Tools/ChangeLog

   1 2012-04-23  Raphael Kubo da Costa  <rakuco@webkit.org>
   2
   3         [EFL] Pass a WTF::String directly from DRT to ewebkit when possible
   4         https://bugs.webkit.org/show_bug.cgi?id=84634
   5
   6         Reviewed by Gustavo Noronha Silva.
   7
   8         Drop unnecessary const char* -> WTF::String conversions by
   9         creating String objects directly in LayoutTestController out of
  10         JSStringRef's and passing them to DumpRenderTreeSupportEfl.
  11
  12         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
  13         (LayoutTestController::queueLoad):
  14         (LayoutTestController::setValueForUser):
  15         (LayoutTestController::findString):
  16         (LayoutTestController::addUserStyleSheet):
  17
  18 2012-05-08  Alexander Shalamov  <alexander.shalamov@intel.com>
  19
  20         [EFL][DRT] EventSender needs implementation of leapForward function
  21         https://bugs.webkit.org/show_bug.cgi?id=83486
  22
  23         Reviewed by Eric Seidel.
  24
  25         Implemented leapForward function in Efl's DRT. This function is required
  26         to delay and queue event execution and simulate real user behavior.
  27
  28         * DumpRenderTree/efl/EventSender.cpp:
  29         (DelayedEvent):
  30         (MouseEventInfo::MouseEventInfo):
  31         (MouseEventInfo):
  32         (sendClick):
  33         (mouseDownCallback):
  34         (mouseUpCallback):
  35         (mouseMoveToCallback):
  36         (leapForwardCallback):
  37         (mouseScrollByCallback):
  38         (feedOrQueueMouseEvent):
  39         (feedMouseEvent):
  40         (feedQueuedMouseEvents):
  41
  42 2012-05-08  Ryuan Choi  <ryuan.choi@samsung.com>
  43
  44         [EFL][DRT] WebKit/DRT should not paint webview for the pixel tests.
  45         https://bugs.webkit.org/show_bug.cgi?id=79853
  46
  47         Reviewed by Eric Seidel.
  48
  49         Screen buffer can be different from rendered results.
  50         For example, WebKit/Efl just moves tiles for ewk_view_tiled when scrolling
  51         contents.
  52         It means that pixel tests can't check regression of ewk_view_tiled.
  53
  54         This patch changes a way to provide bitmap from painting webview to capturing
  55         evas.
  56
  57         * DumpRenderTree/efl/PixelDumpSupportEfl.cpp:
  58         (createBitmapContextFromWebView):
  59
  60 2012-05-08  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
  61
  62         [GTK] Memory leak in DRT EventSender.cpp
  63         https://bugs.webkit.org/show_bug.cgi?id=85886
  64
  65         Reviewed by Martin Robinson.
  66
  67         Fix a memory leak in gdkModifersFromJSValue() by using JSRetainPtr for
  68         JSString created with JSStringCreateWithUTF8CString.
  69
  70         * DumpRenderTree/gtk/EventSender.cpp:
  71         (gdkModifersFromJSValue):
  72
  73 2012-05-08  Christophe Dumez  <christophe.dumez@intel.com>
  74
  75         [EFL] glib dependency needs to be updated in jhbuild
  76         https://bugs.webkit.org/show_bug.cgi?id=85878
  77
  78         Reviewed by Martin Robinson.
  79
  80         Bump glib version from 2.31.8 to 2.32.2 in jhbuild.modules. This
  81         addresses flakiness in media tests due to possible deadlock with the
  82         following output on stderr:
  83         GStreamer-WARNING **: wrong STREAM_LOCK count 0
  84
  85         The bug was fixed upstream by:
  86         http://git.gnome.org/browse/glib/commit/?id=265f265c677bc5399dbb1cfa449cb4b915a6459f
  87
  88         * efl/jhbuild.modules:
  89
  90 2012-05-07  Alexis Menard  <alexis.menard@openbossa.org>
  91
  92         [Qt] Unbreak debugging of WebKit2.
  93         https://bugs.webkit.org/show_bug.cgi?id=85839
  94
  95         Reviewed by Simon Hausmann.
  96
  97         When you attach GDB to a running process, it stops it.
  98         http://trac.webkit.org/changeset/115958 introduced a pause()
  99         call to wait the debugger to be attached to then continue
 100         the execution of the WebProcess. Unfortunately the pause()
 101         function does not return unless a signal handler is called.
 102         This patch introduce an event handler to exit from the paused
 103         state when the debugger send the signal SIGCONT. The old code
 104         works with older version of GDB (<7.0) but not with newer
 105         versions where the behavior of pause() is correct.
 106
 107         * WebKitTestRunner/qt/main.cpp:
 108         (sigcontHandler):
 109         (main):
 110
 111 2012-05-08  Christophe Dumez  <christophe.dumez@intel.com>
 112
 113         [EFL][DRT] EFL's DRT does not fully support page visibility
 114         https://bugs.webkit.org/show_bug.cgi?id=85032
 115
 116         Reviewed by Antonio Gomes.
 117
 118         * DumpRenderTree/LayoutTestController.cpp:
 119         (resetPageVisibilityCallback): Fix incorrect argumentCount check.
 120         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 121         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): Properly
 122         reset page visibility setting between tests.
 123         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 124         (LayoutTestController::setPageVisibility): Add support for "preview"
 125         visibility state.
 126
 127 2012-05-08  Gavin Peters  <gavinp@chromium.org>
 128
 129         [webkit-patch] Return meaningful results from single-result queries.
 130         https://bugs.webkit.org/show_bug.cgi?id=85367
 131
 132         Reviewed by Dirk Pranke.
 133
 134         Sometimes a quicksearch returns a single result by just redirecting to the page for that single
 135         bug. This patch hacks around that by noticing the redirect target is a single page, and constructing
 136         a result list consisting of that one bug.
 137
 138         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
 139         (BugzillaQueries._fetch_bugs_from_advanced_query):
 140
 141 2012-05-08  Gavin Peters  <gavinp@chromium.org>
 142
 143         [webkit-patch] Add new chrome-channels command to track down current chrome release channels for a committed bug.
 144         https://bugs.webkit.org/show_bug.cgi?id=85368
 145
 146         Reviewed by Eric Seidel.
 147
 148         * Scripts/webkitpy/common/config/urls.py:
 149         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
 150         (MockBugzillaQueries.fetch_bugs_matching_quicksearch):
 151         * Scripts/webkitpy/common/net/omahaproxy.py: Added.
 152         (OmahaProxy):
 153         (OmahaProxy.__init__):
 154         (OmahaProxy.set_url):
 155         (OmahaProxy._json_url):
 156         (OmahaProxy._get_json):
 157         (OmahaProxy.get_revisions):
 158         * Scripts/webkitpy/common/net/omahaproxy_unittest.py: Added.
