAPI/mac/PlatformUtilitiesMac.mm: (TestWebKitAPI::Util::run): Made the call to -runMode:beforeDate: return immediately if there are no sources by passing a date in the distant past. 2011-11-02 Adam Roben Add git-add-reviewer I've been using this script for a while to add reviewers to git commits in my local repository before pushing them to the Subversion repository. Basic usage is: git-add-reviewer HEAD^ "Joe Schmoe" or git-add-reviewer -i trunk Fixes Would like an easy way to add reviewer names to existing git commits Reviewed by Sam Weinig. * Scripts/git-add-reviewer: Added. (top level): Parse options, check that the working tree is clean, then run in either interactive or non-interactive mode as appropriate. (interactive): Write out a temp file that contains each commit with "NOBODY" next to it and open it in the user's editor. The user can replace "NOBODY" with reviewer names. When the editor exits, we check out the commit just before the earliest one we're modifying. Then for each commit, we check it out, update the reviewer, and amend the commit to contain the new reviewer. Finally, we update the branch that was originally checked out to point to the last amended commit. (nonInteractive): Check out the commit to modify, write the current commit message to the MERGE_MSG file so addReviewer will update it, add the reviewer to the ChangeLog(s), amend the commit, and rebase the original branch on top of the amended commit. (usage): Print out a usage message and exit. (requireCleanWorkTree): Do the same steps git-rebase does to ensure that the working tree and index are clean. (fail): Print out the specified error message, if any, and return false. This function is used as a return value in various error cases throughout the script. (cherryPick): Cherry-pick the specified commit and return 1 if we succeeded. (addReviewer): Add the specified reviewer to all the ChangeLogs modified in this commit, and to .git/MERGE_MSG, which will be used as the commit message for the next commit. (commit): Call git-commit and use .git/MERGE_MSG as the commit message. (addReviewerToChangeLog): (addReviewerToCommitMessage): These just call through to addReviewerToFile. (addReviewerToFile): Read in the file, replacing any "NOBODY" text with the reviewer name(s) and replacing "Reviewed" with "Rubber-stamped" if specified. Write the resulting text to a temp file, then move the temp file over the original file and stage it for the next commit. (head): Returns the name of the currently-checked-out branch. (isAncestor): Returns true if the first commit is an ancestor of the second. (toCommit): Converts a commitish to a commit ID using git-rev-parse. (changeLogsForCommit): Returns a list of all the ChangeLogs modified in the given commit. (resetToCommit): Checks out the given commit, throwing away any local changes. (writeCommitMessageToFile): Gets the commit message for the current commit and writes it to the specified file. (rebaseOntoHead): Rebases the specified branch onto HEAD. (checkout): Checks out the given commit. (getConfigValue): Reads the specified config variable from git-config. 2011-11-02 Dean Jackson Add ENABLE_CSS_SHADERS flag https://bugs.webkit.org/show_bug.cgi?id=71394 Reviewed by Sam Weinig. * Scripts/build-webkit: 2011-10-31 Eric Seidel BaselineOptimizer tests should use mocks instead of real Executive/FileSystem objects https://bugs.webkit.org/show_bug.cgi?id=71237 Reviewed by Adam Barth. Calling the static version of factory.get() with proper mocking requires passsing an explict filesystem, executive, etc. So instead, we use a PortFactory instance and pass it a Host pointer. I had to add a MockHost since we'd not needed a non-host tool before now. * Scripts/webkitpy/common/checkout/baselineoptimizer.py: * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py: * Scripts/webkitpy/tool/commands/rebaseline.py: * Scripts/webkitpy/tool/mocktool.py: 2011-11-02 Anders Carlsson Update for the WebKit2 API fixage. * MiniBrowser/mac/BrowserWindowController.m: (-[BrowserWindowController awakeFromNib]): * WebKitTestRunner/TestController.cpp: (WTR::TestController::initialize): 2011-11-02 Anders Carlsson Fix build. Add new functions to the page loader client. * MiniBrowser/mac/BrowserWindowController.m: (-[BrowserWindowController awakeFromNib]): * WebKitTestRunner/TestController.cpp: (WTR::TestController::initialize): 2011-11-02 Dan Bernstein Add API to get rendered text image without having to select it https://bugs.webkit.org/show_bug.cgi?id=71407 Reviewed by Simon Fraser. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/mac/RenderedImageFromDOMRange.mm: Added. (-[RenderedImageFromDOMRangeFrameLoadDelegate webView:didFinishLoadForFrame:]): (TestWebKitAPI::TEST): 2011-11-02 Benjamin Poulain Make the main frame's base path more explicit in _drt_descriptionSuitableForTestResult https://bugs.webkit.org/show_bug.cgi?id=71351 Reviewed by Darin Adler. Add the delimiter "/" at the end of the path instead of adding +1 when using the path. * DumpRenderTree/mac/ResourceLoadDelegate.mm: (-[NSURL _drt_descriptionSuitableForTestResult]): 2011-11-02 Tom Sepez XSSAuditor is silent https://bugs.webkit.org/show_bug.cgi?id=70973 Reviewed by Adam Barth. * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::didDetectXSS): * DumpRenderTree/chromium/WebViewHost.h: * DumpRenderTree/mac/FrameLoadDelegate.mm: (-[FrameLoadDelegate webView:didDetectXSS:]): * MiniBrowser/mac/BrowserWindowController.m: (didDetectXSSForFrame): (-[BrowserWindowController awakeFromNib]): * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::InjectedBundlePage::InjectedBundlePage): (WTR::InjectedBundlePage::didDetectXSSForFrame): * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h: 2011-11-02 Simon Hausmann [Qt] REGRESSION: 5 editing tests fail https://bugs.webkit.org/show_bug.cgi?id=71117 Reviewed by Andreas Kling. Make sure that the view is properly focused, in order to get the didBeginEditing DRT calls. This requires activating the window (the call will result in a FocusIn event to the window as well as setting QGuiApplication::focusWindow()) as well as enabling the focus on the view item. The signature of setFocus has changed to take a boolean. * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::WrapperWindow::handleStatusChanged): 2011-11-02 Matt Falkenhagen [chromium] Add per-script font preferences support in overridePreference https://bugs.webkit.org/show_bug.cgi?id=71110 This allows a layout test to do something like: // Set standard font for Arabic to Ahem. overridePreference("WebKitStandardFontMap", [ "Arab", "Ahem" ]); This should make writing a layout test for per-script fonts easier. This change only affects the Chromium DumpRenderTree, but the other platforms should probably also have a similar change eventually. Reviewed by Tony Chang. * DumpRenderTree/chromium/LayoutTestController.cpp: add per-script font preferences to overridePreference (LayoutTestController::cppVariantToWebStringArray): (setFontMap): (LayoutTestController::overridePreference): * DumpRenderTree/chromium/LayoutTestController.h: * DumpRenderTree/chromium/WebPreferences.cpp: (setStandardFontFamilyWrapper): (setFixedFontFamilyWrapper): (setSerifFontFamilyWrapper): (setSansSerifFontFamilyWrapper): (setCursiveFontFamilyWrapper): (setFantasyFontFamilyWrapper): (applyFontMap): (WebPreferences::applyTo): * DumpRenderTree/chromium/WebPreferences.h: add per-script font preference members (WebPreferences::UScriptCodeHashTraits::emptyValue): (WebPreferences::UScriptCodeHashTraits::constructDeletedValue): (WebPreferences::UScriptCodeHashTraits::isDeletedValue): 2011-11-02 Elliot Poger until use_skia=1 by default on mac, default port should be chromium-cg-mac https://bugs.webkit.org/show_bug.cgi?id=71312 Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: * Scripts/webkitpy/layout_tests/port/chromium_mac.py: * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py: 2011-11-02 Stephen Chenney printing layout test failures after r98634 https://bugs.webkit.org/show_bug.cgi?id=71080 Reviewed by Kent Tamura * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::paintPagesWithBoundaries): Added transforms invert the CG Bitmap so the coord system matches that for Skia. 2011-11-01 Sam Weinig Can't complete ChangeLog when using resolve-Changelogs and webkit-tools-completion.sh https://bugs.webkit.org/show_bug.cgi?id=71346 Reviewed by Dan Bernstein. * Scripts/webkit-tools-completion.sh: Add "-o default" which allows for completing files in addition to the word list. 2011-11-02 Simon Hausmann [Qt][WK2] Add support for touch event testing to WebKitTestRunner https://bugs.webkit.org/show_bug.cgi?id=71310 Reviewed by Kenneth Rohde Christiansen. * WebKitTestRunner/DerivedSources.pro: Build system foo to ensure the ENABLE_TOUCH_EVENTS define works in EventSendingController.idl * WebKitTestRunner/EventSenderProxy.h: Add declarations for Qt implementations of touch handlers. * WebKitTestRunner/InjectedBundle/Bindings/EventSendingController.idl: Add touch functions required by tests. * WebKitTestRunner/InjectedBundle/EventSendingController.cpp: Forward touch messages to the ui process. (WTR::EventSendingController::addTouchPoint): (WTR::EventSendingController::updateTouchPoint): (WTR::EventSendingController::setTouchModifier): (WTR::EventSendingController::touchStart): (WTR::EventSendingController::touchMove): (WTR::EventSendingController::touchEnd): (WTR::EventSendingController::clearTouchPoints): (WTR::EventSendingController::releaseTouchPoint): * WebKitTestRunner/InjectedBundle/EventSendingController.h: * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: Include features.pri for feature defines. * WebKitTestRunner/TestController.cpp: (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle): Forward incoming touch messages to the EventSenderProxy. * WebKitTestRunner/qt/EventSenderProxyQt.cpp: Implement touch test harness functions. (WTR::EventSenderProxy::EventSenderProxy): (WTR::EventSenderProxy::addTouchPoint): (WTR::EventSenderProxy::updateTouchPoint): (WTR::EventSenderProxy::setTouchModifier): (WTR::EventSenderProxy::touchStart): (WTR::EventSenderProxy::touchMove): (WTR::EventSenderProxy::touchEnd): (WTR::EventSenderProxy::clearTouchPoints): (WTR::EventSenderProxy::releaseTouchPoint): (WTR::EventSenderProxy::sendTouchEvent): * WebKitTestRunner/qt/WebKitTestRunner.pro: Include features.pri for feature defines. 2011-11-01 Sam Weinig resolve-ChangeLogs --help should be faster. Reviewed by Adam Roben. * Scripts/resolve-ChangeLogs: (usageAndExit): Add a call to usageAndExit() before doing work to find unmerged changelogs if --help or something illegal is provided on the command line. 2011-11-01 Benjamin Poulain [Mac] _drt_descriptionSuitableForTestResult generate an exception for absolute paths above the main frame https://bugs.webkit.org/show_bug.cgi?id=71342 Reviewed by Darin Adler. [NSURL _drt_descriptionSuitableForTestResult] was generating a NSRangeException due to the call to [NSString substringFromIndex:] with an index out of the string range. The source of this path is the test-loading-archive-subresource-null-mimetype.html which load a subresource from the root of the disk. The problem was hidden because the exception are catched when calling the delegate. This patch changes _drt_descriptionSuitableForTestResult to return absolute string if the resource is not in a subdirectory of the main frame. * DumpRenderTree/mac/ResourceLoadDelegate.mm: (-[NSURL _drt_descriptionSuitableForTestResult]): 2011-11-01 Alok Priyadarshi [chromium] --enable-hardware-gpu flag is not recognized by run-webkit-tests https://bugs.webkit.org/show_bug.cgi?id=71318 Reviewed by James Robinson. Fixed a typo for --enable-hardware-gpu flag. * Scripts/webkitpy/layout_tests/port/chromium.py: 2011-11-01 Elliot Poger chromium-gpu port should default to chromium-gpu-mac, not chromium-gpu-cg-mac https://bugs.webkit.org/show_bug.cgi?id=71286 Reviewed by Stephen White. * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: 2011-11-01 Simon Hausmann [Qt][WK2] Fix build of WebKitTestRunner with ELF symbol visibility https://bugs.webkit.org/show_bug.cgi?id=71299 Reviewed by Andreas Kling. * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: Use the properly exported symbol for the Qt build to enable WebCoreTestSupport::injectInternalsObject. (WTR::InjectedBundlePage::didClearWindowForFrame): * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: Fix include path to DRTSupportQt 2011-11-01 Jocelyn Turcotte [Qt] Merge common desktop and touch view APIs under a base class. https://bugs.webkit.org/show_bug.cgi?id=71280 Reviewed by Andreas Kling. Remove unneded forwarding logic for TouchWebView since those properties are now in the view just as the desktop web view. * MiniBrowser/qt/qml/TouchView.qml: 2011-11-01 Jocelyn Turcotte check-webkit-style: Allow names starting with "_q_". https://bugs.webkit.org/show_bug.cgi?id=70625 Reviewed by Tor Arne Vestbø. Used by the Qt port as the standard prefix for private slots. * Scripts/webkitpy/style/checkers/cpp.py: * Scripts/webkitpy/style/checkers/cpp_unittest.py: 2011-11-01 Simon Hausmann [Qt][WK2] Fix valgrind error about uninitialized variable https://bugs.webkit.org/show_bug.cgi?id=71273 Reviewed by Kenneth Christiansen. * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::notify): Initialize isPrimary. 2011-10-31 Adam Roben Make Leaks Viewer less strict when parsing leaks-related output from NRWT/ORWT Fixes REGRESSION (r98639): Leaks Viewer doesn't show recent leaky builds Reviewed by Anders Carlsson. * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/RecentBuildsLoader.js: (RecentBuildsLoader.prototype.start): Relax a regex to gloss over differences in ORWT's vs. NRWT's leaks-related output (i.e., "M total leaks found!" vs. "M total leaks found for a total of N bytes!"). 2011-10-31 Raphael Kubo da Costa [EFL] Rename ewk_view_setting_scripts_window_open_{get,set} after r93833 https://bugs.webkit.org/show_bug.cgi?id=71228 Reviewed by Antonio Gomes. * DumpRenderTree/efl/DumpRenderTreeChrome.cpp: (DumpRenderTreeChrome::createNewWindow): (DumpRenderTreeChrome::resetDefaultsToConsistentValues): 2011-10-31 Eric Seidel webkitpy tests depend too much on the user's environment https://bugs.webkit.org/show_bug.cgi?id=71234 Reviewed by Dirk Pranke. This change just makes a bunch of our older tests use more modern mocking to avoid trying to launch processes or read from the user's filesystem during unittesting. I found many of these by adding an assert in Executive.run_command that we were not unittesting. I can't add that assert always as there are some valid uses of Executive during unittesting. Once I fix more of these, I may find a way to add such an assert conditionally. * Scripts/webkitpy/common/checkout/baselineoptimizer.py: * Scripts/webkitpy/common/net/credentials_unittest.py: * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py: * Scripts/webkitpy/layout_tests/port/base.py: * Scripts/webkitpy/layout_tests/port/base_unittest.py: 2011-10-31 Kenneth Rohde Christiansen [Qt] MiniBrowser doesn't resize as the size is always overridden https://bugs.webkit.org/show_bug.cgi?id=71212 Reviewed by Simon Hausmann. Remove call overriden the size. * MiniBrowser/qt/main.cpp: (main): 2011-10-30 Sheriff Bot Unreviewed, rolling out r98819, r98823, r98825, and r98830. http://trac.webkit.org/changeset/98819 http://trac.webkit.org/changeset/98823 http://trac.webkit.org/changeset/98825 http://trac.webkit.org/changeset/98830 https://bugs.webkit.org/show_bug.cgi?id=71199 Broke Chromium Windows bot (Requested by rniwa on #webkit). * 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: 2011-10-30 Eric Seidel new-run-webkit-tests is locale dependent https://bugs.webkit.org/show_bug.cgi?id=68691 Unreviewed. This is a temporary hack until someone from the cr-win port can take a look and determine which environment variable lighttpd is missing. * Scripts/webkitpy/layout_tests/port/chromium_win.py: 2011-10-30 Eric Seidel new-run-webkit-tests is locale dependent https://bugs.webkit.org/show_bug.cgi?id=68691 Unreviewed. I would have preferred to have this reviewed, but relevant reviewers are asleep and bots are broken. This was a regression from moving to a clean environment. ChromiumWin (and possibly other ports), need the "PATH" environment copied over. This wasn't caught in my testing because although we had unittests to cover this, they weren't being run on anything but windows. The vast majority of this change is just fixing the unittests to use our modern MockFileSystem/MockUser/MockExecutive so they can be run on any system (and removing the platform checks from the unittests so they are run everywhere). The actual fix is the single line "PATH" string added to base.py. The rest of this change is just fixing the Chromium port unittests to run on all systems (including changing the Chromium port to use FileSystem.path_to_module instead of __file__). * 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/google_chrome_unittest.py: 2011-10-29 Eric Seidel new-run-webkit-tests is locale dependent https://bugs.webkit.org/show_bug.cgi?id=68691 Unreviewed. Fix the Qt port and unittest the function so we don't break it again. * Scripts/webkitpy/layout_tests/port/qt.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: 2011-10-29 Eric Seidel new-run-webkit-tests is locale dependent https://bugs.webkit.org/show_bug.cgi?id=68691 Reviewed by Adam Barth. Make NRWT use a clean environment just like ORWT did. I've manually validated that NRWT now copies over the same environment variables that ORWT did. This new code is exercised by existing setup_environ_for_server tests. * Scripts/webkitpy/layout_tests/port/base.py: * Scripts/webkitpy/layout_tests/port/efl.py: * Scripts/webkitpy/layout_tests/port/mac.py: * Scripts/webkitpy/layout_tests/port/qt.py: * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/win.py: 2011-10-29 Eric Seidel new-run-webkit-tests: MASTER BUG: Switch all webkit.org bots over https://bugs.webkit.org/show_bug.cgi?id=34984 Reviewed by Adam Barth. Move run-webkit-tests from using a white-list of supported NRWT ports to a blacklist of ports requiring ORWT. I've also added the ability to opt-in to parallel testing on a per-port basis. With these changes I'm going to officially declare this bug dead and work on remaining polish issues and individual developer complaints. * Scripts/run-webkit-tests: (useNewRunWebKitTests): (platformIsReadyForParallelTesting): 2011-10-29 Sheriff Bot Unreviewed, rolling out r98780. http://trac.webkit.org/changeset/98780 https://bugs.webkit.org/show_bug.cgi?id=71173 Broke Qt bots (Requested by rniwa on #webkit). * Scripts/run-webkit-tests: (useNewRunWebKitTests): 2011-10-28 Rafael Weinstein webkit.py gdb visualizer is broken after change to StringImpl https://bugs.webkit.org/show_bug.cgi?id=71154 Reviewed by Tony Chang. r98624 changed StringImpl m_data to m_data32. This changes updates webkit.py to match. * gdb/webkit.py: 2011-10-28 Joseph Pecoraro Fix typo in check-for-inappropriate-objc-class-names https://bugs.webkit.org/show_bug.cgi?id=71144 Reviewed by Dan Bernstein. * Scripts/check-for-inappropriate-objc-class-names: 2011-10-28 Sadrul Habib Chowdhury Add support for sending scroll-update events from EventSender. https://bugs.webkit.org/show_bug.cgi?id=66272 Reviewed by Adam Barth. * DumpRenderTree/chromium/EventSender.cpp: (EventSender::EventSender): (EventSender::reset): (EventSender::gestureScrollUpdate): (EventSender::gestureEvent): * DumpRenderTree/chromium/EventSender.h: 2011-10-28 Eric Seidel new-run-webkit-tests: MASTER BUG: Switch all webkit.org bots over https://bugs.webkit.org/show_bug.cgi?id=34984 Reviewed by Adam Barth. Move run-webkit-tests from using a white-list of supported NRWT ports to a blacklist of ports requiring ORWT. I've also added the ability to opt-in to parallel testing on a per-port basis. With these changes I'm going to officially declare this bug dead and work on remaining polish issues and individual developer complaints. * Scripts/run-webkit-tests: (useNewRunWebKitTests): (platformIsReadyForParallelTesting): 2011-10-28 Eric Seidel NRWT crashes when parsing leaks output due to UTF-8 decoding error https://bugs.webkit.org/show_bug.cgi?id=71112 Reviewed by Adam Barth. Use read_binary_file instead of read_text_file and add a test case with some non-utf8 data to make sure we don't break this in the future. * Scripts/webkitpy/layout_tests/port/leakdetector.py: * Scripts/webkitpy/layout_tests/port/leakdetector_unittest.py: 2011-10-28 David Levin watchlist should not CC the bug reporter https://bugs.webkit.org/show_bug.cgi?id=71079 Reviewed by Adam Barth. Tested by test_apply_watch_list_local (webkitpy.tool.steps.applywatchlist_unittest.ApplyWatchListTest) which remains upchanged and thus verifies that the reporter doesn't get added to the cc line. * Scripts/webkitpy/tool/commands/applywatchlistlocal_unittest.py: Add the new email in the output. * Scripts/webkitpy/tool/mocktool.py: Change the reporter and make them one of the people to cc. * Scripts/webkitpy/tool/steps/applywatchlist.py: Remove the reporter from the cc list. 2011-10-28 Zoltan Horvath Add watchlist for QtWebKit2 API and for WK2's Qt specific files. https://bugs.webkit.org/show_bug.cgi?id=71116 Reviewed by David Levin. * Scripts/webkitpy/common/config/watchlist: 2011-10-28 Simon Hausmann [Qt] Adapt to QSG* to QQuick* API change in qtdeclarative https://bugs.webkit.org/show_bug.cgi?id=70494 Reviewed by Kenneth Christiansen. Use QQuick* instead of QSG* as well as QStandardPaths instead of QDesktopLocation. * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): * MiniBrowser/qt/BrowserWindow.h: * QtTestBrowser/cookiejar.cpp: (TestBrowserCookieJar::TestBrowserCookieJar): * QtTestBrowser/launcherwindow.cpp: (LauncherWindow::setDiskCache): * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::WrapperWindow::WrapperWindow): (WTR::WrapperWindow::handleStatusChanged): 2011-10-28 Jocelyn Turcotte [Qt] Fix the Ctrl behavior for touch mocking in MiniBrowser. https://bugs.webkit.org/show_bug.cgi?id=71106 Reviewed by Simon Hausmann. Holding Ctrl allows multiple touch points to be held on the screen using mouse buttons. It would previously only prevent TouchReleased to be sent and would require another MouseButtonRelease to be sent without holding Ctrl. This patch makes sure that all held touch points are released when Ctrl is released if MouseButtonRelease was received. It also removes the touch QEvent::Type logic since it's currently handled by QtGui by observing the modified touch points. * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::notify): (MiniBrowserApplication::sendTouchEvent): * MiniBrowser/qt/MiniBrowserApplication.h: 2011-10-28 Kenneth Rohde Christiansen [Qt] MiniBrowser needs a -window-size option https://bugs.webkit.org/show_bug.cgi?id=70999 Reviewed by Simon Hausmann. Add an -window-size option useful for testing the viewport handling. Also clean up the option handling. * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::handleUserOptions): * MiniBrowser/qt/MiniBrowserApplication.h: (WindowOptions::WindowOptions): (WindowOptions::setRequestedWindowSize): (WindowOptions::requestedWindowSize): * MiniBrowser/qt/utils.cpp: (takeOptionValue): * MiniBrowser/qt/utils.h: 2011-10-28 Jochen Eisinger Add allowScriptFromSource callback to FrameLoaderClient https://bugs.webkit.org/show_bug.cgi?id=71013 Reviewed by Darin Fisher. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (LayoutTestController::setScriptsAllowed): * DumpRenderTree/chromium/LayoutTestController.h: * DumpRenderTree/chromium/WebPermissions.cpp: (WebPermissions::allowScriptFromSource): (WebPermissions::setScriptsAllowed): (WebPermissions::reset): * DumpRenderTree/chromium/WebPermissions.h: 2011-10-28 Alexandru Chiculita Added myself as a committer. * Scripts/webkitpy/common/config/committers.py: 2011-10-27 Adam Barth Remove WinCairo from the set of core builders. WinCairo rarely builds and is never green. Rubber-stamped by Eric Seidel. * Scripts/webkitpy/common/net/buildbot/buildbot.py: * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py: 2011-10-27 Yuta Kitamura WebSocket: Connecting to localhost:8880 takes one second on Windows https://bugs.webkit.org/show_bug.cgi?id=64788 Reviewed by Dirk Pranke. * Scripts/webkitpy/layout_tests/servers/websocket_server.py: Bind to "localhost" instead of "127.0.0.1" to let pywebsocket listen on both IPv4 and IPv6 addresses. This should prevent the test reserved-opcodes.html from timing out on Windows, because this test tries to open a lot of connections to localhost and each attempt takes one second to fall back from IPv6 to IPv4 on Windows (I have no idea why Windows works like this, though). 2011-10-27 Adam Barth Dis-integrate run-bindings-tests with webkitpy https://bugs.webkit.org/show_bug.cgi?id=71092 Reviewed by Ryosuke Niwa. These tests aren't really adding any value. * Scripts/webkitpy/common/config/ports.py: * Scripts/webkitpy/tool/commands/download_unittest.py: * Scripts/webkitpy/tool/mocktool.py: * Scripts/webkitpy/tool/steps/runtests.py: * Scripts/webkitpy/tool/steps/runtests_unittest.py: * Scripts/webkitpy/tool/steps/steps_unittest.py: 2011-10-27 Eric Seidel REGRESSION (r98639): NRWT crashes when parsing leaks output https://bugs.webkit.org/show_bug.cgi?id=71087 Unreviewed. Sadly we have no good way to test logging. * Scripts/webkitpy/layout_tests/port/leakdetector.py: 2011-10-27 David Levin Test expectation errors should include the file name and platform in both the log and exception info. https://bugs.webkit.org/show_bug.cgi?id=71067 Reviewed by Ojan Vafai. * Scripts/webkitpy/layout_tests/models/test_expectations.py: * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: 2011-10-27 Eric Seidel new-run-webkit-tests doesn't understand --qt or --gtk https://bugs.webkit.org/show_bug.cgi?id=63970 Reviewed by Ojan Vafai. Add support for --qt and --gtk to NRWT. I also removed some incorrect comments in the option parsing code (options which we already implement, but said we didn't) as well as moved --platform parsing out of the "results_options" section into the top configuration section. Now run-webkit-tests no longer needs NRWT-specific code for passing --platform=qt, etc. * Scripts/run-webkit-tests: * Scripts/webkitpy/layout_tests/run_webkit_tests.py: 2011-10-27 Eric Seidel REGRESSION (NRWT): build.webkit.org doesn't show the total number of leaks found during a test run on the Leaks bot https://bugs.webkit.org/show_bug.cgi?id=66227 Reviewed by Adam Roben. I believe this should fix the bug. * Scripts/run-webkit-tests: make NRWT default for --leaks * Scripts/webkitpy/layout_tests/port/leakdetector.py: (LeakDetector._parse_leaks_output): removed the (unneeded) process_pid argument, and made the regexp use named groups (even though we don't ever grab them by name) (LeakDetector.count_total_bytes_and_unique_leaks): renamed from parse_leak_files (LeakDetector.count_total_leaks): new file (the guts of this change) which is used to re-parse the leaks output during the summarize leaks phase. * Scripts/webkitpy/layout_tests/port/leakdetector_unittest.py: Changes to reflect the rename of count_total_bytes_and_unique_leaks and a new test for count_total_leaks. * Scripts/webkitpy/layout_tests/port/mac.py: Use count_total_leaks to spit out the total leak count like ORWT did, and remove the FIXME on the subject. 2011-10-27 Stephen Chenney [Chromium] Need setPrinting https://bugs.webkit.org/show_bug.cgi?id=46152 Reviewed by Hajime Morita. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): Added logic for handling setPrinting. (LayoutTestController::reset): Clearing flags for setPrinting. (LayoutTestController::setPrinting): Set flags for setPrinting. * DumpRenderTree/chromium/LayoutTestController.h: (LayoutTestController::setIsPrinting): Added the setPrinting methods and flags. (LayoutTestController::isPrinting): Method to return the setPrinting status. * DumpRenderTree/chromium/TestShell.cpp: (dumpFramesAsPrintedText): A method for creating a text dump in printed layout. (TestShell::dump): Modified to dump setPrinting-specific output. * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::paintPagesWithBoundaries): Handling for pixel tests when setPrinting is active. * DumpRenderTree/chromium/WebViewHost.h: Prototype for paintPagesWithBoundaries. 2011-10-27 Sam Weinig Add allowsPlugIns property to WKBrowsingContextGroup https://bugs.webkit.org/show_bug.cgi?id=70987 Reviewed by Anders Carlsson. * TestWebKitAPI/Tests/WebKit2ObjC/WKBrowsingContextGroupTest.mm: Add basic test for WKBrowsingContextGroup.allowsPlugIns. 2011-10-27 Adam Roben Test WKBundlePageGetBackingScaleFactor Test for REGRESSION (r97191): Clients can't reliably determine the scale factor of snapshots returned via WKBundlePageCreateSnapshot* API Reviewed by Sam Weinig. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added new files. * TestWebKitAPI/Tests/WebKit2/mac/GetBackingScaleFactor.mm: Added. (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Record the passed-back scale factor. (TestWebKitAPI::setInjectedBundleClient): Hook up our didReceiveMessage callback. (TestWebKitAPI::createWindow): Conjure up a window. (TestWebKitAPI::TEST): Ask the injected bundle code to call the new WKBundlePageGetBackingScaleFactor API when the WKView is in windows with various scale factors and when using a custom scale factor, and check that we get those same scale factors back. * TestWebKitAPI/Tests/WebKit2/mac/GetBackingScaleFactor_Bundle.mm: Added. (TestWebKitAPI::GetBackingScaleFactorTest::GetBackingScaleFactorTest): Just initialize our members. (TestWebKitAPI::GetBackingScaleFactorTest::didCreatePage): Store the page for later. (TestWebKitAPI::GetBackingScaleFactorTest::didReceiveMessage): Get the backing scale factor from the page using the new API and send it back. 2011-10-27 Balazs Kelemen [Qt][WK2] MiniBrowser crashes when started with 0 arguments Rubber-stamped by Simon Hausmann. * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::handleUserOptions): Fix typo error. 2011-10-27 Takashi Toyoshima [WebSocket] update pywebsocket to 0.7. https://bugs.webkit.org/show_bug.cgi?id=70992 Reviewed by Kent Tamura. pywebsocket 0.7 fixed the bug on dual stack support on OS X. * Scripts/webkitpy/thirdparty/__init__.py: 2011-10-27 Elliot Poger make lion observe MAC lines in test_expectations.txt https://bugs.webkit.org/show_bug.cgi?id=70941 Reviewed by Ojan Vafai. * Scripts/webkitpy/layout_tests/port/chromium.py: * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: 2011-10-27 John Knottenbelt [Chromium] Allow DRT to be built without notifications enabled. https://bugs.webkit.org/show_bug.cgi?id=70919 Reviewed by Tony Gentilcore. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::grantDesktopNotificationPermission): (LayoutTestController::simulateDesktopNotificationClick): * DumpRenderTree/chromium/NotificationPresenter.cpp: * DumpRenderTree/chromium/TestShell.cpp: (TestShell::TestShell): (TestShell::resetTestController): 2011-10-21 Jocelyn Turcotte [Qt] Improve command line arguments handling for MiniBrowser. https://bugs.webkit.org/show_bug.cgi?id=70616 Reviewed by Kenneth Rohde Christiansen. - Use two dashes for word arguments. - Return with an error when an unsupported argument is given. - Allow the URL to be placed before option arguments. * MiniBrowser/qt/MiniBrowserApplication.cpp: (printHelp): (MiniBrowserApplication::handleUserOptions): * MiniBrowser/qt/utils.cpp: (takeOptionFlag): (takeOptionValue): * MiniBrowser/qt/utils.h: 2011-10-27 Csaba Osztrogonác [Qt] NRWT doesn't work on qt-mac platform https://bugs.webkit.org/show_bug.cgi?id=64445 Reviewed by Adam Barth. * Scripts/run-webkit-tests: Switch qt-mac platform to NRWT. (useNewRunWebKitTests): * Scripts/webkitpy/layout_tests/port/qt.py: Fix library path on Mac. 2011-10-26 Sam Weinig Add initial setting to WKBrowsingContextGroup https://bugs.webkit.org/show_bug.cgi?id=70942 Reviewed by Anders Carlsson. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/WebKit2ObjC/WKBrowsingContextGroupTest.mm: Added. Add basic test that setting the javaScriptEnabled setting takes. It is not possible to test its effect quite yet. 2011-10-26 Alejandro G. Castro [GTK] [WK2] Add WebKit2 distcheck support https://bugs.webkit.org/show_bug.cgi?id=70933 Reviewed by Martin Robinson. * WebKitTestRunner/GNUmakefile.am: Add missing files to the sources and EXTRA_DIST. 2011-10-26 Simon Hausmann [Qt][WK2] Fix failing assertion on startup of MiniBrowser -touch. Reviewed by Tor Arne Vestbø. We're using QScroller since r98460, which requires QWidget/QApplication. So use QApplication and QtWidgets for the time being. * MiniBrowser/qt/MiniBrowser.pro: * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::MiniBrowserApplication): (MiniBrowserApplication::notify): * MiniBrowser/qt/MiniBrowserApplication.h: 2011-10-25 Zeno Albisser [Qt][WK2] Rewrite MiniBrowser in QML https://bugs.webkit.org/show_bug.cgi?id=70315 Add the first QML based implementation of MiniBrowser. Patch by Jocelyn Turcotte , Zeno Albisser Reviewed by Tor Arne Vestbø. * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): (BrowserWindow::webView): (BrowserWindow::load): (BrowserWindow::keyPressEvent): (BrowserWindow::~BrowserWindow): * MiniBrowser/qt/BrowserWindow.h: * MiniBrowser/qt/MiniBrowser.pro: * MiniBrowser/qt/MiniBrowser.qrc: * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::MiniBrowserApplication): (MiniBrowserApplication::handleUserOptions): * MiniBrowser/qt/MiniBrowserApplication.h: (WindowOptions::WindowOptions): * MiniBrowser/qt/UrlLoader.cpp: (UrlLoader::UrlLoader): * MiniBrowser/qt/icons/next.png: Added. * MiniBrowser/qt/icons/previous.png: Added. * MiniBrowser/qt/icons/refresh.png: Added. * MiniBrowser/qt/icons/stop.png: Added. * MiniBrowser/qt/main.cpp: (main): * MiniBrowser/qt/qml/BrowserWindow.qml: Added. 2011-10-26 Zeno Albisser [Qt][WK2] Remove QtWidgets dependency in MiniBrowser https://bugs.webkit.org/show_bug.cgi?id=70529 Remove obsolete code for a new QML based implementation of MiniBrowser. Change BrowserWindow to derive from QSGView instead of QMainWindow. QSGView is used as a top level window that provides the container for placing QML content. Make MiniBrowserApplication derive from QGuiApplication instead of QApplication. Patch by Jocelyn Turcotte , Zeno Albisser Reviewed by Simon Hausmann. * MiniBrowser/qt/BrowserView.cpp: Removed. * MiniBrowser/qt/BrowserView.h: Removed. * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): (BrowserWindow::load): (BrowserWindow::screenshot): (BrowserWindow::loadURLListFromFile): (BrowserWindow::updateUserAgentList): (BrowserWindow::~BrowserWindow): * MiniBrowser/qt/BrowserWindow.h: * MiniBrowser/qt/MiniBrowser.pro: * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::MiniBrowserApplication): (MiniBrowserApplication::notify): * MiniBrowser/qt/MiniBrowserApplication.h: * MiniBrowser/qt/UrlLoader.cpp: (UrlLoader::UrlLoader): * MiniBrowser/qt/main.cpp: 2011-10-25 Zeno Albisser [Qt][WK2] Remove QAction from MiniBrowser https://bugs.webkit.org/show_bug.cgi?id=70525 In order to rewrite MiniBrowser in QML we need a suitable mechanism for triggering navigation actions. Due to the QtWidgets dependency of QAction the current solution is not suitable anymore. Therefore we introduce invokable methods and export properties in QWebNavigationController. Patch by Jocelyn Turcotte , Zeno Albisser Reviewed by Simon Hausmann. * MiniBrowser/qt/BrowserView.cpp: * MiniBrowser/qt/BrowserView.h: * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): 2011-10-26 Gyuyoung Kim [EFL] Unregister viewport,changed signal in EWebLauncher. https://bugs.webkit.org/show_bug.cgi?id=70478 Reviewed by Eric Seidel. 