 159         (MockOmahaProxy):
 160         (MockOmahaProxy.__init__):
 161         (OmahaProxyTest):
 162         (OmahaProxyTest.test_get_revisions):
 163         * Scripts/webkitpy/tool/commands/__init__.py:
 164         * Scripts/webkitpy/tool/commands/bugsearch.py:
 165         (BugSearch):
 166         * Scripts/webkitpy/tool/commands/chromechannels.py: Added.
 167         (ChromeChannels):
 168         (__init__):
 169         (execute):
 170         * Scripts/webkitpy/tool/commands/chromechannels_unittest.py: Added.
 171         (MockOmahaProxy):
 172         (MockOmahaProxy.get_revisions):
 173         (TestableChromeChannels):
 174         (TestableChromeChannels.__init__):
 175         (ChromeChannelsTest):
 176         (ChromeChannelsTest.test_single_bug):
 177         (ChromeChannelsTest.test_with_query):
 178
 179 2012-05-08  Carlos Garcia Campos  <cgarcia@igalia.com>
 180
 181         Unreviewed. Fix GTK+ unit tests.
 182
 183         * gtk/common.py:
 184         (pkg_config_file_variable):
 185
 186 2012-05-08  Carlos Garcia Campos  <cgarcia@igalia.com>
 187
 188         [GTK] Google tests that time out are leaked
 189         https://bugs.webkit.org/show_bug.cgi?id=84973
 190
 191         Reviewed by Philippe Normand.
 192
 193         Move Tools/Scripts/run-gtk-tests to Tools/gtk/run-api-tests
 194         leaving run-gtk-tests as simple wrapper that runs run-api-tests in
 195         a jhbuild environment when jhbuild is available.
 196
 197         * Scripts/run-gtk-tests: Run Tools/gtk/run-api-tests directly or
 198         with jhbuild if present.
 199         * gtk/common.py:
 200         (pkg_config_file_variable): Generic function to get a variable
 201         from a pkg-config file.
 202         (prefix_of_pkg_config_file): Use pkg_config_file_variable with 'prefix'.
 203         * gtk/run-api-tests: Copied from Tools/Scripts/run-gtk-tests.
 204         (TestRunner.__init__): Use common.build_path() to get the programs path.
 205         (TestRunner._lookup_atspi2_binary): Use
 206         common.pkg_config_file_variable() to get the 'exec_prefix'.
 207         (TestRunner._start_accessibility_daemons): Use subprocess.Popen()
 208         directly instead of helper _create_process that has been removed.
 209         (TestRunner._setup_testing_environment): Use subprocess.Popen()
 210         directly instead of helper _create_process that has been
 211         removed. Use common.top_level_path() and common.build_path()
 212         instead of custom methods that have been removed too.
 213         (TestRunner._run_test_command): Use subprocess.Popen() directly
 214         instead of helper _create_process that has been removed.
 215
 216 2012-05-08  Thiago Marcos P. Santos  <thiago.santos@intel.com>
 217
 218         [EFL] [DRT] Clear extra views before setting a blank page
 219         https://bugs.webkit.org/show_bug.cgi?id=85608
 220
 221         Reviewed by Kenneth Rohde Christiansen.
 222
 223         We need to clear the extra views before setting a blank page,
 224         otherwise it will cause more events interactions and sometimes
 225         trigger a unload(). This was causing frame-crash-with-page-cache.html
 226         to leak events to the next test causing any subsequent test to
 227         fail.
 228
 229         * DumpRenderTree/efl/DumpRenderTree.cpp:
 230         (runTest):
 231         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 232         (DumpRenderTreeChrome::clearExtraViews):
 233         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
 234         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
 235         (DumpRenderTreeChrome):
 236
 237 2012-05-08  Balazs Kelemen  <kbalazs@webkit.org>
 238
 239         [Qt] X11 plugins need to be reworked for Qt5+WK1
 240         https://bugs.webkit.org/show_bug.cgi?id=80691
 241
 242         Reviewed by Simon Hausmann.
 243
 244         Implement basic windowless plugin support with Qt5.
 245
 246         * qmake/mkspecs/features/features.prf:
 247         Enable NPAPI plugins if X11 libraries are available.
 248         * qmake/mkspecs/features/functions.prf:
 249         Added a convenience function to determine availability
 250         of X11 libraries.
 251
 252 2012-05-07  Raphael Kubo da Costa  <rakuco@webkit.org>
 253
 254         [webkitpy] Remove Python 2.5-specific workaround from http_server_base.py.
 255         https://bugs.webkit.org/show_bug.cgi?id=85852
 256
 257         Reviewed by Dirk Pranke.
 258
 259         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
 260         (HttpServerBase._is_server_running_on_all_ports): Revert the part
 261         of r89412 which was a workaround for an issue with Python 2.5 on
 262         Mac 10.5.
 263
 264 2012-05-07  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
 265
 266         webkitdirs.pm: Make determineNumberOfCPUs work on FreeBSD.
 267         https://bugs.webkit.org/show_bug.cgi?id=81481
 268
 269         Reviewed by Dirk Pranke.
 270
 271         FreeBSD has the same sysctl interface present in Darwin to query
 272         the number of CPUs available, so use that too.
 273
 274         * Scripts/webkitdirs.pm:
 275         (determineNumberOfCPUs): Share the sysctl call with Darwin.
 276         (isFreeBSD): Added.
 277
 278 2012-05-07  Alexis Menard  <alexis.menard@openbossa.org>
 279
 280         Unreviewed Qt build fix in Debug.
 281
 282         We need some system includes here for getpid() to be recognized.
 283
 284         * WebKitTestRunner/qt/main.cpp:
 285
 286 2012-05-07  Ojan Vafai  <ojan@chromium.org>
 287
 288         When embedding the flakiness dashboard, hide popups when the frame is blurred.
 289         https://bugs.webkit.org/show_bug.cgi?id=85684
 290
 291         Reviewed by Dirk Pranke.
 292
 293         If you show a popup, then click elsewhere in garden-o-matic, it will now
 294         properly hide it.
 295
 296         * TestResultServer/static-dashboards/dashboard_base.js:
 297         (showPopup):
 298         * TestResultServer/static-dashboards/flakiness_dashboard.js:
 299         (showPopupForBuild):
 300         (generatePageForIndividualTests.if):
 301         (generatePageForIndividualTests):
 302         * TestResultServer/static-dashboards/flakiness_dashboard_embedded_unittests.js: Added.
 303         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
 304         * TestResultServer/static-dashboards/run-embedded-unittests.html: Added.