'viewport,changed' signal was registered by EWebLauncher in order to test viewport meta tag. However, it sets wrong layout to some web sites which have viewport meta tag. Because, host pc doesn't support proper DPI value though viewport is using DPI value to compute layout values. This signal is for mobile device. * EWebLauncher/main.c: (browserCreate): 2011-10-25 Anders Carlsson Plug-ins have no way to find out when the device scale factor changes https://bugs.webkit.org/show_bug.cgi?id=67226 Reviewed by Sam Weinig. Keep a cached copy of the contents scale and update it when it changes. * DumpRenderTree/TestNetscapePlugIn/Tests/mac/ContentsScaleFactor.cpp: (ContentsScaleFactor::ContentsScaleFactor): (ContentsScaleFactor::cachedContentsScaleFactor): (ContentsScaleFactor::ScriptableObject::hasProperty): (ContentsScaleFactor::ScriptableObject::getProperty): (ContentsScaleFactor::NPP_New): (ContentsScaleFactor::NPP_SetValue): 2011-10-25 Sam Weinig Add WKBrowsingContextLoadDelegateTest test for a failed load https://bugs.webkit.org/show_bug.cgi?id=70859 Reviewed by Anders Carlsson. * TestWebKitAPI/Tests/WebKit2ObjC/WKBrowsingContextLoadDelegateTest.mm: (TEST_F): Add TEST_F(WKBrowsingContextLoadDelegateTest, SimpleLoadFail). (-[SimpleLoadFailDelegate browsingContextControllerDidFailProvisionalLoad:withError:]): Test that we are getting the correct NSError code and domain. 2011-10-25 Erik Arvidsson JS Test Harness: Make successfullyParsed optional https://bugs.webkit.org/show_bug.cgi?id=70784 Reviewed by Ojan Vafai. * Scripts/make-new-script-test: (writeTestFile): Update the template for script tests. 2011-10-25 Anders Carlsson Plug-ins have to use JavaScript to find out the current device scale factor https://bugs.webkit.org/show_bug.cgi?id=67225 Reviewed by Darin Adler. Add a plug-in test for getting the contents scale factor. * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: * DumpRenderTree/TestNetscapePlugIn/Tests/mac/ContentsScaleFactor.cpp: Added. (ContentsScaleFactor::ContentsScaleFactor): (ContentsScaleFactor::contentsScaleFactor): (ContentsScaleFactor::ScriptableObject::hasProperty): (ContentsScaleFactor::ScriptableObject::getProperty): (ContentsScaleFactor::ScriptableObject::pluginTest): (ContentsScaleFactor::NPP_GetValue): 2011-10-25 Sam Weinig Flesh out WKBrowsingContextLoadDelegate a bit https://bugs.webkit.org/show_bug.cgi?id=70846 Reviewed by Anders Carlsson. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/WebKit2ObjC: Added. * TestWebKitAPI/Tests/WebKit2ObjC/WKBrowsingContextLoadDelegateTest.mm: Added. (WKBrowsingContextLoadDelegateTest::WKBrowsingContextLoadDelegateTest): (WKBrowsingContextLoadDelegateTest::SetUp): (WKBrowsingContextLoadDelegateTest::TearDown): (-[SimpleLoadDelegate browsingContextControllerDidFinishLoad:]): (TEST_F): Add basic testing for WKBrowsingContextLoadDelegate. 2011-10-25 Alexey Proskuryakov make-script-test-wrappers should not look outside fast/js https://bugs.webkit.org/show_bug.cgi?id=66357 Reviewed by Maciej Stachowiak. * Scripts/make-script-test-wrappers: Don't look for template files everywhere, only support fast/js. 2011-10-25 Adam Barth Add a --make option to update-webkit to engage the Makefile-based build system https://bugs.webkit.org/show_bug.cgi?id=70847 Reviewed by Dimitri Glazkov. I can never remember what crazy environment variables I'm supposed to set to tell GYP what to do. * Scripts/update-webkit: 2011-10-25 Raphael Kubo da Costa [EFL] DRT: Force layout when sending events. https://bugs.webkit.org/show_bug.cgi?id=70357 Reviewed by Antonio Gomes. Some tests such as editing/input/page-up-down-scrolls.html rely on the scrollbars being properly set up during the onload events. Document::implicitClose(), however, dispatches the onload events before calling FrameView::layout(), so these tests usually fail. We now do the same as the GTK+ and Chromium ports and manually force the layout when dispatching events. * DumpRenderTree/efl/EventSender.cpp: (sendMouseEvent): (keyDownCallback): 2011-10-25 Daniel Bates build-webkit --clean should only clean build directory for CMake-based ports https://bugs.webkit.org/show_bug.cgi?id=70834 Reviewed by Antonio Gomes. Currently build-webkit --clean both cleans the build directory and builds WebKit for CMake-based ports. Instead, build-webkit --clean should only clean up the build directory as per its description in the usage message of build-webkit. * Scripts/webkitdirs.pm: (buildCMakeProjectOrExit): 2011-10-25 Anna Cavender Enable running of track layout tests for Chromium https://bugs.webkit.org/show_bug.cgi?id=70803 Reviewed by Adam Barth. Make DumpRenderTree/TestShell always run with track enabled (equivalent of --enable-video-track). * DumpRenderTree/chromium/TestShell.cpp: (TestShell::TestShell): 2011-10-25 Vamshikrishna Yellenki [WK2]Main window never gets focus using TAB key in WebKit2 GTK+. https://bugs.webkit.org/show_bug.cgi?id=70603 Reviewed by Gustavo Noronha Silva. * MiniBrowser/gtk/BrowserWindow.c: (focus): (unFocus): (browserWindowUIClientInit): 2011-10-25 Dan Bernstein DOMRangeOfString:relativeTo:options has problems with -webkit-user-select: none Reviewed by Adam Roben. * TestWebKitAPI/Tests/mac/DOMRangeOfString.mm: (TestWebKitAPI::TEST): 2011-10-25 Eric Seidel WIN: editing tests fail under NRWT because editing delegate callbacks aren't stripped https://bugs.webkit.org/show_bug.cgi?id=64471 Reviewed by Adam Roben. Speculative fix. * Scripts/webkitpy/layout_tests/port/win.py: * Scripts/webkitpy/layout_tests/port/win_unittest.py: 2011-10-24 Mikhail Naganov Fix bug summary when rolling Chromium DEPS to LKGR. https://bugs.webkit.org/show_bug.cgi?id=69917 Reviewed by Adam Barth. * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/sheriff.py: * Scripts/webkitpy/tool/commands/roll.py: * Scripts/webkitpy/tool/commands/roll_unittest.py: 2011-10-24 Ivan Briano [EFL] Build fix. Add include path to find npapi.h https://bugs.webkit.org/show_bug.cgi?id=70730 Unreviewed build fix. * DumpRenderTree/efl/CMakeLists.txt: Add include path for npapi.h 2011-10-24 Eric Seidel Plugin tests are failing after switching WK2 to NRWT https://bugs.webkit.org/show_bug.cgi?id=70760 Reviewed by Simon Fraser. Perpetuate this horrible hack. * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-10-24 Eric Seidel NRWT doesn't support ORWT's --root option, but should https://bugs.webkit.org/show_bug.cgi?id=70416 Reviewed by Dirk Pranke. Add support for ORWT's --root option which is used for running tests with a nightly build of WebKit. * Scripts/old-run-webkit-tests: * Scripts/webkitpy/layout_tests/port/base.py: - webkit_build_path was never called by any code, removed. * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/run_webkit_tests.py: 2011-10-24 David Levin check-webkit-style complains about WEBKIT_EXPORT in test/WebUnitTests.h https://bugs.webkit.org/show_bug.cgi?id=70747 Reviewed by Tony Chang. * Scripts/webkitpy/style/checkers/cpp.py: Allow WEBKIT_EXPORT in tests dir. * Scripts/webkitpy/style/checkers/cpp_unittest.py: Corresponding test. 2011-10-24 Balazs Kelemen [Qt][WK2] Crash when running pixel tests https://bugs.webkit.org/show_bug.cgi?id=70725 Reviewed by Chang Shu. * WebKitTestRunner/qt/TestInvocationQt.cpp: (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): 2011-10-24 Carlos Garcia Campos [GTK] Remove g_thread_init() calls https://bugs.webkit.org/show_bug.cgi?id=70717 Reviewed by Xan Lopez. With newer glib it's not needed anymore and with previous versions it was already called by g_type_init(). * DumpRenderTree/gtk/DumpRenderTree.cpp: (main): * GtkLauncher/main.c: (main): * MiniBrowser/gtk/main.c: (main): 2011-10-24 Alejandro G. Castro [WK2] [GTK] WebKitTestRunner crashes with heap corruption https://bugs.webkit.org/show_bug.cgi?id=69403 We did not added autotoolsconfig.h to the config.h file and we were adding Platform.h without the config.h in the WebKitTestRunnerPrefix.h. Added the autotoolsconfig.h include in the Prefix file. Reviewed by Martin Robinson. * WebKitTestRunner/WebKitTestRunnerPrefix.h: 2011-10-23 Tomasz Morawski [EFL] Remove unused code from EWebLauncher https://bugs.webkit.org/show_bug.cgi?id=70590 Reviewed by Andreas Kling. Removed unused function on_closeWindow from EWebLauncher * EWebLauncher/main.c: 2011-10-23 Luke Macpherson Add watchlist for WebCore/css and put myself on it. https://bugs.webkit.org/show_bug.cgi?id=70701 Reviewed by Adam Barth. * Scripts/webkitpy/common/config/watchlist: Add watchlist for WebCore/css and put myself on it. 2011-10-23 Kaustubh Atrawalkar [GTK] Implement KeyDown function for WebKit2 EventSender. https://bugs.webkit.org/show_bug.cgi?id=69410 Reviewed by Martin Robinson. Implement the KeyDown function on Gtk platform. * WebKitTestRunner/GNUmakefile.am: * WebKitTestRunner/TestController.cpp: (WTR::TestController::TestController): (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle): * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp: Added. (WTR::EventSenderProxy::EventSenderProxy): (WTR::dispatchEvent): (WTR::getModifiers): (WTR::getGDKKeySymForKeyRef): (WTR::EventSenderProxy::keyDown): 2011-10-22 Zan Dobersek [WK2][GTK] run-launcher script fails after testing WK2 tests https://bugs.webkit.org/show_bug.cgi?id=70676 Reviewed by Andreas Kling. Pass the '-2' flag instead of '--webkit-test-runner' when executing run-launcher script after testing with WebKitTestRunner. * Scripts/webkitpy/layout_tests/port/gtk.py: 2011-10-22 Zan Dobersek [WK2][GTK] Missing env variables cause WKTR to crash https://bugs.webkit.org/show_bug.cgi?id=70674 Reviewed by Martin Robinson. Set two environment variables that the Gtk's port of WebKitTestRunner depends on and crashes if they're not explicitly set by the user. Also modify the setting of the WEBKIT_INSPECTOR_PATH variable so it properly uses the _build_path method. * Scripts/webkitpy/layout_tests/port/gtk.py: Set TEST_RUNNER_INJECTED_BUNDLE_FILENAME and TEST_RUNNER_TEST_PLUGIN_PATH environment variables. 2011-10-21 Filip Pizlo Bencher script doesn't measure GC times accurately https://bugs.webkit.org/show_bug.cgi?id=70588 Reviewed by Geoff Garen. Added two new options which allow different ways of measuring GC times: --measure-gc, which omits calls to gc() between benchmark invocations. This option takes an optional argument, which is the name of the VM in which to enable this feature. This allows comparing a single VM against itself, with and without GC. --rerun , which causes each sample measurement to include N invocations which do not have gc() calls between them. The default is N = 1, which results in the same behavior as before. You can use either --measure-gc or --rerun for N > 1 (preferably N >= 3) to get more of a contribution from GC to the measured times. --rerun results in tighter confidence intervals than --measure-gc, since it amortizes GC effects in each sample, while with --measure-gc some samples will see GC and some won't leading to a higher standard devation and thus requiring more samples to reduce confidence intervals to managable levels. * Scripts/bencher: 2011-10-21 Sam Weinig Remove ability to create a WKView without a WKContextRef and WKPageGroupRef https://bugs.webkit.org/show_bug.cgi?id=70653 Reviewed by Simon Fraser. * MiniBrowser/mac/AppDelegate.h: * MiniBrowser/mac/AppDelegate.m: (-[BrowserAppDelegate init]): (-[BrowserAppDelegate newWindow:]): (-[BrowserAppDelegate openPanelDidEnd:returnCode:contextInfo:]): * MiniBrowser/mac/BrowserWindowController.h: * MiniBrowser/mac/BrowserWindowController.m: (-[BrowserWindowController initWithContext:pageGroup:]): (-[BrowserWindowController windowWillClose:]): (createNewPage): (-[BrowserWindowController awakeFromNib]): * TestWebKitAPI/mac/WebKitAgnosticTest.mm: (TestWebKitAPI::WebKitAgnosticTest::runWebKit2Test): Update testing code to pass a PageGroup as necessary. 2011-10-21 Leandro Pereira webkitpy: Teach NRWT about the EFL port https://bugs.webkit.org/show_bug.cgi?id=70637 Allows using ``efl'' as a platform when executing NRWT. Reviewed by Eric Seidel. * Scripts/webkitpy/layout_tests/port/efl.py: Added. * Scripts/webkitpy/layout_tests/port/efl_unittest.py: Added. * Scripts/webkitpy/layout_tests/port/factory.py: Adjust factory to make EflPort objects when using PortFactory.get(port_name='efl'). 2011-10-21 Devdatta Deshpande [Gtk] mousemove event always has metaKey == true https://bugs.webkit.org/show_bug.cgi?id=35299 Reviewed by Martin Robinson. * DumpRenderTree/gtk/EventSender.cpp: (gdkModifierFromJSValue): GDK_META_MASK is used to represent metaKey for platform GTK mouse and keyboard events. So making it in sync with WebCore files. (mouseMoveToCallback): Sending modifier keys state for mouse move event. 2011-10-21 Carlos Garcia Campos watchlist: Add a new entry for WebKit2 GTK+ public API https://bugs.webkit.org/show_bug.cgi?id=70601 Reviewed by Philippe Normand. Subscribe myself to WebKit2 GTK+ API patches. Also add a message for patches that might include new API that points to the WebKit2 GTK+ API guidelines. * Scripts/webkitpy/common/config/watchlist: 2011-10-21 Simon Hausmann Remove QtScript source code from WebKit. https://bugs.webkit.org/show_bug.cgi?id=64088 Reviewed by Tor Arne Vestbø. Removed dead code that isn't developed anymore. * Scripts/webkitpy/style/checker_unittest.py: 2011-10-20 Eric Seidel Switch webkit2 bot to NRWT https://bugs.webkit.org/show_bug.cgi?id=56729 Reviewed by Adam Barth. Move the WK2 bot to using NRWT instead of ORWT, as announced on webkit-dev: https://lists.webkit.org/pipermail/webkit-dev/2011-October/018337.html * Scripts/run-webkit-tests: (runningOnBuildBot): (useNewRunWebKitTests): 2011-10-20 Eric Seidel NRWT is asserting on DRT due to lack of newline when processing line from DumpRenderTree https://bugs.webkit.org/show_bug.cgi?id=70585 Reviewed by Adam Barth. This really should be an assert, as this inidicates a programming error in either DRT or NRWT. However since DRT was functioning fine before in Qt, we'll make this an error log for now and once we find the root cause flip this back to an assert. * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-10-20 Yuta Kitamura Unreviewed, rolling out r98064. http://trac.webkit.org/changeset/98064 https://bugs.webkit.org/show_bug.cgi?id=64788 Broke Mac bots. * Scripts/webkitpy/layout_tests/servers/websocket_server.py: 2011-10-20 Yuta Kitamura WebSocket: Connecting to localhost:8880 takes one second on Windows https://bugs.webkit.org/show_bug.cgi?id=64788 Reviewed by Dirk Pranke. * Scripts/webkitpy/layout_tests/servers/websocket_server.py: Bind to "localhost" instead of "127.0.0.1" to let pywebsocket listen on both IPv4 and IPv6 addresses. This should prevent the test reserved-opcodes.html from timing out on Windows, because this test tries to open a lot of connections to localhost and each attempt takes one second to fall back from IPv6 to IPv4 on Windows (I have no idea why Windows works like this, though). 2011-10-20 Ryosuke Niwa nrwt: newly generated results are put in cross-platform directory https://bugs.webkit.org/show_bug.cgi?id=68931 Reviewed by Dirk Pranke. The bug was caused by SingleTestRunner._add_missing_baselines's always calling _save_baseline_data with generate_new_baseline set to False. Fixed the bug by always passing True when .png file is missing (because png images are typically different on each platform), and passing True when .txt file is missing and the actual result's first line matches the regular expression "layer at \(\d+,\d+\) size \d+x\d+". * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py: * Scripts/webkitpy/layout_tests/port/test.py: * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: Changed the expectation and added a test case. 2011-10-20 Kent Tamura Unreviewed. Adding myself to watchlist. * Scripts/webkitpy/common/config/watchlist: Add "ChromiumDumpRenderTree" and "Forms" definitions, and cc them to me. 2011-10-20 Nico Weber [chromium/mac] Add support for building with make https://bugs.webkit.org/show_bug.cgi?id=70455 Reviewed by Adam Barth. Depends on http://codereview.chromium.org/8352029/ `GYP_GENERATORS=make Tools/Scripts/update-webkit --chromium` will create a toplevel Makefile.chromium. If that file exists and is newer than Source/WebKit/chromium/WebKit.xcodeproj, `Tools/Scripts/build-webkit --chromium` will build with make instead of xcodebuild. Also let nrwt use a binary from out/ if it exists. * Scripts/webkitdirs.pm: (determineBaseProductDir): (isChromiumMacMake): (determineIsChromiumMacMake): (buildChromium): 2011-10-20 Eric Seidel Possible REGRESSION(97879): NRWT fails when DumpRenderTree crashes https://bugs.webkit.org/show_bug.cgi?id=70524 Reviewed by Adam Barth. This is a speculative fix, since I do not use a platform which outputs crashlogs over stderr. * Scripts/webkitpy/layout_tests/port/server_process.py: * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-10-20 Eric Seidel REGRESSION(97879): Pixel tests no longer work with NRWT on Mac https://bugs.webkit.org/show_bug.cgi?id=70492 Reviewed by Adam Barth. The bug turned out to be that I was assuming the block.content would be empty before the binary content following Content-Length was read inside _read_block. Turns out its not, due to extra newlines and "ExpectedHash" header. In the process of trying to figure out what was going wrong I ended up cleaning up our newline usage in DumpRenderTree a little. Moved two error messages from stdout to stderr, and fixed a little code indent/whitespace. I also fixed ServerProcess to use "deadline" everywhere instead of timeout per Adam's request in the original bug. * DumpRenderTree/PixelDumpSupport.cpp: (dumpWebViewAsPixelsAndCompareWithExpected): * DumpRenderTree/cg/ImageDiffCG.cpp: (main): * DumpRenderTree/mac/DumpRenderTree.mm: (dump): * DumpRenderTree/mac/PixelDumpSupportMac.mm: (restoreMainDisplayColorProfile): (setupMainDisplayColorProfile): * Scripts/webkitpy/layout_tests/port/server_process.py: * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: 2011-10-20 Tony Chang [chromium] Remove from ImageDiff and use unsigned int instead of uint32_t. Unreviewed, fixing the chromium win build. * DumpRenderTree/chromium/ImageDiff.cpp: (Image::pixelAt): (Image::setPixelAt): (maxOf3): (getRedComponent): (getGreenComponent): (getBlueComponent): (weightedPercentageDifferent): (createImageDiff): 2011-10-20 Hao Zheng [Chromium] Reduce dependencies of ImageDiff to compile it for Android. https://bugs.webkit.org/show_bug.cgi?id=69997 Reviewed by Tony Chang. This change is required to make ImageDiff for Android, and can benefit other platforms, too. On Android, most targets are built as toolsets:target. while ImageDiff needs to be built as toolsets:host. Currently, building the standalone target of ImageDiff depends on about 85 other targets (try make ImageDiff'), which is inefficient. After the change, ImageDiff only depends on webkit_support_gfx, which in turn depends on libpng and zlib. * DumpRenderTree/chromium/ImageDiff.cpp: (main): 2011-10-20 Leandro Pereira [EFL] Unreviewed DumpRenderTree build fix. * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::addChromeInputField): Add stub. (LayoutTestController::removeChromeInputField): Ditto. (LayoutTestController::focusWebView): Ditto. (LayoutTestController::setBackingScaleFactor): Ditto. 2011-10-20 Leandro Pereira [EFL] Plug ImageDiff/DumpRenderTree on CMake build system https://bugs.webkit.org/show_bug.cgi?id=70142 Reviewed by Gustavo Noronha Silva. * CMakeListsEfl.txt: Include DRT CMakeLists. * DumpRenderTree/efl/CMakeLists.txt: Added. 2011-10-20 Philippe Normand [style] Allow usage of NULL in gst_* https://bugs.webkit.org/show_bug.cgi?id=70498 Reviewed by David Levin. * Scripts/webkitpy/style/checkers/cpp.py: Simplified the detection of gst_ calls. Now just ignore NULL in all of them. * Scripts/webkitpy/style/checkers/cpp_unittest.py: Test for above change. 2011-10-20 Leandro Pereira [EFL] Unreviewed. Build fix after r97043. * DumpRenderTree/efl/DumpRenderTreeChrome.cpp: (DumpRenderTreeChrome::resetDefaultsToConsistentValues): Use ewk_view_scale_set() instead of ewk_view_page_scale(). * DumpRenderTree/efl/EventSender.cpp: (scalePageByCallback): Ditto. 2011-10-20 Philippe Normand Unreviewed. Adding myself to watchlists. * Scripts/webkitpy/common/config/watchlist: 2011-10-19 Eric Seidel Reviewed by Adam Barth. Teach new-run-webkit-tests about #CRASHED and #CRASHED - WebProcess https://bugs.webkit.org/show_bug.cgi?id=63683 * Scripts/webkitpy/common/net/layouttestresults.py: * Scripts/webkitpy/common/net/resultsjsonparser.py: * Scripts/webkitpy/layout_tests/layout_package/manager.py: * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py: * Scripts/webkitpy/layout_tests/layout_package/test_failures.py: * Scripts/webkitpy/layout_tests/layout_package/test_failures_unittest.py: * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py: * Scripts/webkitpy/layout_tests/layout_package/worker.py: * Scripts/webkitpy/layout_tests/port/base.py: * Scripts/webkitpy/layout_tests/port/server_process.py: * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: 2011-10-19 Jochen Eisinger Chromium doesn't support git branch builds, so don't add the branch to the product dir https://bugs.webkit.org/show_bug.cgi?id=70449 Reviewed by Eric Seidel. * Scripts/webkitdirs.pm: (determineBaseProductDir): 2011-10-19 Ojan Vafai Unreviewed. Adding myself to watchlists. * Scripts/webkitpy/common/config/watchlist: 2011-10-19 Tom Zakrajsek Add a suggest-nominations command to webkit-patch for computing potential committer/reviewer nominations https://bugs.webkit.org/show_bug.cgi?id=62166 Reviewed by Eric Seidel. Included options to control committer/reviewer patch count requirements, an age-limit on patches, and verbose output for more in-depth analysis. * Scripts/webkitpy/common/checkout/changelog.py: * Scripts/webkitpy/tool/commands/__init__.py: * Scripts/webkitpy/tool/commands/suggestnominations.py: Added. * Scripts/webkitpy/tool/commands/suggestnominations_unittest.py: Added. 2011-10-19 Eric Seidel new-run-webkit-tests -2 was crashing due to both "java" and "java/" being listed in Skipped files. Make NRWT normalize directory names in Skipped files during parsing. Unreviewed. Very simple (tested) change. * Scripts/webkitpy/layout_tests/port/mac_unittest.py: * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-10-19 David Levin watchlist: Should be more robust to bad regex. https://bugs.webkit.org/show_bug.cgi?id=69486 Reviewed by Adam Barth. * Scripts/webkitpy/common/config/watchlist: Change the instructions due to watchlist being checked by check-webkit-style (bug 69487) and remove the .* from file patterns since they are no longer anchored on the right hand side. * Scripts/webkitpy/common/watchlist/amountchangedpattern.py: Change *pattern to take a compiled regex directly. * Scripts/webkitpy/common/watchlist/amountchangedpattern_unittest.py: Ditto. * Scripts/webkitpy/common/watchlist/changedlinepattern.py: Ditto. * Scripts/webkitpy/common/watchlist/changedlinepattern_unittest.py: Ditto. * Scripts/webkitpy/common/watchlist/filenamepattern.py: Ditto. * Scripts/webkitpy/common/watchlist/filenamepattern_unittest.py: Ditto and change a test now that filenames are no longer anchored on the right hand side. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Catch regex errors and log them as errors. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Add some tests to verify that bad regexes don't kill everything. 2011-10-19 David Levin watchlist: Add a stylecheck to do validity checks for the watchlist config. https://bugs.webkit.org/show_bug.cgi?id=69487 Reviewed by Adam Barth. * Scripts/webkitpy/style/checker.py: Add the watchlist file type. * Scripts/webkitpy/style/checkers/watchlist.py: Added. * Scripts/webkitpy/style/checkers/watchlist_unittest.py: Added. 2011-10-19 Eric Seidel NRWT ServerProcess can't read lines from stderr and stdio separately https://bugs.webkit.org/show_bug.cgi?id=70435 Reviewed by Adam Barth. This is the underlying bug behind our lack of WK2 support, and timeout detection on Windows. DRT expects RWT to read one line at a time from stderr and stdout independently. This allows DRT to communicate errors over stderr and have them responded to immediately regardless of what state the stdio stream may be in. Previously NRWT's ServerProcess class only exposed a read_line function which would read a line from stdout and batch all output from stderr into an error buffer. Callers would only be able to respond to commands over stderr if they happened to correspond with a stdout line boundary. * Scripts/webkitpy/layout_tests/port/server_process.py: * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-10-19 Chang Shu [Qt] Support WK1-only Skipped list https://bugs.webkit.org/show_bug.cgi?id=70254 Reviewed by Csaba Osztrogonác. Added a new search path qt-wk1 for skipping WK1-only tests. * Scripts/webkitpy/layout_tests/port/qt.py: NRWT * Scripts/old-run-webkit-tests: ORWT 2011-10-19 Gabor Rapcsanyi [GTK] Enable to use built-product-archive on GTK platform https://bugs.webkit.org/show_bug.cgi?id=70319 Reviewed by Csaba Osztrogonác. * BuildSlaveSupport/built-product-archive: 2011-10-18 David Levin watchlist: Should try to run if it can and not throw on mistakes. https://bugs.webkit.org/show_bug.cgi?id=70358 Reviewed by Adam Barth. * Scripts/webkitpy/common/system/outputcapture.py: Add the ability to capture log output. * Scripts/webkitpy/common/watchlist/watchlistloader_unittest.py: Adapt to the logging of errors. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Change to log problems and fix problems when found. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Adapt to the logging of errors. * Scripts/webkitpy/common/watchlist/watchlistrule.py: Expose a way to remove instructions. * Scripts/webkitpy/common/watchlist/watchlistrule_unittest.py: Test the new function. 2011-10-18 Sam Weinig Move uses of C-SPI out of WKView.h and into WKViewPrivate.h https://bugs.webkit.org/show_bug.cgi?id=70387 Reviewed by Dan Bernstein. * MiniBrowser/mac/BrowserWindowController.m: * TestWebKitAPI/Tests/mac/DeviceScaleFactorOnBack.mm: * TestWebKitAPI/mac/JavaScriptTestMac.mm: * TestWebKitAPI/mac/PlatformWebViewMac.mm: * TestWebKitAPI/mac/WebKitAgnosticTest.mm: * WebKitTestRunner/mac/PlatformWebViewMac.mm: Update to include WKViewPrivate.h where necessary. 2011-10-18 Joseph Pecoraro Web Inspector: Some localizedStrings.js Updates https://bugs.webkit.org/show_bug.cgi?id=70365 Reviewed by Darin Adler. * Scripts/check-inspector-strings: Updated to the new webkitpy way to determine checkout root. * Scripts/webkitdirs.pm: (copyInspectorFrontendFiles): Made it so build-webkit --inspector-frontend copies over the localizedStrings.js file as well as the frontend files. 2011-10-18 Raphael Kubo da Costa [EFL] DRT: Clear list of visited pages before each test. https://bugs.webkit.org/show_bug.cgi?id=70355 Reviewed by Antonio Gomes. r97596 uncovered a bug in the implementation: the list of visited links was not cleared between tests, which caused tests such as fast/repaint/outline-repaint-glitch.html to be flaky (the link color changed depending on whether fast/repaint/border-repaint-glitch.html was visited before or not). Call ewk_history_clear to make sure the visited links list does not persist between tests. * DumpRenderTree/efl/DumpRenderTreeChrome.cpp: (DumpRenderTreeChrome::resetDefaultsToConsistentValues): 2011-10-18 Dimitri Glazkov Split TestExpectationsEditor into its own file. https://bugs.webkit.org/show_bug.cgi?id=70348 Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor.py: Copied from Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py. * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py. * Scripts/webkitpy/layout_tests/models/test_expectations.py: Adjusted for change. * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Ditto. * Scripts/webkitpy/tool/servers/gardeningserver.py: Ditto. 2011-10-18 Ryosuke Niwa [Chromium] buildbot runs nrwt with --chromium instead of --chromium-cg-mac https://bugs.webkit.org/show_bug.cgi?id=70313 Reviewed by Adam Barth. Fixed the bug by passing --platform --chromium-cg-mac in RunWebKitTests. We can't always add this parameter because build-webkit and other scripts don't accept it. * BuildSlaveSupport/build.webkit.org-config/master.cfg: 2011-10-18 Adam Barth Subscribe myself to WebIDL changes https://bugs.webkit.org/show_bug.cgi?id=70341 Reviewed by David Levin. I've also added some documentation and refactored my previous subscriptions to make it eaiser for others to subscribe as well. * Scripts/webkitpy/common/config/watchlist: 2011-10-18 Dimitri Glazkov Add myself to the newfangled watchlist plumbing. https://bugs.webkit.org/show_bug.cgi?id=70338 Reviewed by David Levin. * Scripts/webkitpy/common/config/watchlist: Added sheepishly. 2011-10-18 Dimitri Glazkov Fix a unit test in TestFailures. https://bugs.webkit.org/show_bug.cgi?id=70337 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/NewBugForm_unittests.js: Turns out, the value of a method property should always be lowercase. 2011-10-18 Adam Barth Always enable ENABLE(XPATH) https://bugs.webkit.org/show_bug.cgi?id=70217 Reviewed by Eric Seidel. * Scripts/build-webkit: 2011-10-18 Dimitri Glazkov garden-o-matic should not report flaking in expected failures as failures. https://bugs.webkit.org/show_bug.cgi?id=70279 Reviewed by Adam Barth. On the bots, flakes that are already expected to fail aren't reported as failures (they keep the bot green). For example, if TEXT failure is expected, a TEXT TIMEOUT result will not cause the redness. We should do the same for garden-o-matic. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js: Spuriously changed to use Array iterators rather than jQuery APIs, rewrote the logic for determining expected or unexpected failures as a central analysis tool. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js: Added unit tests. 2011-10-18 Zan Dobersek [Gtk] Support for client-based geolocation https://bugs.webkit.org/show_bug.cgi?id=64970 Reviewed by Martin Robinson. Add support for testing client-based geolocation tests in DumpRenderTree. * DumpRenderTree/gtk/DumpRenderTree.cpp: (resetDefaultsToConsistentValues): Reset GeolocationClientMock. (createWebView): Create web view after declaring DumpRenderTree mode. That way a mock client for geolocation is used instead of a GeoClue client if a web view is created when testing. * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: Implement functions for geolocation layout tests. (LayoutTestController::setMockGeolocationPosition): (LayoutTestController::setMockGeolocationError): (LayoutTestController::setGeolocationPermission): (LayoutTestController::numberOfPendingGeolocationPermissionRequests): * Scripts/build-webkit: Note that the Gtk port is using client-based geolocation, even though changing this flag has no effect on compilation. 2011-10-07 Andrey Kosyakov [Chromium] suppress console output from DevTools window https://bugs.webkit.org/show_bug.cgi?id=69632 Reviewed by Pavel Feldman. * DumpRenderTree/chromium/TestShell.cpp: (TestShell::showDevTools): * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::didAddMessageToConsole): (WebViewHost::reset): (WebViewHost::setLogConsoleOutput): * DumpRenderTree/chromium/WebViewHost.h: 2011-10-18 Sheriff Bot Unreviewed, rolling out r97735. http://trac.webkit.org/changeset/97735 https://bugs.webkit.org/show_bug.cgi?id=70317 some people apparently use build-jsc with the qt port without building webkit first (Requested by torarne on #webkit). * Scripts/build-webkit: * Scripts/webkitdirs.pm: (buildQMakeProject): (buildQMakeQtProject): 2011-10-18 Tor Arne Vestbø [Qt] Remove dead code in build-webkit build-webkit will iterate over a list of subdirectories to build, but this feature is only used by the Apple ports. For QtWebKit we always pass an empty directory to buildQMakeProject, which makes the logic for dealing with sub-projects in buildQMakeProject just added noise. If we do want this feature at some point, we should redo it in the context of the upcoming build system changes for Qt. Reviewed by Simon Hausmann. * Scripts/build-webkit: * Scripts/webkitdirs.pm: (buildQt): 2011-10-18 Tor Arne Vestbø Remove traces of Symbian support from the build scripts Reviewed by Simon Hausmann. * Scripts/build-webkit: * Scripts/webkitdirs.pm: (determineBaseProductDir): (argumentsForConfiguration): (usesPerConfigurationBuildDirectory): (buildQMakeProject): 2011-10-18 Ryosuke Niwa Build fix after r97672. * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: 2011-10-17 Kent Tamura A follow-up of r97716. https://bugs.webkit.org/show_bug.cgi?id=68803 * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::reset): Initialize m_dumpCreateView. 2011-10-17 Jochen Eisinger [chromium] add LayoutTestController.dumpCreateView() to monitor calls to WebViewClient::createView https://bugs.webkit.org/show_bug.cgi?id=68803 Reviewed by Nate Chapin. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (LayoutTestController::dumpCreateView): * DumpRenderTree/chromium/LayoutTestController.h: (LayoutTestController::shouldDumpCreateView): * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::createView): 2011-10-17 Nate Chapin Add myself to watchlists, and add a watch for Source/WebCore/loader/. https://bugs.webkit.org/show_bug.cgi?id=70263 Reviewed by David Levin. * Scripts/webkitpy/common/config/watchlist: 2011-10-17 Raphael Kubo da Costa [EFL] DRT: Make non-recognized keys work in events as well. https://bugs.webkit.org/show_bug.cgi?id=69860 Reviewed by Antonio Gomes. fast/forms/select-type-ahead-non-latin.html is the canonical case we fix here: it has a select box with a few non-latin options, and one of them is selected via the keyboard. In this specific case, we need to make EventHandler::keyEvent send a keyPressEvent to SelectElement, so that SelectElement::defaultEventHandler calls typeAheadFind(). This happens only when PlatformKeyEvent::text() is not empty, which was not the case so far. * DumpRenderTree/efl/EventSender.cpp: (keyDownCallback): 2011-10-17 Alok Priyadarshi Add layout tests for WebPlugin compositor path https://bugs.webkit.org/show_bug.cgi?id=69027 Reviewed by James Robinson. * DumpRenderTree/DumpRenderTree.gypi: * DumpRenderTree/chromium/TestWebPlugin.cpp: Added. (premultiplyAlpha): (TestWebPlugin::TestWebPlugin): (TestWebPlugin::~TestWebPlugin): (TestWebPlugin::mimeType): (TestWebPlugin::initialize): (TestWebPlugin::destroy): (TestWebPlugin::updateGeometry): (TestWebPlugin::parsePrimitive): (TestWebPlugin::parseColor): (TestWebPlugin::parseOpacity): (TestWebPlugin::initScene): (TestWebPlugin::drawScene): (TestWebPlugin::destroyScene): (TestWebPlugin::initProgram): (TestWebPlugin::initPrimitive): (TestWebPlugin::drawPrimitive): (TestWebPlugin::loadShader): (TestWebPlugin::loadProgram): * DumpRenderTree/chromium/TestWebPlugin.h: Added. (TestWebPlugin::scriptableObject): (TestWebPlugin::paint): (TestWebPlugin::updateFocus): (TestWebPlugin::updateVisibility): (TestWebPlugin::acceptsInputEvents): (TestWebPlugin::handleInputEvent): (TestWebPlugin::didReceiveResponse): (TestWebPlugin::didReceiveData): (TestWebPlugin::didFinishLoading): (TestWebPlugin::didFailLoading): (TestWebPlugin::didFinishLoadingFrameRequest): (TestWebPlugin::didFailLoadingFrameRequest): (TestWebPlugin::Scene::Scene): * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::createPlugin): 2011-10-17 Ryosuke Niwa [nrwt] Add new category for MISSING https://bugs.webkit.org/show_bug.cgi?id=69990 Reviewed by Dirk Pranke. Treat missing results as a new category in summerize_results. Because Chromium port wants to turn bots red when there are tests with missing results, extracted the logic to compute the exit code as exit_code_from_summarized_results in base and chromium ports. * Scripts/webkitpy/layout_tests/controllers/manager.py: * Scripts/webkitpy/layout_tests/port/base.py: * Scripts/webkitpy/layout_tests/port/chromium.py: * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: 2011-10-17 Dimitri Glazkov garden-o-matic operations fail because of CSP. https://bugs.webkit.org/show_bug.cgi?id=70249 * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: Added gardening server to the policy. 2011-10-17 Ada Chan Add a test to call WKPageLoadAlternateHTMLString() with a non-directory URL. https://bugs.webkit.org/show_bug.cgi?id=70168 Reviewed by Alexey Proskuryakov. * TestWebKitAPI/Tests/WebKit2/LoadAlternateHTMLStringWithNonDirectoryURL.cpp: Added. (TestWebKitAPI::didFinishLoadForFrame): (TestWebKitAPI::TEST): Call WKPageLoadAlternateHTMLString() with a file URL and make sure it finishes loading. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/win/TestWebKitAPI.vcproj: Add the new file. 2011-10-17 Darin Adler Add some more *.pyc ignores in directories with Python scripts. * QueueStatusServer: Added property svn:ignore. * Scripts/webkitpy/common/watchlist: Added property svn:ignore. * Scripts/webkitpy/layout_tests/controllers: Added property svn:ignore. * Scripts/webkitpy/layout_tests/views: Added property svn:ignore. 2011-10-17 Alexis Menard [Qt][WK2][meta] Fix failing API tests https://bugs.webkit.org/show_bug.cgi?id=70236 Reviewed by Csaba Osztrogonác. In order to run WebKit2 API tests we need the WebProcess to be in the path so let's add the bin directory when running the API tests of Qt. * Scripts/run-qtwebkit-tests: 2011-10-17 Kristóf Kosztyó [Qt][WK2] Build dependency problems https://bugs.webkit.org/show_bug.cgi?id=68456 Reviewed by Csaba Osztrogonác. * Scripts/webkitdirs.pm: (buildQMakeProject): 2011-10-17 Holger Hans Peter Freyther [build] Introduce Qt/SH4 buildbot https://bugs.webkit.org/show_bug.cgi?id=70227 Reviewed by Adam Roben. Add buildslave for Linux/Qt/SH4. * BuildSlaveSupport/build.webkit.org-config/config.json: 2011-10-16 Adam Barth Always enable ENABLE(DOM_STORAGE) https://bugs.webkit.org/show_bug.cgi?id=70189 Reviewed by Eric Seidel. * Scripts/build-webkit: * waf/build/settings.py: 2011-10-15 Adam Barth Rename ENABLE(TILED_BACKING_STORE) to USE(TILED_BACKING_STORE) https://bugs.webkit.org/show_bug.cgi?id=70194 Reviewed by Daniel Bates. * Scripts/build-webkit: 2011-10-15 Ojan Vafai run-webkit-tests should accept --child-processes https://bugs.webkit.org/show_bug.cgi?id=69713 Reviewed by Adam Barth. Only set --child-processes=1 if --child-processes it not passed on the commandline. * Scripts/run-webkit-tests: 2011-10-15 Daniel Bates Extract common gdb code into its own function; Remove script gdb-safari https://bugs.webkit.org/show_bug.cgi?id=68499 Reviewed by David Kilzer. Almost identical code is used in webkitdirs::{debugMiniBrowser, debugWebKitTestRunner, runSafari}() and in the script gdb-safari to launch gdb to debug an application. We should extract the common parts into a function that can be shared by all callers so as to remove duplicate code. As a side effect of removing duplicate code we can remove the script gdb-safari since its code has been incorporated into webkitdirs::debugSafari() which is called by the script debug-safari. * Scripts/debug-safari: Modified to call webkitdirs::debugSafari(). * Scripts/gdb-safari: Removed. * Scripts/webkitdirs.pm: (execMacWebKitAppForDebugging): Added; Extracted common debugging code from debugMiniBrowser(), debugWebKitTestRunner(), runSafari() and script gdb-safari to here. (debugSafari): Added. (runSafari): Extracted debugging code to debugSafari() and debugMacWebKitApp(); Also, return the exit status of WebKit launcher (WebKit.exe) on Windows. Currently, runSafari() always returns 1 on Windows and the script run-safari exits using this value as its exit status. Hence, the script run-safari always exits with a nonzero exit status (i.e. abnormal termination) on Windows. Instead, we should return the exit status of the WebKit launcher application. (debugMiniBrowser): Modified to call debugMacWebKitApp(). (debugWebKitTestRunner): Ditto. 2011-10-15 Robert Hogan [webkit-patch] Put Source/Tools changes at the top of patches, LayoutTests at the bottom https://bugs.webkit.org/show_bug.cgi?id=70056 Reviewed by Adam Barth. The preferred file-order for patches is LayoutTests at the end and ChangeLogs before everything else in a subdirectory. Specify this in a git 'orderfile' and use it when creating patches for review and landing. * Scripts/webkitpy/common/checkout/scm/git.py: * Scripts/webkitpy/common/checkout/scm/scm_unittest.py: * Scripts/webkitpy/common/config/orderfile: Added. 2011-10-15 Laszlo Gombos [Qt] [Symbian] Remove support for the Symbian platform for the QtWebKit port https://bugs.webkit.org/show_bug.cgi?id=69920 Reviewed by Kenneth Rohde Christiansen. * DumpRenderTree/qt/DumpRenderTree.pro: * DumpRenderTree/qt/DumpRenderTreeQt.cpp: (WebCore::DumpRenderTree::open): * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro: * QtTestBrowser/QtTestBrowser.pro: * QtTestBrowser/launcherwindow.cpp: (LauncherWindow::screenshot): (LauncherWindow::showFPS): (LauncherWindow::updateFPS): * QtTestBrowser/launcherwindow.h: (WindowOptions::WindowOptions): * QtTestBrowser/mainwindow.cpp: (MainWindow::buildUI): * Scripts/webkitpy/style/checker.py: Remove rules for the WebKit/qt/symbian/platformplugin directory. * Scripts/webkitpy/style/checker_unittest.py: * Tools.pro: * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: * WebKitTestRunner/InjectedBundle/qt/LayoutTestControllerQt.cpp: (WTR::LayoutTestController::platformInitialize): * WebKitTestRunner/qt/TestControllerQt.cpp: (WTR::isExistingLibrary): * WebKitTestRunner/qt/WebKitTestRunner.pro: 2011-10-15 Adam Barth Fix typo in webkitpy help https://bugs.webkit.org/show_bug.cgi?id=70169 Reviewed by Eric Seidel. * Scripts/webkitpy/tool/commands/upload.py: 2011-10-15 Tony Chang fix another typo in treemap.html https://bugs.webkit.org/show_bug.cgi?id=70149 Reviewed by Ojan Vafai. Fixes a bug where #treemapfocus=LayoutTests%2Fhttp didn't focus the http tests. * TestResultServer/static-dashboards/treemap.html: 2011-10-14 Adam Barth Hide status bubbles from queues which are very far behind https://bugs.webkit.org/show_bug.cgi?id=70133 Reviewed by Eric Seidel. The mac-ews is so far behind that we shouldn't show its status bubble. The status bubble just makes contributors sad. * QueueStatusServer/handlers/statusbubble.py: 2011-10-14 Dimitri Glazkov Write unit test to ensure style-checker-filter option is passed properly in webkit-patch https://bugs.webkit.org/show_bug.cgi?id=70143 Reviewed by Adam Barth. * Scripts/webkitpy/tool/commands/download_unittest.py: Added a test for land-cowboy. Not so cowboy after all. * Scripts/webkitpy/tool/steps/checkstyle.py: Fixed a bug, discovered by the newly added test. 2011-10-14 David Levin watchlist: Add a validation check for the email names in the list. https://bugs.webkit.org/show_bug.cgi?id=70154 Reviewed by Adam Barth. If an email is listed which doesn't have a bugzilla log in, then any emails added to the bug along with that email will fail to happen, so try to prevent this from happening by validating that emails added to the watchlist are known to be contributors. * Scripts/webkitpy/common/config/committers.py: Add my other email aliases for bugzilla. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Add the check. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: A unit test for the check. 2011-10-14 David Levin watchlist: If the style check fails, then the watchlist will not be run. https://bugs.webkit.org/show_bug.cgi?id=69484 Reviewed by Adam Barth. * Scripts/webkitpy/tool/commands/queues.py: Run the watch list even if the style part fails and don't allow watch list failures turn the bot run red. * Scripts/webkitpy/tool/commands/queues_unittest.py: Appropriate unit tests. * Scripts/webkitpy/tool/mocktool.py: Add support to make an executive command throw. 2011-10-14 Dimitri Glazkov Plumb style-checker filter up to command options and make land-cowboy use it. https://bugs.webkit.org/show_bug.cgi?id=70119 Reviewed by Adam Barth. * Scripts/webkitpy/style/checker.py: Added ChangeLogChecker categories to the big list. * Scripts/webkitpy/style/checkers/changelog.py: Declared categories, used in ChangeLogChecker. * Scripts/webkitpy/tool/commands/download.py: Made LandCowboy use filter. * Scripts/webkitpy/tool/commands/download_unittest.py: Tweaked the test. * Scripts/webkitpy/tool/commands/upload_unittest.py: Ditto. * Scripts/webkitpy/tool/steps/checkstyle.py: Made check_style_filter work. * Scripts/webkitpy/tool/steps/options.py: Added new option, check-style-filter. 2011-10-14 Sam Weinig Remove the Leopard slaves. Reviewed by Adam Roben. * BuildSlaveSupport/build.webkit.org-config/config.json: 2011-10-14 Peter Beverloo [Chromium] Inherit settings from Chromium's envsetup.sh, address a NDK todo https://bugs.webkit.org/show_bug.cgi?id=70028 Reviewed by Adam Barth. * Scripts/update-webkit-chromium: * Scripts/webkitdirs.pm: (buildChromiumMakefile): (buildChromium): 2011-10-14 Leandro Pereira Unreviewed; add missing #includes. * DumpRenderTree/efl/DumpRenderTreeChrome.cpp: * DumpRenderTree/efl/EventSender.cpp: 2011-10-14 Raphael Kubo da Costa [EFL] Correctly get the absolute URL in LayoutTestController::queueLoad() https://bugs.webkit.org/show_bug.cgi?id=69770 Reviewed by Antonio Gomes. We were just appending the given URL to the absolute URI of the main frame, which resulted in "file:///foo/bar/baz.htmltheother/url.html" instead of "file:///foo/bar/theother/url.html". This should make fast/dom/navigation-type-navigate.html behave better (it still fails, but later). * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::queueLoad): 2011-10-13 Arko Saha Microdata: Basic implementation of document.getItems() method. https://bugs.webkit.org/show_bug.cgi?id=68610 Reviewed by Ryosuke Niwa. Added ENABLE(MICRODATA) feature flag. Implement document.getItems() DOM API. Spec : http://www.whatwg.org/specs/web-apps/current-work/complete/microdata.html * Scripts/build-webkit: 2011-10-14 Csaba Osztrogonác new-run-webkit-tests does not support qt-4.8 results https://bugs.webkit.org/show_bug.cgi?id=64071 Roll-back second part of r97252. :) * Scripts/webkitpy/layout_tests/port/qt.py: 2011-10-14 Raphael Kubo da Costa [EFL] DRT: Do not use OwnFastMallocPtr to manage char*'s. https://bugs.webkit.org/show_bug.cgi?id=70106 Reviewed by Antonio Gomes. OwnFastMallocPtr was being used as a smart pointer that automatically called free() on the strings returned by the EFL or by ewk. However, when WTF is built in release mode, it uses its own memory management code instead of using the system malloc(), free() and friends. This means bad things will happen when one uses WTF's free() on memory allocated with system malloc() by the EFL or ewk. The easiest way to solve this is to call free() ourselves. * DumpRenderTree/efl/DumpRenderTree.cpp: (dumpFramesAsText): (getFinalTestURL): 2011-10-14 Raphael Kubo da Costa [EFL] Add DumpRenderTreeSupportEfl https://bugs.webkit.org/show_bug.cgi?id=68458 Reviewed by Kenneth Rohde Christiansen. Make the calls in DumpRenderTree use DumpRenderTreeSupportEfl. * DumpRenderTree/efl/DumpRenderTreeChrome.cpp: (DumpRenderTreeChrome::initialize): Turn on mock scrollbars. (DumpRenderTreeChrome::resetDefaultsToConsistentValues): (DumpRenderTreeChrome::onWindowObjectCleared): (DumpRenderTreeChrome::onDocumentLoadFinished): * DumpRenderTree/efl/GCControllerEfl.cpp: (GCController::collect): (GCController::collectOnAlternateThread): (GCController::getJSObjectCount): * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::counterValueForElementById): (LayoutTestController::pageNumberForElementById): (LayoutTestController::numberOfPages): (LayoutTestController::workerThreadCount): (LayoutTestController::findString): (LayoutTestController::pauseAnimationAtTimeOnElementWithId): (LayoutTestController::pauseTransitionAtTimeOnElementWithId): (LayoutTestController::sampleSVGAnimationForElementAtTime): (LayoutTestController::numberOfActiveAnimations): (LayoutTestController::suspendAnimations): (LayoutTestController::resumeAnimations): * DumpRenderTree/efl/PixelDumpSupportEfl.cpp: (createBitmapContextFromWebView): 2011-10-14 Csaba Osztrogonác [Qt] Enable NRWT for Qt 4.8 version too, because it is supported after r97461. Unreviewed. * Scripts/run-webkit-tests: (useNewRunWebKitTests): 2011-10-14 Csaba Osztrogonác new-run-webkit-tests does not support qt-4.8 results https://bugs.webkit.org/show_bug.cgi?id=64071 Roll-back r97252 with unit test fix. Reviewed by Csaba Osztrogonác. * Scripts/webkitpy/layout_tests/port/qt.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: 2011-10-13 Ojan Vafai Make filter-build-webkit exit 1 if the build fails. https://bugs.webkit.org/show_bug.cgi?id=70080 Reviewed by Eric Seidel. This only works if you redirect stderr to stdout, but it at least makes it possible. This is useful if you want to run a followup command (e.g. run-webkit-tests) but only if the build succeeds. * Scripts/filter-build-webkit: 2011-10-13 Dimitri Glazkov land-cowboy should at least check style. https://bugs.webkit.org/show_bug.cgi?id=70073 Reviewed by Adam Barth. * Scripts/webkitpy/tool/commands/download.py: Added CheckStyle step to LandCowboy command. 2011-10-13 Tony Chang fix regex in layout test times dashboard https://bugs.webkit.org/show_bug.cgi?id=70059 This fixes the treemapfocus query param. Reviewed by Ojan Vafai. * TestResultServer/static-dashboards/treemap.html: 2011-10-13 Beth Dakin https://bugs.webkit.org/show_bug.cgi?id=70050 DRT and WRT should have HiDPI testing capabilities Reviewed by Darin Adler. New layoutTestController function setBackingScaleFactor takes a double for a scale factor and a callback function to call once the backing scale factor has successfully been set for the view. * DumpRenderTree/LayoutTestController.cpp: (setBackingScaleFactorCallback): (LayoutTestController::staticFunctions): On Mac, setBackingScaleFactor calls WebView _setCustomBackingScaleFactor. Other platforms do nothing. * DumpRenderTree/LayoutTestController.h: * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::setBackingScaleFactor): * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setBackingScaleFactor): * DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::setBackingScaleFactor): Reset the backing scale factor by calling _setCutsomBackingScaleFactor with a value of 0, which resets to the system default. * DumpRenderTree/mac/DumpRenderTree.mm: (resetWebViewToConsistentStateBeforeTesting): When generating the pixel dump, create a Bitmap of the appropriate size based on the backing scale factor. If it is a HiDPI view, paint into the context using displayRectIgnoringOpacity. * DumpRenderTree/mac/PixelDumpSupportMac.mm: (createBitmapContextFromWebView): And now the same stuff for WK2. Again, new layoutTestController function setBackingScaleFactor takes a double for a scale factor and a callback function. * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::LayoutTestController::setBackingScaleFactor): (WTR::LayoutTestController::callSetBackingScaleFactorCallback): * WebKitTestRunner/InjectedBundle/LayoutTestController.h: The InjectedBundle sends a message to the UI process to set the backing scale factor. It also processes a message the UI process will send back once it completes this task, and at that time, the InjectedBundle calls the callback. * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::didReceiveMessage): (WTR::InjectedBundle::postSetBackingScaleFactor): * WebKitTestRunner/InjectedBundle/InjectedBundle.h: Again, reset the backing scale factor by calling WKPageSetCustomBackingScaleFactor with a value of 0, which resets to the system default. * WebKitTestRunner/TestController.cpp: (WTR::TestController::resetStateToConsistentValues): Upon receiving the SetBackingScaleFactor message, call WKPageSetCustomBackingScaleFactor to the given scale factor, and post a message back to InjectedBundle to say it has been done. * WebKitTestRunner/TestInvocation.cpp: (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): Don't use window snapshotting for the HiDPI tests. We don't have a way to fake a window's scale factor so we have to avoid that. * WebKitTestRunner/cg/TestInvocationCG.cpp: (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): 2011-10-13 John Knottenbelt [Chromium] Allow building without speech input enabled. https://bugs.webkit.org/show_bug.cgi?id=69867 Reviewed by Darin Fisher. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::addMockSpeechInputResult): 2011-10-13 Kent Tamura [Chromium] A DRT fix for r97348. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::overridePreference): Ignore overridePreference() for "WebKitWebAudioEnabled", and don't print an error message. 2011-10-12 Lucas Forschler Update kill-old-processes logic. https://bugs.webkit.org/show_bug.cgi?id=63651 Reviewed by Ryosuke Niwa. Remove unused mac platform specific folder. Update master.config to use new kill-old-processes. * BuildSlaveSupport/build.webkit.org-config/master.cfg: * BuildSlaveSupport/kill-old-processes: * BuildSlaveSupport/mac: Removed. 2011-10-12 Ryosuke Niwa [NRWT] New tests without expected files are reported as flakey tests https://bugs.webkit.org/show_bug.cgi?id=67268 Reviewed by Dirk Pranke. The bug was caused by summarize_results's regarding all tests that have an entry in result_summary and no entry in retry_summary as flaky. Since we don't retry missing tests, there will be no entries for missing results in retry_summary. Fixed the bug by treating missing results separately. * Scripts/webkitpy/layout_tests/controllers/manager.py: * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: 2011-10-12 Adam Barth Remove ENABLE(XHTMLMP) and associated code https://bugs.webkit.org/show_bug.cgi?id=69729 Reviewed by David Levin. * Scripts/build-webkit: * Scripts/old-run-webkit-tests: * Scripts/webkitperl/features.pm: * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: 2011-10-12 Ryosuke Niwa Kill old run-webkit-tests processes in kill-process on Mac https://bugs.webkit.org/show_bug.cgi?id=69976 Reviewed by Dirk Pranke. Add a system call to remove all python processes running run-webkit-tests. * BuildSlaveSupport/kill-old-processes: * BuildSlaveSupport/mac: Removed. 2011-10-12 Simon Fraser Many WK2 snapshots are blank https://bugs.webkit.org/show_bug.cgi?id=69953 Reviewed by Beth Dakin. Two fixes: 1. negating a size_t in CGContextTranslateCTM() fails, because size_t is unsigned. Fix by casing to CGFloat before negation. 2. Turn on window autodisplay, so that compositing tests snapshot correctly. Autodisplay is necessary to kick off Core Animation rendering. * WebKitTestRunner/cg/TestInvocationCG.cpp: (WTR::createCGContextFromImage): * WebKitTestRunner/mac/PlatformWebViewMac.mm: (WTR::PlatformWebView::PlatformWebView): 2011-10-12 Adam Barth commit-queue doesn't have a friendly error message when the reviewer line is messed up https://bugs.webkit.org/show_bug.cgi?id=69979 Reviewed by Eric Seidel. Rather than combining the ChangeLog validation with a more complicated command, this patch has the commit-queue run it as a separate command, which will give us more control over the error message. * Scripts/webkitpy/tool/bot/commitqueuetask.py: * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py: * Scripts/webkitpy/tool/commands/download.py: * Scripts/webkitpy/tool/commands/queues_unittest.py: * Scripts/webkitpy/tool/steps/validatechangelogs.py: * Scripts/webkitpy/tool/steps/validatereviewer.py: 2011-10-12 Eric Seidel Layout tests asserting in LayoutTestController::pathToLocalResource() https://bugs.webkit.org/show_bug.cgi?id=69898 Reviewed by Simon Fraser. old-run-webkit-tests just didn't know to set LOCAL_RESOURCE_ROOT, oops. * Scripts/old-run-webkit-tests: (openDumpTool): 2011-10-12 Adam Barth The commit-queue doesn't tell me which tests my patch broke https://bugs.webkit.org/show_bug.cgi?id=69975 Reviewed by Eric Seidel. We now print the list of newly failing tests, like we do for the EWS. * Scripts/webkitpy/tool/commands/queues.py: * Scripts/webkitpy/tool/commands/queues_unittest.py: 2011-10-10 Alice Boxhall Added myself as a committer. * Scripts/webkitpy/common/config/committers.py: 2011-10-12 Adam Barth _run_tests is a confusing variable name in webkitpy https://bugs.webkit.org/show_bug.cgi?id=69971 Reviewed by Eric Seidel. Eric and I thought this was a function. * Scripts/webkitpy/tool/bot/earlywarningsystemtask.py: 2011-10-12 Tony Chang Fix test-webkitpy after r97307 and r97293. * Scripts/webkitpy/layout_tests/port/chromium.py: options.time_out_ms is None during tests so add back the check. * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py: The output directory is now build/Debug. 2011-10-12 Tony Chang [chromium] remove old build dirs on chromium-win https://bugs.webkit.org/show_bug.cgi?id=69961 Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/chromium_win.py: Output is now always in build and never in chrome or webkit. 2011-10-12 Dimitri Glazkov garden-o-matic's "Expect Failure" button does not work correctly, so we should remove it until the plumbing is fixed to avoid landmines. https://bugs.webkit.org/show_bug.cgi?id=69954 Also renamed the action and associated event name to address an FIXME. Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: Renamed event name. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js: Renamed the action. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: Removed action temporarily. 2011-10-12 Tony Chang stop spamming 'stopping test driver timed out, killing it' when running rwt https://bugs.webkit.org/show_bug.cgi?id=69938 Reviewed by Adam Barth. In http://trac.webkit.org/changeset/95875 , we made the timeout for stopping the driver based on time_out_ms, but we based it on the wrong default times. * Scripts/webkitpy/layout_tests/port/chromium.py: 2011-10-12 Simon Fraser 3D transforms are flattened in WebKit2 snapshots https://bugs.webkit.org/show_bug.cgi?id=68276 Reviewed by Adam Roben. Fix WKImage leak noticed by Adam. * WebKitTestRunner/mac/PlatformWebViewMac.mm: (WTR::PlatformWebView::windowSnapshotImage): 2011-10-12 Dimitri Glazkov garden-o-matic should color-code test failure types. https://bugs.webkit.org/show_bug.cgi?id=69945 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: Added an extra span for styling. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: Updated tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Ditto. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: Added color-coding. 2011-10-12 Adam Barth webkitpy tests are failing on Mac bots https://bugs.webkit.org/show_bug.cgi?id=69929 Reviewed by Simon Fraser. Admit that Lion exists. * Scripts/webkitpy/layout_tests/port/chromium.py: * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: 2011-10-12 Simon Fraser Improve output of run-api-tests to make time-outs more obvious https://bugs.webkit.org/show_bug.cgi?id=69788 Reviewed by Adam Roben. List the tests that failed and timed out at the bottom of the output, if run with --verbose. * Scripts/run-api-tests: (runAllTests): (runTest): 2011-10-12 Sheriff Bot Unreviewed, rolling out r97252. http://trac.webkit.org/changeset/97252 https://bugs.webkit.org/show_bug.cgi?id=69944 Broke test-webkpy (Requested by abarth on #webkit). * Scripts/webkitpy/layout_tests/port/qt.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: 2011-10-12 Kristóf Kosztyó new-run-webkit-tests does not support qt-4.8 results https://bugs.webkit.org/show_bug.cgi?id=64071 Reviewed by Csaba Osztrogonác. * Scripts/webkitpy/layout_tests/port/qt.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: 2011-10-10 Martin Robinson [GTK] [WebKit2] Make adding another unit test easier https://bugs.webkit.org/show_bug.cgi?id=69409 Reviewed by Gustavo Noronha Silva. * Scripts/run-gtk-tests: Update run-gtk-tests to be able to execute new WebKit2 tests. 2011-10-11 Takashi Toyoshima [WebSocket] update pywebsocket to 0.6b6. https://bugs.webkit.org/show_bug.cgi?id=69824 Reviewed by Kent Tamura. WebKit supports WebSocket with version 8 based protocol for now. We should update its protocol to version 13 based one like hybi-17. pywebsocket 0.6b6 supports both of version 8 and 13. * Scripts/webkitpy/thirdparty/__init__.py: 2011-10-11 Simon Fraser WebKitTestRunner needs to link with WebKit for NSURL category methods https://bugs.webkit.org/show_bug.cgi?id=69892 Reviewed by Sam Weinig. WebDragClient::declareAndWriteDragImage() was throwing an Obj-C exception on 10.6 because the +_web_originalDataAsString selector on NSURL was not recognized. Link with WebKit.framework to fix this. * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: 2011-10-11 Adam Barth sheriffbot takes too long to acknowledge rollout commands https://bugs.webkit.org/show_bug.cgi?id=69871 Reviewed by Eric Seidel. We used to update the working copy before acknowledging the command because we wanted to ping all the relevant IRC nicks. That's caused a bunch of frustration because folks don't know whether the bot has heard their commands. This patch makes the bot reply immediately before updating the working copy. All the relevenat folks are still pinged when the bot finishes preparing the rollout. * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py: 2011-10-11 Adam Barth test-webkitpy fails on Lion https://bugs.webkit.org/show_bug.cgi?id=69873 Reviewed by Eric Seidel. This patch fixes one of the two failures. I'll need to track down the other failure another time. * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py: 2011-10-11 Adam Barth garden-o-matic should warn me when run-webkit-tests errors out (e.g., due to python screw-ups) https://bugs.webkit.org/show_bug.cgi?id=65230 Reviewed by Dimitri Glazkov. This patch causes us to treat crashed or hung run-webkit-tests runs similarly to how we treat compile failures. In some sense, they're similar because they prevent us from getting test coverage. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js: 2011-10-11 Adam Barth Failing builder display in garden-o-matic should have the same visual style as the failure grid https://bugs.webkit.org/show_bug.cgi?id=69876 Reviewed by Dimitri Glazkov. This patch generalizes the view we use for failing builders in the FailureGrid so that we can use it for the list of builders that aren't compiling. One downside of this patch is that it makes it harder to tell when debug builders are failing (as opposed to release builders) because that information isn't presented in this view. A future patch will improve this aspect. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/failures.css: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: 2011-10-11 Filip Pizlo Some API tests are timing out in Release https://bugs.webkit.org/show_bug.cgi?id=69787 Reviewed by Mark Rowe. NDEBUG must be set in release builds. Otherwise the world breaks. Also added some magic to ensure that GTest builds in release mode in clang with NDEBUG set. * TestWebKitAPI/Configurations/Base.xcconfig: * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/config.h: 2011-10-11 Adam Barth garden-o-matic links to flakiness dashboard should open in a new tab https://bugs.webkit.org/show_bug.cgi?id=69864 Reviewed by Dimitri Glazkov. This bug was obvious in the first five minutes of using this new feature. :) * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: 2011-10-11 Adam Barth Tweak the Content-Security-Policy for garden-o-matic. It turns out that * doesn't really mean "all". I've emailed the W3C working group about changing the spec. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: 2011-10-11 Adam Barth garden-o-matic should link to the flakiness dashboard https://bugs.webkit.org/show_bug.cgi?id=69862 Reviewed by Dimitri Glazkov. The flakiness dashboard has lots of great information about the failure history of each test. This patch adds a link from garden-o-matic to the flakiness dashboard so the gardener can look at this information. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: 2011-10-11 David Levin watchlist: assertRaisesRegexp implementation should be compatible with python 2.7. https://bugs.webkit.org/show_bug.cgi?id=69789 Reviewed by Eric Seidel. Made the signature of _assertRaisesRegexp match that of Python 2.7. * Scripts/webkitpy/common/watchlist/watchlistloader_unittest.py: Adjusted to be compatible with Python 2.7. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Ditto. * Scripts/webkitpy/common/webkitunittest.py: Ditto. * Scripts/webkitpy/tool/commands/applywatchlistlocal_unittest.py: Ditto. 2011-10-11 Raphael Kubo da Costa [EFL] DRT: Do not abort() when an unknown key is pressed. https://bugs.webkit.org/show_bug.cgi?id=69856 Reviewed by Antonio Gomes. A lot of tests crash due to the previous behaviour. Just using the received name of the pressed key works most of the time, and when it does not the failing tests should tell us. * DumpRenderTree/efl/EventSender.cpp: (keyPadNameFromJSValue): (keyNameFromJSValue): (keyDownCallback): 2011-10-11 Kristóf Kosztyó [Qt] [WK2] NRWT failed to launch MiniBrowser after test run https://bugs.webkit.org/show_bug.cgi?id=68016 Reviewed by Chang Shu. * Scripts/webkitpy/layout_tests/port/qt.py: 2011-10-11 Raphael Kubo da Costa [EFL] DRT: Return the right window count in LayoutTestController. https://bugs.webkit.org/show_bug.cgi?id=69764 Reviewed by Antonio Gomes. Follow-up to r96943: now that we have DumpRenderTreeChrome and DumpRenderTreeView in the tree, we can correctly count the number of open windows in DRT. * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::windowCount): 2011-10-11 Jer Noble REGRESSION (r96770-r96777): fast/dom/Window/window-properties.html, fast/dom/prototype-inheritance-2.html, fast/js/global-constructors.html failing on SnowLeopard Intel Release (Tests) https://bugs.webkit.org/show_bug.cgi?id=69800 Reviewed by Csaba Osztrogonác. Enable WEB_AUDIO at runtime. * DumpRenderTree/mac/DumpRenderTree.mm: (resetDefaultsToConsistentValues): 2011-10-11 Raphael Kubo da Costa [EFL] DRT: Make sure IconDatabase is closed before trying to open it again. https://bugs.webkit.org/show_bug.cgi?id=69769 Reviewed by Antonio Gomes. * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::setIconDatabaseEnabled): 2011-10-11 Tor Arne Vestbø [Qt] Remove all references to QTDIR_build and standalone_package Qt is now modularized, which means we no longer import WebKit into the Qt source tree. Instead we use git submodules, and building QtWebKit as "part of Qt" is really building QtWebKit as from trunk. To decrease the number of buildsystem configurations we also remove the standalone_package code-path used when we were providing tarballs with the derived sources pre-generated. Reviewed by Simon Hausmann. * MiniBrowser/qt/MiniBrowser.pro: * QtTestBrowser/QtTestBrowser.pro: * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: 2011-10-11 Dominic Mazzoni Update Chromium accessibility support in DRT. https://bugs.webkit.org/show_bug.cgi?id=69676 Reviewed by Chris Fleizach. Adds several missing properties and methods to enable more tests to pass, and in particular adds support for notification callbacks. * DumpRenderTree/chromium/AccessibilityController.cpp: (AccessibilityController::AccessibilityController): (AccessibilityController::reset): (AccessibilityController::getFocusedElement): (AccessibilityController::shouldLogAccessibilityEvents): (AccessibilityController::notificationReceived): (AccessibilityController::logAccessibilityEventsCallback): * DumpRenderTree/chromium/AccessibilityController.h: * DumpRenderTree/chromium/AccessibilityUIElement.cpp: (getHelpText): (getStringValue): (getOrientation): (AccessibilityUIElement::AccessibilityUIElement): (AccessibilityUIElement::getChildAtIndex): (AccessibilityUIElement::isEqual): (AccessibilityUIElement::notificationReceived): (AccessibilityUIElement::roleGetterCallback): (AccessibilityUIElement::titleGetterCallback): (AccessibilityUIElement::descriptionGetterCallback): (AccessibilityUIElement::helpTextGetterCallback): (AccessibilityUIElement::stringValueGetterCallback): (AccessibilityUIElement::xGetterCallback): (AccessibilityUIElement::yGetterCallback): (AccessibilityUIElement::widthGetterCallback): (AccessibilityUIElement::heightGetterCallback): (AccessibilityUIElement::intValueGetterCallback): (AccessibilityUIElement::minValueGetterCallback): (AccessibilityUIElement::maxValueGetterCallback): (AccessibilityUIElement::childrenCountGetterCallback): (AccessibilityUIElement::insertionPointLineNumberGetterCallback): (AccessibilityUIElement::selectedTextRangeGetterCallback): (AccessibilityUIElement::isEnabledGetterCallback): (AccessibilityUIElement::isRequiredGetterCallback): (AccessibilityUIElement::isFocusedGetterCallback): (AccessibilityUIElement::isFocusableGetterCallback): (AccessibilityUIElement::isSelectedGetterCallback): (AccessibilityUIElement::isSelectableGetterCallback): (AccessibilityUIElement::isMultiSelectableGetterCallback): (AccessibilityUIElement::isExpandedGetterCallback): (AccessibilityUIElement::isCheckedGetterCallback): (AccessibilityUIElement::isVisibleGetterCallback): (AccessibilityUIElement::isOffScreenGetterCallback): (AccessibilityUIElement::isCollapsedGetterCallback): (AccessibilityUIElement::hasPopupGetterCallback): (AccessibilityUIElement::isValidGetterCallback): (AccessibilityUIElement::orientationGetterCallback): (AccessibilityUIElement::showMenuCallback): (AccessibilityUIElement::pressCallback): (AccessibilityUIElement::isEqualCallback): (AccessibilityUIElement::addNotificationListenerCallback): (AccessibilityUIElement::removeNotificationListenerCallback): (AccessibilityUIElement::takeFocusCallback): (AccessibilityUIElement::fallbackCallback): (RootAccessibilityUIElement::getChildAtIndex): (AccessibilityUIElementList::getOrCreate): * DumpRenderTree/chromium/AccessibilityUIElement.h: * DumpRenderTree/chromium/CppVariant.cpp: (CppVariant::invokeDefault): * DumpRenderTree/chromium/CppVariant.h: * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (LayoutTestController::reset): * DumpRenderTree/chromium/LayoutTestController.h: * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::postAccessibilityNotification): 2011-10-10 Adam Klein Support Distributed-CompileC as an action name in filter-build-webkit https://bugs.webkit.org/show_bug.cgi?id=69783 Reviewed by Ojan Vafai. * Scripts/filter-build-webkit: 2011-10-10 Simon Fraser 3D transforms are flattened in WebKit2 snapshots https://bugs.webkit.org/show_bug.cgi?id=68276 Reviewed by Anders Carlsson. Use a snapshot of the window, rather than an image from the web process for pixel testing in WebKitTestRunner. This correctly captures compositing layers, so works for 3D transform tests. * WebKitTestRunner/PlatformWebView.h: Add a windowSnapshotImage() method. * WebKitTestRunner/cg/TestInvocationCG.cpp: (WTR::createCGContextFromImage): Add a param to optionally flip the image when drawing. (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): Call windowSnapshotImage(), and if it returns an image, use that instead of the image we get from the web process. * WebKitTestRunner/gtk/PlatformWebViewGtk.cpp: (WTR::PlatformWebView::windowSnapshotImage): Stub out the new method. * WebKitTestRunner/mac/PlatformWebViewMac.mm: (WTR::PlatformWebView::windowSnapshotImage): Ditto. * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::windowSnapshotImage): Ditto. * WebKitTestRunner/win/PlatformWebViewWin.cpp: (WTR::PlatformWebView::windowSnapshotImage): Ditto. 2011-10-10 David Levin Quick fix for watchlist unit test failures when using Python 2.7+ Reviewed by Adam Barth. Filed bug https://bugs.webkit.org/show_bug.cgi?id=69789 about a better fix. * Scripts/webkitpy/common/webkitunittest.py: Always use our version of assertRaisesRegexp. 