 305
 306 2012-05-07  Ian Vollick  <vollick@chromium.org>
 307
 308         [chromium] WebViewHost should setCompositorSurfaceReady
 309         https://bugs.webkit.org/show_bug.cgi?id=85418
 310
 311         Reviewed by James Robinson.
 312
 313         * DumpRenderTree/chromium/WebViewHost.cpp:
 314         (WebViewHost::setWebWidget):
 315
 316 2012-05-07  Jon Lee  <jonlee@apple.com>
 317
 318         WebKitTestRunner needs layoutTestController.dumpFrameLoadCallbacks
 319         https://bugs.webkit.org/show_bug.cgi?id=42331
 320         <rdar://problem/8193641>
 321
 322         Reviewed by Darin Adler.
 323
 324         This patch sets up the framework for supporting dumpFrameLoadCallbacks, as well as
 325         outputting some of the required strings. As tests become unskipped the other callbacks
 326         can be supported.
 327
 328         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: Add dumpFrameLoadCallbacks()
 329         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
 330         (WTR::InjectedBundle::didReceiveMessage): pass in the dictionary containing keys needed to initialize
 331         dump callbacks for the layoutTestController.
 332         (WTR::InjectedBundle::booleanForKey): Helper function to extract boolean value from key in WKDictionaryRef.
 333         Output some warning if we are trying to extract a value that is not a boolean.
 334         (WTR::InjectedBundle::beginTesting): Update to take in the dictionary passed in by the TestController.
 335         We cannot set the bits on the layoutTestController until it has been created in beginTesting().
 336         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
 337         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
 338         (WTR::InjectedBundle::didReceiveMessage):
 339         (WTR):
 340         (WTR::InjectedBundle::beginTesting):
 341         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
 342         (InjectedBundle):
 343         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
 344         (WTR):
 345         (WTR::dumpFrameDescriptionSuitableForTestResult): Static function similar to
 346         -[WebFrame _drt_descriptionSuitableForTestResult] in DumpRenderTree.
 347         (WTR::InjectedBundlePage::didStartProvisionalLoadForFrame): Output string.
 348         (WTR::InjectedBundlePage::didCommitLoadForFrame): Output string.
 349         (WTR::InjectedBundlePage::didFinishLoadForFrame): Output string.
 350         (WTR::InjectedBundlePage::didReceiveTitleForFrame): Output string.
 351         (WTR::InjectedBundlePage::didCancelClientRedirectForFrame): Output string.
 352         (WTR::InjectedBundlePage::willPerformClientRedirectForFrame): Output string. In WK1 it's
 353         "willPerformClientRedirectToURL" so we will have to use that value here.
 354         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame): Output string.
 355         (WTR::InjectedBundlePage::didHandleOnloadEventsForFrame): Output string.
 356         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
 357         (WTR::LayoutTestController::LayoutTestController): Initialize to not dump frame load callbacks.
 358         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
 359         (WTR::LayoutTestController::dumpFrameLoadCallbacks): Change to use setShouldDumpFrameLoadCallbacks.
 360         (WTR::LayoutTestController::setShouldDumpFrameLoadCallbacks): Set bit.
 361         (WTR::LayoutTestController::shouldDumpFrameLoadCallbacks): Return bit.
 362         (LayoutTestController): Added m_dumpFrameLoadCallbacks to track whether to dump the output.
 363         * WebKitTestRunner/TestInvocation.cpp:
 364         (WTR::shouldLogFrameLoadDelegates): Similar to DRT.
 365         (WTR::TestInvocation::invoke): Based on the URL loaded, set the bit in the dictionary that gets passed
 366         to the injected bundle.
 367
 368 2012-05-07  Tommy Widenflycht  <tommyw@google.com>
 369
 370         [chromium] MediaStream API: Enhance WebUserMediaClientMock and add a test for it
 371         https://bugs.webkit.org/show_bug.cgi?id=85618
 372
 373         Reviewed by Adam Barth.
 374
 375         Adding the capability to control which MediaStreamTracks actually get created,
 376         and adding a test that exercises the JS api using that capability.
 377
 378         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
 379         (WebKit::WebUserMediaClientMock::requestUserMedia):
 380         * DumpRenderTree/chromium/WebUserMediaClientMock.h:
 381         * DumpRenderTree/chromium/WebViewHost.cpp:
 382         (WebViewHost::testMediaStreamClient):
 383         * DumpRenderTree/chromium/WebViewHost.h:
 384         (WebViewHost):
 385
 386 2012-05-07  Zan Dobersek  <zandobersek@gmail.com>
 387
 388         [Gtk] WebGL feature is not built anymore through build-webkit after r116251
 389         https://bugs.webkit.org/show_bug.cgi?id=85788
 390
 391         Reviewed by Martin Robinson.
 392
 393         Enable WebGL by default for the Gtk port. The feature is already enabled by
 394         default in configure.ac, so there's no reason it shouldn't be enabled when
 395         building through build-webkit.
 396
 397         * Scripts/webkitperl/FeatureList.pm:
 398
 399 2012-05-07  Csaba Osztrogonác  <ossy@webkit.org>
 400
 401         Unittest fix after r116288, because now there are more than 5 hit for the 'Eric' word.
 402         We have only 2 'Gavin', so it is a better choice than 'Eric' for unit testing.
 403
 404         Reviewed by Tor Arne Vestbø.
 405
 406         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
 407         (IRCCommandTest.test_whois):
 408
 409 2012-05-07  Sheriff Bot  <webkit.review.bot@gmail.com>
 410
 411         Unreviewed, rolling out r116299, r116301, and r116303.
 412         http://trac.webkit.org/changeset/116299
 413         http://trac.webkit.org/changeset/116301
 414         http://trac.webkit.org/changeset/116303
 415         https://bugs.webkit.org/show_bug.cgi?id=85795
 416
 417         Build is still broken (Requested by Ossy on #webkit).
 418
 419         * DumpRenderTree/qt/DumpRenderTree.pro:
 420         * DumpRenderTree/qt/main.cpp:
 421         (main):
 422         * QtTestPlatformPlugin/QtTestPlatformPlugin.pro: Removed.
 423         * QtTestPlatformPlugin/TestIntegration.cpp: Removed.
 424         * QtTestPlatformPlugin/TestIntegration.h: Removed.
 425         * QtTestPlatformPlugin/mac/TestFontDatabase.h: Removed.
 426         * QtTestPlatformPlugin/mac/TestFontDatabase.mm: Removed.
 427         * QtTestPlatformPlugin/mac/TestIntegrationMac.mm: Removed.
 428         * QtTestPlatformPlugin/main.cpp: Removed.
 429         * QtTestPlatformPlugin/testplatform.json: Removed.
 430         * Tools.pro:
 431         * WebKitTestRunner/Target.pri:
 432         * WebKitTestRunner/qt/main.cpp:
 433         (main):
 434
 435 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
 436
 437         Unreviewed Qt build fix.
 438
 439         One more try to learn how to use
 440         const with a pointer array.