2011-10-10 Sheriff Bot Unreviewed, rolling out r97082. http://trac.webkit.org/changeset/97082 https://bugs.webkit.org/show_bug.cgi?id=69784 broke Windows build (Requested by smfr on #webkit). * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: * WebKitTestRunner/cg/TestInvocationCG.cpp: Renamed from Tools/WebKitTestRunner/cg/TestInvocationCG.mm. (WTR::createCGContextFromImage): (WTR::computeMD5HashStringForContext): (WTR::dumpBitmap): (WTR::paintRepaintRectOverlay): (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): * WebKitTestRunner/mac/PlatformWebViewMac.mm: (WTR::PlatformWebView::PlatformWebView): 2011-10-10 Simon Fraser 3D transforms are flattened in WebKit2 snapshots https://bugs.webkit.org/show_bug.cgi?id=68276 Reviewed by Anders Carlsson. The WKImage passed from the WebProcess when pixel dumping is created by painting compositing layers flattened, which flattens 3D transforms. Instead, use CGWindowListCreateImage() to get a snapshot of the window when doing pixel tests. Also allow the window to auto-display, so that Core Animation layers are rendered, and animate. * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: TestInvocationCG became a .mm file * WebKitTestRunner/cg/TestInvocationCG.mm: Renamed from Tools/WebKitTestRunner/cg/TestInvocationCG.cpp. (WTR::createBitmapCGContext): Utility function to create a bitmap context. (WTR::computeMD5HashStringForContext): (WTR::dumpBitmap): (WTR::paintRepaintRectOverlay): (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): Use CGWindowListCreateImage() to grap a snapshot of the window, rather than relying on the WKImage which was passed from the web process. * WebKitTestRunner/mac/PlatformWebViewMac.mm: (WTR::PlatformWebView::PlatformWebView): Remove the line that turned off autoDisplay for the NSWindow. We need the window to autodisplay for Core Animation to render layers and start animations. 2011-10-10 Simon Fraser WebKitTestRunner needs layoutTestController.setWindowIsKey https://bugs.webkit.org/show_bug.cgi?id=42688 Reviewed by Anders Carlsson. Implement layoutTestController.setWindowIsKey() in WebKitTestRunner. This sets a flag that the platform layer uses to indicate that the window is the key window, which fixes focus ring drawing in pixel results. Also focus the web view when we get a 'didCommitLoad' for the main frame, as DumpRenderTree does. * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::postSetWindowIsKey): * WebKitTestRunner/InjectedBundle/InjectedBundle.h: * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::cacheLayoutTestControllerCallback): Replace some magic numbers with an enum. (WTR::LayoutTestController::addChromeInputField): (WTR::LayoutTestController::removeChromeInputField): (WTR::LayoutTestController::focusWebView): (WTR::LayoutTestController::setWindowIsKey): (WTR::LayoutTestController::callAddChromeInputFieldCallback): (WTR::LayoutTestController::callRemoveChromeInputFieldCallback): (WTR::LayoutTestController::callFocusWebViewCallback): * WebKitTestRunner/InjectedBundle/LayoutTestController.h: * WebKitTestRunner/PlatformWebView.h: Add an NSWindow subclass so that we can override -isKeyWindow. Use a boolean member variable to track window key status. (WTR::PlatformWebView::setWindowIsKey): (WTR::PlatformWebView::windowIsKey): * WebKitTestRunner/TestController.cpp: (WTR::TestController::runModal): Change the window key status when showing a modal alert, as DRT does. (WTR::closeOtherPage): Use the same cast as other functions. (WTR::focus): Implement to call setWindowIsKey(true) (WTR::unfocus): Implement to call setWindowIsKey(false) (WTR::TestController::createOtherPage): Add page callbacks for focus and unfocus. (WTR::TestController::initialize): (WTR::TestController::didCommitLoadForFrame): Hook up callback to focus the web view, as DRT does. * WebKitTestRunner/TestController.h: * WebKitTestRunner/TestInvocation.cpp: (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle): * WebKitTestRunner/gtk/PlatformWebViewGtk.cpp: (WTR::PlatformWebView::PlatformWebView): * WebKitTestRunner/mac/PlatformWebViewMac.mm: (-[WebKitTestRunnerWindow isKeyWindow]): (WTR::PlatformWebView::PlatformWebView): Make a WebKitTestRunnerWindow. (WTR::PlatformWebView::~PlatformWebView): Clean up the back pointer on the NSWindow subclass, in case it outlives us. (WTR::PlatformWebView::focus): Set the view as the first responder, and the window as the key window, as DRT does. * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::PlatformWebView): * WebKitTestRunner/win/PlatformWebViewWin.cpp: (WTR::PlatformWebView::PlatformWebView): 2011-10-10 Raphael Kubo da Costa check-webkit-style: Add readability exception for Ewk_* data types. https://bugs.webkit.org/show_bug.cgi?id=69763 Reviewed by David Levin. Many EFL-related files (such as EditorClientEfl.h or DumpRenderTreeSupportEfl.h) need to have declarations such as "typedef struct _Ewk_Foo Ewk_Foo", which currently produce false positives in check-webkit-style. Add an exception for them, just like there are exceptions for Evas_*, Ecore_* and others. * Scripts/webkitpy/style/checkers/cpp.py: * Scripts/webkitpy/style/checkers/cpp_unittest.py: 2011-10-10 Dominic Mazzoni [Chromium] Get rid of WebAccessibilityCache. https://bugs.webkit.org/show_bug.cgi?id=68224 Reviewed by Dimitri Glazkov. * DumpRenderTree/chromium/AccessibilityController.cpp: (AccessibilityController::bindToJavascript): 2011-10-09 Adam Barth Remove "near miss" XSS vulnerabilities in garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=69708 Reviewed by David Levin. Collin Jackson did a security audit of garden-o-matic. He hasn't found any actual vulnerabilities yet, but he did identify a couple "near misses" where a slight change in garden-o-matic could lead to a vulnerability. This patch removes those potential issues. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/Bugzilla.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: 2011-10-07 Adam Barth Add a Content-Security-Policy to garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=69668 Reviewed by David Levin. This will help mitigate cross-site script and will also help us dogfood CSP. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: 2011-10-07 Raphael Kubo da Costa [EFL] DRT: Add DumpRenderTreeChrome.{cpp,h} https://bugs.webkit.org/show_bug.cgi?id=66380 Reviewed by Antonio Gomes. This encapsulates EWebKit into an object that manages multiple windows and provides a cleaner interface to EFL's DumpRenderTree. * DumpRenderTree/efl/DumpRenderTreeChrome.cpp: Added. (DumpRenderTreeChrome::create): (DumpRenderTreeChrome::DumpRenderTreeChrome): (DumpRenderTreeChrome::~DumpRenderTreeChrome): (DumpRenderTreeChrome::createNewWindow): (DumpRenderTreeChrome::createView): (DumpRenderTreeChrome::removeWindow): (DumpRenderTreeChrome::initialize): (DumpRenderTreeChrome::extraViews): (DumpRenderTreeChrome::mainFrame): (DumpRenderTreeChrome::mainView): (DumpRenderTreeChrome::resetDefaultsToConsistentValues): (DumpRenderTreeChrome::onWindowObjectCleared): (DumpRenderTreeChrome::onLoadStarted): (DumpRenderTreeChrome::processWork): (DumpRenderTreeChrome::onLoadFinished): (DumpRenderTreeChrome::onStatusbarTextSet): (DumpRenderTreeChrome::onTitleChanged): (DumpRenderTreeChrome::onDocumentLoadFinished): * DumpRenderTree/efl/DumpRenderTreeChrome.h: Added. * DumpRenderTree/efl/EventSender.cpp: (sendClick): Use browser->mainFrame() instead of mainFrame. (mouseDownCallback): Ditto. (mouseUpCallback): Ditto. (mouseMoveToCallback): Ditto. (mouseScrollByCallback): Ditto. (keyDownCallback): Ditto. (scalePageByCallback): Ditto. * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::counterValueForElementById): Use browser->mainFrame() instead of mainFrame. (LayoutTestController::pageNumberForElementById): Ditto. (LayoutTestController::numberOfPages): Ditto. (LayoutTestController::pauseAnimationAtTimeOnElementWithId): Ditto. (LayoutTestController::pauseTransitionAtTimeOnElementWithId): Ditto. (LayoutTestController::sampleSVGAnimationForElementAtTime): Ditto. (LayoutTestController::numberOfActiveAnimations): Ditto. (LayoutTestController::suspendAnimations): Ditto. (LayoutTestController::resumeAnimations): Ditto. (LayoutTestController::setWebViewEditable): Ditto. (LayoutTestController::webHistoryItemCount): Use browser->mainView() instead of browser. (LayoutTestController::queueLoad): Ditto. (LayoutTestController::setAcceptsEditing): Ditto. (LayoutTestController::setUserStyleSheetEnabled): Ditto. (LayoutTestController::setPrivateBrowsingEnabled): Ditto. (LayoutTestController::setFrameFlatteningEnabled): Ditto. (LayoutTestController::setSpatialNavigationEnabled): Ditto. (LayoutTestController::setPluginsEnabled): Ditto. (LayoutTestController::findString): Ditto. (LayoutTestController::setDeveloperExtrasEnabled): Ditto. * DumpRenderTree/efl/PixelDumpSupportEfl.cpp: (createBitmapContextFromWebView): Use browser->mainFrame() instead of mainFrame. * DumpRenderTree/efl/WorkQueueItemEfl.cpp: (LoadItem::invoke): Ditto. (LoadHTMLStringItem::invoke): Ditto. (ScriptItem::invoke): Ditto. (ReloadItem::invoke): Use browser->mainView() instead of browser. 2011-10-07 Simon Hausmann [Qt][WK2] Touch mocking is broken with Qt 5 post refactor merge https://bugs.webkit.org/show_bug.cgi?id=69617 Reviewed by Andreas Kling. Injecting fake touch events using qt_translateRawTouchEvent does not work anymore, as it sends the touch events to widgets only. The QML view however is a QWindow. Therefore the fake touch events have to be dispatched through QWindowSystemInterface::handleTouchEvent. As a result the events appear to be spontaenous, which required adjusting the are-we-on-a-real-touch-device logic. * MiniBrowser/qt/MiniBrowserApplication.cpp: (MiniBrowserApplication::MiniBrowserApplication): (MiniBrowserApplication::notify): * MiniBrowser/qt/MiniBrowserApplication.h: 2011-10-06 Adam Barth Port PixelZoomer to work in garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=69577 Reviewed by David Levin. This implementation of PixelZoomer is basically a fork of the implementation used by results.html. These two apps have different requirements on where this code lives. results.html wants to be self-contained whereas garden-o-matic wants to live in the public_html folder of the buildbot. We could do something fancy here to share the code, but that's not likely to be of much benefit. I had to change a few things about pixelzoomer to accounter for the differences in DOM between results.html and garden-o-matic, but surprisingly little. I also moved all the code into an anonymous function, which is the style we're suing for module separation in garden-o-matic. Finally, some jQuery-ims snuck in because I didn't want to drag along too many extra functions from results.html. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/pixelzoomer.css: Added. (.pixel-zoom-container): (.pixel-zoom-container > *): (.pixel-zoom-container .scaled-image-container): (.scaled-image-container > img): 2011-10-06 Gyuyoung Kim [EFL] Change data type of parameter in ewk_view_viewport_attributes_get(). https://bugs.webkit.org/show_bug.cgi?id=69505 Reviewed by Andreas Kling. * EWebLauncher/main.c: Use *int* variables for width and height of viewport meta tag. (on_viewport_changed): 2011-10-06 Brent Fulgham [WinCairo] Unreviewed test correction. * Scripts/webkitdirs.pm: (argumentsForConfiguration): Add 'wincairo' option to the set of configuration flags used by run-javascript-core tests, so that it does not attempt to run Apple's release version. 2011-10-06 Brent Fulgham [WinCairo] Unreviewed test correction. * BuildSlaveSupport/test-result-archive: Add 'wincairo' to the set of known build slave types so that test-result-archive won't fail when running tests. 2011-10-06 Martin Robinson The GTK+ WebKit2 headers produce a lot of style warnings https://bugs.webkit.org/show_bug.cgi?id=69481 Reviewed by David Levin. Prevent emitting so many style warnings for GTK+ API. We skip header files in the WebKit2 GTK+ API directory and also avoid warnings about identifier names that begin with "webkit_" in files that contain the string "gtk". * Scripts/webkitpy/style/checker.py: Do not check header files in Source/WebKit2/UIProcess/API/gtk that do not end in Private.h. This required adding the ability to specify a regular expression in the skip list. Remove a few files from the skipped list that no longer exist. * Scripts/webkitpy/style/checker_unittest.py: Added a test for this behavior. * Scripts/webkitpy/style/checkers/cpp.py: If a path contains "gtk" don't warn about identifiers that begin with "webkit_". * Scripts/webkitpy/style/checkers/cpp_unittest.py: Added a test for this behavior. 2011-10-06 Brent Fulgham [WinCairo] Correct config.json for WinCairo Test builds. https://bugs.webkit.org/show_bug.cgi?id=69272 Reviewed by Ryosuke Niwa. * BuildSlaveSupport/build.webkit.org-config/config.json: Remove old 'WinCairo Debug (Build)' label and replace with correct 'WinCairo Release' label. 2011-10-06 David Levin Add a style checker watchlist definition and add myself to it. https://bugs.webkit.org/show_bug.cgi?id=69564 Reviewed by Daniel Bates. * Scripts/webkitpy/common/config/watchlist: 2011-10-06 Raphael Kubo da Costa [EFL] DRT: Create icon database path in LayoutTestController::setIconDatabaseEnabled. https://bugs.webkit.org/show_bug.cgi?id=69450 Reviewed by Ryosuke Niwa. This should avoid stderr messages when running webarchive/test-link-rel-icon.html. * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::setIconDatabaseEnabled): 2011-10-06 Alejandro G. Castro [WK2] [GTK] Add Skipped file for Webkit2 test runner https://bugs.webkit.org/show_bug.cgi?id=69517 Use the shared WK2 skipped file also in GTK port. Reviewed by Martin Robinson. * Scripts/old-run-webkit-tests: (readSkippedFiles): 2011-10-06 Elliot Poger allow new-run-webkit-tests to run on Lion https://bugs.webkit.org/show_bug.cgi?id=69429 Add fallback lists for chromium-mac-lion, and add chromium-mac-snowleopard directory to fallback lists (even though that directory doesn't exist yet) Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/chromium_mac.py: * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py: 2011-10-03 Hans Wennborg IndexedDB: Remove SQLite-LevelDB migration code https://bugs.webkit.org/show_bug.cgi?id=69264 Reviewed by Tony Chang. Remove setOverrideIndexedDBBackingStore. It was used in the layout test for migration. * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): * DumpRenderTree/chromium/LayoutTestController.h: 2011-10-05 James Robinson [chromium] Add a command line option to DumpRenderTree and run_webkit_tests.py to enable threaded compositing mode https://bugs.webkit.org/show_bug.cgi?id=69498 Reviewed by Adam Barth. * DumpRenderTree/chromium/DumpRenderTree.cpp: (main): * DumpRenderTree/chromium/TestShell.cpp: (TestShell::TestShell): (TestShell::resetWebSettings): * DumpRenderTree/chromium/TestShell.h: (TestShell::setThreadedCompositingEnabled): * DumpRenderTree/chromium/WebPreferences.cpp: (WebPreferences::reset): (WebPreferences::applyTo): * DumpRenderTree/chromium/WebPreferences.h: * Scripts/webkitpy/layout_tests/port/chromium.py: * Scripts/webkitpy/layout_tests/run_webkit_tests.py: 2011-10-05 Erik Arvidsson [QT] Fix DRT after r96779 unreviewed. * DumpRenderTree/qt/DumpRenderTreeQt.cpp: (WebCore::dumpHistoryItem): 2011-10-05 Erik Arvidsson window.location.href and others needlessly decodes URI-encoded characters https://bugs.webkit.org/show_bug.cgi?id=30225 Reviewed by Darin Adler. * DumpRenderTree/chromium/TestShell.cpp: (normalizeLayoutTestURLInternal): Remove extra encode. 2011-10-05 Darin Adler Fix build. * TestWebKitAPI/Tests/WebKit2/CanHandleRequest.cpp: (TestWebKitAPI::TEST): Remove underscore prefix from call to WKContextRegisterURLSchemeAsEmptyDocument. 2011-10-05 Brent Fulgham [WinCairo] Switch to building Release target, and activate tests. https://bugs.webkit.org/show_bug.cgi?id=69272 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/config.json: Change from debug to release target. Switch from "Build" to "BuildAndTest" 2011-10-05 Simon Fraser In WebKitTestRunner, text has font smoothing in pixel snapshots https://bugs.webkit.org/show_bug.cgi?id=69396 Reviewed by Darin Adler. Call the new WKContext method that disables font smoothing in WebKitTestRunner, so that pixel snapshots don't have font smoothing enabled. Remove leading underscore from a WKContext function call. * WebKitTestRunner/TestController.cpp: (WTR::TestController::resetStateToConsistentValues): 2011-10-05 David Levin Improve the watchlist for threading entries. https://bugs.webkit.org/show_bug.cgi?id=69463 Reviewed by Adam Barth. * Scripts/webkitpy/common/config/watchlist: 2011-10-05 Adam Roben Add tests for using RetainPtrs inside HashMap and HashSet Fixes Using RetainPtr as the key type in HashMap/HashSet fails to compile Reviewed by John Sullivan. * TestWebKitAPI/Tests/WTF/cf/RetainPtrHashing.cpp: Added. (TestWebKitAPI::TEST): Show that RetainPtr can be used inside HashSet and as both the key and value type of HashMap. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/win/TestWebKitAPI.vcproj: Added new file. 2011-10-05 Adam Barth Mark as Expected in garden-o-matic is confusing https://bugs.webkit.org/show_bug.cgi?id=69452 Reviewed by David Levin. krit didn't understand the difference between this button and the Rebaseline button because both of them seem to indicate that the new behavior is expected. This patch renames "Mark as Expected" to "Expect Failure" to make it clear that we're just expecting the test to fail rather than accepting the new results as passing. We might need to iterate on these names a bit more if folks continue to find them confusing. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js: 2011-10-05 Adam Barth Add watchlist entries for abarth https://bugs.webkit.org/show_bug.cgi?id=69461 Reviewed by David Levin. I might have gone a bit overboard. We'll see if this is too much spam for me. * Scripts/webkitpy/common/config/watchlist: 2011-10-05 Brent Fulgham [WinCairo] Teach old-run-webkit-tests about wincairo. https://bugs.webkit.org/show_bug.cgi?id=69280 Reviewed by Adam Roben. * Scripts/old-run-webkit-tests: Handle wincairo case. (buildPlatformResultHierarchy): (readSkippedFiles): Ditto * Scripts/webkitdirs.pm: Deal with wincairo build case. (determineConfiguration): 2011-10-05 Chang Shu [WK2] Support setMarkedText/hasMarkedText/unmarkText/insertText for WTR TextInputController https://bugs.webkit.org/show_bug.cgi?id=68924 Added TextInputController files and JS interfaces for the above functions. Reviewed by Darin Adler. * WebKitTestRunner/DerivedSources.make: * WebKitTestRunner/DerivedSources.pro: * WebKitTestRunner/GNUmakefile.am: * WebKitTestRunner/InjectedBundle/Bindings/TextInputController.idl: Added. * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): * WebKitTestRunner/InjectedBundle/InjectedBundle.h: (WTR::InjectedBundle::textInputController): * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::InjectedBundlePage::didClearWindowForFrame): * WebKitTestRunner/InjectedBundle/TextInputController.cpp: Added. (WTR::TextInputController::create): (WTR::TextInputController::TextInputController): (WTR::TextInputController::~TextInputController): (WTR::TextInputController::wrapperClass): (WTR::TextInputController::makeWindowObject): (WTR::TextInputController::setMarkedText): (WTR::TextInputController::hasMarkedText): (WTR::TextInputController::unmarkText): (WTR::TextInputController::insertText): * WebKitTestRunner/InjectedBundle/TextInputController.h: Added. * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: * WebKitTestRunner/win/InjectedBundle.vcproj: 2011-10-04 Oliver Hunt Add rudimentary filtering to write barriers https://bugs.webkit.org/show_bug.cgi?id=69392 Reviewed by Filip Pizlo. When we're recording gc phase times Heap.o picks up some exit time destructors, so we'll just ignore Heap.o in this check. * Scripts/check-for-exit-time-destructors: 2011-10-05 Chang Shu [WK2] WebKitTestRunner needs LayoutTestController.dumpConfigurationForViewport https://bugs.webkit.org/show_bug.cgi?id=69365 Reviewed by Darin Adler. Added js API in WTR LayoutTestController. * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::LayoutTestController::dumpConfigurationForViewport): * WebKitTestRunner/InjectedBundle/LayoutTestController.h: 2011-10-05 Raphael Kubo da Costa [EFL] ImageDiff: Do not clear the image before it is processed and compared https://bugs.webkit.org/show_bug.cgi?id=69378 Reviewed by Antonio Gomes. Commit 61836 to Evas added a few cleanup calls to some evas_object_image functions which effectively made ImageDiff consider all images to be equal. It turns out that now evas_object_image_colorspace_set clears any data set, so the call to evas_object_image_data_set was not having any effect. * DumpRenderTree/efl/ImageDiff.cpp: (readImageFromStdin): Call evas_object_image_data_set after evas_object_image_colorspace_set. 2011-10-05 Raphael Kubo da Costa [EFL] Fix Evas_Object_Image reference counts and file descriptor leakage in ImageDiff. https://bugs.webkit.org/show_bug.cgi?id=68449 Reviewed by Antonio Gomes. evas_render() must be called to close up temporary files and unmap unused resources. evas_object_image_data_set() must be called after an evas_object_image_data_get() to balance an internal reference count. * DumpRenderTree/efl/ImageDiff.cpp: (differenceImageFromDifferenceBuffer): (calculateDifference): (printImage): (printImageDifferences): 2011-10-05 Balazs Kelemen [Qt][WK2] Unreviewed build fix. Don't use WebCore directly in WTR. * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::addChromeInputField): (WTR::PlatformWebView::removeChromeInputField): (WTR::PlatformWebView::makeWebViewFirstResponder): 2011-10-05 David Levin watchlist: Add more rules for chromium public api and sort the file. https://bugs.webkit.org/show_bug.cgi?id=69323 Reviewed by Adam Barth. * Scripts/webkitpy/common/config/watchlist: 2011-10-05 David Levin watchlist: Don't add the same message to a bug more than once. https://bugs.webkit.org/show_bug.cgi?id=69303 Reviewed by Adam Barth. * Scripts/webkitpy/common/net/bugzilla/bug.py: Added a way to determine if a message is in the comments already. * Scripts/webkitpy/common/net/bugzilla/bug_unittest.py: A test for the above. * Scripts/webkitpy/tool/commands/applywatchlistlocal_unittest.py: Pick a bug supplied by the mock. Change the test due to different output from the mock watch list tool. * Scripts/webkitpy/tool/mocktool.py: Change the mock watch list to return another email so it will be filtered out, fix bugs to have the cc and comment fields, and fix fetch_bug to handle bug_id's which are text (because that took me way too long to debug). * Scripts/webkitpy/tool/steps/applywatchlist.py: Change to filter out comments and/or cc's that are already in the bug. * Scripts/webkitpy/tool/steps/applywatchlist_unittest.py: Pick a bug supplied by the mock, and remove a comment that is filtered out. 2011-10-05 Balazs Kelemen [Qt][WK2] Unreviewed build fix after r96643. * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::addChromeInputField): (WTR::PlatformWebView::removeChromeInputField): (WTR::PlatformWebView::makeWebViewFirstResponder): 2011-10-04 Kent Tamura Introduce feature flags for incomplete input types https://bugs.webkit.org/show_bug.cgi?id=68971 Reviewed by Hajime Morita. * Scripts/build-webkit: Add --input-type-{date,datetime,datetimelocal,month,time,week} flags. They are disabled by default. 2011-10-04 Adam Barth garden-o-matic doesn't launch it's UI on Lion https://bugs.webkit.org/show_bug.cgi?id=69349 Reviewed by Darin Adler. It turns out that Lion is a bit more picky about file URLs. This patch switches us from using a path to using an actual file URL, silencing the warning and making the tool actually launch on Lion. * Scripts/webkitpy/tool/servers/gardeningserver.py: 2011-10-04 Daniel Bates Fix Perl undefined variable warning when using run-safari after (https://bugs.webkit.org/show_bug.cgi?id=68662) Rubber-stamped by Alexey Proskuryakov. * Scripts/webkitdirs.pm: (runMacWebKitApp): Check that the optional function argument $useOpenCommand is defined before using it in a comparison operation. 2011-10-04 Scott Graham Add GAMEPAD feature flag https://bugs.webkit.org/show_bug.cgi?id=66859 Reviewed by Darin Fisher. * Scripts/build-webkit: 2011-10-04 Jon Lee REGRESSION (WK2): (Shift-)option-tabbing skips over elements when transitioning from chrome to webview https://bugs.webkit.org/show_bug.cgi?id=68412 Reviewed by Darin Adler. In order to create a test for the bug, I had to update DRT and WKTR to create some widget that allows first responder status to move away from the main web view. Three methods were added to layoutTestController: addChromeInputField, removeChromeInputField, and focusWebView. addChromeInputField adds a text field that is a sibling to the web view, and sets up the key event loop between the two. removeChromeInputField removes that field. focusWebView moves first responder status to the web view. The test makes the call via layoutTestController and passes a callback that it assumes will be executed once the task is completed. In DRT the callback is called synchronously. In WKTR this is handled with message passing between the two processes. * DumpRenderTree/LayoutTestController.cpp: (addChromeInputFieldCallback): (removeChromeInputFieldCallback): (focusWebViewCallback): (LayoutTestController::staticFunctions): * DumpRenderTree/LayoutTestController.h: * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::addChromeInputField): (LayoutTestController::removeChromeInputField): (LayoutTestController::focusWebView): * DumpRenderTree/mac/DumpRenderTree.mm: (resetWebViewToConsistentStateBeforeTesting): When resetting for the next test, make sure to remove the chrome input field. * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::addChromeInputField): (LayoutTestController::removeChromeInputField): (LayoutTestController::focusWebView): * DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::addChromeInputField): (LayoutTestController::removeChromeInputField): (LayoutTestController::focusWebView): * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::didReceiveMessage): (WTR::InjectedBundle::postAddChromeInputField): (WTR::InjectedBundle::postRemoveChromeInputField): (WTR::InjectedBundle::postFocusWebView): * WebKitTestRunner/InjectedBundle/InjectedBundle.h: * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::callbackMap): Create a hash map that keeps track of the callbacks provided through JS. (WTR::cacheLayoutTestControllerCallback): (WTR::callLayoutTestControllerCallback): (WTR::LayoutTestController::addChromeInputField): (WTR::LayoutTestController::removeChromeInputField): (WTR::LayoutTestController::focusWebView): (WTR::LayoutTestController::callAddChromeInputFieldCallback): (WTR::LayoutTestController::callRemoveChromeInputFieldCallback): (WTR::LayoutTestController::callFocusWebViewCallback): * WebKitTestRunner/InjectedBundle/LayoutTestController.h: * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/TestController.cpp: (WTR::TestController::resetStateToConsistentValues): * WebKitTestRunner/TestInvocation.cpp: (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): * WebKitTestRunner/gtk/PlatformWebViewGtk.cpp: (WTR::PlatformWebView::addChromeInputField): (WTR::PlatformWebView::removeChromeInputField): (WTR::PlatformWebView::makeWebViewFirstResponder): * WebKitTestRunner/mac/PlatformWebViewMac.mm: (WTR::PlatformWebView::addChromeInputField): (WTR::PlatformWebView::removeChromeInputField): (WTR::PlatformWebView::makeWebViewFirstResponder): * WebKitTestRunner/win/PlatformWebViewWin.cpp: (WTR::PlatformWebView::addChromeInputField): (WTR::PlatformWebView::removeChromeInputField): (WTR::PlatformWebView::makeWebViewFirstResponder): * DumpRenderTree/mac/LayoutTestControllerMac.mm: These functions have nothing to do with the patch-- just cleaning up style. (LayoutTestController::addDisallowedURL): (originsArrayToJS): (LayoutTestController::queueLoad): (LayoutTestController::setMockDeviceOrientation): (LayoutTestController::setIconDatabaseEnabled): (LayoutTestController::setEditingBehavior): 2011-10-04 Simon Fraser Move font-fixup code in WebKitTestRunner to a better place https://bugs.webkit.org/show_bug.cgi?id=69356 Reviewed by Sam Weinig. Move code that swizzles NSFontManager methods to ActivateFonts.mm, which already contains font-related code. * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::initialize): * WebKitTestRunner/InjectedBundle/mac/ActivateFonts.mm: (WTR::allowedFontFamilySet): (WTR::drt_NSFontManager_availableFontFamilies): (WTR::drt_NSFontManager_availableFonts): (WTR::swizzleNSFontManagerMethods): (WTR::activateFonts): * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm: (WTR::InjectedBundle::platformInitialize): 2011-10-04 Raphael Kubo da Costa [EFL] DRT: Add DumpRenderTree.cpp https://bugs.webkit.org/show_bug.cgi?id=62034 Reviewed by Antonio Gomes. This is the main implementation file for EFL's DumpRenderTree. * DumpRenderTree/efl/DumpRenderTree.cpp: Added. (dumpFramesAsText): (dumpFrameScrollPosition): (shouldLogFrameLoadDelegates): (shouldDumpAsText): (sendPixelResultsEOF): (invalidateAnyPreviousWaitToDumpWatchdog): (onEcoreEvasResize): (onCloseWindow): (useLongRunningServerMode): (parseCommandLineOptions): (getFinalTestURL): (getExpectedPixelHash): (createLayoutTestController): (runTest): (runTestingServerLoop): (adjustOutputTypeByMimeType): (dumpFrameContentsAsText): (shouldDumpFrameScrollPosition): (shouldDumpPixelsAndCompareWithExpected): (shouldDumpBackForwardList): (initEfl): (shutdownEfl): (displayWebView): (dump): (initEcoreEvas): (main): 2011-10-03 Adam Barth garden-o-matic should remove builders from the failure grid once they start to pass https://bugs.webkit.org/show_bug.cgi?id=69309 Unreviewed. dglazkov is on vacation. I need to interest someone else in reviewing these patches. This patch follows the update/purge model from UpdateTracker. This patch is less awesome than it could be because it causes the UI to flash slightly. In a future patch, we'll want to use these update/purge notifications to update the UI without flashing. However, that's not such a big deal because this UI is hidden in the default view. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: 2011-10-03 David Levin webkitpy: Expose cc_emails and comments from bug.py https://bugs.webkit.org/show_bug.cgi?id=69308 Reviewed by Adam Barth. * Scripts/webkitpy/common/net/bugzilla/bug.py: Expose cc_emails and comments . * Scripts/webkitpy/common/net/bugzilla/bugzilla.py: Add parsing support for comments. * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py: Fix the unit test due to comments being exposed. 2011-10-03 James Robinson Add myself to the watchlist for platform/graphics/chromium https://bugs.webkit.org/show_bug.cgi?id=69297 Reviewed by David Levin. * Scripts/webkitpy/common/config/watchlist: 2011-10-03 Anders Carlsson Remove custom scrollbar painting hooks https://bugs.webkit.org/show_bug.cgi?id=69163 Reviewed by Alexey Proskuryakov. * DumpRenderTree/chromium/WebPreferences.cpp: (WebPreferences::applyTo): No need to enable custom scrollbar painting. 2011-10-03 David Levin watchlist: Make watchlist run on a bot. https://bugs.webkit.org/show_bug.cgi?id=68974 Reviewed by Adam Barth. * Scripts/webkitpy/tool/commands/queues.py: Add the watchlist to the style bot. * Scripts/webkitpy/tool/commands/queues_unittest.py: Change to log what is run so that we can see the style command and the watchlist command being run. 2011-10-03 David Levin watchlist: Allow specified a bug to attach the watchlist info to. https://bugs.webkit.org/show_bug.cgi?id=69288 Reviewed by Eric Seidel. * Scripts/webkitpy/tool/commands/applywatchlistlocal.py: Add support for bugid. * Scripts/webkitpy/tool/commands/applywatchlistlocal_unittest.py: Add tests and fix an existing test because the unit tests do not go through the normal command line processing (so the -g argument remained when we determined bug ids). * Scripts/webkitpy/tool/commands/commandtest.py: Change the derivation to get the assert raises regex method. * Scripts/webkitpy/tool/commands/download.py: Fix the command help to be more unixy. 2011-10-03 Leandro Pereira [EFL] DRT: Add JSStringUtils.{cpp,h} https://bugs.webkit.org/show_bug.cgi?id=66397 Reviewed by Ryosuke Niwa. Adds helper functions to compare the equality of C-style strings and JSStringRef objects. * DumpRenderTree/efl/JSStringUtils.cpp: Added. (equals): * DumpRenderTree/efl/JSStringUtils.h: Added. 2011-10-03 Adam Barth garden-o-matic should work in Safari 5.1 https://bugs.webkit.org/show_bug.cgi?id=69290 Reviewed by Sam Weinig. My old implementation of bind was too clever by half. This one seems to work better, at least according to this test. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: 2011-09-29 Ademar de Souza Reis Jr. Unreviewed: change my e-mail in commiters.py (I'm leaving the company and my @openbossa.org will be no more) * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: 2011-10-01 Brent Fulgham Correct post-build step for WinCairo test runs. https://bugs.webkit.org/show_bug.cgi?id=69191 Reviewed by Brian Weinstein. * DumpRenderTree/win/DumpRenderTreePostBuild.cmd: Add additional shell case to handle the CFLite build. 2011-10-01 Filip Pizlo Added myself to reviewers, because I'm now a reviewer. * Scripts/webkitpy/common/config/committers.py: 2011-10-01 Filip Pizlo Bencher script makes it difficult to do automated performance testing https://bugs.webkit.org/show_bug.cgi?id=69207 Reviewed by Sam Weinig. This adds two new features: The ability to disable automatic VM detection, which is flaky if any profiling features are enabled in jsc. The ability to compute, and report, a scaled result for all benchmark suites. It is the geometric mean of three numbers: SunSpider's arithmetic mean, V8's geometric mean, and Kraken's arithmetic mean. It is also possible to turn off all other output from bencher and just get this number with the --brief option. * Scripts/bencher: 2011-10-01 Sam Weinig WTR is not successfully changing the NSUserDefaults https://bugs.webkit.org/show_bug.cgi?id=69204 Reviewed by Dan Bernstein. * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm: (WTR::InjectedBundle::platformInitialize): Use -setVolatileDomain:forName:NSArgumentDomain instead of registerDefaults to get our overrides picked up. 2011-09-30 Daniel Bates Cleanup: Extract code to launch a Mac WebKit application into a common function https://bugs.webkit.org/show_bug.cgi?id=68662 Reviewed by Adam Roben. Extracts common code to launch a WebKit application on Mac OS X into a function so that we can remove duplicate code. * Scripts/run-webkit-app: Modified to use webkitdirs::runMacWebKitApp(). * Scripts/webkitdirs.pm: Sorted @EXPORT list and added constant USE_OPEN_COMMAND. (runMacWebKitApp): Added. (runSafari): Modified to use webkitdirs::runMacWebKitApp(). (runMiniBrowser): Ditto. (runWebKitTestRunner): Ditto. (runTestWebKitAPI): Ditto. 2011-09-29 David Levin watchlist: Add webkit-patch command to run watchlist. https://bugs.webkit.org/show_bug.cgi?id=68973 Reviewed by Eric Seidel. * Scripts/webkitpy/common/host.py: Exposed the watch_list. * Scripts/webkitpy/common/watchlist/watchlist.py: Changed to return the cc and messages as sorted lists. To allow for deterministic ordering. * Scripts/webkitpy/common/watchlist/watchlist_unittest.py: Ditto. * Scripts/webkitpy/tool/commands/__init__.py: Add ApplyWatchListLocal so that it will be exposed as a command by webkit-patch. * Scripts/webkitpy/tool/commands/applywatchlistlocal.py: Added. * Scripts/webkitpy/tool/commands/applywatchlistlocal_unittest.py: Added. * Scripts/webkitpy/tool/commands/download.py: Added ApplyWatchList which mimics CheckStyle. * Scripts/webkitpy/tool/commands/download_unittest.py: Added an appropriate test. * Scripts/webkitpy/tool/mocktool.py: Mock out the watch list. * Scripts/webkitpy/tool/steps/__init__.py: Added ApplyWatchList. * Scripts/webkitpy/tool/steps/applywatchlist.py: Added. * Scripts/webkitpy/tool/steps/applywatchlist_unittest.py: Added. 