 441
 442         * DumpRenderTree/qt/main.cpp:
 443         (initializeTestPlatformPlugin):
 444
 445 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
 446
 447         Unreviewed build fix after r116299.
 448
 449         * DumpRenderTree/qt/main.cpp:
 450         (initializeTestPlatformPlugin):
 451
 452 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
 453
 454         [Qt] Add test specific platform plugin to achieve unified layout test results
 455         https://bugs.webkit.org/show_bug.cgi?id=80996
 456
 457         Reviewed by Simon Hausmann.
 458
 459         Added QtTestPlatformPlugin as a new project under Tools.
 460         This is a Qt5-ish platform plugin that can be used to tweak the
 461         platform support interfaces in order to unify layout test results.
 462         For now it only overrides the font database on Mac and redirects
 463         everything else to the real platform plugin. The font database it
 464         provides mimics the way how we set up test fonts with fontconfig on Linux.
 465         Make DumpRenderTree and WebKitTestRunner use this platform plugin.
 466
 467         * DumpRenderTree/qt/DumpRenderTree.pro:
 468         * DumpRenderTree/qt/main.cpp:
 469         (initializeTestPlatformPlugin):
 470         (main):
 471         * QtTestPlatformPlugin/QtTestPlatformPlugin.pro: Added.
 472         * QtTestPlatformPlugin/TestIntegration.cpp: Added.
 473         (TestIntegration::TestIntegration):
 474         (TestIntegration::fontDatabase):
 475         * QtTestPlatformPlugin/TestIntegration.h: Added.
 476         (TestIntegration):
 477         (TestIntegration::hasCapability):
 478         (TestIntegration::createPlatformPixmap):
 479         (TestIntegration::createPlatformWindow):
 480         (TestIntegration::createPlatformBackingStore):
 481         (TestIntegration::createPlatformOpenGLContext):
 482         (TestIntegration::createPlatformSharedGraphicsCache):
 483         (TestIntegration::guiThreadEventDispatcher):
 484         (TestIntegration::clipboard):
 485         (TestIntegration::drag):
 486         (TestIntegration::inputContext):
 487         (TestIntegration::accessibility):
 488         (TestIntegration::nativeInterface):
 489         (TestIntegration::services):
 490         (TestIntegration::styleHint):
 491         (TestIntegration::platformTheme):
 492         * QtTestPlatformPlugin/mac/TestFontDatabase.h: Added.
 493         (TestFontDatabase):
 494         * QtTestPlatformPlugin/mac/TestFontDatabase.mm: Added.
 495         (TestFontDatabase::populateFontDatabase):
 496         * QtTestPlatformPlugin/mac/TestIntegrationMac.mm: Added.
 497         (TestIntegration::fontDatabase):
 498         * QtTestPlatformPlugin/main.cpp: Added.
 499         (TestIntegrationPlugin::keys):
 500         (TestIntegrationPlugin::create):
 501         (TestIntegrationPlugin::initialize):
 502         * QtTestPlatformPlugin/testplatform.json: Added.
 503         * Tools.pro:
 504         * WebKitTestRunner/Target.pri:
 505         * WebKitTestRunner/qt/main.cpp:
 506         (main):
 507
 508 2012-05-07  Christophe Dumez  <christophe.dumez@intel.com>
 509
 510         [EFL] media/video-poster-blocked-by-willsendrequest.html is flaky
 511         https://bugs.webkit.org/show_bug.cgi?id=85609
 512
 513         Reviewed by Simon Hausmann.
 514
 515         Flakiness was caused by the fact that
 516         media/crash-closing-page-with-media-as-plugin-fallback.html and
 517         media/video-poster-blocked-by-willsendrequest.html use the same PNG
 518         file as poster attribute in the video tag. The PNG file gets cached
 519         when the first test is executed and this somehow causes the second
 520         test to fail.
 521
 522         The flakiness was addressed by clearing the memory and page caches
 523         between tests.
 524
 525         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 526         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
 527
 528 2012-05-07  Eric Seidel  <eric@webkit.org>
 529
 530         Add values for all features to Qt's features.pri
 531         https://bugs.webkit.org/show_bug.cgi?id=85746
 532
 533         Reviewed by Tor Arne Vestbø.
 534
 535         When generate-feature-files generates features.pri it provides a value
 536         for every possible ENABLE_.  In prepration for landing generate-feature-files
 537         it seems appropriate to add a value for all possible ENABLEs to the existing
 538         features.pri.  This lets me make sure that I'm getting them all right!
 539
 540         * qmake/mkspecs/features/features.pri:
 541
 542 2012-05-07  Adam Bergkvist  <adam.bergkvist@ericsson.com>
 543
 544         Added myself as a committer.
 545
 546         Not reviewed.
 547
 548         * Scripts/webkitpy/common/config/committers.py:
 549
 550 2012-05-07  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
 551
 552         [EFL][DRT] Implement LayoutTestController::execCommand
 553         https://bugs.webkit.org/show_bug.cgi?id=82286
 554
 555         Reviewed by Antonio Gomes.
 556
 557         Add missing implementation execCommand and isCommandEnabled to EFL's
 558         LayoutTestController.
 559
 560         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 561         (LayoutTestController::execCommand):
 562         (LayoutTestController::isCommandEnabled):
 563
 564 2012-05-06  Hayato Ito  <hayato@chromium.org>
 565
 566         [prepare-ChangeLog] Excludes a reference file used by reftests from test files.
 567         https://bugs.webkit.org/show_bug.cgi?id=74268
 568
 569         Reviewed by Ryosuke Niwa.
 570
 571         * Scripts/prepare-ChangeLog:
 572         (generateFileList):
 573
 574 2012-05-06  Mary Wu  <mary.wu@torchmobile.com.cn>
 575
 576         Add a contributor to committers.py
 577         https://bugs.webkit.org/show_bug.cgi?id=85761
 578
 579         Reviewed by Unreviewed.
 580
 581         Add myself to committers.py contributor.
 582
 583         * Scripts/webkitpy/common/config/committers.py:
 584
 585 2012-05-06  Dan Bernstein  <mitz@apple.com>
 586
 587         Tools part of: Building and debugging WebKit in the Xcode IDE requires a lot of setup
 588         https://bugs.webkit.org/show_bug.cgi?id=85739
 589
 590         Reviewed by Daniel Bates.
 591
 592         * Scripts/build-webkit: Moved code that copies from WebKitLibraries to the product directory
 593         from here...
 594         * Scripts/copy-webkitlibraries-to-product-directory: ...to this new script.
 595
 596 2012-05-06  Andy Estes  <aestes@apple.com>
 597
 598         Fix the build after r116246.
 599
 600         * DumpRenderTree/LayoutTestController.cpp: Remove an unused function.
 601
 602 2012-05-06  Eric Seidel  <eric@webkit.org>
 603
 604         Update several build-webkit options to match the #defines they toggle
 605         https://bugs.webkit.org/show_bug.cgi?id=85744
 606
 607         Reviewed by Adam Barth.
 608
 609         Most notable here is fixing --3d-canvas to be --webgl these days --
 610         the define was renamed 15 months ago!
 611
 612         * Scripts/webkitperl/FeatureList.pm:
 613
 614 2012-05-06  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
 615
 616         [EFL] EFL's LayoutTestController setJavaScriptCanAccessClipboard implementation
 617         https://bugs.webkit.org/show_bug.cgi?id=83687
 618
 619         Reviewed by Antonio Gomes.
 620
 621         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 622         (LayoutTestController::setJavaScriptCanAccessClipboard):
 623
 624 2012-05-06  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
 625
 626         Convert isPageBoxVisible to use Internals interface.
 627         https://bugs.webkit.org/show_bug.cgi?id=85692
 628
 629         Reviewed by Darin Adler.
 630
 631         Remove isPageBoxVisible functions, because it is able to work in the
 632         cross-port way through the Internals interface.
 633
 634         * DumpRenderTree/LayoutTestController.cpp:
 635         (LayoutTestController::staticFunctions):
 636         * DumpRenderTree/LayoutTestController.h:
 637         (LayoutTestController):
 638         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
 639         * DumpRenderTree/chromium/LayoutTestController.cpp:
 640         (LayoutTestController::LayoutTestController):
 641         * DumpRenderTree/chromium/LayoutTestController.h:
 642         (LayoutTestController):
 643         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 644         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
 645         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
 646         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
 647         * DumpRenderTree/qt/LayoutTestControllerQt.h:
 648         (LayoutTestController):
 649         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
 650         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
 651
 652 2012-05-06  Tom Zakrajsek  <tomz@codeaurora.org>
 653
 654         sheriffbot isn't reopening patches after it lands rollouts
 655         https://bugs.webkit.org/show_bug.cgi?id=64418
 656
 657         Reviewed by Adam Barth.
 658
 659         * Scripts/webkitpy/tool/commands/download_unittest.py:
 660         * Scripts/webkitpy/tool/steps/createbug.py:
 661         (CreateBug.run):
 662
 663 2012-05-04  Raphael Kubo da Costa  <rakuco@webkit.org>
 664
 665         webkitpy: Use os.pathsep instead of manually finding out the path separator in NRWT.
 666         https://bugs.webkit.org/show_bug.cgi?id=85697
 667
 668         Reviewed by Eric Seidel.
 669
 670         * Scripts/new-run-webkit-tests: Instead of manually detecting whether
 671         the path separator should be ':' or ';' by checking sys.platform, rely
 672         on os.pathsep which has exactly the same information (os.pathsep is
 673         ':' for Cygwin, from what I could check in Python's sources).
 674
 675 2012-05-05  Sheriff Bot  <webkit.review.bot@gmail.com>
 676
 677         Unreviewed, rolling out r116223.
 678         http://trac.webkit.org/changeset/116223
 679         https://bugs.webkit.org/show_bug.cgi?id=85718
 680
 681         The bots need to set XDG_DATA_DIRS (Requested by rakuco on
 682         #webkit).
 683
 684         * efl/jhbuildrc:
 685         * gtk/jhbuildrc:
 686
 687 2012-05-05  Raphael Kubo da Costa  <rakuco@webkit.org>
 688
 689         [jhbuild] Do not set XDG_DATA_DIRS in jhbuildrc.
 690         https://bugs.webkit.org/show_bug.cgi?id=85717
 691
 692         Reviewed by Martin Robinson.
 693
 694         After r116209, the XDG_DATA_DIRS environment variable is now
 695         passed by webkitpy's layout_tests code to run-with-jhbuild, so
 696         instead of setting a few defaults in jhbuildrc we can now rely on
 697         the values set and really used by each system.
 698
 699         * efl/jhbuildrc:
 700         * gtk/jhbuildrc:
 701
 702 2012-05-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
 703
 704         [Qt] Buildfix for --minimal build after r116107.
 705
 706         Reviewed by Csaba Osztrogonác.
 707
 708         * qmake/mkspecs/features/features.prf:
 709
 710 2012-05-05  Zan Dobersek  <zandobersek@gmail.com>
 711
 712         [Gtk] GtkDriver is not required anymore after r116134
 713         https://bugs.webkit.org/show_bug.cgi?id=85699
 714
 715         Reviewed by Csaba Osztrogonác.
 716
 717         Remove GtkDriver as it was replaced by the XvfbDriver and is
 718         no longer needed.
 719
 720         * Scripts/webkitpy/layout_tests/port/gtk.py:
 721
 722 2012-05-05  Zan Dobersek  <zandobersek@gmail.com>
 723
 724         Increase the display ID for the Xvfb process to use if pixel_tests argument
 725         is true rather than if pixel_tests option is present. This fixes the
 726         XvfbDriver in the same way r115825 fixed the now-redundant GtkDriver.
 727
 728         Rubber-stamped by Csaba Osztrogonác.
 729
 730         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py:
 731         (XvfbDriver._start):
 732
 733 2012-05-04  Raphael Kubo da Costa  <rakuco@webkit.org>
 734
 735         webkitpy: Preserve XDG_DATA_DIRS in the environment passed to ServerProcess.
 736         https://bugs.webkit.org/show_bug.cgi?id=85694
 737
 738         Reviewed by Martin Robinson.
 739
 740         This is sort of a follow-up to r100674: $XDG_DATA_DIRS should be
 741         preserved, as it is particularly useful for Linux/Unix
 742         environments.  As a real use case, if shared-mime-info is
 743         installed into a non-trivial prefix, Glib and other libraries will
 744         have trouble to properly identify mimetypes (in ResourceResponses,
 745         for example) as the base freedesktop.org mimetype file will not be
 746         found.
 747
 748         * Scripts/webkitpy/layout_tests/port/base.py:
 749         (Port.to.setup_environ_for_server):
 750
 751 2012-05-04  Gustavo Noronha Silva  <gns@gnome.org>
 752
 753         [GTK] Simplify how libWebCoreModules is linked in, and fix WebKit2 build
 754         https://bugs.webkit.org/show_bug.cgi?id=85691
 755
 756         * GNUmakefile.am: no longer link libWebCoreModules to DumpRenderTree.
 757
 758 2012-05-04  Eric Seidel  <eric@webkit.org>
 759
 760         Make capitalization and descriptions in FeatureList.pm more consistent to make easier to autogenerate
 761         https://bugs.webkit.org/show_bug.cgi?id=85583
 762
 763         Reviewed by Daniel Bates.
 764
 765         * Scripts/webkitperl/FeatureList.pm:
 766
 767 2012-05-04  Eric Seidel  <eric@webkit.org>
 768
 769         Sort entries in FeaturesList.pm to make them easier to autogenerate
 770         https://bugs.webkit.org/show_bug.cgi?id=85584
 771
 772         Reviewed by Daniel Bates.
 773
 774         No functional changes, only sorting.
 775
 776         * Scripts/webkitperl/FeatureList.pm:
 777
 778 2012-05-04  Andy Estes  <aestes@apple.com>
 779
 780         REGRESSION (r115752): WebKit2.NewFirstVisuallyNonEmptyLayout API test times out
 781         https://bugs.webkit.org/show_bug.cgi?id=85675
 782
 783         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout.cpp:
 784         (TestWebKitAPI::TEST): Disable the test.