2011-09-30 David Levin watchlist: Add a way to detect a net increase or decrease of a pattern (in a file). https://bugs.webkit.org/show_bug.cgi?id=69031 Reviewed by Adam Barth. * Scripts/webkitpy/common/config/watchlist: Added usage of the new pattern, a comment to explain something important about the email addresses, and a comment to let emacs know to treat the file as a python file. * Scripts/webkitpy/common/watchlist/amountchangedpattern.py: Added. Detects increases or decreases in a pattern. * Scripts/webkitpy/common/watchlist/amountchangedpattern_unittest.py: Added. * Scripts/webkitpy/common/watchlist/watchlist_unittest.py: Added unit tests for watchlist which has "more" or "less". * Scripts/webkitpy/common/watchlist/watchlistparser.py: Added support for "more" or "less". 2011-09-30 Sam Weinig Add support for eventSender.mouseScrollBy in WTR https://bugs.webkit.org/show_bug.cgi?id=69189 Add implementation of eventSender.mouseScrollBy for the mac. Reviewed by Simon Fraser. * WebKitTestRunner/EventSenderProxy.h: * WebKitTestRunner/InjectedBundle/Bindings/EventSendingController.idl: * WebKitTestRunner/InjectedBundle/EventSendingController.cpp: (WTR::EventSendingController::keyDown): (WTR::EventSendingController::mouseScrollBy): * WebKitTestRunner/InjectedBundle/EventSendingController.h: * WebKitTestRunner/TestController.cpp: (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle): * WebKitTestRunner/mac/EventSenderProxy.mm: (WTR::EventSenderProxy::mouseScrollBy): 2011-09-30 Simon Fraser fast/forms/text-control-intrinsic-widths.html fails on Mac if Word/Office fonts are installed https://bugs.webkit.org/show_bug.cgi?id=35273 Reviewed by Dan Bernstein. There are two types of font issue that can cause layout test failure. First, the user may have non-standard fonts installed that affect the layout of some tests. For example, some tests use the MS Gothic font if installed. Secondly, the user may have duplicate fonts installed in ~/Library/Fonts, which override the system fonts (e.g. a copy of Arial from MS Office). For some tests, this can affect font metrics or fallback behavior. This change addresses the first problem only. It limits the set of fonts available through NSFontManager, by swizzling some methods to return a list of fonts or font families which only include fonts from a hardcoded whitelist. * DumpRenderTree/mac/DumpRenderTree.mm: (allowedFontFamilySet): (drt_NSFontManager_availableFontFamilies): (drt_NSFontManager_availableFonts): (swizzleNSFontManagerMethods): (activateTestingFonts): (adjustFonts): (createWebViewAndOffscreenWindow): (prepareConsistentTestingEnvironment): * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm: (WTR::allowedFontFamilySet): (WTR::drt_NSFontManager_availableFontFamilies): (WTR::drt_NSFontManager_availableFonts): (WTR::swizzleNSFontManagerMethods): (WTR::InjectedBundle::platformInitialize): 2011-09-30 David Levin Need to skip webkitpy.common.checkout.baselineoptimizer_unittest.BaselineOptimizerTest on Windows.. https://bugs.webkit.org/show_bug.cgi?id=69039 Reviewed by Adam Barth. * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py: Actually skip only one specific test that is failing on Windows. 2011-09-30 Tor Arne Vestbø [Qt] Prevent qDebug() output from DRT and WTR unless --verbose For DRT we didn't install the message handler early enough to catch output while constructing the QApplication. For WTR we didn't even install a message handler. Since the UI process will forward any output from the web process we set an environment variable QT_WEBKIT_SUPPRESS_WEB_PROCESS_OUTPUT in WTR before the web process is started. This is picked up by the web process which installs its own message handler. The environment variable can be overriden on the command line if you want to see output from the web process, or you can pass --verbose to WTR to see output from both processes. https://bugs.webkit.org/show_bug.cgi?id=69132 Reviewed by Andreas Kling. * DumpRenderTree/qt/main.cpp: * WebKitTestRunner/qt/main.cpp: 2011-09-30 Tor Arne Vestbø [Qt] Make sure WTR sizes the window and item correctly Revision 96345 changed the logic for how the view and window was created, but missed a vital part, setting the size. We now use a QSGView for the window, that has a simple item as its root object that is always resized to fit within the window. The webview is then parented to the root object and set to anchors.fill: parent. That way any window geometry changes will propagate to the web view. https://bugs.webkit.org/show_bug.cgi?id=69134 Reviewed by Andreas Kling. * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/qt/PlatformWebViewQt.cpp: 2011-09-30 Raphael Kubo da Costa [EFL] Only save the current viewport in PixelDumpSupportEfl. https://bugs.webkit.org/show_bug.cgi?id=68450 Reviewed by Antonio Gomes. The previous implementation saved the whole page contents in the final PNG, however the baselines and expectations in the tree only want the current viewport (with a scrollbar, if needed), so we now only save the currently visible contents. * DumpRenderTree/efl/PixelDumpSupportEfl.cpp: (createBitmapContextFromWebView): Use ewk_frame_visible_content_geometry_get instead of ewk_frame_contents_size_get. 2011-09-30 Raphael Kubo da Costa [EFL] Support and keep track of extra windows (views). https://bugs.webkit.org/show_bug.cgi?id=68454 Reviewed by Antonio Gomes. So far, window.open and related JavaScript code would not work, as DRTView did not implementation Ewk_View_Smart_Class::window_{open,close}. The default implementation in ewk just returns the current window/view, which broke any layouttest which relied on window.open. We now properly create and remove these new views. Unfortunately, the current way ewk works (and due to the fact that we cannot pass class methods as function pointers to the EFL functions), DRTView needs to know that DRTChrome exists and call `browser' directly in some points. * DumpRenderTree/efl/DumpRenderTreeView.cpp: (onWindowCreate): (onWindowCloseDelayed): (onWindowClose): (drtViewTiledAdd): 2011-09-30 Qi Zhang WebKitTestRunner needs layoutTestController.setAppCacheMaximumSize https://bugs.webkit.org/show_bug.cgi?id=42698 Reviewed by Andreas Kling. Implemented layoutTestController.setAppCacheMaximumSize and layoutTestController.clearApplicationCache, and call clearApplicationCache when layoutTest start up. * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::LayoutTestController::clearAllApplicationCaches): (WTR::LayoutTestController::setAppCacheMaximumSize): * WebKitTestRunner/InjectedBundle/LayoutTestController.h: 2011-09-30 Zeno Albisser [Qt][WK2] Change WebKitTestRunner to use QMouseEvents instead of QGraphicsSceneMouseEvents https://bugs.webkit.org/show_bug.cgi?id=69145 This is a follow up to https://bugs.webkit.org/show_bug.cgi?id=69105 Due to refactorings in Qt5, QMouseEvent now supports floating point coordinates and is perfectly suitable for SceneGraph as well. GraphicsScene events will not be available in QtGui anymore in future. Therefore we also have to change QtWebKit to use QMouseEvents instead of QGraphicsSceneMouseEvents. Reviewed by Chang Shu. * WebKitTestRunner/EventSenderProxy.h: * WebKitTestRunner/qt/EventSenderProxyQt.cpp: (WTR::EventSenderProxy::createMouseEvent): (WTR::EventSenderProxy::mouseDown): (WTR::EventSenderProxy::mouseUp): (WTR::EventSenderProxy::mouseMoveTo): 2011-09-30 James Robinson [chromium] Add WebKit API for sending input events to the compositor thread https://bugs.webkit.org/show_bug.cgi?id=69117 Reviewed by Darin Fisher. Adds support for creating and setting a compositor thread in DRT. * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::WebViewHost): * DumpRenderTree/chromium/WebViewHost.h: 2011-09-29 David Levin watchlist: Add cross-checks for WatchList once it is filled. https://bugs.webkit.org/show_bug.cgi?id=68975 Reviewed by Eric Seidel. * Scripts/webkitpy/common/watchlist/watchlist.py: Made the data members public instead of having trivial getter and setters. * Scripts/webkitpy/common/watchlist/watchlist_unittest.py: Fix the unit tests to pass the validation checks. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Add validation checks and fix a few style nits. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Add tests for the validation checks. * Scripts/webkitpy/common/watchlist/watchlistrule.py: Make definitions_to_match public. 2011-09-29 Xianzhu Wang run-api-tests fails on chromium-win bot https://bugs.webkit.org/show_bug.cgi?id=69121 On Windows running a perl script needs the interpreter. Reviewed by Adam Barth. * Scripts/run-api-tests: (buildTestTool): 2011-09-29 Adam Barth Attempt to make this tool work on Windows. * Scripts/run-api-tests: (buildTestTool): 2011-09-29 Tor Arne Vestbø [Qt] Try to fix the minial build * QtTestBrowser/launcherwindow.cpp: * QtTestBrowser/webpage.cpp: 2011-09-23 Tor Arne Vestbø [Qt] Fix build against Qt5 after refactor of widgets out of QtGUi QWidget and friends now live in the QtWidgets library. We update includes in implementation files and private headers to us the non-module-prefixed path, and leave the lookup for the include path. For public headers we have to ifdef the includes as the user might now have the modules we need in his QT config. Finally, QSGCanvas is no longer a QWidget but a QWindow, so we have to update our code and use windowHandle() for setting the parent relationships. https://bugs.webkit.org/show_bug.cgi?id=68687 Reviewed by Andreas Kling. * DumpRenderTree/qt/DumpRenderTree.pro: * DumpRenderTree/qt/ImageDiff.pro: * MiniBrowser/qt/BrowserView.cpp: * MiniBrowser/qt/BrowserView.h: * MiniBrowser/qt/BrowserWindow.cpp: * MiniBrowser/qt/BrowserWindow.h: * MiniBrowser/qt/MiniBrowser.pro: * MiniBrowser/qt/MiniBrowserApplication.h: * MiniBrowser/qt/main.cpp: * QtTestBrowser/QtTestBrowser.pro: * QtTestBrowser/launcherwindow.cpp: * QtTestBrowser/launcherwindow.h: * QtTestBrowser/locationedit.h: * QtTestBrowser/main.cpp: * QtTestBrowser/mainwindow.cpp: * QtTestBrowser/mainwindow.h: * QtTestBrowser/webinspector.h: * QtTestBrowser/webpage.cpp: * QtTestBrowser/webview.cpp: * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/qt/PlatformWebViewQt.cpp: 2011-09-29 Ravi Phaneendra Kasibhatla Initial implementation of WebInspector for WebKit2 GTK port. https://bugs.webkit.org/show_bug.cgi?id=68235 Reviewed by Martin Robinson. * MiniBrowser/gtk/BrowserWindow.c: (browserWindowConstructed): Set the developer extra flags to true to enable inspect element. 2011-09-29 Adam Barth Unbreak the "update expectatations" button by using the existing machinery to create the failureInfo objects. Unbreak the "close" button on the progress dialog by using |this| instead of the old name of the object. Teach the controllers unit test not to talk to the actual network. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: 2011-09-29 Hans Wennborg IndexedDB: Use LevelDB also for in-memory databases https://bugs.webkit.org/show_bug.cgi?id=68903 Reviewed by Steve Block. Remove the temporary dir that was necessary before LevelDB supported in-memory databases. * DumpRenderTree/chromium/TestShell.cpp: (TestShell::TestShell): * DumpRenderTree/chromium/TestShell.h: 2011-09-28 Xianzhu Wang Run TestWebKitAPI on Chromium buildbots https://bugs.webkit.org/show_bug.cgi?id=68678 Reviewed by Tony Chang. * BuildSlaveSupport/build.webkit.org-config/master.cfg: 2011-09-28 Dimitri Glazkov garden-o-matic results should anticipate 0 unexpected failures. https://bugs.webkit.org/show_bug.cgi?id=69036 Also fixes controllers: UnexpectedFailures test. Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: Tweaked. 2011-09-28 Dirk Pranke Modify new-run-webkit-tests to return the percentage difference in image diffs where possible. https://bugs.webkit.org/show_bug.cgi?id=67253 Reviewed by Eric Seidel. Nearly all of the work in this was done by Simon Fraser; I'm just repackaging it and cleaning it up a bit. This change modifies port.diff_image() to return a tuple of (pass/fail, %age), adds the value to the FailureImageHashMismatch, and writes the value into the full_results.json files to be displayed in the results page. * Scripts/webkitpy/layout_tests/controllers/manager.py: * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py: * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py: * Scripts/webkitpy/layout_tests/models/test_failures.py: * 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/port_testcase.py: * Scripts/webkitpy/layout_tests/port/test.py: * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: 2011-09-25 Kentaro Hara Unreviewed. Adding myself to committers.py. * Scripts/webkitpy/common/config/committers.py: 2011-09-28 Adam Barth garden-o-matic shouldn't depend on test-results.appspot.com https://bugs.webkit.org/show_bug.cgi?id=69019 Reviewed by Dimitri Glazkov. This patch moves garden-o-matic from getting full_results.json from test-results.appspot.com to getting that information directly from the buildbot. This patch improve the load time and accuracy of garden-o-matic because test-results.appspot.com is slow and is dropping a significant number of writes (see https://bugs.webkit.org/show_bug.cgi?id=68748). * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/net.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js: 2011-09-28 David Levin watchlist: Add a way to load the watchlist from config. https://bugs.webkit.org/show_bug.cgi?id=68991 Reviewed by Eric Seidel. * Scripts/webkitpy/common/config/watchlist: Added. * Scripts/webkitpy/common/watchlist/watchlistloader.py: Added. Support for loading the watch list. * Scripts/webkitpy/common/watchlist/watchlistloader_unittest.py: Added. Tests for the loading. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Adjusted to use the common base class. * Scripts/webkitpy/common/webkitunittest.py: Added. Made a common base class for functionality used in more than one. 2011-09-28 David Levin Attempt fix for Leopard python unit test run. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: 2011-09-27 Dimitri Glazkov REGRESSION(r95573): Crash when loading SVG documents in a flattened frame or any SVG document in Chromium/Mac. https://bugs.webkit.org/show_bug.cgi?id=68938 Made Chromium/Mac DRT better match how Chromium/Mac queries WebKit. Reviewed by David Hyatt. * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::didUpdateLayout): Added width/height queries. * DumpRenderTree/chromium/WebViewHost.h: Updated defs. 2011-09-28 Mihai Parparita Get rebaseline server running again https://bugs.webkit.org/show_bug.cgi?id=69020 Reviewed by Adam Barth. Make rebaseline server run again after r90337 (missing format_date_time import, results_json wasn't outputting anything). * Scripts/webkitpy/tool/servers/rebaselineserver.py: * Scripts/webkitpy/tool/servers/reflectionhandler.py: 2011-09-28 David Grogan Move dgrogan from contributor to committer https://bugs.webkit.org/show_bug.cgi?id=69021 * Scripts/webkitpy/common/config/committers.py: 2011-09-28 Mihai Parparita Add Mac Skia GPU bots to builders.py https://bugs.webkit.org/show_bug.cgi?id=69016 Reviewed by Eric Seidel. This allows wwebkit-patch rebaseline-expectations to work with Mac Skia GPU tests. * Scripts/webkitpy/layout_tests/port/builders.py: * Scripts/webkitpy/tool/commands/rebaseline_unittest.py: 2011-09-28 Gustavo Noronha Silva Second try at fixing the GTK+ build. * GNUmakefile.am: 2011-09-28 Xianzhu Wang Let TestWebKitAPI and webkit_unit_tests work for chromium-win https://bugs.webkit.org/show_bug.cgi?id=68680 Reviewed by Tony Chang. * Scripts/run-chromium-webkit-unit-tests: * Scripts/webkitdirs.pm: (determineBaseProductDir): 2011-09-28 Kaustubh Atrawalkar [Gtk] Implement textInputController.doCommand https://bugs.webkit.org/show_bug.cgi?id=66496 Reviewed by Martin Robinson. * DumpRenderTree/gtk/TextInputController.cpp: (doCommandCallback): Callback function for textInputController.doCommand 2011-09-28 David Levin watchlist: Suggest corrections for typos and improve error message consistency. https://bugs.webkit.org/show_bug.cgi?id=68976 Reviewed by Adam Barth. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Add suggestions for typos and improve consistency of the error messages. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Add tests and improve the regex that were already present. 2011-09-28 David Levin watchlist: Add support for matching added or deleted lines. https://bugs.webkit.org/show_bug.cgi?id=68972 Reviewed by Adam Barth. * Scripts/webkitpy/common/watchlist/changedlinepattern.py: Added. * Scripts/webkitpy/common/watchlist/changedlinepattern_unittest.py: Added. * Scripts/webkitpy/common/watchlist/watchlist.py: Comment fix up and fix input to the pattern match to only have the diff lines instead of the DiffFile. * Scripts/webkitpy/common/watchlist/watchlist_unittest.py: Add tests for the new patterns and combinations of the patterns. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Sort imports. Add changeline support. 2011-09-28 Carlos Garcia Campos [GTK] Add WebKitError to GTK+ WebKit2 API https://bugs.webkit.org/show_bug.cgi?id=68508 Reviewed by Martin Robinson. * GNUmakefile.am: Add path to webkitgtk derived sources to the include path. 2011-09-28 Philippe Normand [GTK] missing WebKit2 support in run-gtk-tests https://bugs.webkit.org/show_bug.cgi?id=68992 Rubber-stamped by Andreas Kling. * Scripts/run-gtk-tests: Get the WebKit2 tests to run from the webkit2 directory. 2011-09-27 Philippe Normand [GTK] build-webkit --no-webkit2 still builds WebKit2 https://bugs.webkit.org/show_bug.cgi?id=68892 Reviewed by Martin Robinson. Take the --no-webkit2 option in account when building WebKitGTK+. * Scripts/build-webkit: * Scripts/webkitdirs.pm: (buildAutotoolsProject): (buildGtkProject): 2011-09-28 Ryosuke Niwa Make Chromium Win Release (Tests) a core builder https://bugs.webkit.org/show_bug.cgi?id=68906 Reviewed by Adam Barth. Add Chromium Win Release (Tests) to the list of core builders. * Scripts/webkitpy/common/net/buildbot/buildbot.py: * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py: 2011-09-27 Adam Barth garden-o-matic should load faster and inform the user what is happening https://bugs.webkit.org/show_bug.cgi?id=68954 Reviewed by Dimitri Glazkov. This test disables the expected failures tab. The web inspector showed that loading the data for this tab was slowing down the loading time for the whole app. We need to find a way to load that data lazily. This patch also improves the update message to better describe the progress the app is making analyzing data from the buildbot. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: 2011-09-27 Adam Barth garden-o-matic should show the first result when examining results https://bugs.webkit.org/show_bug.cgi?id=68960 Reviewed by Dimitri Glazkov. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: 2011-09-27 David Levin watchlist: Add support for cc and message rules. https://bugs.webkit.org/show_bug.cgi?id=68950 Reviewed by Adam Barth. * Scripts/webkitpy/common/watchlist/watchlist.py: Added support to get the cc's and messages for a patch. * Scripts/webkitpy/common/watchlist/watchlist_unittest.py: Tests for the above. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Parsing support for the rules. * Scripts/webkitpy/common/watchlist/watchlistrule.py: Copied from Tools/Scripts/webkitpy/common/watchlist/watchlist.py. A generic encapsulation of either a message list or a cc list. * Scripts/webkitpy/common/watchlist/watchlistrule_unittest.py: Copied from Tools/Scripts/webkitpy/common/watchlist/watchlist.py. 2011-09-27 Tom Zakrajsek webkit-patch doesn't like UTF-8 characters in reviewers names https://bugs.webkit.org/show_bug.cgi?id=63452 Reviewed by Eric Seidel. * Scripts/webkit-patch: 2011-09-27 Adam Barth garden-o-matic examine buttons shows both expected and unexpected failures https://bugs.webkit.org/show_bug.cgi?id=68918 Reviewed by Dimitri Glazkov. This was a copy/paste error when I refactored this classes to share more code. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: 2011-09-27 Ryosuke Niwa Add Kaustubh Atrawalkar to the list of contributors. * Scripts/webkitpy/common/config/committers.py: 2011-09-27 Mihai Parparita [Chromium] Fixed layout mode tests break subsequent tests https://bugs.webkit.org/show_bug.cgi?id=68923 Reviewed by Adam Barth. Reset fixed layout mode between tests, so that tests that run after platform/chromium/fast/repaint/fixed-layout-360x240.html still have the regular 800x600 non-fixed layout. * DumpRenderTree/chromium/TestShell.cpp: (TestShell::resetTestController): 2011-09-27 David Levin watchlist: Add the filename pattern for definitions. https://bugs.webkit.org/show_bug.cgi?id=68917 Reviewed by Adam Barth. * Scripts/webkitpy/common/watchlist/filenamepattern.py: Added. * Scripts/webkitpy/common/watchlist/filenamepattern_unittest.py: Added. * Scripts/webkitpy/common/watchlist/watchlist.py: Added the filename pattern for definitions. * Scripts/webkitpy/common/watchlist/watchlist_unittest.py: Added tests. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Typo fix. 2011-09-27 Adam Barth garden-o-matic results view should sort test and builder names https://bugs.webkit.org/show_bug.cgi?id=68488 Reviewed by Andy Estes. Previously, the test and builder names were displayed in an arbitrary order that changed from time to time. That confused one user study participant. This patch sorts the lists so that they occur in a predictable order. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: 2011-09-27 Sheriff Bot Unreviewed, rolling out r96108, r96111, r96113, and r96116. http://trac.webkit.org/changeset/96108 http://trac.webkit.org/changeset/96111 http://trac.webkit.org/changeset/96113 http://trac.webkit.org/changeset/96116 https://bugs.webkit.org/show_bug.cgi?id=68913 Wait for working Qt5 (Requested by ossy on #webkit). * DumpRenderTree/qt/DumpRenderTree.pro: * DumpRenderTree/qt/ImageDiff.pro: * MiniBrowser/qt/BrowserView.cpp: (BrowserView::BrowserView): * MiniBrowser/qt/BrowserView.h: * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): (BrowserWindow::urlChanged): (BrowserWindow::~BrowserWindow): * MiniBrowser/qt/BrowserWindow.h: * MiniBrowser/qt/MiniBrowser.pro: * MiniBrowser/qt/MiniBrowserApplication.h: * MiniBrowser/qt/main.cpp: * QtTestBrowser/QtTestBrowser.pro: * QtTestBrowser/launcherwindow.cpp: * QtTestBrowser/launcherwindow.h: * QtTestBrowser/locationedit.h: * QtTestBrowser/main.cpp: * QtTestBrowser/mainwindow.cpp: * QtTestBrowser/mainwindow.h: * QtTestBrowser/webinspector.h: * QtTestBrowser/webpage.cpp: * QtTestBrowser/webview.cpp: * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/qt/EventSenderProxyQt.cpp: * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::WebView::wkView): (WTR::WebView::pageRef): (WTR::WebView::~WebView): (WTR::WebView::WebView): (WTR::PlatformWebView::PlatformWebView): (WTR::PlatformWebView::sendEvent): (WTR::PlatformWebView::postEvent): * WebKitTestRunner/qt/WebKitTestRunner.pro: 2011-09-27 Carlos Garcia Campos [GTK] Rename WebKit2 GTK+ API main header as webkit2.h https://bugs.webkit.org/show_bug.cgi?id=65178 Reviewed by Martin Robinson. * GNUmakefile.am: * GtkLauncher/main.c: Include . 2011-09-27 Csaba Osztrogonác [Qt][WK2] Buildfix after r96108. Rubber-stamped by Andreas Kling. * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro: * WebKitTestRunner/qt/EventSenderProxyQt.cpp: * WebKitTestRunner/qt/WebKitTestRunner.pro: 2011-09-27 David Levin watchlist: Change watchlistparser.py to be class based. https://bugs.webkit.org/show_bug.cgi?id=68869 Reviewed by Eric Seidel. * Scripts/webkitpy/common/watchlist/watchlistparser.py: * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: 2011-09-27 David Levin watchlist: Break out the diff boilerplate to allow for re-use. https://bugs.webkit.org/show_bug.cgi?id=68871 Reviewed by Eric Seidel. * Scripts/webkitpy/common/checkout/diff_parser_unittest.py: Break out the diff into a new file. * Scripts/webkitpy/common/checkout/diff_test_data.py: Added. 2011-09-27 Ilya Tikhonovsky Web Inspector: UI performance: introduce heap size tracking stats. https://bugs.webkit.org/show_bug.cgi?id=68901 It is interesting how much the heap memory is used by Inspector in order of running the test. Reviewed by Yury Semikhatsky. * DumpRenderTree/chromium/TestShell.cpp: (TestShell::showDevTools): (TestShell::closeDevTools): 2011-09-23 Tor Arne Vestbø [Qt] Fix build against Qt5 after refactor of widgets out of QtGUi QWidget and friends now live in the QtWidgets library. We update includes in implementation files and private headers to us the non-module-prefixed path, and leave the lookup for the include path. For public headers we have to ifdef the includes as the user might now have the modules we need in his QT config. Finally, QSGCanvas is no longer a QWidget but a QWindow, so we have to update our code and use windowHandle() for setting the parent relationships. https://bugs.webkit.org/show_bug.cgi?id=68687 Reviewed by Andreas Kling. * DumpRenderTree/qt/DumpRenderTree.pro: * DumpRenderTree/qt/ImageDiff.pro: * MiniBrowser/qt/BrowserView.cpp: * MiniBrowser/qt/BrowserView.h: * MiniBrowser/qt/BrowserWindow.cpp: * MiniBrowser/qt/BrowserWindow.h: * MiniBrowser/qt/MiniBrowser.pro: * MiniBrowser/qt/MiniBrowserApplication.h: * MiniBrowser/qt/main.cpp: * QtTestBrowser/QtTestBrowser.pro: * QtTestBrowser/launcherwindow.cpp: * QtTestBrowser/launcherwindow.h: * QtTestBrowser/locationedit.h: * QtTestBrowser/main.cpp: * QtTestBrowser/mainwindow.cpp: * QtTestBrowser/mainwindow.h: * QtTestBrowser/webinspector.h: * QtTestBrowser/webpage.cpp: * QtTestBrowser/webview.cpp: * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/qt/PlatformWebViewQt.cpp: 2011-09-26 Caio Marcelo de Oliveira Filho [Qt][WK2] Add support for hover API in Qt WebKit2 https://bugs.webkit.org/show_bug.cgi?id=68369 Reviewed by Andreas Kling. Change the statusbar to show the link URL when hovering links in MiniBrowser using QDesktopWebView. * MiniBrowser/qt/BrowserWindow.cpp: (BrowserWindow::BrowserWindow): (BrowserWindow::onLinkHovered): * MiniBrowser/qt/BrowserWindow.h: 2011-09-26 Dimitri Glazkov garden-o-matic's commit data on summary page should not crowd itself or twitch when hovered over. https://bugs.webkit.org/show_bug.cgi?id=68864 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary-mock.js: Updated mocks to work. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: Changed the structure of commit data to keep commit revision apart from the rest of details. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Adjusted unit tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: Made things look shiny. 2011-09-26 Csaba Osztrogonác [Qt][WK2] One more unreviewed buildfix after r96005. * WebKitTestRunner/qt/TestInvocationQt.cpp: Missing include added. 2011-09-26 Csaba Osztrogonác [Qt][WK2] Unreviewed buildfix after r96005. * WebKitTestRunner/qt/TestInvocationQt.cpp: (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): 2011-09-26 Dimitri Glazkov garden-o-matic should be pretty in Open Sans. https://bugs.webkit.org/show_bug.cgi?id=68845 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: Tweaked to bring back teh fonts. 2011-09-26 Dimitri Glazkov garden-o-matic should not say "Just now" when it doesn't actually know when an event happened. https://bugs.webkit.org/show_bug.cgi?id=68841 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: Changed ui.RelativeTime to not initialize with current time. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Adjusted unit tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js: Ditto. 2011-09-26 David Levin watchlist: Add parsing for definition section. https://bugs.webkit.org/show_bug.cgi?id=68850 Reviewed by Adam Barth. * Scripts/webkitpy/common/watchlist/watchlist.py: * Scripts/webkitpy/common/watchlist/watchlistparser.py: * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: 2011-09-26 Simon Fraser Fix WebKitTestRunner builds for Cairo, Windows and Qt. * WebKitTestRunner/cairo/TestInvocationCairo.cpp: (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): * WebKitTestRunner/win/TestInvocationWin.cpp: (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): 2011-09-26 Simon Fraser Repaint tests don't work in WebKit2 https://bugs.webkit.org/show_bug.cgi?id=68453 Reviewed by Sam Weinig. Convert DumpRenderTree to use use the repaint rects exposed via WebKit API when generating the repaint test overlay. Fix WebKitTestRunner to work with repaint tests, again using the repaint rects exposed from WK2. In both cases, layoutTestController.display() now just forces a display of the web view, and turns on repaint rect tracking on the main FrameView. Later, when pixel results are requested, the alpha overlay that shows the repaint rects is generated using the list of rects supplied by WebCore. * DumpRenderTree/mac/DumpRenderTree.mm: (resetDefaultsToConsistentValues): (resetWebViewToConsistentStateBeforeTesting): (displayWebView): * DumpRenderTree/mac/PixelDumpSupportMac.mm: (paintRepaintRectOverlay): (createBitmapContextFromWebView): * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): (WTR::InjectedBundle::done): * WebKitTestRunner/InjectedBundle/InjectedBundle.h: (WTR::InjectedBundle::setRepaintRects): * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::InjectedBundlePage::reset): (WTR::InjectedBundlePage::dump): * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::LayoutTestController::display): * WebKitTestRunner/TestInvocation.cpp: (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): * WebKitTestRunner/TestInvocation.h: * WebKitTestRunner/cg/TestInvocationCG.cpp: (WTR::paintRepaintRectOverlay): (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): 2011-09-26 Nayan Kumar K Get hover'ed element URL from HitTest. mouseDidMoveOverElement now gets the hover'ed element link using WKHitTestResult API's. https://bugs.webkit.org/show_bug.cgi?id=68426 Reviewed by Anders Carlsson. * MiniBrowser/gtk/BrowserWindow.c: (mouseDidMoveOverElement): 2011-09-26 David Levin Add skeleton parsing for a WatchList. https://bugs.webkit.org/show_bug.cgi?id=68823 Reviewed by Adam Barth. * Scripts/webkitpy/common/watchlist/watchlistparser.py: Added. Parses the top level watch list sections -- none are defined yet -- and throws an exception if an invalid one is listed. * Scripts/webkitpy/common/watchlist/watchlistparser_unittest.py: Added. Verify the exception for an invalid section in a watch list. * Scripts/webkitpy/common/watchlist/watchlist.py: Added. 2011-09-26 Adam Roben Clean up code imported from WebKitAPITest Fixes WebViewDestruction tests and related code don't match TestWebKitAPI conventions Reviewed by David Levin. * TestWebKitAPI/Tests/WebKit/win/WebViewDestruction.cpp: Changed tests to use gtest macros and to share functionality via test fixtures. Prefixed Win32 API calls with ::. Updated namespace name. (TestWebKitAPI::WebViewDestruction::SetUp): (TestWebKitAPI::WebViewDestruction::webViewCount): (TestWebKitAPI::WebViewDestructionWithHostWindow::SetUp): (TestWebKitAPI::WebViewDestruction::runMessagePump): (TestWebKitAPI::WebViewDestruction::TearDown): (TestWebKitAPI::WebViewDestructionWithHostWindow::TearDown): Moved functionality from free functions into these new test fixtures. * TestWebKitAPI/win/HostWindow.cpp: * TestWebKitAPI/win/HostWindow.h: Prefixed Win32 API calls with ::. Updated namespace name. 2011-09-23 Adam Roben Merge WebKitAPITest into TestWebKitAPI This also means that our formerly-WebKitAPITest tests will be run on the bots! I made the minimal changes necessary to get the tests to run in TestWebKitAPI. I'll make them more idiomatic in a separate pass. Fixes WebKitAPITest and TestWebKitAPI have duplicate functionality (and names!) Reviewed by Anders Carlsson. * TestWebKitAPI/Tests/WebKit/win/WebViewDestruction.cpp: Renamed from Tools/WebKitAPITest/tests/WebViewDestruction.cpp. * TestWebKitAPI/win/HostWindow.cpp: Renamed from Tools/WebKitAPITest/HostWindow.cpp. * TestWebKitAPI/win/HostWindow.h: Renamed from Tools/WebKitAPITest/HostWindow.h. * TestWebKitAPI/win/TestWebKitAPI.vcproj: Added new files. * TestWebKitAPI/win/main.cpp: (main): Added a call to ::OleInitialize to make WebKit1 work. This came from WebKitAPITest. * WebKitAPITest: Removed this directory and its contents. 2011-09-23 Adam Roben Remove two failing WebKitAPITest tests The failures are covered by and . Reviewed by Anders Carlsson. * WebKitAPITest/tests/WebViewDestruction.cpp: Removed the NoCloseOrDestroyViewWindow and MainFrameAfterClose tests. 2011-09-25 Adam Barth Finish removing PLATFORM(BREWMP) by removing associated code https://bugs.webkit.org/show_bug.cgi?id=68779 Reviewed by Sam Weinig. * Scripts/webkitpy/common/config/build.py: * waf/build/settings.py: 2011-09-21 Ai Makabi Provides a simple LRU cache class in Python. https://bugs.webkit.org/show_bug.cgi?id=67823 Reviewed by Tony Chang. * Scripts/webkitpy/common/lru_cache.py: * Scripts/webkitpy/common/lru_cache_unittest.py: 2011-09-25 Adam Barth Remove PLATFORM(HAIKU) and associated code https://bugs.webkit.org/show_bug.cgi?id=68774 Reviewed by Sam Weinig. * Scripts/webkitpy/common/config/build.py: * waf/build/settings.py: 2011-09-24 Adam Barth Always enable ENABLE(OFFLINE_WEB_APPLICATIONS) https://bugs.webkit.org/show_bug.cgi?id=68767 Reviewed by Eric Seidel. * Scripts/build-webkit: 2011-09-24 Adam Barth Remove ENABLE(WCSS) and associated code https://bugs.webkit.org/show_bug.cgi?id=68759 Reviewed by Darin Adler. * Scripts/build-webkit: * Scripts/old-run-webkit-tests: * Scripts/webkitperl/features.pm: (hasFeature): * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: 2011-09-24 Alejandro G. Castro Fixed GTK compilation after r95878, the operator== was defined twice when USE_WEBPROCESS_EVENT_SIMULATION is defined. * WebKitTestRunner/InjectedBundle/EventSendingController.cpp: 2011-09-23 Raphael Kubo da Costa [DRT] Include the right config file for EFL's DRT. https://bugs.webkit.org/show_bug.cgi?id=67042 Reviewed by Martin Robinson. Ports which use CMake as their buildsystem (such as the EFL one) also have config.h files, but they are named differently, so include the right one depending on the buildsystem being used. * DumpRenderTree/config.h: 2011-09-23 Adam Klein Add ENABLE_MUTATION_OBSERVERS feature flag https://bugs.webkit.org/show_bug.cgi?id=68732 Reviewed by Ojan Vafai. This flag will guard an implementation of the "Mutation Observers" proposed in http://lists.w3.org/Archives/Public/public-webapps/2011JulSep/1622.html * Scripts/build-webkit: 2011-09-23 Chang Shu [WK2] [Qt] Implement MouseDown/MouseUp/MouseMoveTo functions for WebKit2 EventSender https://bugs.webkit.org/show_bug.cgi?id=68556 Implement the MouseDown/MouseUp/MouseMoveTo functions on Qt platform. Reviewed by Darin Adler. * WebKitTestRunner/EventSenderProxy.h: * WebKitTestRunner/InjectedBundle/EventSendingController.h: * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/mac/EventSenderProxy.mm: (WTR::EventSenderProxy::EventSenderProxy): (WTR::EventSenderProxy::leapForward): * WebKitTestRunner/qt/EventSenderProxyQt.cpp: (WTR::EventSenderProxy::EventSenderProxy): (WTR::getMouseButton): (WTR::getModifiers): (WTR::EventSenderProxy::updateClickCountForButton): (WTR::EventSenderProxy::createGraphicsSceneMouseEvent): (WTR::EventSenderProxy::mouseDown): (WTR::EventSenderProxy::mouseUp): (WTR::EventSenderProxy::mouseMoveTo): (WTR::EventSenderProxy::leapForward): (WTR::EventSenderProxy::sendOrQueueEvent): (WTR::EventSenderProxy::replaySavedEvents): * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::postEvent): * WebKitTestRunner/qt/WebKitTestRunner.pro: 2011-09-23 Lei Zhang [chromium] Make the layout test script's kill timeout proportional to --time-out-ms https://bugs.webkit.org/show_bug.cgi?id=68026 Reviewed by Dirk Pranke. * Scripts/webkitpy/layout_tests/port/chromium.py: 2011-09-23 Elliot Poger update layout_tests to account for new default of use_skia=1 https://bugs.webkit.org/show_bug.cgi?id=68698 Rolls Source/WebKit/chromium/DEPS to pick up new default use_skia=1 from http://src.chromium.org/viewvc/chrome?view=rev&revision=102532 (Chromium-on-Mac now uses the Skia graphics library instead of Core Graphics). Also updates layout_test code to work with that change. Reviewed by Mihai Parparita. * Scripts/webkitpy/layout_tests/port/chromium_mac.py: * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py: 2011-09-23 Mark Rowe Fix the build. * TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Define operator<<(std::ostream&, const WTF::String&) inside the WTF namespace. 