 785
 786 2012-04-30  Jer Noble  <jer.noble@apple.com>
 787
 788         Full screen will exit during a provisional load of a non-ancestor iframe.
 789         https://bugs.webkit.org/show_bug.cgi?id=85230
 790
 791         Reviewed by Maciej Stachowiak .
 792
 793         Add support for the new UIDelegate webView:closeFullScreenWithListener: method:
 794         * DumpRenderTree/mac/UIDelegate.mm:
 795         (-[UIDelegate webView:closeFullScreenWithListener:]):
 796
 797         Add support for the new closeFullScreen injected bundle method:
 798         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
 799         (WTR::InjectedBundlePage::InjectedBundlePage):
 800         (WTR::InjectedBundlePage::closeFullScreen):
 801         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
 802
 803 2012-05-04  Ojan Vafai  <ojan@chromium.org>
 804
 805         Fix test naming to be camel-case like the methods they are testing.
 806
 807         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
 808         (test):
 809
 810 2012-05-04  Ojan Vafai  <ojan@chromium.org>
 811
 812         Show the results links in the embedded flakiness dashboard
 813         https://bugs.webkit.org/show_bug.cgi?id=85660
 814
 815         Reviewed by Adam Barth.
 816
 817         Turns out these links don't add much visual noise and it's annoying
 818         to have to go to the dashboard from garden-o-matic just to view the
 819         results for a test.
 820
 821         This is useful for cases where garden-o-matic doesn't properly show the
 822         results (e.g. missing tests). If we address all those cases, then
 823         we could easily hide the results links again if we wanted to.
 824
 825         * TestResultServer/static-dashboards/flakiness_dashboard.js:
 826         (showUpdateInfoForTest):
 827         (htmlForIndividualTestOnAllBuildersWithResultsLinks):
 828         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
 829         (test):
 830
 831 2012-05-04  Dirk Pranke  <dpranke@chromium.org>
 832
 833         Revert to using chromium --test-shell mode on SL in NRWT
 834
 835         https://bugs.webkit.org/show_bug.cgi?id=83076
 836
 837         Unreviewed, build fix.
 838
 839         We seem to be seeing a lot more flakiness on the bot since
 840         I switched NRWT to "DRT" mode. We'll try reverting back and
 841         see if this helps things.
 842
 843         * Scripts/webkitpy/layout_tests/port/chromium.py:
 844         (ChromiumDriver.__init__):
 845
 846 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
 847
 848         [EFL] Implement layoutTestController.dumpResourceResponseMIMETypes
 849         https://bugs.webkit.org/show_bug.cgi?id=84941
 850
 851         Reviewed by Martin Robinson.
 852
 853         Print MIME type detection messages in EFL's DumpRenderTree if
 854         LayoutTestController's dumpResourceResponseMIMETypes() returns true.
 855
 856         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 857         (DumpRenderTreeChrome::onResponseReceived):
 858
 859 2012-05-04  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
 860
 861         [GTK] DRT needs an implementation of LayoutTestController::setDefersLoading and ::goBack
 862         https://bugs.webkit.org/show_bug.cgi?id=85134
 863
 864         Reviewed by Martin Robinson.
 865
 866         Add missing implementation setDefersLoading and goBack to GTK's
 867         LayoutTestController so that the test case rely on these methods
 868         can be enabled.
 869
 870         * DumpRenderTree/gtk/DumpRenderTree.cpp:
 871         (resetDefaultsToConsistentValues):
 872         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
 873         (LayoutTestController::goBack):
 874         (LayoutTestController::setDefersLoading):
 875
 876 2012-05-04  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
 877
 878         [Qt][NRWT] Run each DRT in it's own xvfb
 879         https://bugs.webkit.org/show_bug.cgi?id=77335
 880
 881         Reviewed by Dirk Pranke.
 882
 883         * Scripts/webkitpy/layout_tests/port/gtk.py:
 884         (GtkPort._driver_class):
 885         * Scripts/webkitpy/layout_tests/port/qt.py:
 886         (QtPort._driver_class):
 887         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Added.
 888         (XvfbDriver):
 889         (XvfbDriver._start):
 890         (XvfbDriver._start.x_filter):
 891         (XvfbDriver.stop):
 892
 893 2012-05-04  Zan Dobersek  <zandobersek@gmail.com>
 894
 895         Use destructuring assingment for the return values of the
 896         port.diff_image() method call.
 897
 898         Rubber-stamped by Dirk Pranke.
 899
 900         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
 901         (write_test_result):
 902
 903 2012-05-03  Ojan Vafai  <ojan@chromium.org>
 904
 905         Store fewer runs by default in the flakiness dashboard
 906         https://bugs.webkit.org/show_bug.cgi?id=85568
 907
 908         Reviewed by Dirk Pranke.
 909
 910         This will greatly shrink the size of the json files we load,
 911         resulting in a much faster initial load. People rarely care
 912         about more than 100 runs. For the cases they do, they can still click
 913         the show all runs checkbox to show 500 runs.
 914
 915         * TestResultServer/model/jsonresults.py:
 916
 917 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
 918
 919         [EFL] Frame load callbacks output is missing for redirections
 920         https://bugs.webkit.org/show_bug.cgi?id=85173
 921
 922         Reviewed by Antonio Gomes.
 923
 924         Catch the new ewk_frame signals to print the expected output in case
 925         of redirections, when LayoutTestController's dumpFrameLoadCallbacks()
 926         returns true.
 927
 928         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 929         (DumpRenderTreeChrome::createView):
 930         (DumpRenderTreeChrome::onWillSendRequest):
 931         (DumpRenderTreeChrome::onFrameCreated):
 932         (DumpRenderTreeChrome::onFrameRedirectForProvisionalLoad):
 933         (DumpRenderTreeChrome::onFrameRedirectRequested):
 934         (DumpRenderTreeChrome::onResponseReceived):
 935         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
 936         (DumpRenderTreeChrome):
 937
 938 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
 939
 940         [Qt] Fix up warning about missing GLib/Gio/GStreamer
 941
 942         We don't actually fall back.
 943
 944         Rubber-stamped by Csaba Osztrogonác.
 945
 946         * qmake/mkspecs/features/features.prf:
 947
 948 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
 949
 950         [Qt] Clarify warning about missing GLib/Gio/GStreamer for media support
 951
 952         And only print it once.
 953
 954         Reviewed by Csaba Osztrogonác.
 955
 956         * qmake/mkspecs/features/features.prf:
 957
 958 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
 959
 960         [Qt] Clean up and split features.prf into a static list of defaults
 961
 962         The static list of feature defaults is used as a fallback for any
 963         feature that's not dynamically detected or overriden on the command
 964         line (though build-webkit or passing DEFINES+= to qmake).
 965
 966         The static list is complete, which allows for auto-generation based
 967         on Features.py (see bug https://bugs.webkit.org/show_bug.cgi?id=85456)
 968
 969         https://bugs.webkit.org/show_bug.cgi?id=85611
 970
 971         Reviewed by Simon Hausmann.
 972
 973         * Scripts/build-webkit:
 974         * Tools.pro:
 975         * qmake/mkspecs/features/features.prf:
 976         * qmake/mkspecs/features/features.pri: Added.
 977
 978 2012-05-04  Sheriff Bot  <webkit.review.bot@gmail.com>
 979
 980         Unreviewed, rolling out r116085, r116091, and r116095.
 981         http://trac.webkit.org/changeset/116085
 982         http://trac.webkit.org/changeset/116091
 983         http://trac.webkit.org/changeset/116095
 984         https://bugs.webkit.org/show_bug.cgi?id=85628
 985
 986         We are not ready with dependencies on all platform yet (mac) +
 987         problems with debug builds. (Requested by Zoltan on #webkit).
 988
 989         * qmake/mkspecs/features/features.prf:
 990
 991 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
 992
 993         [EFL] LayoutTestController does not support overriding WebKitLoadSiteIconsKey
 994         https://bugs.webkit.org/show_bug.cgi?id=85171
 995
 996         Reviewed by Andreas Kling.
 997
 998         Implement support for overriding "WebKitLoadSiteIconsKey" setting in
 999         EFL's LayoutTestController.
1000
1001         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1002         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1003         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1004         (LayoutTestController::overridePreference):
1005
1006 2012-05-03  Alexis Menard  <alexis.menard@openbossa.org>
1007
1008         [Qt] Enable fullscreen API for WebKit2.
1009         https://bugs.webkit.org/show_bug.cgi?id=85498
1010
1011         Reviewed by Simon Hausmann.
1012
1013         Enable the fullscreen API on the MiniBrowser and turn
1014         it default on WebKit2, disable it on WK1.
1015
1016         * MiniBrowser/qt/BrowserWindow.cpp:
1017         (BrowserWindow::BrowserWindow):
1018         * MiniBrowser/qt/qml/BrowserWindow.qml:
1019         * Scripts/webkitperl/FeatureList.pm:
1020         * qmake/mkspecs/features/features.prf:
1021
1022 2012-05-04  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1023
1024         REGRESSION(r116008) old-run-webkit-tests broken
1025         https://bugs.webkit.org/show_bug.cgi?id=85597
1026
1027         Reviewed by Simon Hausmann.
1028
1029         Reinstate libraryContainsSymbol.
1030
1031         * Scripts/webkitperl/features.pm:
1032         (libraryContainsSymbol):
1033
1034 2012-05-04  Zoltan Horvath  <zoltan@webkit.org>
1035
1036         [Qt] Set WebCore imagedecoders as default and add fallback to QImageDecoder
1037         https://bugs.webkit.org/show_bug.cgi?id=80400
1038
1039         Remove WTF_USE_QT_IMAGE_DECODER macro.
1040
1041         Reviewed by Simon Hausmann.
1042
1043         * qmake/mkspecs/features/features.prf:
1044
1045 2012-05-03  Christophe Dumez  <christophe.dumez@intel.com>
1046
1047         [EFL] DRT should obey layoutTestController's addURLToRedirect()
1048         https://bugs.webkit.org/show_bug.cgi?id=82722
1049
1050         Reviewed by Maciej Stachowiak.
1051
1052         Implement support for layoutTestController's addURLToRedirect()
1053         in EFL's DRT so that it redirects properly when indicated.
1054
1055         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1056         (DumpRenderTreeChrome::onWillSendRequest):
1057
1058 2012-05-03  Christophe Dumez  <christophe.dumez@intel.com>
1059
1060         [EFL] DRT should support LayoutTestController's willSendRequestReturnsNullOnRedirect()
1061         https://bugs.webkit.org/show_bug.cgi?id=82704
1062
1063         Reviewed by Antonio Gomes.
1064
1065         EFL's DRT needs to obey LayoutTestController's
1066         willSendRequestReturnsNullOnRedirect().
1067
1068         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1069         (DumpRenderTreeChrome::onWillSendRequest):
1070
1071 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1072
1073         git-add-reviewer: Do not put the bug title and its URL in the same line in the commit message.
1074         https://bugs.webkit.org/show_bug.cgi?id=85570
1075
1076         Reviewed by Adam Roben.
1077
1078         The standard format of commit messages and ChangeLog entries has a
1079         short summary in the first line, followed by the bug URL in
1080         Bugzilla. Passing "%s" in the pretty format line to git-rev-list
1081         merges these two lines into a single one separated by a space
1082         character, which is not what we want.
1083
1084         Since "%s%n%n%b" is in practice "%B" (raw message), use it
1085         instead, as the first two lines do not end up being merged
1086         together.
1087
1088         * Scripts/git-add-reviewer:
1089         (writeCommitMessageToFile):
1090
1091 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1092
1093         Improve UI for garden-o-matic examine view when there are fewer than 5 tests.
1094         https://bugs.webkit.org/show_bug.cgi?id=85566
1095
1096         Reviewed by Adam Barth.
1097
1098         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1099         (.):
1100         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
1101         (.):
1102         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1103
1104 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1105
1106         webkitpy: Recognize FreeBSD as a valid platform.
1107         https://bugs.webkit.org/show_bug.cgi?id=81467
1108
1109         Reviewed by Adam Barth.
1110
1111         Add PlatformInfo.is_freebsd() so that trying to use webkit-patch
1112         on FreeBSD does not raise an "unrecognized platform" exception.
1113
1114         Most of this patch contains some unit tests for the mentioned
1115         method, as currently platform-specific code is needed only on
1116         Windows.
1117
1118         * Scripts/webkitpy/common/system/platforminfo.py:
1119         (PlatformInfo.__init__):
1120         (PlatformInfo.is_freebsd):
1121         (PlatformInfo._determine_os_name):
1122         * Scripts/webkitpy/common/system/platforminfo_mock.py:
1123         (MockPlatformInfo.is_freebsd):
1124         * Scripts/webkitpy/common/system/platforminfo_unittest.py:
1125         (fake_platform):
1126         (fake_platform.FakePlatformModule.release):
1127         (TestPlatformInfo.test_real_code):
1128         (TestPlatformInfo.test_os_name_and_wrappers):
1129         (TestPlatformInfo.test_os_version):
1130         (TestPlatformInfo.test_display_name):
1131         (TestPlatformInfo.test_total_bytes_memory):
1132         (TestPlatformInfo.test_free_bytes_memory):
1133         * Scripts/webkitpy/layout_tests/port/factory.py:
1134         (PortFactory._default_port):
1135
1136 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1137
1138         Make the table headers white-space:nowrap. This decreases the vertical size,
1139         which makes for more minimal embedding in garden-o-matic.