2011-09-22 Dirk Pranke enable NRWT for run-webkit-tests on Lion https://bugs.webkit.org/show_bug.cgi?id=68673 Reviewed by Eric Seidel. * Scripts/run-webkit-tests: (useNewRunWebKitTests): 2011-09-22 Xianzhu Wang Add unit test for existing StringBuilder https://bugs.webkit.org/show_bug.cgi?id=67080 Reviewed by Darin Adler. * TestWebKitAPI/TestWebKitAPI.gypi: * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Added. (operator<<): (TestWebKitAPI::expectBuilderContent): (TestWebKitAPI::expectEmpty): (TestWebKitAPI::TEST): 2011-09-22 Raphael Kubo da Costa [EFL] Use ewk_view to navigate in history instead of ewk_history. https://bugs.webkit.org/show_bug.cgi?id=68455 Reviewed by Antonio Gomes. BackForwardItem::invoke() was using ewk_history functions to navigate in history; however, this will only call WebCore::BackForwardListImpl and not move between pages at all. We now call ewk_view_navigate and its siblings to make sure the navigation actually happens. This should make tests like fast/dom/navigation-type-back-forward.html stop timing out. * DumpRenderTree/efl/WorkQueueItemEfl.cpp: (BackForwardItem::invoke): 2011-09-22 Dean Jackson Add ENABLE_CSS_FILTERS https://bugs.webkit.org/show_bug.cgi?id=68652 Reviewed by Simon Fraser. * Scripts/build-webkit: 2011-09-22 Anders Carlsson EventSenderProxy::mouseUp should call -[WKView mouseUp:] instead of -[WKView mouseDown:] https://bugs.webkit.org/show_bug.cgi?id=68660 Reviewed by Adam Barth. * WebKitTestRunner/mac/EventSenderProxy.mm: (WTR::EventSenderProxy::mouseUp): 2011-09-22 Adam Roben Remove FindSafari It isn't used or built anymore. Fixes Reviewed by Darin Adler. * FindSafari: Removed this directory and its contents. 2011-09-22 Adam Roben Remove FindSafari from all our .sln files It isn't used anymore, so there's no point in building it. Part of Remove FindSafari Reviewed by Steve Falkenburg. * DumpRenderTree/DumpRenderTree.sln: * WebKitTestRunner/WebKitTestRunner.sln: 2011-09-22 Adam Roben Remove all uses of FindSafari It isn't really needed anymore. Fixes . Reviewed by Darin Adler. * Scripts/old-run-webkit-tests: Moved read/writeRegistryString from here to webkitdirs.pm. * Scripts/run-webkit-nightly.cmd: Removed. This used to be used to launch nightly builds of WebKit, but now we use WebKit.exe for that. * Scripts/webkitdirs.pm: (installedSafariPath): Changed to use readRegistryString instead of invoking FindSafari to do the same thing. (readRegistryString): (writeRegistryString): Moved here from old-run-webkit-tests. 2011-09-22 Adam Roben Use the "direct object" form of system() to run WebKit.exe This prevents the shell from parsing (and possibly splitting, if it contains spaces) the path to WebKit.exe. Fixes run-safari doesn't work if there are spaces in the path to %WEBKITOUTPUTDIR% Reviewed by Steve Falkenburg. * Scripts/webkitdirs.pm: (runSafari): 2011-09-22 Adam Roben Quote the WEBKITLIBRARIESDIR path before passing it to the shell/cygpath Fixes build-webkit goes totally haywire on Windows if there are spaces in the path to the WebKit source tree Reviewed by John Sullivan. * Scripts/webkitdirs.pm: (setupCygwinEnv): 2011-09-22 Tor Arne Vestbø [Qt] Use same DPI for application font as rest of app in DRT and WTR QApplication will initialize the default application font based on the application DPI at construction time, but we then override the application DPI using QX11Info (hard-coding it to 96 for consistency). This hard-coding is not reflected in the application font, so we explicitly have update the font ourselves. The 6 test results that are updated were originally produced with a DPI of 75, as this is the default fallback DPI when a QFont is constructed before QApplication. This was wrong, and the results are updated to match a DPI of 96. Reviewed by Csaba Osztrogonác. * DumpRenderTree/qt/main.cpp: * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp: 2011-09-21 Anna Cavender Update committers.py with a new contributor contact https://bugs.webkit.org/show_bug.cgi?id=68565 Reviewed by Eric Carlson. * Scripts/webkitpy/common/config/committers.py: 2011-09-21 Chang Shu [WK2] [Mac] Implement a more-complete MouseDown/MouseUp/MouseMoveTo functions for WebKit2 EventSender https://bugs.webkit.org/show_bug.cgi?id=68108 This patch implements the event sender mouse events through WebKit2 UIProcess which is closer to the real simulation than a WebProcess-only approach. The patch only supports Mac platform as the first step so the existing code is still kept for other platforms. Reviewed by Darin Adler. * WebKitTestRunner/EventSenderProxy.h: (WTR::EventSenderProxy::EventSenderProxy): (WTR::EventSenderProxy::leapForward): (WTR::EventSenderProxy::currentEventTime): * WebKitTestRunner/InjectedBundle/EventSendingController.cpp: (WTR::EventSendingController::EventSendingController): (WTR::EventSendingController::mouseDown): (WTR::EventSendingController::mouseUp): (WTR::EventSendingController::mouseMoveTo): (WTR::EventSendingController::leapForward): (WTR::EventSendingController::keyDown): * WebKitTestRunner/InjectedBundle/EventSendingController.h: * WebKitTestRunner/TestController.cpp: (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle): * WebKitTestRunner/mac/EventSenderProxy.mm: (WTR::eventTypeForMouseButtonAndAction): (WTR::operator==): (WTR::EventSenderProxy::updateClickCountForButton): (WTR::EventSenderProxy::mouseDown): (WTR::EventSenderProxy::mouseUp): (WTR::EventSenderProxy::mouseMoveTo): (WTR::EventSenderProxy::keyDown): * WebKitTestRunner/qt/EventSenderProxyQt.cpp: (WTR::EventSenderProxy::keyDown): (WTR::EventSenderProxy::mouseDown): (WTR::EventSenderProxy::mouseUp): (WTR::EventSenderProxy::mouseMoveTo): 2011-09-21 Dan Bernstein Tools part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes https://bugs.webkit.org/show_bug.cgi?id=68451 Reviewed by Darin Adler. * Scripts/check-for-inappropriate-objc-class-names: Added. Checks for Objective-C classes with names not having one of a list of prefixes passed on the command line. 2011-09-21 Xianzhu Wang Ensure TestWebKitAPI works on mac, win, chromium-mac and chromium-linux https://bugs.webkit.org/show_bug.cgi?id=68206 Reviewed by Tony Chang. * Scripts/webkitdirs.pm: (determineBaseProductDir): add support for chromium-mac * TestWebKitAPI/Tests/WTF/MetaAllocator.cpp: * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp: 2011-09-21 Adam Barth garden-o-matic command should explain that it is starting a local server https://bugs.webkit.org/show_bug.cgi?id=68486 Reviewed by Dimitri Glazkov. In the user study, both participants didn't understand why the garden-o-matic command appeared to hang. This patch adds an explaination of what's going on. It also adds a confirmation prompt before launching the browser to give the user a chance to read the text. If that's too much of a speed bump, we can iterate. * Scripts/webkitpy/tool/commands/gardenomatic.py: 2011-09-21 Chang Shu [Qt] Implement KeyDown function for WebKit2 EventSender. https://bugs.webkit.org/show_bug.cgi?id=56485 Implement the KeyDown function on Qt platform. Reviewed by Andreas Kling. * WebKitTestRunner/PlatformWebView.h: * WebKitTestRunner/TestController.cpp: (WTR::TestController::TestController): (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle): * WebKitTestRunner/qt/EventSenderProxyQt.cpp: Added. (WTR::getModifiers): (WTR::EventSenderProxy::keyDown): (buildModifierFlags): (EventSenderProxy::keyDown): * WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::sendEvent): * WebKitTestRunner/qt/WebKitTestRunner.pro: 2011-09-20 Adam Barth garden-o-matic unit test fail because Safari 5.1 doesn't enumerate dictionaries in the same order as V8 https://bugs.webkit.org/show_bug.cgi?id=68496 Reviewed by Dimitri Glazkov. This patch adds a bunch of sorting to make sure things happen in a predictable order in both JavaScriptCore and V8. We've talked before about whether these two JavaScript engines should enumerate dictionaries in the same order, but the consensus has been that they shouldn't. Personally, I disagree with that consensus, but such is life. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js: 2011-09-20 Adam Barth garden-o-matic doesn't work in Safari 5.1 because Safari 5.1 lacks Function.prototype.bind https://bugs.webkit.org/show_bug.cgi?id=68495 Reviewed by Dimitri Glazkov. This patch adds an implementation of Function.prototype.bind if the browser lacks one. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: 2011-09-20 Eric Seidel [NRWT] REGRESSION: Local loader tests are failing on machines that lost /tmp/LayoutTests symlink https://bugs.webkit.org/show_bug.cgi?id=65781 Reviewed by Ryosuke Niwa. Instead of making NRWT create the symlink, I made DumpRenderTree smart enough to resolve the passed in url relative to the absolute url for the test. For http tests, since the test url is an http url, we can't resolve relative to the test path, and thus use a new LOCAL_RESOURCE_ROOT environment variable for resolving. I believe this is a better approach than the on used in the Qt and Chromium DRT's (which resolves the path relative to the built location of the DRT executable) and we should move this new code into a shared location in a follow-up patch. It turns out that there was a second use for pathToLocalResource, used by one test (http/tests/plugins/post-url-file.html) used for getting a path to /tmp. To support this test I made the new pathToLocalResource smart enough to map /tmp to DUMPRENDERTREE_TEMP (which is already defined for all ports). * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::pathToLocalResource): * Scripts/webkitpy/layout_tests/port/webkit.py: Set LOCAL_RESOURCE_ROOT for use by DRT. 2011-09-20 Tom Zakrajsek Need to specify a django version, instead of using default https://bugs.webkit.org/show_bug.cgi?id=68398 Reviewed by Eric Seidel. * QueueStatusServer/__init__.py: * QueueStatusServer/main.py: 2011-09-20 Tom Zakrajsek Update committers.py with a few more contributor contacts https://bugs.webkit.org/show_bug.cgi?id=68487 Reviewed by Eric Seidel. * Scripts/webkitpy/common/config/committers.py: 2011-09-20 Leandro Pereira Fix webkitpy unit tests after 95547 and 95549 https://bugs.webkit.org/show_bug.cgi?id=68483 Reviewed by Adam Barth. Fix tests for ChangeLog.touched_files() and ChangeLog.parse_entries_from_file(). * Scripts/webkitpy/common/checkout/changelog_unittest.py: While testing parse_entries_from_file(), use StringIO and entry.reviewer_text(). While testing parse_latest_entry_from_file(), expect the correct touched_files() list. 2011-09-20 Adam Barth garden-o-matic makes it hard to compare images when scroll bars appear https://bugs.webkit.org/show_bug.cgi?id=68482 Reviewed by Eric Seidel. Because we autoscale image width, the images rescale if we overflow vertically and get a scrollbar. That makes it hard to compare two sets of images. This patch reserves space for the scrollbar so we won't have the layout change. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: 2011-09-20 Sheriff Bot Unreviewed, rolling out r95565. http://trac.webkit.org/changeset/95565 https://bugs.webkit.org/show_bug.cgi?id=68470 Broke builds (Requested by andersca on #webkit). * MiniBrowser/gtk/BrowserWindow.c: (mouseDidMoveOverElement): 2011-09-20 Nayan Kumar K Get hover'ed element URL from HitTest. mouseDidMoveOverElement now gets the hover'ed element link using WKHitTestResult API's. https://bugs.webkit.org/show_bug.cgi?id=68426 Reviewed by Anders Carlsson. * MiniBrowser/gtk/BrowserWindow.c: (mouseDidMoveOverElement): 2011-09-20 Leandro Pereira Changelog class should have a method to return all entries https://bugs.webkit.org/show_bug.cgi?id=68399 Implement ChangeLog.parse_entries_from_file(). This method returns a generator of ChangeLogEntry objects, ordered from the latest to the oldest entry in the file. Reviewed by Ryosuke Niwa. * Scripts/webkitpy/common/checkout/changelog.py: Copy parse_latest_entry_from_file() and adapt it to become a generator. * Scripts/webkitpy/common/checkout/changelog_unittest.py: Add test case. 2011-09-20 Jarred Nicholls [Qt] Permit qrc resources to load in QWebSettings::setUserStyleSheetUrl() https://bugs.webkit.org/show_bug.cgi?id=51159 Permit any URL with a local scheme to set a user style sheet. Allow qrc resource URLs as arguments to QWebSettings::setUserStyleSheetUrl(). Add DRT/WKTR qrc resource to test loading user style sheets from a qrc URL. Reviewed by Kenneth Rohde Christiansen. * DumpRenderTree/qt/DumpRenderTree.pro: * DumpRenderTree/qt/DumpRenderTree.qrc: Added. * DumpRenderTree/qt/resources/user.css: Added. (body): * WebKitTestRunner/qt/WebKitTestRunner.pro: * WebKitTestRunner/qt/WebKitTestRunner.qrc: Added. * WebKitTestRunner/qt/resources/user.css: Added. (body): 2011-09-20 Leandro Pereira ChangeLogEntry class should contain a list of touched files https://bugs.webkit.org/show_bug.cgi?id=68115 Reviewed by Eric Seidel. Add method to return the list of files touched in a ChangeLog entry. * Scripts/webkitpy/common/checkout/changelog.py: Add touched_files_regexp and touched_files() method. * Scripts/webkitpy/common/checkout/changelog_unittest.py: Implement unit tests for touched_files() method. 2011-09-20 Sheriff Bot Unreviewed, rolling out r95526 and r95541. http://trac.webkit.org/changeset/95526 http://trac.webkit.org/changeset/95541 https://bugs.webkit.org/show_bug.cgi?id=68443 It broke 6 tests (Requested by ossy_ on #webkit). * DumpRenderTree/qt/main.cpp: (main): * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp: (WTR::activateFonts): 2011-09-20 Tom Zakrajsek Fix patch and bug numbering in the mocktool simulation code https://bugs.webkit.org/show_bug.cgi?id=68374 Reviewed by Adam Barth. * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py: * Scripts/webkitpy/tool/bot/feeders_unittest.py: * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py: * Scripts/webkitpy/tool/bot/queueengine_unittest.py: * Scripts/webkitpy/tool/commands/download_unittest.py: * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py: * Scripts/webkitpy/tool/commands/queries_unittest.py: * Scripts/webkitpy/tool/commands/queues_unittest.py: * Scripts/webkitpy/tool/commands/queuestest.py: * Scripts/webkitpy/tool/commands/upload_unittest.py: * Scripts/webkitpy/tool/mocktool.py: * Scripts/webkitpy/tool/steps/steps_unittest.py: * Scripts/webkitpy/tool/steps/updatechangelogswithreview_unittest.py: 2011-09-20 Tor Arne Vestbø [Qt] Don't override default QApplication font for DRT and WTR Revision 95526 moved the relevant code below the QApplication constructor, to fix a potential crash, and that made the code take effect, breaking 6 tests. We should rely on the QWebSettings to set default fonts, and this QApplication::setFont() is a relic from when we didn't have that option, so we remove the code. Reviewed by Csaba Osztrogonác. * DumpRenderTree/qt/main.cpp: * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp: 2011-09-20 Tor Arne Vestbø WebKitTestRunner: Provide usage if run without arguments If WTR is run without either one or more filenames, or a '-' to indicate server mode, we don't need to start up the test machinery. Doing so might actually crash, as we'll exit the application immedeatly after anyways. Reviewed by Simon Hausmann. * WebKitTestRunner/TestController.cpp: (WTR::TestController::initialize): 2011-09-20 Tor Arne Vestbø [Qt] Don't use QFont before initializing QApplication Could potentially result in crash. Reviewed by Simon Hausmann. * DumpRenderTree/qt/main.cpp: (main): 2011-09-19 Adam Barth Always enable ENABLE(EVENTSOURCE) https://bugs.webkit.org/show_bug.cgi?id=68414 Reviewed by Eric Seidel. * Scripts/build-webkit: 2011-09-19 David Levin Sheriffbot rollout should be more intuitive. https://bugs.webkit.org/show_bug.cgi?id=68415 Reviewed by Adam Barth. * Scripts/webkitpy/tool/bot/irc_command.py: Add support for revert and comma separated args. * Scripts/webkitpy/tool/bot/irc_command_unittest.py: Add parsing tests for comma separated args and a few others cases. * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py: Verify that revert works. 2011-09-19 David Levin check-webkit-style generates bogus warning for StructuredExceptionHandlerSupressor.h https://bugs.webkit.org/show_bug.cgi?id=68391 Reviewed by Darin Adler. * Scripts/webkitpy/style/checkers/cpp.py: Skip __asm lines when doing style checks on a line. * Scripts/webkitpy/style/checkers/cpp_unittest.py: Added a test for this. 2011-09-19 Alexis Menard Style checker should not check public Qt WK2 API for parameter_name. https://bugs.webkit.org/show_bug.cgi?id=68359 Reviewed by Andreas Kling. The public Qt WK2 APIs have different rules than WebKit for parameters of methods in h files. Add an exception in the style checker to disable the check. Also added the test for it. * Scripts/webkitpy/style/checker.py: * Scripts/webkitpy/style/checker_unittest.py: 2011-09-19 Leandro Pereira GTK DRT leaks memory while comparing history items https://bugs.webkit.org/show_bug.cgi?id=68086 Reviewed by Martin Robinson. webkit_web_history_item_get_target() returns a g_strdup()'d string, so memory must be freed before compareHistoryItems() returns. * DumpRenderTree/gtk/DumpRenderTree.cpp: (compareHistoryItems): Use GOwnPtr to manage memory automatically. 2011-09-19 Kristóf Kosztyó Buildbot marks a nrwt bot red when tests are missing results https://bugs.webkit.org/show_bug.cgi?id=64812 Reviewed by Ryosuke Niwa. * Scripts/webkitpy/layout_tests/controllers/manager.py: * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: 2011-09-19 Jessie Berlin Sites that use history pushState or replaceState are recorded in history in Private Browsing mode. https://bugs.webkit.org/show_bug.cgi?id=68208 Reviewed by Brady Eidson. Add a test that the WKContextDidNavigateWithNavigationDataCallback (which is invoked by updateGlobalHistory) is not called in Private Browsing when the page does history.pushState. * TestWebKitAPI/Tests/WebKit2/PrivateBrowsingPushStateNoHistoryCallback.cpp: Added. (TestWebKitAPI::didNavigateWithNavigationData): This is a context history client callback, so it should not be called when in Private Browsing. Add a call to FAIL(). (TestWebKitAPI::didSameDocumentNavigationForFrame): This is a page load client callback that happens after the didNavigateWithNavigationData callback. End the test. (TestWebKitAPI::TEST): Enable Private Browsing and load a page that uses history.pushState. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add PrivateBrowsingPushStateNoHistoryCallback.cpp and push-state.html. * TestWebKitAPI/Tests/WebKit2/push-state.html: Added. Ditto. * TestWebKitAPI/win/TestWebKitAPI.vcproj: Ditto. * TestWebKitAPI/win/copy-resources.cmd: Ditto. 2011-09-16 Elliot Poger add non-CG Webkit-Mac builds to flakiness dashboard https://bugs.webkit.org/show_bug.cgi?id=68265 Reviewed by Adam Barth. As Chromium-on-Mac approaches the switchover from Core Graphics to Skia, we need to add the Skia version to the flakiness dashboard. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js: * TestResultServer/static-dashboards/builders.js: * TestResultServer/static-dashboards/flakiness_dashboard_tests.js: 2011-09-16 Jochen Eisinger Rename FrameLoaderClient::allowImages to FrameLoaderClient::allowImage and include the image URL as parameter https://bugs.webkit.org/show_bug.cgi?id=68071 Reviewed by Adam Barth. * DumpRenderTree/chromium/WebPermissions.cpp: (WebPermissions::allowImage): * DumpRenderTree/chromium/WebPermissions.h: 2011-09-16 Tom Zakrajsek update-webkit-chromium should retry the gclient call if it fails once https://bugs.webkit.org/show_bug.cgi?id=68252 Reviewed by Adam Barth. * Scripts/update-webkit-chromium: 2011-09-16 Rob Buis Add Eli's rim address and change the nick. * Scripts/webkitpy/common/config/committers.py: 2011-09-16 Anders Carlsson Add back the WKHitTestResultRef parameter to mouseDidMoveOverElement in an ABI preserving way https://bugs.webkit.org/show_bug.cgi?id=68269 Reviewed by Geoffrey Garen. Update client function signatures. * MiniBrowser/gtk/BrowserWindow.c: (browserWindowUIClientInit): * MiniBrowser/mac/BrowserWindowController.m: (mouseDidMoveOverElement): (-[BrowserWindowController awakeFromNib]): * MiniBrowser/win/BrowserView.cpp: (BrowserView::create): * WebKitTestRunner/TestController.cpp: (WTR::TestController::createOtherPage): (WTR::TestController::initialize): 2011-09-16 Anders Carlsson Temporarily remove the WKHitTestResultRef parameter to avoid breaking the WK2 ABI/API https://bugs.webkit.org/show_bug.cgi?id=68266 Reviewed by John Sullivan. * MiniBrowser/mac/BrowserWindowController.m: (mouseDidMoveOverElement): 2011-09-16 Vincent Scheib Add Vincent Scheib to the committers list. * Scripts/webkitpy/common/config/committers.py: 2011-09-16 Sheriff Bot Unreviewed, rolling out r95285. http://trac.webkit.org/changeset/95285 https://bugs.webkit.org/show_bug.cgi?id=68247 It broke a webkitpy test (Requested by ossy on #webkit). * Scripts/webkitpy/layout_tests/controllers/manager.py: 2011-09-16 Kristóf Kosztyó Buildbot marks a nrwt bot red when tests are missing results https://bugs.webkit.org/show_bug.cgi?id=64812 Reviewed by Dirk Pranke. * Scripts/webkitpy/layout_tests/controllers/manager.py: 2011-09-16 Martin Robinson [GTK] Disable WebKit2 by default in configure.ac, but enable it with build-webkit https://bugs.webkit.org/show_bug.cgi?id=68178 Reviewed by Xan Lopez. Add a work-around to enable WebKit2 when using build-webkit. * Scripts/webkitdirs.pm: (buildAutotoolsProject): 2011-09-15 James Robinson Add Antoine Labour and Shawn Singh to the contributors-but-not-committers list. * Scripts/webkitpy/common/config/committers.py: 2011-09-15 Tom Zakrajsek UpdateChangeLogsWithReviewer should be able to determine reviewer of latest reviewed patch in a bug https://bugs.webkit.org/show_bug.cgi?id=68003 Reviewed by Adam Barth. * Scripts/webkitpy/tool/mocktool.py: * Scripts/webkitpy/tool/steps/updatechangelogswithreview_unittest.py: * Scripts/webkitpy/tool/steps/updatechangelogswithreviewer.py: 2011-09-15 Tom Zakrajsek Python version check is confusing in test-webkitpy https://bugs.webkit.org/show_bug.cgi?id=68004 Reviewed by Adam Barth. * Scripts/test-webkitpy: 2011-09-15 Eric Seidel Reviewed by Adam Barth. webkit-patch should be able to find users and add them to bugzilla groups https://bugs.webkit.org/show_bug.cgi?id=63351 These are both very basic commands. But it's now possible to find all users matching a regexp, as well as add all users matching a regexp to a set of groups. bugzilla.py already knew how to find users (for validate-committer-lists) but now it has the ability to modify the user records. I split some of the logic out into a new EditUsersParser class to try and reduce the amount of code in Bugzilla/BugzillaQueries. * Scripts/webkitpy/common/net/bugzilla/bugzilla.py: * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py: * Scripts/webkitpy/tool/commands/__init__.py: * Scripts/webkitpy/tool/commands/adduserstogroups.py: Added. * Scripts/webkitpy/tool/commands/findusers.py: Added. 2011-09-15 Eric Seidel Remove ENABLE(SVG_AS_IMAGE) since all major ports have it on by default https://bugs.webkit.org/show_bug.cgi?id=68182 Reviewed by Adam Barth. * Scripts/build-webkit: * waf/build/settings.py: 2011-09-15 Eric Seidel Unreviewed. Make contributor email look-up case-insensitive. validate-committer-list was incorrectly reporting that "chang.shu@nokia.com" was missing from committers.py due to case sensitivity. This also includes a test expectations update to committers_unittest.py after my previous change. * Scripts/webkitpy/common/config/committers.py: * Scripts/webkitpy/common/config/committers_unittest.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: 2011-09-15 Eric Seidel Unreviewed. Updated this based on webkit-patch suggest-nominations and validate-committer-list output. Add a bunch of email aliases to committers.py for regular contributors who are committing using other email addresses than they have listed in the file. * Scripts/webkitpy/common/config/committers.py: 2011-09-15 Eric Seidel Remove ENABLE_SVG_ANIMATION as all major ports have it on by default https://bugs.webkit.org/show_bug.cgi?id=68022 Reviewed by Ryosuke Niwa. * Scripts/build-webkit: * waf/build/settings.py: 2011-09-15 Anders Carlsson Fix Mac build by including config.h where needed. * TestWebKitAPI/Tests/TestWebKitAPI/mac/InstanceMethodSwizzler.mm: * TestWebKitAPI/Tests/WTF/MetaAllocator.cpp: * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp: * TestWebKitAPI/Tests/WebKit2/WebArchive.cpp: * TestWebKitAPI/Tests/WebKit2/WebArchive_Bundle.cpp: * TestWebKitAPI/Tests/mac/DOMRangeOfString.mm: * TestWebKitAPI/Tests/mac/DeviceScaleFactorOnBack.mm: * TestWebKitAPI/Tests/mac/DynamicDeviceScaleFactor.mm: * TestWebKitAPI/Tests/mac/StringByEvaluatingJavaScriptFromString.mm: * TestWebKitAPI/mac/InstanceMethodSwizzler.mm: * TestWebKitAPI/mac/JavaScriptTestMac.mm: * TestWebKitAPI/mac/PlatformUtilitiesMac.mm: * TestWebKitAPI/mac/PlatformWebViewMac.mm: * TestWebKitAPI/mac/SyntheticBackingScaleFactorWindow.m: * TestWebKitAPI/mac/WebKitAgnosticTest.mm: * TestWebKitAPI/mac/main.mm: 2011-09-15 Xianzhu Wang Let TestWebKitAPI work for chromium https://bugs.webkit.org/show_bug.cgi?id=67756 Reviewed by Sam Weinig. * Scripts/build-api-tests: * Scripts/run-api-tests: (isSupportedPlatform): (runTest): (populateTests): * Scripts/webkitdirs.pm: (determineBaseProductDir): * TestWebKitAPI/Configurations/InjectedBundle.xcconfig: * TestWebKitAPI/Configurations/TestWebKitAPI.xcconfig: * TestWebKitAPI/Configurations/TestWebKitAPICommon.vsprops: * TestWebKitAPI/InjectedBundleController.cpp: * TestWebKitAPI/InjectedBundleMain.cpp: * TestWebKitAPI/JavaScriptTest.cpp: * TestWebKitAPI/PlatformUtilities.cpp: * TestWebKitAPI/TestWebKitAPI.gypi: Added. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/TestWebKitAPIPrefix.h: Removed. * TestWebKitAPI/Tests/WTF/CheckedArithmeticOperations.cpp: * TestWebKitAPI/Tests/WTF/StringOperators.cpp: * TestWebKitAPI/Tests/WTF/VectorBasic.cpp: * TestWebKitAPI/Tests/WTF/VectorReverse.cpp: * TestWebKitAPI/Tests/WebKit2/AboutBlankLoad.cpp: * TestWebKitAPI/Tests/WebKit2/CanHandleRequest.cpp: * TestWebKitAPI/Tests/WebKit2/CanHandleRequest_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/CookieManager.cpp: * TestWebKitAPI/Tests/WebKit2/DocumentStartUserScriptAlertCrash.cpp: * TestWebKitAPI/Tests/WebKit2/DocumentStartUserScriptAlertCrash_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/DownloadDecideDestinationCrash.cpp: * TestWebKitAPI/Tests/WebKit2/EvaluateJavaScript.cpp: * TestWebKitAPI/Tests/WebKit2/FailedLoad.cpp: * TestWebKitAPI/Tests/WebKit2/Find.cpp: * TestWebKitAPI/Tests/WebKit2/ForceRepaint.cpp: * TestWebKitAPI/Tests/WebKit2/FrameMIMETypeHTML.cpp: * TestWebKitAPI/Tests/WebKit2/FrameMIMETypePNG.cpp: * TestWebKitAPI/Tests/WebKit2/HitTestResultNodeHandle.cpp: * TestWebKitAPI/Tests/WebKit2/HitTestResultNodeHandle_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/InjectedBundleBasic.cpp: * TestWebKitAPI/Tests/WebKit2/InjectedBundleBasic_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/LoadCanceledNoServerRedirectCallback.cpp: * TestWebKitAPI/Tests/WebKit2/LoadCanceledNoServerRedirectCallback_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/MouseMoveAfterCrash.cpp: * TestWebKitAPI/Tests/WebKit2/MouseMoveAfterCrash_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp: * TestWebKitAPI/Tests/WebKit2/PageLoadDidChangeLocationWithinPageForFrame.cpp: * TestWebKitAPI/Tests/WebKit2/PreventEmptyUserAgent.cpp: * TestWebKitAPI/Tests/WebKit2/ResponsivenessTimerDoesntFireEarly.cpp: * TestWebKitAPI/Tests/WebKit2/ResponsivenessTimerDoesntFireEarly_Bundle.cpp: * TestWebKitAPI/Tests/WebKit2/RestoreSessionStateContainingFormData.cpp: * TestWebKitAPI/Tests/WebKit2/SpacebarScrolling.cpp: * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp: * TestWebKitAPI/Tests/WebKit2/WKString.cpp: * TestWebKitAPI/Tests/WebKit2/WKStringJSString.cpp: * TestWebKitAPI/Tests/WebKit2/win/AltKeyGeneratesWMSysCommand.cpp: * TestWebKitAPI/Tests/WebKit2/win/DoNotCopyANullCFURLResponse.cpp: * TestWebKitAPI/Tests/WebKit2/win/HideFindIndicator.cpp: * TestWebKitAPI/Tests/WebKit2/win/ResizeViewWhileHidden.cpp: * TestWebKitAPI/Tests/WebKit2/win/TranslateMessageGeneratesWMChar.cpp: * TestWebKitAPI/Tests/WebKit2/win/WMCloseCallsUIClientClose.cpp: * TestWebKitAPI/Tests/WebKit2/win/WMPrint.cpp: * TestWebKitAPI/TestsController.cpp: * TestWebKitAPI/config.h: Copied from TestWebKitAPI/TestWebKitAPIPrefix.h. * TestWebKitAPI/win/InjectedBundleControllerWin.cpp: * TestWebKitAPI/win/PlatformUtilitiesWin.cpp: * TestWebKitAPI/win/PlatformWebViewWin.cpp: * TestWebKitAPI/win/TestWebKitAPI.vcproj: * TestWebKitAPI/win/main.cpp: 2011-09-15 Jochen Eisinger [chromium] extract normalization of LayoutTests URLs to a method of TestShell https://bugs.webkit.org/show_bug.cgi?id=68145 This will allow for other classes in DumpRenderTree to reuse this functionality Reviewed by David Levin. * DumpRenderTree/chromium/TestShell.cpp: (normalizeLayoutTestURLInternal): (dumpHistoryItem): (TestShell::normalizeLayoutTestURL): * DumpRenderTree/chromium/TestShell.h: 2011-09-14 Ada Chan Add test for WKBundleFrameCopyWebArchive(). https://bugs.webkit.org/show_bug.cgi?id=67857 Reviewed by Anders Carlsson. * TestWebKitAPI/Tests/WebKit2/WebArchive.cpp: Added. (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Do some basic sanity checks to make sure the returned data is a valid webarchive. (TestWebKitAPI::setInjectedBundleClient): (TestWebKitAPI::didFinishLoadForFrame): (TestWebKitAPI::TEST): Load simple.html in a WebView. Wait for it to finish loading, and then ask the injected bundle for the webarchive and check the returned webarchive. * TestWebKitAPI/Tests/WebKit2/WebArchive_Bundle.cpp: Added. (TestWebKitAPI::WebArchiveTest::WebArchiveTest): (TestWebKitAPI::WebArchiveTest::didReceiveMessage): Handle the "GetWebArchive" message and return the web archive. Add files to project. * TestWebKitAPI/win/TestWebKitAPI.vcproj: Also let Visual Studio reorder a file. * TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj: * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: 2011-09-14 Ryosuke Niwa committers.py should support multiple IRC nicknames https://bugs.webkit.org/show_bug.cgi?id=68110 Reviewed by Eric Seidel. Add multiple IRC nickname support to committers.py * Scripts/webkitpy/common/config/committers.py: * Scripts/webkitpy/common/config/committers_unittest.py: * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/sheriff.py: 2011-09-14 Jon Lee Implement globalFlag for WebKitTestRunner https://bugs.webkit.org/show_bug.cgi?id=68033 Reviewed by Sam Weinig. * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: Porting globalFlag from DRT * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: (WTR::LayoutTestController::LayoutTestController): initialize m_globalFlag to false * WebKitTestRunner/InjectedBundle/LayoutTestController.h: add m_globalFlag (WTR::LayoutTestController::globalFlag): simple getter (WTR::LayoutTestController::setGlobalFlag): simple setter 2011-09-13 Eric Seidel Remove ENABLE_SVG_USE as is required by HTML5 https://bugs.webkit.org/show_bug.cgi?id=68019 Reviewed by Ryosuke Niwa. * Scripts/build-webkit: * waf/build/settings.py: 2011-09-14 Ryosuke Niwa Remove Jan Erik Hanssen (added in my r95078) from the list for now since I can't confirm acceptable his committership. * Scripts/webkitpy/common/config/committers.py: 2011-09-14 Jochen Eisinger [chromium] Remove obsolete WebViewClient methods from DumpRenderTree. https://bugs.webkit.org/show_bug.cgi?id=68066 Reviewed by Adam Barth. * DumpRenderTree/chromium/WebViewHost.cpp: * DumpRenderTree/chromium/WebViewHost.h: 2011-09-13 Ryosuke Niwa committers.py is missing some contributors https://bugs.webkit.org/show_bug.cgi?id=68059 Reviewed by Adam Barth. Add David Barr, Jan Erik Hanssen, and Kim Grönholm. Also add nicks of Alejandro G. Castro, Collin Jackson, and Matt Lilek. * Scripts/webkitpy/common/config/committers.py: 2011-09-13 Ryosuke Niwa Fix webkit python test after r95061. * Scripts/webkitpy/tool/bot/irc_command_unittest.py: 2011-09-13 Xianzhu Wang Add myself as a contributor. No review needed. * Scripts/webkitpy/common/config/committers.py: 2011-09-13 Adam Barth garden-o-matic has unfriendly error messages when the local server is not available https://bugs.webkit.org/show_bug.cgi?id=68042 Reviewed by Dimitri Glazkov. This patch moves us from using alert to showing the error inline in the progress dialog. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: 2011-09-13 Adam Barth FailureGrid in garden-o-matic should link to the builder's waterfall page https://bugs.webkit.org/show_bug.cgi?id=68036 Reviewed by Dimitri Glazkov. One of the users in the user study thought the configurations names in the failure grid would link to the underlying bots. That makes sense, so this patch adds the behavior. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: 2011-09-13 Adam Barth garden-o-matic should be able to mark failures as expected https://bugs.webkit.org/show_bug.cgi?id=68027 Reviewed by Dimitri Glazkov. All the back-end infrastructure exists for this operation. This patch just adds some UI to the front-end and wires that UI up to the back end. This exact UI will likely need some more polish, but this should be a reasonable first cut. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: 2011-09-13 Adam Barth garden-o-matic should tell you about the progress of your rebaseline https://bugs.webkit.org/show_bug.cgi?id=67967 Reviewed by Dimitri Glazkov. This patch adds ui.MessageBox, which is a light wrapper around jQuery UI's dialog widget. We then use ui.MessageBox to show status messages about in-progress rebaselines. As each baseline is downloaded, the MessageBox updates. This UI replaces our temporary "alert"-based UI. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css: 2011-09-13 Leandro Pereira GTK DumpRenderTree uses inefficient idioms to iterate over G[S]Lists https://bugs.webkit.org/show_bug.cgi?id=68024 Reviewed by Gustavo Noronha Silva. Using g_list_count() and g_list_nth_data() together on a loop is inneficient since they're both O(n). Iterate over lists in a saner way. * DumpRenderTree/gtk/DumpRenderTree.cpp: (dumpHistoryItem): Reduce the scope for the 'kids' variable, and iterate on it using g_list_next(). Free the list after done with it. (dumpBackForwardListForWebView): Instead of appending (which is expensive in GLists) history items and then iterating from the tail of the itemsToPrint list, prepend items and walk forwards as usual. (dumpBackForwardListForAllWebViews): Walk the list in a saner way, remove the (unneeded) viewList variable. 2011-09-13 Ryosuke Niwa Add Eric's IRC nick. * Scripts/webkitpy/common/config/committers.py: 2011-09-13 Anders Carlsson Disable C++ exceptions when building with clang https://bugs.webkit.org/show_bug.cgi?id=68031 Reviewed by Mark Rowe. * MiniBrowser/Configurations/Base.xcconfig: * TestWebKitAPI/Configurations/Base.xcconfig: * WebKitTestRunner/Configurations/Base.xcconfig: 2011-09-13 Ojan Vafai [chromium] fix flakiness dashboard for chromium interactive tests https://bugs.webkit.org/show_bug.cgi?id=68029 Reviewed by Tony Chang. * TestResultServer/static-dashboards/flakiness_dashboard.html: * TestResultServer/static-dashboards/flakiness_dashboard_tests.js: 2011-09-13 Ryosuke Niwa Add Alice and Aryeh to the list of contributors. * Scripts/webkitpy/common/config/committers.py: 2011-09-13 Ryosuke Niwa sheriffbot whois should also tell us email addresses https://bugs.webkit.org/show_bug.cgi?id=67975 Reviewed by Eric Seidel and David Levin. Taught sheriffbot how to tell us contributor's email addresses. * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: 2011-09-13 Eric Seidel Remove ENABLE_SVG_FOREIGN_OBJECT as it is a required part of HTML5 https://bugs.webkit.org/show_bug.cgi?id=68018 Reviewed by Ryosuke Niwa. * Scripts/build-webkit: * waf/build/settings.py: 2011-09-13 Chang Shu [WK2] [Mac] Implement KeyDown function for WebKit2 EventSender. https://bugs.webkit.org/show_bug.cgi?id=57515 The code change in WebKitTestRunner first sends keydown event from InjectedBundle to TestController synchronously. The latter then sends simulated event to WebKit2 module, which is synchronous, too. Reviewed by Darin Adler. * WebKitTestRunner/EventSenderProxy.h: Added. (WTR::EventSenderProxy::EventSenderProxy): * WebKitTestRunner/InjectedBundle/Bindings/EventSendingController.idl: * WebKitTestRunner/InjectedBundle/EventSendingController.cpp: (WTR::EventSendingController::keyDown): * WebKitTestRunner/InjectedBundle/EventSendingController.h: * WebKitTestRunner/PlatformWebView.h: (WTR::PlatformWebView::platformWindow): * WebKitTestRunner/TestController.cpp: (WTR::TestController::TestController): (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle): * WebKitTestRunner/TestController.h: * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: * WebKitTestRunner/mac/EventSenderProxy.mm: Added. (WTR::buildModifierFlags): (WTR::EventSenderProxy::keyDown): 2011-09-12 Leandro Pereira [EFL] DRT: Add DumpRenderTreeEfl.h https://bugs.webkit.org/show_bug.cgi?id=63993 Reviewed by Eric Seidel. This header contains some global variables used by EFL's DumpRenderTree implementation. * DumpRenderTree/DumpRenderTree.h: Include DumpRenderTreeEfl.h when appropriate. * DumpRenderTree/efl/DumpRenderTreeEfl.h: Added. 2011-09-12 Raphael Kubo da Costa [EFL] Send the right key names for PageUp and PageDown. https://bugs.webkit.org/show_bug.cgi?id=67044 Reviewed by Kenneth Rohde Christiansen. PlatformKeyboardEventEfl uses "Prior" and "Next" to refer to PageUp and PageDown (which is close to the defines in WindowsKeyboardCodes.h). However, EventSender was sending "Page_Up" and "Page_Down", so tests such as editing/input/page-up-down-scrolls.html were timing out due to the right key down events not being sent. * DumpRenderTree/efl/EventSender.cpp: (keyPadNameFromJSValue): (keyNameFromJSValue): 2011-09-12 Eric Seidel sheriffbot whois is case sensitive https://bugs.webkit.org/show_bug.cgi?id=67972 Reviewed by Adam Barth. I also fixed a typo in the "More than 5 contributors" message from my previous commit. * Scripts/webkitpy/common/config/committers.py: * Scripts/webkitpy/common/config/committers_unittest.py: * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: 2011-09-12 Eric Seidel sheriffbot whois should be smarter https://bugs.webkit.org/show_bug.cgi?id=67971 Reviewed by Adam Barth. This makes sheriffbot whois basically grep the committers list making it much easier to find a contributor by name. Unfortunately the current unittests depend on the real committers.py file, so as we add more eric's to the project, this unittest result will change. I think that's OK for now. We'll fix it when it breaks. * Scripts/webkitpy/common/config/committers.py: * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: 2011-09-12 Sheriff Bot Unreviewed, rolling out r94976. http://trac.webkit.org/changeset/94976 https://bugs.webkit.org/show_bug.cgi?id=67953 Broke a bunch of security/load& tests (Requested by rniwa on #webkit). * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::pathToLocalResource): 2011-09-12 Eric Seidel Rubber-stamped by Adam Barth. Remove the "the tree is burning!" warning previously shown when landing patches. It was a dream. But it's just noise now. Some bots will require a restart to notice the removal of the option so I've left the --ignore-builders option in for now. * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py: * Scripts/webkitpy/tool/bot/patchanalysistask.py: * Scripts/webkitpy/tool/commands/download.py: * Scripts/webkitpy/tool/commands/download_unittest.py: * Scripts/webkitpy/tool/commands/queues_unittest.py: * Scripts/webkitpy/tool/steps/__init__.py: * Scripts/webkitpy/tool/steps/commit.py: * Scripts/webkitpy/tool/steps/ensurebuildersaregreen.py: Removed. * Scripts/webkitpy/tool/steps/postdiffforrevert.py: 2011-09-12 Eric Seidel [NRWT] REGRESSION: Local loader tests are failing on machines that lost /tmp/LayoutTests symlink https://bugs.webkit.org/show_bug.cgi?id=65781 Reviewed by Ryosuke Niwa. Instead of making NRWT create the symlink, I just made DumpRenderTree smart enough to resolve the passed in url relative to the absolute url for the test. I believe this is a better approach than the on used in the Qt and Chromium DRT's (which resolves the path relative to the built location of the DRT executable) and we should move this new code into a shared location in a follow-up patch. * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::pathToLocalResource): 2011-09-12 Eric Seidel Reshuffle some code in WebKitDriver._read_block in preparation for reading stderr/stdout separately https://bugs.webkit.org/show_bug.cgi?id=67530 Reviewed by Adam Barth. I believe I've fixed the bug in the original patch which prompted the rollout. The previous patch was using the wrong deadline for the initial read, subtracting time.time() twice from the deadline value. * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: 2011-09-12 Balazs Kelemen [Qt][WK2] WebKitTestRunner does not produce crash logs https://bugs.webkit.org/show_bug.cgi?id=67714 Reviewed by Chang Shu. Added a simple way of generating backtrace on crash to the web process. The implementation is similar what we have in DRT. It depends on GNU libc functionality so it is only enabled where we are running in such an environment. * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp: (WTR::printBacktrace): (WTR::crashHandler): (WTR::InjectedBundle::platformInitialize): 2011-09-11 Filip Pizlo Added my IRC nick to the committers file. * Scripts/webkitpy/common/config/committers.py: 2011-09-01 Filip Pizlo The executable allocator makes it difficult to free individual chunks of executable memory https://bugs.webkit.org/show_bug.cgi?id=66363 Reviewed by Oliver Hunt. Introduced a best-fit, balanced-tree based allocator. The allocator required a balanced tree that does not allocate memory and that permits the removal of individual nodes directly (as opposed to by key); neither AVLTree nor WebCore's PODRedBlackTree supported this. Changed all references to executable code to use a reference counted handle. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/WTF/MetaAllocator.cpp: Added. (TestWebKitAPI::TEST_F): * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp: Added. (TestWebKitAPI::Pair::findExact): (TestWebKitAPI::Pair::remove): (TestWebKitAPI::Pair::findLeastGreaterThanOrEqual): (TestWebKitAPI::Pair::assertFoundAndRemove): (TestWebKitAPI::Pair::assertEqual): (TestWebKitAPI::Pair::assertSameValuesForKey): (TestWebKitAPI::Pair::testDriver): (TestWebKitAPI::TEST_F): 2011-09-10 Andy Estes Move myself from committers_unable_to_review to reviewers_list. * Scripts/webkitpy/common/config/committers.py: 2011-09-10 Kevin Ollivier [wx] Unreviewed build fix. MSW build fixes. * DumpRenderTree/config.h: * waf/build/wxpresets.py: 2011-09-10 Andreas Kling Crash beneath WKURLIsEqual() when reloading in MiniBrowser. https://bugs.webkit.org/show_bug.cgi?id=59328 Reviewed by Kenneth Rohde Christiansen. * MiniBrowser/mac/BrowserWindowController.m: (-[BrowserWindowController updateProvisionalURLForFrame:]): Add missing null check. 2011-09-10 Adam Barth Build fix for Android. DumpRenderTree is a client of the WebKit API and does not have access to notImplemented(). * DumpRenderTree/chromium/TestShellAndroid.cpp: (openStartupDialog): 2011-09-09 Alexey Proskuryakov Add a script to create a monolithic script-test https://bugs.webkit.org/show_bug.cgi?id=67747 Reviewed by Darin Adler. * Scripts/make-new-script-test: Added. 2011-09-09 Luiz Agostini Adding myself to the reviewers list. * Scripts/webkitpy/common/config/committers.py: 2011-09-09 Ryosuke Niwa Buildbot may use RunWebKitTest's evaluateCommand on bots that run nrwt https://bugs.webkit.org/show_bug.cgi?id=67855 Reviewed by Dirk Pranke. The bug was caused by the fact some ports use RunWebKitTests instead of NewRunWebKitTests to run tests, and using RunWebKitTests.evaluateCommand after calling _parseNewRunWebKitTestsOutput. Fixed the bug by merging evaluateCommand of RunWebKitTests and NewRunWebKitTests. * BuildSlaveSupport/build.webkit.org-config/master.cfg: 2011-09-09 Mark Hahnenberg Add myself to committers.py https://bugs.webkit.org/show_bug.cgi?id=67853 Reviewed by Darin Adler. * Scripts/webkitpy/common/config/committers.py: 2011-09-09 Laszlo Gombos [Qt] Remove common.pri https://bugs.webkit.org/show_bug.cgi?id=67814 Reviewed by Andreas Kling. * Tools.pro: 2011-09-09 Kristóf Kosztyó [NRWT] the nrwt check httpd with --no-http option https://bugs.webkit.org/show_bug.cgi?id=67835 Reviewed by Tony Chang. * Scripts/webkitpy/layout_tests/controllers/manager.py: * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py: 2011-09-09 Hao Zheng [chromium] Implement DRT support for Android. https://bugs.webkit.org/show_bug.cgi?id=67590 Reviewed by Tony Chang. Add 2 mechanism to make DRT interaction more robust on Android: - Send 'QUIT' to DRT to make it stop. - Wait until receiving '#READY' from DRT, so that DRT won't miss any input commands before it starts. And LayoutTestController.cpp needs to include for 'tolower' and for 'LC_ALL' to compile on Android. Also change to use sigaction instead of signal for TestShellGtk.cpp, as signal() man page recommends using sigaction(). * DumpRenderTree/DumpRenderTree.gypi: * DumpRenderTree/chromium/DumpRenderTree.cpp: (main): * DumpRenderTree/chromium/LayoutTestController.cpp: * DumpRenderTree/chromium/TestShellAndroid.cpp: Added. (AlarmHandler): (TestShell::waitTestFinished): (platformInit): (openStartupDialog): (checkLayoutTestSystemDependencies): * DumpRenderTree/chromium/TestShellGtk.cpp: (AlarmHandler): (TestShell::waitTestFinished): 2011-09-09 Kevin Ollivier [wx] Unreviewed build fix. Add Source/JavaScriptCore/dfg directory to the build. * waf/build/settings.py: 2011-09-09 Sheriff Bot Unreviewed, rolling out r94851. http://trac.webkit.org/changeset/94851 https://bugs.webkit.org/show_bug.cgi?id=67847 It broke a webkitpy test (Requested by ossy on #webkit). * Scripts/webkitpy/layout_tests/controllers/manager.py: 2011-09-09 Kristóf Kosztyó [NRWT] the nrwt check httpd with --no-http option https://bugs.webkit.org/show_bug.cgi?id=67835 Reviewed by Csaba Osztrogonác. * Scripts/webkitpy/layout_tests/controllers/manager.py: 2011-09-08 Brian Weinstein WebKit2: Assertion when calling didPerform*Redirect on null source/destination URL string https://bugs.webkit.org/show_bug.cgi?id=67794 Add a test that we don't send a didPerformServerRedirect when a load is cancelled from willSendRequest. Reviewed by Brady Eidson. * TestWebKitAPI/Tests/WebKit2/LoadCanceledNoServerRedirectCallback.cpp: Added. (TestWebKitAPI::didFinishLoadForFrame): Set flags on whether we've loaded the main frame, iframe, or both. (TestWebKitAPI::didPerformServerRedirect): Set a flag for whether we've performed a server redirect. (TestWebKitAPI::TEST): Set up a WKPageLoaderClient and WKContextHistoryClient, and load simple-iframe.html. Once both frames have loaded (main frame and blocked iframe), make sure that we have never called didPerformServerRedirect. * TestWebKitAPI/Tests/WebKit2/LoadCanceledNoServerRedirectCallback_Bundle.cpp: Added. (TestWebKitAPI::LoadCanceledNoServerRedirectCallbackTest::LoadCanceledNoServerRedirectCallbackTest): Initialize our InjectedBundleTest. (TestWebKitAPI::LoadCanceledNoServerRedirectCallbackTest::willSendRequestForFrame): If we are dealing with a request from the main frame (the main resource in simple-iframe.html), allow it. If not (dealing with the iframe), block the request. (TestWebKitAPI::LoadCanceledNoServerRedirectCallbackTest::didCreatePage): Set up the WKBundlePageResourceLoadClient. * TestWebKitAPI/Tests/WebKit2/simple-iframe.html: Added. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add new files. * TestWebKitAPI/win/TestWebKitAPI.vcproj: Ditto. * TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj: Ditto. * TestWebKitAPI/win/copy-resources.cmd: Ditto. 2011-09-08 David Levin check-webkit-style: Ref|Own Ptr& should be allowed as a parameter. https://bugs.webkit.org/show_bug.cgi?id=67797 Reviewed by Eric Seidel. * Scripts/webkitpy/style/checkers/cpp.py: Allow Ref/Own Ptr when it ends with &. * Scripts/webkitpy/style/checkers/cpp_unittest.py: Added test cases for this. 2011-09-08 Fady Samuel Fixed Layout Mode should be adjustable from layoutTestController for testing on Chromium platforms https://bugs.webkit.org/show_bug.cgi?id=67723 Reviewed by Adam Barth. * DumpRenderTree/chromium/EventSender.h: * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (LayoutTestController::enableFixedLayoutMode): (LayoutTestController::setFixedLayoutSize): * DumpRenderTree/chromium/LayoutTestController.h: 2011-09-07 Matthew Delaney Add standalone script that filters the output of build-webkit to be more human-readable https://bugs.webkit.org/show_bug.cgi?id=44081 Reviewed by Daniel Bates. * Scripts/filter-build-webkit: Added. Just need to feed it the output of a build script (e.g. build-webkit) and it spits out a prettier version. * Scripts/VCSUtils.pm: Refactored "possiblyColored" out of run-api-tests into VCSUtils. * Scripts/run-api-tests: Now includes VCSUtils to use "possiblyColored" 2011-09-07 Noel Gordon Add myself to committers.py https://bugs.webkit.org/show_bug.cgi?id=67696 Reviewed by Kent Tamura. * Scripts/webkitpy/common/config/committers.py: 2011-09-07 Adam Barth rebaseline button in garden-o-matic details view should work https://bugs.webkit.org/show_bug.cgi?id=67738 Reviewed by Dimitri Glazkov. This patch wires up the rebaseline button in garden-o-matic results details view. The button is current per-test, so there isn't a way to rebaseline individual bots. This is a place to start. If we decide we want per-bot rebaselining, we'll need to think of some better UI. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css: 2011-09-07 Eric Seidel Teach the PortFactory object how to pass along executive/user/filesystem to port objects (which fixes the failing rebaseline_test on bots) https://bugs.webkit.org/show_bug.cgi?id=67737 Reviewed by Adam Barth. I got a little over-eager when deprecating host.port_factory in the previous commit. Turns out that Host.port_factory was already the "modern" port/factory.py it was just the Module object. I un-deprecated the variable, and changed it from being the module to being an actual PortFactory, passing it a host object. I also taught PortFactory how to set user/executive/filesystem in the kwargs when instantiating a new Port object. This magically fixed the test_rebaseline test, since its now no longer talking to the real filesystem. * Scripts/webkitpy/common/host.py: * Scripts/webkitpy/layout_tests/port/factory.py: * Scripts/webkitpy/tool/commands/queries.py: * Scripts/webkitpy/tool/mocktool.py: 2011-09-07 Eric Seidel Add a new PortFactory class for creating new-style ports and deprecate the old PortFactory https://bugs.webkit.org/show_bug.cgi?id=67734 Reviewed by Adam Barth. Right now every place that creates a Port object has to be very careful to pass in a MockSCM/MockUser/MockFilesystem when appropriate. The correct fix is for port/factory.py (which creates most port objects) to know how to pass these in automatically. However, currently all port/factory.py functions are free functions instead of being on an object. This patch adds a new PortFactory object and deprecates the PortFactory for the old-style Port objects. In a follow-up patch I will add PortFactory to the Host object and teach PortFactory how to pass along the proper scm/user/filesystem to the Port object. * Scripts/webkitpy/common/host.py: * Scripts/webkitpy/layout_tests/port/factory.py: * Scripts/webkitpy/tool/commands/queries.py: * Scripts/webkitpy/tool/mocktool.py: 2011-09-07 Sheriff Bot Unreviewed, rolling out r94441. http://trac.webkit.org/changeset/94441 https://bugs.webkit.org/show_bug.cgi?id=67731 "MacEWSes are backed up due to new machines not able to reach git.webkit.org. Disabling tests for now" (Requested by eseidel2 on #webkit). * Scripts/webkitpy/tool/commands/earlywarningsystem.py: 2011-09-07 Elliot Poger filter test_expectations properly for chromium-mac vs chromium-cg-mac https://bugs.webkit.org/show_bug.cgi?id=67324 Reviewed by Dirk Pranke. * Scripts/webkitpy/layout_tests/port/chromium_gpu.py: * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py: * Scripts/webkitpy/layout_tests/port/chromium_mac.py: * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py: * Scripts/webkitpy/layout_tests/port/factory_unittest.py: 2011-09-07 Eric Boren [Chromium] Modify WebTouchEvent structure to match WebCore::TouchEvent https://bugs.webkit.org/show_bug.cgi?id=66800 Reviewed by Darin Fisher. * DumpRenderTree/chromium/EventSender.cpp: (EventSender::sendCurrentTouchEvent): 2011-09-07 Jarred Nicholls [Qt][DRT] Normalize file:///tmp/LayoutTests in LayoutTestController::pathToLocalResource() https://bugs.webkit.org/show_bug.cgi?id=67254 Reviewed by Csaba Osztrogonác. Translate file:///tmp/LayoutTests/* urls into the repository LayoutTests directory, which is derived from the running location of DumpRenderTree binary. * DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::pathToLocalResource): (LayoutTestController::setUserStyleSheetLocation): pass the url through pathToLocalResource 2011-09-07 Csaba Osztrogonác REGRESSION(r94288) build-webkit doesn't work on Symbian https://bugs.webkit.org/show_bug.cgi?id=67706 * Scripts/webkitdirs.pm: Disable automatic clean build on Symbian, because it doesn't support shadow build. (buildQMakeProject): 2011-09-07 Csaba Osztrogonác REGRESSION(r94288) build-webkit doesn't work on Windows https://bugs.webkit.org/show_bug.cgi?id=67646 Roll back r94655 with a small fix. * Scripts/webkitdirs.pm: (buildQMakeProject): 2011-09-07 Sheriff Bot Unreviewed, rolling out r94635. http://trac.webkit.org/changeset/94635 https://bugs.webkit.org/show_bug.cgi?id=67704 Linux Touch build is failing to compile this change. (Requested by loislo on #webkit). * DumpRenderTree/chromium/EventSender.cpp: (EventSender::sendCurrentTouchEvent): 2011-09-07 Sheriff Bot Unreviewed, rolling out r94655. http://trac.webkit.org/changeset/94655 https://bugs.webkit.org/show_bug.cgi?id=67703 It broke Leopard debug bot (Requested by ossy on #webkit). * Scripts/webkitdirs.pm: 2011-09-07 Kristóf Kosztyó REGRESSION(r94288) build-webkit doesn't work on Windows https://bugs.webkit.org/show_bug.cgi?id=67646 Reviewed by Csaba Osztrogonác. * Scripts/webkitdirs.pm: 2011-09-06 Eric Boren [Chromium] Modify WebTouchEvent structure to match WebCore::TouchEvent https://bugs.webkit.org/show_bug.cgi?id=66800 Reviewed by Darin Fisher. * DumpRenderTree/chromium/EventSender.cpp: (EventSender::sendCurrentTouchEvent): 2011-09-06 Adam Barth garden-o-matic details view should having working rebaseline and next/previous buttons https://bugs.webkit.org/show_bug.cgi?id=67659 Reviewed by Dimitri Glazkov. This patch wires up basic back/forward buttons that let you traverse through the results we're examining in the details view. This ended up being more code than I expected, but I wanted to keep all the state information in the DOM itself. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css: 2011-09-01 Dirk Pranke fix MockFileSystem.glob(), refactor filesystem tests https://bugs.webkit.org/show_bug.cgi?id=67462 Reviewed by Eric Seidel. In debugging why my proposed change for MockFileSystem.glob() in bug 66228 didn't work, I stumbled across two real bugs that cancelled each other out for the unit tests. I fixed those (glob shouldn't recurse into subdirs, and we weren't populating self.dirs from self.files properly). I have also created a "mixin" class for tests that can be shared between the real filesystem and the mock filesystem - MockFileSystem is being used enough it needs its own tests. * Scripts/webkitpy/common/system/filesystem_mock.py: * Scripts/webkitpy/common/system/filesystem_mock_unittest.py: Added. * Scripts/webkitpy/common/system/filesystem_unittest.py: 2011-09-06 Anders Carlsson Move NPAPI headers in bridge to plugins https://bugs.webkit.org/show_bug.cgi?id=67661 Update forwarding headers. * DumpRenderTree/unix/TestNetscapePlugin/ForwardingHeaders/WebKit/npapi.h: * DumpRenderTree/unix/TestNetscapePlugin/ForwardingHeaders/WebKit/npruntime.h: 2011-09-06 Tony Chang [chromium] Check in 'build' for compiled binaries first https://bugs.webkit.org/show_bug.cgi?id=67549 Reviewed by Dirk Pranke. I'm working on moving the windows build output to a common location rather than relative to the sln file. Because of that, we need to check those build locations first for DumpRenderTree.exe. * Scripts/webkitpy/layout_tests/port/chromium_win.py: 2011-09-06 Adam Barth Use a hierarchial structure for the garden-o-matic results view https://bugs.webkit.org/show_bug.cgi?id=67620 Reviewed by Dimitri Glazkov. This patch re-organizes the results view to use a hierachial structure for selection. At the top-level are sections for each test. Expanding the test section shows a tab selector for builders. This structure prepares us for wiring up various actions to this view. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: (.): * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js: (): * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: (.): * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css: (.results-view>.toolbar): (.results-view>.toolbar ul.actions): (.results-view>.toolbar ul.actions li): (.results-grid table): (.results-grid table td, .results-view table th): (.results-grid table th): (.results-grid .text-result): (.results-grid .image-result): 2011-09-06 Adam Barth garden-o-matic's base.probe generates spam in the JavaScript console https://bugs.webkit.org/show_bug.cgi?id=67622 Reviewed by Dimitri Glazkov. Now that we have CORS set up on the buildbot, we can use XMLHttpRequest instead of HTMLScriptElement to test whether results exist. This reduces the console.log spam and saves bandwidth because we can use a HEAD request instead of a GET request. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/net.js: 2011-09-05 Adam Barth Add a tab to garden-o-matic for viewing expected failures https://bugs.webkit.org/show_bug.cgi?id=67619 Reviewed by Dimitri Glazkov. This patch adds a basic tab to garden-o-matic that shows all the tests failing on the bots, including the expected failures. This patch mostly involves refactoring existing classes to allow new subclasses. This view is useful for when gardeners mark tests as expected to fail and then rebaseline them later. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary-mock.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: 2011-09-05 Adam Barth Fix global variable leak in garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=67617 Reviewed by Dimitri Glazkov. Caught by running the unit tests with noglobals turned on. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: 2011-09-05 Balazs Kelemen [Qt][WK2] Add pixel test support https://bugs.webkit.org/show_bug.cgi?id=66283 Reviewed by Andreas Kling. Implemented pixel result generation by reusing the logic that we have in DumpRenderTree. * WebKitTestRunner/qt/TestInvocationQt.cpp: (WTR::dumpImage): (WTR::TestInvocation::dumpPixelsAndCompareWithExpected): 2011-09-04 Robin Dunn [wx] Enable wxWebKit to run using the wxGC Cairo backend on platforms other than GTK. https://bugs.webkit.org/show_bug.cgi?id=67577 Reviewed by Kevin Ollivier. * waf/build/settings.py: 2011-09-03 Kevin Ollivier [wx] Unreviewed build fix. Add wtf/dtoa directory to build. * waf/build/settings.py: 2011-09-03 Dan Bernstein 32-bit build fix. * TestWebKitAPI/mac/PlatformUtilitiesMac.mm: (TestWebKitAPI::Util::toSTD): 2011-09-03 Dan Bernstein Move testStringByEvaluatingJavaScriptFromString() from DumpRenderTree to TestWebKitAPI https://bugs.webkit.org/show_bug.cgi?id=67559 Reviewed by Darin Adler. * DumpRenderTree/mac/DumpRenderTree.mm: (dumpRenderTree): Moved testStringByEvaluatingJavaScriptFromString() from here and removed the call to it. * TestWebKitAPI/PlatformUtilities.h: Declared Util::toSTD(NSString *) * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added StringByEvaluatingJavaScriptFromString.mm. * TestWebKitAPI/Tests/mac/StringByEvaluatingJavaScriptFromString.mm: Added. (TestWebKitAPI::TEST): Added. Moved testStringByEvaluatingJavaScriptFromString() to here. * TestWebKitAPI/mac/PlatformUtilitiesMac.mm: (TestWebKitAPI::Util::toSTD): Added. 2011-09-03 Adam Barth garden-o-matic results view should have a better results selector https://bugs.webkit.org/show_bug.cgi?id=67542 Reviewed by Dimitri Glazkov. This patch adds a basic grid-based results selector to the results view. Currently, this grid just shows how the tests fail on the various bots. A future patch will let you select which failure to view and to mark the failures for rebaselining. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: - Add some more basic functions for manipulating dictionaries. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js: 2011-09-03 Adam Barth garden-o-matic results view should show Actual and Expected results for text failures https://bugs.webkit.org/show_bug.cgi?id=67536 Reviewed by Eric Seidel. In the past, we didn't show actual and expected text results (we just showed the diff) because we didn't have the awesome six-by-two results grid. Now that we do, we can move show these results. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: - Fixed some typos that were preventing these unit tests from running. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js: (.): * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js: - Report -actual.txt and -expected.txt as possible results URLs. 2011-09-03 Adam Barth garden-o-matic shows "ghost" failures that are already fixed https://bugs.webkit.org/show_bug.cgi?id=67535 Reviewed by Eric Seidel. About 10% of writes fail to stick on test-results.appspot.com. This patch causes us to fetch the full_results.json blob directly from build.chromium.org rather than relying upon AppEngine. Unforuntately, we still need to get historical information from AppEngine. That will take several patches to fix becuase I need to change the bots to write the full_results.json file to the build-specific directory in addition to the zip file. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js: 2011-09-03 Sheriff Bot Unreviewed, rolling out r94458. http://trac.webkit.org/changeset/94458 https://bugs.webkit.org/show_bug.cgi?id=67558 Caused appcache test to fail on various bots (Requested by rniwa on #webkit). * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: 2011-09-03 Sheriff Bot Unreviewed, rolling out r94453. http://trac.webkit.org/changeset/94453 https://bugs.webkit.org/show_bug.cgi?id=67557 Caused appcache test to fail on various bots (Requested by rniwa on #webkit). * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-09-02 Michael Saboff Add JSC:RegExp functional tests https://bugs.webkit.org/show_bug.cgi?id=67339 New perl script to build JavaScriptCore and run the RegExp tests. Reviewed by Gavin Barraclough. * Scripts/run-regexp-tests: Added. 2011-09-02 Michael Saboff Add JSC:RegExp functional tests https://bugs.webkit.org/show_bug.cgi?id=67339 New perl script to build JavaScriptCore and run the RegExp tests. Reviewed by Gavin Barraclough. * Scripts/run-regexp-tests: Added. 2011-09-02 Eric Seidel Reshuffle some code in WebKitDriver._read_block in preparation for reading stderr/stdout separately https://bugs.webkit.org/show_bug.cgi?id=67530 Unreviewed. Fixing typo from previous commit. Turns out there was *no* unittesting of WebKitDriver. Added a basic test of _read_block which exercises the code I previously made a typo in. * Scripts/webkitpy/layout_tests/port/webkit.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: 2011-09-02 Eric Seidel Reshuffle some code in WebKitDriver._read_block in preparation for reading stderr/stdout separately https://bugs.webkit.org/show_bug.cgi?id=67530 Reviewed by Adam Barth. No functional change, just reshuffling code. * Scripts/webkitpy/layout_tests/port/webkit.py: 2011-09-02 Darin Adler Added Subversion ignore for .pyc generated files. * Scripts/webkitpy/layout_tests/models: Added property svn:ignore. * Scripts/webkitpy/layout_tests/servers: Added property svn:ignore. * Scripts/webkitpy/tool/servers: Added property svn:ignore. 2011-09-02 Adam Barth Enable testing on mac-ews https://bugs.webkit.org/show_bug.cgi?id=67505 Reviewed by Eric Seidel. This patch enables testing on the Mac EWS, which should help contributors not turn the Mac bots red. * Scripts/webkitpy/tool/commands/earlywarningsystem.py: 2011-09-01 Beth Dakin https://bugs.webkit.org/show_bug.cgi?id=67451 Add a test for going back to a cached page after changing the device scale factor Reviewed by Adam Roben. New test DeviceScaleFactorOnBack.mm inherits from WebKitAgnosticTest. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/mac/DeviceScaleFactorOnBack.mm: Added. (TestWebKitAPI::DeviceScaleFactorOnBack::url): (TestWebKitAPI::DeviceScaleFactorOnBack::didLoadURL): (TestWebKitAPI::DeviceScaleFactorOnBack::createWindow): (TestWebKitAPI::DeviceScaleFactorOnBack::initializeView): (TestWebKitAPI::DeviceScaleFactorOnBack::runTest): (TestWebKitAPI::TEST_F): Added new function to WebKitAgnosticTest to goBack. Separated waitForLoadToFinish() into its own function (instead of being a part of loadAndWaitUntilFinished()) so that it can be called from goBack as well. Also added initializeView() so we can set the cache model. * TestWebKitAPI/mac/WebKitAgnosticTest.h: (TestWebKitAPI::WebKitAgnosticTest::initializeView): * TestWebKitAPI/mac/WebKitAgnosticTest.mm: (TestWebKitAPI::WebKitAgnosticTest::WebKitAgnosticTest): (TestWebKitAPI::WebKitAgnosticTest::runWebKit1Test): (TestWebKitAPI::WebKitAgnosticTest::runWebKit2Test): (TestWebKitAPI::WebKitAgnosticTest::loadURL): (TestWebKitAPI::WebKitAgnosticTest::goBack): (TestWebKitAPI::WebKitAgnosticTest::waitForLoadToFinish): 2011-09-02 Anders Carlsson Build fix. * DumpRenderTree/TestNetscapePlugIn/Tests/PrivateBrowsing.cpp: (PrivateBrowsing::ScriptableObject::pluginTest): 2011-09-02 Anders Carlsson Move private browsing test to a PluginTest subclass https://bugs.webkit.org/show_bug.cgi?id=67498 Reviewed by Darin Adler. * DumpRenderTree/DumpRenderTree.gypi: * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: * DumpRenderTree/TestNetscapePlugIn/PluginObject.cpp: (pluginGetProperty): * DumpRenderTree/TestNetscapePlugIn/PluginObject.h: * DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp: (PluginTest::NPP_SetValue): * DumpRenderTree/TestNetscapePlugIn/PluginTest.h: * DumpRenderTree/TestNetscapePlugIn/Tests/PrivateBrowsing.cpp: Added. (PrivateBrowsing::PrivateBrowsing): (PrivateBrowsing::privateBrowsingEnabled): (PrivateBrowsing::cachedPrivateBrowsingEnabled): (PrivateBrowsing::ScriptableObject::hasProperty): (PrivateBrowsing::ScriptableObject::getProperty): (PrivateBrowsing::ScriptableObject::pluginTest): (PrivateBrowsing::NPP_New): (PrivateBrowsing::NPP_GetValue): (PrivateBrowsing::NPP_SetValue): * DumpRenderTree/TestNetscapePlugIn/main.cpp: (NPP_New): (NPP_SetValue): * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj: * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro: * GNUmakefile.am: 2011-09-02 Csaba Osztrogonác [Qt] Build dependency problems https://bugs.webkit.org/show_bug.cgi?id=38054 Trivial fix after r94288. Rubber-stamped by Andras Becsi. * Scripts/webkitdirs.pm: We should read the whole defaults.txt, not only the first two lines. (buildQMakeProject): 2011-09-02 Andras Becsi Remove my old e-mail address and add my new one. Rubber-stamped by Tor Arne Vestbø. * Scripts/webkitpy/common/config/committers.py: 2011-09-02 Philippe Normand [GTK] run-gtk-tests leaves a zombie Xvfb process https://bugs.webkit.org/show_bug.cgi?id=67389 Reviewed by Martin Robinson. * Scripts/run-gtk-tests: Properly close STDIN and STDERR on the child process instead of doing shell redirection. 2011-09-01 Mark Rowe Fix the build by adding some more missing return types. * DumpRenderTree/mac/ResourceLoadDelegate.mm: (-[ResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]): 2011-09-01 Eric Seidel REGRESSION (NRWT): Leaks Viewer can't load leaks from test runs that used NRWT https://bugs.webkit.org/show_bug.cgi?id=66228 Reviewed by Dirk Pranke. ORWT used $(PROCESS)$(NUMBER)-leaks.txt for leaks filenames. The LeaksViewer tool globs for all *-leaks.txt files. This patch makes NRWT output $(PROCESS)-$(PID)-leaks.txt which should match. In order to test this I had to fix a limitation in our MockFileSystem.glob method. However, doing so uncovered a typo and bug in the integration tests. I've disabled the offending integration tests. * Scripts/webkitpy/common/system/filesystem_mock.py: * Scripts/webkitpy/layout_tests/port/leakdetector.py: * Scripts/webkitpy/layout_tests/port/leakdetector_unittest.py: 2011-09-01 Dominic Mazzoni Adds a LayoutTestController method to make it possible to test starting speech input via an API call. https://bugs.webkit.org/show_bug.cgi?id=60170 Reviewed by Dimitri Glazkov. * DumpRenderTree/LayoutTestController.h: * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (LayoutTestController::startSpeechInput): * DumpRenderTree/chromium/LayoutTestController.h: * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: (LayoutTestController::startSpeechInput): * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::startSpeechInput): * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::startSpeechInput): * DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::startSpeechInput): * DumpRenderTree/qt/LayoutTestControllerQt.h: * DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::startSpeechInput): * DumpRenderTree/wx/LayoutTestControllerWx.cpp: (LayoutTestController::startSpeechInput): 2011-09-01 Tim Horton Fix TestWebKitAPI build on clang-2.0 * TestWebKitAPI/Tests/WTF/CheckedArithmeticOperations.cpp: 2011-09-01 Adam Barth Remove unused "extension" parameter from rebaseline gardeningserver handler https://bugs.webkit.org/show_bug.cgi?id=67384 Reviewed by James Robinson. This parameter hasn't done anything in a while. * Scripts/webkitpy/tool/commands/rebaseline.py: * Scripts/webkitpy/tool/servers/gardeningserver.py: * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: 2011-09-01 Eric Seidel Unreviewed. Fix a typo in unittests. Unfortunately fixing it caused several assertions. Just disabling the test instead, since it was providing no value before due to the typo. * Scripts/webkitpy/layout_tests/port/test.py: 2011-09-01 Adam Barth Add a "rebaseline" button to the garden-o-matic summary page https://bugs.webkit.org/show_bug.cgi?id=67382 Reviewed by Dimitri Glazkov. As much as we might wish folks to not rebaseline entire test groups at once, the truth is that doing so is a common operation. Adding a button for this operation is likely to make folks happy. This patch also cleans up a bunch of plumbing surrounding failureInfo objects. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: 2011-09-01 Adam Barth garden-o-matic should let the gardener assign blame for a group of test failures https://bugs.webkit.org/show_bug.cgi?id=67379 Reviewed by Dimitri Glazkov. If a group of test failures has a large revision range, the gardener might wish to blame a specific revision for the failure. Currently, this only effect of assigning blame in this way is to hide the other suspect commits. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: 2011-09-01 Adam Barth Hide garden-o-matic failure table until hover https://bugs.webkit.org/show_bug.cgi?id=67376 Reviewed by Dimitri Glazkov. This patch removes the failure table for build failures (because it does not make sense) and hides the failure table for test groups until the user hovers over the test failure block, reducing visual noise. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: 2011-09-01 Adam Barth Switch garden-o-matic from commitData.title to commitData.summary https://bugs.webkit.org/show_bug.cgi?id=67375 Reviewed by Dimitri Glazkov. commitData.title is a string provided by Trac, which isn't great in all cases. commitData.summary is my attempt to find the most important one-line summary of the commit. This patch switches garden-o-matic over to summary. Also, this patch tweaks a bunch of CSS to tighten up the summary page and reduce visual noise. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: 2011-09-01 Adam Barth Group failing tests in garden-o-matic if there are more than three in a folder https://bugs.webkit.org/show_bug.cgi?id=67373 Reviewed by Dimitri Glazkov. Sometimes a failure will cause many tests in a given directory to fail. When that happens, it's better to list the directory as a single item (with a count) rather than spamming all the test names. (This patch also updates a bunch of unit test expected results after my last commit.) * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: 2011-09-01 Adam Barth Lower the visual noise from the failing builders display https://bugs.webkit.org/show_bug.cgi?id=67369 Reviewed by Dimitri Glazkov. This patch ensures that the "Building" row in the failing builders display is always last and dims it by 50% to reduce visual noise. This lets the gardener focus on the actual failures. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/failures.css: Added. 2011-09-01 Kristóf Kosztyó [Qt] Build dependency problems https://bugs.webkit.org/show_bug.cgi?id=38054 Make clean build after the defines are changed. Reviewed by Csaba Osztrogonác. * Scripts/webkitdirs.pm: (buildQMakeProject): 2011-08-31 Adam Barth Move DumpRenderTree to WebKitPlatformSupport https://bugs.webkit.org/show_bug.cgi?id=67331 Reviewed by Darin Fisher. This patch update DumpRenderTree to account for the WebKitClient => WebKitPlatformSupport rename. * DumpRenderTree/chromium/NotificationPresenter.cpp: (NotificationPresenter::show): * DumpRenderTree/chromium/Task.cpp: (postTask): 2011-08-31 Eric Seidel Add queues.webkit.org/active-bots page to make it easy to see which bots are down https://bugs.webkit.org/show_bug.cgi?id=67314 Reviewed by Adam Barth. Very simple page. Just lists all the bots, what queue they belong to and what we last heard from them. If we haven't heard from a bot in the last 500 messages (last day or so), then it won't show up here. This makes it very easy for me to see which of my cr-jail bots might be down at any one time. I also moved a bunch of code into dashboard.css which belonged there. * QueueStatusServer/handlers/activebots.py: Added. * QueueStatusServer/main.py: * QueueStatusServer/stylesheets/dashboard.css: (.queue_bubble): (.queue_name): (.last_heard_from): (.status_text): (.alive): (.behind): (.dead): * QueueStatusServer/templates/activebots.html: Added. * QueueStatusServer/templates/dashboard.html: * QueueStatusServer/templates/recentstatus.html: 2011-08-31 Oliver Hunt Move CheckedArithmeticOperations.cpp to the intended location * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/WTF/CheckedArithmeticOperations.cpp: Renamed from Tools/TestWebKitAPI/Tests/CheckedArithmeticOperations.cpp. 2011-08-31 Elliot Poger add unittest to validate the complete set of Chromium configurations https://bugs.webkit.org/show_bug.cgi?id=67211 Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: 2011-08-31 Nico Weber Make ScrollAnimator(Chromium)?Mac check scrollAnimatorEnabled() https://bugs.webkit.org/show_bug.cgi?id=67297 Reviewed by Tony Chang. Explicitly disable smooth scrolling in chromium's DRT. * DumpRenderTree/chromium/WebPreferences.cpp: (WebPreferences::applyTo): 2011-08-31 Oliver Hunt Add Checked<> tests to windows build * TestWebKitAPI/win/TestWebKitAPI.vcproj: 2011-08-30 Oliver Hunt Add support for checked arithmetic https://bugs.webkit.org/show_bug.cgi?id=67095 Reviewed by Sam Weinig. Add test cases for Checked<> * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/CheckedArithmeticOperations.cpp: Added. 2011-08-31 Adam Barth Move summary.html to garden-o-matic.html. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: Copied from Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js: Copied from Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary.js. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary.js: Removed. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: Removed. 2011-08-31 Noel Gordon [chromium] DRT linux should pass nativeKeyCode to plugins https://bugs.webkit.org/show_bug.cgi?id=66679 Reviewed by Tony Chang. * DumpRenderTree/chromium/EventSender.cpp: (EventSender::keyDown): Call a Chromium Linux GTK helper routine that returns the (simulated) nativeKeyCode for the given input keyboard code. 2011-08-31 Adam Barth Tighen up UI for garden-o-matic tab strip https://bugs.webkit.org/show_bug.cgi?id=67266 Reviewed by Dimitri Glazkov. This patch add CSS to move the tab strip to the edges of the page and to remove some gratuitous border-radius. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: 2011-08-31 Adam Barth garden-o-matic results view shouldn't be horrifically ugly https://bugs.webkit.org/show_bug.cgi?id=67264 Reviewed by Dimitri Glazkov. Minor tweak to the results view CSS to make it fit in better with the new tabbed UI. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css: 2011-08-31 Adam Barth garden-o-matic needs a tabstrip https://bugs.webkit.org/show_bug.cgi?id=67263 Reviewed by Dimitri Glazkov. This patch adds a basic tabstrip to garden-o-matic. The results details view now opens in the "Results" tab, and the user can flip back and forth between the Summary and the Results tab quickly. There's also a stub for a "Commits" tab, which is disabled. This patch isn't the most beautiful in the world, but it gets the ball rolling on a multi-modal UI. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: 2011-08-31 Ai Makabi Extract reference links from reftest test file. https://bugs.webkit.org/show_bug.cgi?id=66838 Reviewed by Shinichiro Hamaji. * Scripts/webkitpy/layout_tests/reftests/__init__.py: Added. * Scripts/webkitpy/layout_tests/reftests/extract_reference_link.py: Added. * Scripts/webkitpy/layout_tests/reftests/extract_reference_link_unittest.py: Added. 2011-08-03 Philippe Normand [webkitpy] missing log import in common/net/resultsjsonparser.py https://bugs.webkit.org/show_bug.cgi?id=65606 Reviewed by Martin Robinson. * Scripts/webkitpy/common/net/resultsjsonparser.py: Import the logging module. 2011-08-30 Philippe Normand [stylechecker] complains about NULL usage in some gst_structure_ calls https://bugs.webkit.org/show_bug.cgi?id=67194 Reviewed by Martin Robinson. Don't warn about NULL in some gst_structure_ calls. 0 can't be used for these, a proper sentinel is required. * Scripts/webkitpy/style/checkers/cpp.py: * Scripts/webkitpy/style/checkers/cpp_unittest.py: 2011-08-30 Adam Barth Remove old-style UI from garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=67257 Reviewed by Dimitri Glazkov. This code is nowhere near as awesome as the new hotness. Let's remove it so it doesn't weigh us down. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: Removed. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/main.js: Removed. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/unexpected-passes.js: Removed. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/unexpected-passes.html: Removed. 2011-08-30 Takashi Toyoshima [WebSocket] update pywebsocket to 0.6b5. https://bugs.webkit.org/show_bug.cgi?id=66924 Reviewed by Kent Tamura. pywebsocket 0.6b2 supported minimum features of Hybi-08 specification. Now, pywebsocket 0.6b5 supports full features of Hybi-10 specification. 0.6b3 or later supports a closing handshake handler and it is useful to verify client's closing handshake implementation. 0.6b4 supports AbortedByUserException. It is useful to eliminate unnecessary server warning messages on intended abnormal closures. 0.6b5 supports Python 2.5.1 or Mac 10.5. * Scripts/webkitpy/thirdparty/__init__.py: Update download URL and directory path. 2011-08-30 Dan Bernstein Test for Add text search API for getting the DOM range of a text match https://bugs.webkit.org/show_bug.cgi?id=67230 Reviewed by Darin Adler. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/mac/DOMRangeOfString.html: Added. * TestWebKitAPI/Tests/mac/DOMRangeOfString.mm: Added. (-[DOMRangeOfStringFrameLoadDelegate webView:didFinishLoadForFrame:]): (TestWebKitAPI::TEST): 2011-08-30 Aaron Colwell Add MediaSource API to HTMLMediaElement https://bugs.webkit.org/show_bug.cgi?id=64731 Reviewed by Eric Carlson. * Scripts/build-webkit: 2011-08-30 Caio Marcelo de Oliveira Filho Emit last progress notification before calling dispatchDidFinishLoad https://bugs.webkit.org/show_bug.cgi?id=28851 Reviewed by Adam Barth. Implement dumpProgressFinishedCallback() in the layoutTestController for Qt, Chromium and GTK ports. The first two were simple changes. Some refactoring was needed in GTK port to assure that the dumps are emitted correctly and the processing after the page load finished is in the right place. To detect the load finished, we use the callback for the "load-status" notifier signal of the top loading frame. And we use the deprecated "load-finished" signal for dumping (if enabled) the progress finished expected output. * DumpRenderTree/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (dumpProgressFinishedCallbackCallback): (LayoutTestController::staticFunctions): * DumpRenderTree/LayoutTestController.h: (LayoutTestController::dumpProgressFinishedCallback): (LayoutTestController::setDumpProgressFinishedCallback): * DumpRenderTree/chromium/LayoutTestController.cpp: (LayoutTestController::LayoutTestController): (LayoutTestController::dumpProgressFinishedCallback): (LayoutTestController::reset): * DumpRenderTree/chromium/LayoutTestController.h: (LayoutTestController::shouldDumpProgressFinishedCallback): (LayoutTestController::setShouldDumpProgressFinishedCallback): * DumpRenderTree/chromium/WebViewHost.cpp: (WebViewHost::didStopLoading): * DumpRenderTree/gtk/DumpRenderTree.cpp: (webViewLoadFinished): (topLoadingFrameLoadFinished): (webFrameLoadStatusNotified): * DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::reset): (LayoutTestController::dumpProgressFinishedCallback): * DumpRenderTree/qt/LayoutTestControllerQt.h: 2011-08-30 Filip Pizlo JavaScriptCore does not have a good mechanism for performance regression testing https://bugs.webkit.org/show_bug.cgi?id=67177 Reviewed by Darin Adler. Added a new script, which allows for testing one, two, or more VMs at once on SunSpider, V8, and Kraken benchmarks. Benchmark/VM runs are interleaved at random to minimize systematics. The results that bencher prints includes all possible relevant information about how bencher was invoked. Bencher can be easily used to measure performance changes down to fractions of a percent. Bencher also includes experimental support for running benchmarks in a DumpRenderTree release build. This works in SunSpider and V8, but not yet in Kraken. Running benchmarks in DumpRenderTree more closely mimics the performance when running in browser. * Scripts/bencher: Added. 2011-08-29 Alexey Proskuryakov DumpRenderTree should begin each test with an empty cookie store https://bugs.webkit.org/show_bug.cgi?id=63545 Reviewed by Darin Adler. * DumpRenderTree/mac/DumpRenderTree.mm: (resetDefaultsToConsistentValues): (setDefaultsToConsistentValuesForTesting): * DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::setAlwaysAcceptCookies): * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): Use new SPIs to manipulate sessions. 2011-08-30 Sheriff Bot Unreviewed, rolling out r94055 and r94057. http://trac.webkit.org/changeset/94055 http://trac.webkit.org/changeset/94057 https://bugs.webkit.org/show_bug.cgi?id=67191 Many failures on Mac 10.5 (Requested by toyoshim on #webkit). * Scripts/webkitpy/thirdparty/__init__.py: 2011-08-30 Adam Barth Chromium Android should not be a core builder https://bugs.webkit.org/show_bug.cgi?id=67184 Reviewed by Ryosuke Niwa. This bot hasn't been consistently green yet because it has never actually run! * Scripts/webkitpy/common/net/buildbot/buildbot.py: * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py: 2011-08-30 Takashi Toyoshima [WebSocket] update pywebsocket to 0.6b4. https://bugs.webkit.org/show_bug.cgi?id=66924 Reviewed by Kent Tamura. pywebsocket 0.6b2 supported minimum features of Hybi-08 specification. Now, pywebsocket 0.6b4 supports full features of Hybi-10 specification. 0.6b3 or later supports a closing handshake handler and it is useful to verify client's closing handshake implementation. 0.6b4 supports AbortedByUserException. It is useful to eliminate unnecessary server warning messages on intended abnormal closures. * Scripts/webkitpy/thirdparty/__init__.py: Update download URL and directory path. 2011-08-29 Tom Zakrajsek sheriffbot "usage" help statements are wrong https://bugs.webkit.org/show_bug.cgi?id=67170 Reviewed by Adam Barth. * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py: 2011-08-29 Adam Barth Add support for chomium-mac (non-CoreGraphics) baselines https://bugs.webkit.org/show_bug.cgi?id=67163 Reviewed by Eric Seidel. This patch teaches the non-CG chromium-mac bots to look at non-CG baselines. However, the patch still leaves CG as the default set of baselines. * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py: * Scripts/webkitpy/common/config/build.py: * Scripts/webkitpy/common/config/build_unittest.py: * Scripts/webkitpy/layout_tests/port/builders.py: * Scripts/webkitpy/layout_tests/port/chromium.py: * Scripts/webkitpy/layout_tests/port/chromium_mac.py: * Scripts/webkitpy/layout_tests/port/factory.py: * Scripts/webkitpy/layout_tests/port/factory_unittest.py: * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py: * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py: * Scripts/webkitpy/tool/commands/rebaseline_unittest.py: 2011-08-29 Shawn Singh Small fix in chromium port of new-run-webkit-tests. https://bugs.webkit.org/show_bug.cgi?id=67078 Reviewed by Dirk Pranke. * Scripts/webkitpy/layout_tests/port/chromium.py: 2011-08-29 Sheriff Bot Unreviewed, rolling out r93987, r93992, r93995, r93998, and r93999. http://trac.webkit.org/changeset/93987 http://trac.webkit.org/changeset/93992 http://trac.webkit.org/changeset/93995 http://trac.webkit.org/changeset/93998 http://trac.webkit.org/changeset/93999 https://bugs.webkit.org/show_bug.cgi?id=67147 Many failing tests (Requested by ap on #webkit). * DumpRenderTree/mac/DumpRenderTree.mm: (setDefaultsToConsistentValuesForTesting): * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): 2011-08-26 Alexey Proskuryakov DumpRenderTree should begin each test with an empty cookie store https://bugs.webkit.org/show_bug.cgi?id=63545 Reviewed by Darin Adler. * DumpRenderTree/mac/DumpRenderTree.mm: (setDefaultsToConsistentValuesForTesting): * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): Tell WebCore to use a temporary session, starting with a clean cookie store. 2011-08-23 Chris Marrin [mac] requestAnimationFrame support for mac port https://bugs.webkit.org/show_bug.cgi?id=59146 Reviewed by Simon Fraser. Add support for request-animation-callback flag * Scripts/build-webkit: 2011-08-27 Peter Beverloo Add a build bot for Chromium with OS(ANDROID) https://bugs.webkit.org/show_bug.cgi?id=66687 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/config.json: 2011-08-27 Peter Beverloo Update the master build bot configuration script to properly deal with the --chromium-android switch. https://bugs.webkit.org/show_bug.cgi?id=66687 The --chromium-android switch is needed when running update-webkit-chromium and build-webkit, as it'll tell the scripts to pull in the right dependencies and cross-compile WebKit using the Android NDK. Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/master.cfg: 2011-08-27 Peter Beverloo Teach the update and build scripts how to handle dependencies and cross-compiling rules for the Chromium port on OS(ANDROID). https://bugs.webkit.org/show_bug.cgi?id=66687 Reviewed by Adam Barth. * Scripts/build-webkit: * Scripts/update-webkit: * Scripts/update-webkit-chromium: * Scripts/webkitdirs.pm: (argumentsForConfiguration): (isChromium): (isChromiumAndroid): (determineIsChromiumAndroid): (buildChromium): 2011-08-26 Adam Barth garden-o-matic should let me know that I need to run the local server https://bugs.webkit.org/show_bug.cgi?id=66985 Reviewed by Dimitri Glazkov. The patch triggers an alert dialog whenever the user tries to use a feature that requires the local server. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout_unittests.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary.js: 2011-08-26 Dimitri Glazkov garden-o-matic summary view should have pretty commit data information. https://bugs.webkit.org/show_bug.cgi?id=67069 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: Added support for styling commit data. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Changed tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: Defined global anchor color. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: Styled commit data. 2011-08-26 Dimitri Glazkov Tweak the use of Open Sans in garden-o-matic summary view to make it look decent on Windows and Linux. https://bugs.webkit.org/show_bug.cgi?id=67066 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary-mock.html: Tweaked. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: Ditto. 2011-08-26 Raphael Kubo da Costa [DRT] Include cairo.h in PixelDumpSupportCairo.h for EFL too. https://bugs.webkit.org/show_bug.cgi?id=67041 Reviewed by Martin Robinson. This is a follow-up to r89053: EFL's PixelDumpSupportEfl.cpp includes PixelDumpSupportCairo.h, but the latter was #including cairo.h only when the GTK+ port was being built. * DumpRenderTree/cairo/PixelDumpSupportCairo.h: 2011-08-26 Dimitri Glazkov Clean up and line up layout in garden-o-matic summary view. https://bugs.webkit.org/show_bug.cgi?id=67057 1) Renamed ui.notifications.TestFailures to ui.notifications.TestsFailing to match BuildersFailing; 2) Rearranged DOM structure a bit to accommodate recent changes; 3) Added minimal CSS to line things up. Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: Renamed TestFailures to TestsFailing. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary-mock.js: Added mocks for BuildersFailing. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: Renamed TestFailures to TestsFailing. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: Updated tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: Changed DOM structure. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Updated tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: Tweaked common appearance. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: Added minimal layout/style. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary-mock.html: Added Google Web Fonts hook. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: Ditto. 2011-08-26 Alice Liu https://bugs.webkit.org/show_bug.cgi?id=66823 Add the request info to the WebPageProxy::CreateNewPage message Reviewed by Anders Carlsson. Update function declarations with additional parameter: * MiniBrowser/gtk/BrowserWindow.c: (createNewPage): * MiniBrowser/mac/BrowserWindowController.m: (createNewPage): * MiniBrowser/win/BrowserView.cpp: (createNewPage): * WebKitTestRunner/TestController.cpp: (WTR::TestController::createOtherPage): * WebKitTestRunner/TestController.h: 2011-08-26 Noel Gordon Make show-pretty-diff work on linux https://bugs.webkit.org/show_bug.cgi?id=67017 Reviewed by Adam Barth. * Scripts/show-pretty-diff: 2011-08-25 Dimitri Glazkov Reorganize garden-o-matic CSS files to match JS-built views. https://bugs.webkit.org/show_bug.cgi?id=66998 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/common.css: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/notifications.css: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/summary.css: Removed. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary-mock.html: Changed to use new CSS files. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: Ditto. 2011-08-25 Adam Barth garden-o-matic should be able to roll out patches https://bugs.webkit.org/show_bug.cgi?id=66982 Reviewed by Dimitri Glazkov. This patch just wires the rollout event through the controller to the local server, which is already set up to perform rollouts. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/checkout.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js: 2011-08-25 Viatcheslav Ostapenko [Qt] [WK2] Assert in computeViewportAttributes when MiniBrowser started with -touch https://bugs.webkit.org/show_bug.cgi?id=66951 Reviewed by Andreas Kling. Function computeViewportAttributes() gets called when height is still 0 because MiniBrowser sets width and height one by one and setWidth cause geometryChange event in QTouchWebView. * MiniBrowser/qt/BrowserView.cpp: (BrowserView::resizeEvent): 2011-08-25 Dimitri Glazkov garden-o-matic summary view should provide information on builders still in flight. https://bugs.webkit.org/show_bug.cgi?id=66992 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js: Added grabbing the list of in-flight builders. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js: Added buildersInFlightForRevision. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js: Added tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: Split off updating builder results to own method. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Added tests. 2011-08-25 Adam Barth webkit-patch rebaseline-expectations don't work when run outside the root https://bugs.webkit.org/show_bug.cgi?id=66966 Reviewed by Eric Seidel. Turns out tool.path() used to be relative, which caused problems when running webkit-patch outside the root directory. This patch makes it absolute. I checked all callers, and they all appear to want this behavior. It was unclear to me how to test this change because we only really have a testing strategy for webkitpy, not for the main script. * Scripts/webkit-patch: 2011-08-25 Dimitri Glazkov garden-o-matic summary needs a mock view that could be used for working on styling https://bugs.webkit.org/show_bug.cgi?id=66971 Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/summary-mock.js: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary-mock.html: Added. 2011-08-25 Anders Carlsson Return the correct value for NPNVsupportsCarbonBool https://bugs.webkit.org/show_bug.cgi?id=66964 Add a test. Reviewed by Adam Roben. * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: * DumpRenderTree/TestNetscapePlugIn/Tests/mac/SupportsCarbonEventModel.cpp: Added. (SupportsCarbonEventModel::SupportsCarbonEventModel): (SupportsCarbonEventModel::testConvert): (SupportsCarbonEventModel::runTest): (SupportsCarbonEventModel::NPP_New): 2011-08-25 Dimitri Glazkov garden-o-matic summary view should not be broken. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/summary.css: (table.comparison): Added more specificity. 2011-08-25 Adam Barth This command has been broken for a while. This patch updates the callsite to the new API. * Scripts/webkitpy/tool/commands/rebaseline.py: 2011-08-25 Dimitri Glazkov garden-o-matic needs a way to report where and how tests are failing in the summary view. https://bugs.webkit.org/show_bug.cgi?id=66955 Introduce FailureGrid, your one-stop failure-indicating destination. Reviewed by Adam Barth. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html: Added new scripts. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js: Added. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: Updated to use FailureGrid. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Updated tests. * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html: Added new scripts. 2011-08-25 Igor Oliveira Add myself to the committers list https://bugs.webkit.org/show_bug.cgi?id=66941 Reviewed by Andreas Kling. * Scripts/webkitpy/common/config/committers.py: 2011-08-25 Sheriff Bot Unreviewed, rolling out r93767. http://trac.webkit.org/changeset/93767 https://bugs.webkit.org/show_bug.cgi?id=66931 Broke Qt bot. (Requested by yutak on #webkit). * Scripts/webkitpy/thirdparty/__init__.py: 2011-08-25 Takashi Toyoshima [WebSocket] update pywebsocket to 0.6b3 which supports closing handshake test https://bugs.webkit.org/show_bug.cgi?id=66924 Reviewed by Kent Tamura. * Scripts/webkitpy/thirdparty/__init__.py: 2011-08-24 Adam Barth The user can't close the details view in garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=66911 Reviewed by Dimitri Glazkov. In addition to adding a close button, this patch changes the test selector to use a 's selection changes https://bugs.webkit.org/show_bug.cgi?id=64504 Reviewed by Alice Liu. * DumpRenderTree/AccessibilityController.h: Added m_notificationsEventHook for addNotificationListener(). m_allEventsHook will now be used for setLogAccessibilityEvents(). * DumpRenderTree/win/AccessibilityControllerWin.cpp: (AccessibilityController::AccessibilityController): Initialize m_notificationsEventHook. (AccessibilityController::~AccessibilityController): Turn off logging of all accessibility events. If m_notificationsEventHook is non-null, unhook it. (logEventProc): Add handling of EVENT_OBJECT_SELECTION. (AccessibilityController::setLogAccessibilityEvents): If the state of logging is not changing, return early. If we're turning off logging, unhook m_allEventsHook, and zero it out. Otherwise, add a hook for all events. (AccessibilityController::addNotificationListener): Use m_notificationsEventHook rather than m_allEventsHook. 2011-07-18 Eric Seidel webkit-patch apply-attachment does not work when not called from the root of the checkout https://bugs.webkit.org/show_bug.cgi?id=64751 Unreviewed. Just updating unit test results after bug 64751. * Scripts/webkitpy/layout_tests/port/chromium_win.py: * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py: 2011-07-18 Adam Barth simplejson has trouble on chromium-linux https://bugs.webkit.org/show_bug.cgi?id=64757 Reviewed by Eric Seidel. Use the native JSON, if available. * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py: 2011-07-18 Ojan Vafai update the flakiness dashboard to understand the new platforms/formats in test_expectations https://bugs.webkit.org/show_bug.cgi?id=64743 Reviewed by Adam Barth. The test expectations format changed a long time ago and the flakiness dashboard was never updated to match. * TestResultServer/static-dashboards/dashboard_base.js: * TestResultServer/static-dashboards/flakiness_dashboard.html: * TestResultServer/static-dashboards/flakiness_dashboard_tests.js: * TestResultServer/static-dashboards/timeline_explorer.html: 2011-07-18 Eric Seidel webkit-patch apply-attachment does not work when not called from the root of the checkout https://bugs.webkit.org/show_bug.cgi?id=64751 Reviewed by Adam Barth. Last week I removed a os.chdir in EnsureWorkingDirectoryClean, which is a step that we run for most commands (as one of the earliest steps). EnsureWorkingDirectoryClean was incorrectly calling os.chdir to change the CWD to the checkout root. This magically made a bunch of otherwise wrong code work. When I realized that apply-attachment no longer worked as expected today, I went and fixed the bug, then realizing that we were not testing what the cwd was when running various commands. I fixed our MockExecutive to always log what the cwd is and fixed a whole bunch of places where we needed to be setting the cwd. Hopefully this will solve our cwd problems once and for-all, and webkit-patch will again correctly work when called from any directory (including outside of a webkit checkout). * Scripts/webkitpy/common/checkout/scm/scm_unittest.py: * Scripts/webkitpy/common/system/executive.py: * Scripts/webkitpy/common/system/workspace_unittest.py: * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py: * Scripts/webkitpy/layout_tests/port/gtk_unittest.py: * Scripts/webkitpy/layout_tests/port/mac_unittest.py: * Scripts/webkitpy/layout_tests/port/qt_unittest.py: * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: * Scripts/webkitpy/tool/bot/irc_command.py: * Scripts/webkitpy/tool/bot/irc_command_unittest.py: * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py: * Scripts/webkitpy/tool/commands/download_unittest.py: * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py: * Scripts/webkitpy/tool/commands/queues.py: * Scripts/webkitpy/tool/commands/queues_unittest.py: * Scripts/webkitpy/tool/commands/queuestest.py: * Scripts/webkitpy/tool/mocktool.py: * Scripts/webkitpy/tool/servers/gardeningserver.py: * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: * Scripts/webkitpy/tool/steps/build.py: * Scripts/webkitpy/tool/steps/checkstyle.py: * Scripts/webkitpy/tool/steps/editchangelog.py: * Scripts/webkitpy/tool/steps/preparechangelog.py: * Scripts/webkitpy/tool/steps/preparechangelogforrevert.py: * Scripts/webkitpy/tool/steps/runtests.py: * Scripts/webkitpy/tool/steps/steps_unittest.py: * Scripts/webkitpy/tool/steps/update.py: 2011-07-18 Adam Barth Increase information garden-o-matic information density by switching to a table https://bugs.webkit.org/show_bug.cgi?id=64642 Reviewed by Dimitri Glazkov. The old layout couldn't handle the large number of failures we had today. This patch switches garden-o-matic to a table-based layout, which can handle many more failures gracefully. (I expect we'll need more UI iterations.) * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css: * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js: * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js: * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js: 2011-07-18 Adam Barth garden-o-matic should be able to determine when compile breaks https://bugs.webkit.org/show_bug.cgi?id=64190 Reviewed by Dimitri Glazkov. This patch adds a red-ish box to the top of the page whenever there is a compile error on the bots. The box automatically opens and closes as appropriate and links to the waterfall display. In the future, we might want to compute a regression range. * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js: - Add the build-only bots to the config. We use these to check whether the build failed, which is faster than waiting for the tester bots to cycle. * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html: - Add DOM for the alert bar. * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css: - CSS to support the alert bar. * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js: - Wiring up events to poll the buildbot to see whether compile has failed. * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js: - Infrastructure for fetching and parsing the buildbot status JSON blob. This code could be better factored for testability. :( * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js: - UI for displaying compile errors. * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js: 2011-07-18 Sheriff Bot Unreviewed, rolling out r91132 and r91135. http://trac.webkit.org/changeset/91132 http://trac.webkit.org/changeset/91135 https://bugs.webkit.org/show_bug.cgi?id=64681 Broke GTK and Chromium debug bots (Requested by rniwa on #webkit). * DumpRenderTree/AccessibilityController.h: * DumpRenderTree/win/AccessibilityControllerWin.cpp: (AccessibilityController::AccessibilityController): (AccessibilityController::~AccessibilityController): (logEventProc): (AccessibilityController::setLogAccessibilityEvents): (AccessibilityController::addNotificationListener): 2011-07-17 Philippe Normand test-webkitpy failing with Python 2.5 https://bugs.webkit.org/show_bug.cgi?id=64594 Reviewed by Eric Seidel. Updated simplejson to version 2.1.6 that fixes an issue when dumping slashes. Also use cgi.parse_qs instead of the urlparse version to be compatible with Python2.5. Verified the fix with test-webkitpy on python 2.5 and python 2.6. * Scripts/webkitpy/thirdparty/simplejson/README.txt: * Scripts/webkitpy/thirdparty/simplejson/__init__.py: * Scripts/webkitpy/thirdparty/simplejson/_speedups.c: (json_PyOS_string_to_double): (_convertPyInt_AsSsize_t): (_convertPyInt_FromSsize_t): (ascii_escape_char): (ascii_escape_unicode): (ascii_escape_str): (raise_errmsg): (join_list_unicode): (join_list_string): (_build_rval_index_tuple): (scanstring_str): (scanstring_unicode): (py_scanstring): (py_encode_basestring_ascii): (scanner_dealloc): (scanner_traverse): (scanner_clear): (_parse_object_str): (_parse_object_unicode): (_parse_array_str): (_parse_array_unicode): (_parse_constant): (_match_number_str): (_match_number_unicode): (scan_once_str): (scan_once_unicode): (scanner_call): (scanner_new): (scanner_init): (encoder_new): (encoder_init): (encoder_call): (_encoded_const): (encoder_encode_float): (encoder_encode_string): (_steal_list_append): (encoder_listencode_obj): (encoder_listencode_dict): (encoder_listencode_list): (encoder_dealloc): (encoder_traverse): (encoder_clear): (init_speedups): * Scripts/webkitpy/thirdparty/simplejson/decoder.py: * Scripts/webkitpy/thirdparty/simplejson/encoder.py: * Scripts/webkitpy/thirdparty/simplejson/ordered_dict.py: Added. * Scripts/webkitpy/thirdparty/simplejson/scanner.py: * Scripts/webkitpy/thirdparty/simplejson/tool.py: Added. * Scripts/webkitpy/tool/servers/reflectionhandler.py: 2011-07-17 Dimitri Glazkov Rename ModifierMatcher to SpecificityCalculator. https://bugs.webkit.org/show_bug.cgi?id=64660 It's a little longer than before, but it is much clearer. Reviewed by Adam Barth. * Scripts/webkitpy/layout_tests/models/test_expectations.py: Renamed and adjusted all callsites, also rewrote the comments. * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Adjusted callsites and renamed tests. 2011-07-15 Adam Barth gardening server should proxy buildbot status for garden-o-matic https://bugs.webkit.org/show_bug.cgi?id=64588 Reviewed by Eric Seidel. This information will be used to detect build breaks. I took this opportunity to make ChromiumBuildBot non-static. * Scripts/webkitpy/common/host.py: * Scripts/webkitpy/tool/commands/rebaseline.py: * Scripts/webkitpy/tool/commands/rebaseline_unittest.py: * Scripts/webkitpy/tool/mocktool.py: * Scripts/webkitpy/tool/servers/gardeningserver.py: * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: 2011-07-15 Julien Chaffraix [NRWT] Add support for --no-http https://bugs.webkit.org/show_bug.cgi?id=64564 Reviewed by Dirk Pranke. Added support for --no-http, which disables both HTTP and websockets tests. It also matches the old-run-webkit-tests behavior if --force is used. * Scripts/webkitpy/layout_tests/controllers/manager.py: Fixed HTTP_SUBDIR and WEBSOCKET_SUBDIR as tests do not start with a leading separator. We check if --no-http is set and add the HTTP / websockets tests to the skipped list prior to looking at the expectation file. Fixed the _test_requires_lock function to use the same code path to determine what is worth have an HTTP lock as --no-http to avoid badness. * Scripts/webkitpy/layout_tests/port/test.py: Added 2 new tests to our mock filesystem to validate that we do skip properly HTTP / websocket tests inside platform/. * Scripts/webkitpy/layout_tests/run_webkit_tests.py: Added tests that we properly skip all the tests. * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py: Added checks for the command line arguments. 2011-07-13 Jon Honeycutt Focus and selection events are not fired when a