1140
1141         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
1142         (.table-header-content *):
1143
1144 2012-05-03  Eric Seidel  <eric@webkit.org>
1145
1146         Split build-webkit's feature option list off into its own module in preparation for autogeneration
1147         https://bugs.webkit.org/show_bug.cgi?id=85548
1148
1149         Reviewed by Daniel Bates.
1150
1151         My Perl-fu is very weak.  This moves the feature option list off into a separate module
1152         file so that I can easily autogenerate just that file with the fancy new generate-feature-files command.
1153         It's a bit odd to have an array which we return out to build-webkit to modify, but it turns out
1154         not to matter.  Also, build-webkit doesn't actually use any of the variables we're using to
1155         store the results from the option parse, so those end up inaccessible from build-webkit (except by reference)
1156         which is actually cleaner.  Callers can't assume that they would get a new array every time, but someone
1157         with more Perl-fu than I could create something fancier than this if we need in the future.
1158
1159         * Scripts/build-webkit:
1160         * Scripts/webkitperl/FeatureList.pm: Added.
1161         (getFeatureOptionList):
1162
1163 2012-05-03  Dirk Pranke  <dpranke@chromium.org>
1164
1165         REGRESSION: run-webkit-tests failing on Chromium Mac
1166         https://bugs.webkit.org/show_bug.cgi?id=85459
1167
1168         Unreviewed, build fix.
1169
1170         Turns out select doesn't throw IOErrors, it has it's own kind of
1171         error :(. Trap that instead.
1172
1173         * Scripts/webkitpy/layout_tests/port/server_process.py:
1174         (ServerProcess._wait_for_data_and_update_buffers_using_select):
1175
1176 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1177
1178         [CMake] Rewrite FindCairo.cmake.
1179         https://bugs.webkit.org/show_bug.cgi?id=84895
1180
1181         Reviewed by Daniel Bates.
1182
1183         The old approach relied on pkg-config for finding Cairo (which
1184         introduced a dependency on pkg-config that could be avoided), used
1185         the LibFindMacros code that we should probably remove in the
1186         future and did not use the FindPackageHandleStandardArguments
1187         module.
1188
1189         Change all that by rewriting the module.
1190         - Use the pkg-config output optionally instead of requiring it
1191         like LibFindMacros did.
1192         - Remove the implicit dependency on FreeType which often found it
1193         the wrong way via pkg-config and without considering
1194         CMAKE_PREFIX_PATH.
1195         - Retrieve the Cairo version by looking at cairo-version.h instead
1196         of relying on pkg-config. It requires some additional code for
1197         checking if the desired version has been found, but that will not
1198         be needed once we start depending on CMake 2.8.3 or later.
1199
1200         The only downside is that FPHSA sets <UPPERCASED_NAME>_FOUND
1201         instead of <Name>_FOUND, and to keep things consistent
1202         Cairo_LIBRARIES and Cairo_INCLUDE_DIRS have become CAIRO_LIBRARIES
1203         and CAIRO_INCLUDE_DIRS.
1204
1205         * DumpRenderTree/efl/CMakeLists.txt: Use CAIRO_FOO instead of
1206         Cairo_FOO.
1207         * EWebLauncher/CMakeLists.txt: Ditto.
1208
1209 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1210
1211         Fix modifier and bug sorting on the flakiness dashboard
1212         https://bugs.webkit.org/show_bug.cgi?id=85554
1213
1214         Reviewed by Dirk Pranke.
1215
1216         The sorting assumes that the column header name matches a field
1217         in the testResults object and uses that field for sorting.
1218         Filter the bugs out of the modifiers list and put them in a bugs field
1219         and only put the non-bug modifiers in the modifiers list.
1220
1221         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1222         (populateExpectationsData):
1223         (htmlForSingleTestRow):
1224         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1225
1226 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1227
1228         Fix some state resetting when running unittests. HtmlForSingleTestRow
1229         was flaky.
1230
1231         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1232         (htmlForSingleTestRow):
1233         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1234         (resetGlobals):
1235
1236 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1237
1238         webkitpy: Use PlatformInfo wherever possible in the User class.
1239         https://bugs.webkit.org/show_bug.cgi?id=85549
1240
1241         Reviewed by Dirk Pranke.
1242
1243         Instead of relying on sys.platform to detect whether the current
1244         platform is a Mac, use the PlatformInfo class instead.
1245
1246         * Scripts/webkitpy/common/system/user.py:
1247         (User.__init__):
1248         (User.edit_changelog):
1249
1250 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1251
1252         Remove the percent failed column from the flakiness dashboard
1253         https://bugs.webkit.org/show_bug.cgi?id=85552
1254
1255         Reviewed by Dirk Pranke.
1256
1257         This column isn't useful enough to justify the visual noise.
1258
1259         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1260         (processMissingAndExtraExpectations):
1261         (tableHeaders):
1262         (htmlForSingleTestRow):
1263         (sortTests):
1264         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1265         (test):
1266
1267 2012-05-03  Eric Seidel  <eric@webkit.org>
1268
1269         --coverage should not not use the "feature" options system in build-webkit, similarly remove special-casing for WTF_URL
1270         https://bugs.webkit.org/show_bug.cgi?id=85542
1271
1272         Reviewed by Daniel Bates.
1273
1274         Remove these last two odd-men-out, so that we can generate the default feature options
1275         for build-webkit using the new generate-feature-files command.
1276         The Mac build system has some special casing for ENABLE_ defines which gets them
1277         defined as -D on the build command, even though initially they start out as
1278         environment variables ENABLE_FOO = ENABLE_FOO.  The Feature options in build-webkit
1279         depend on this support, but it doesn't work for WTF_USE_ defines like WTF_USE_WTF_URL.
1280         I'm told that WTF_URL is not actively being worked on, so just removing it for now.
1281         It should probably come back to life as an ENABLE_, or build-webkit or the Xcode projects
1282         should be made generically smarter about these USE_WTF defines.
1283         (There is still WTF_USE_TILED_BACKING_STORE in the feature list, but it only seems
1284         used by Qt.  If someone tried build-webkit --tiled-backing-store on Mac today it
1285         would not work!)
1286
1287         * Scripts/build-webkit:
1288
1289 2012-05-03  Eric Seidel  <eric@webkit.org>
1290
1291         Remove the last bits of Leopard support from our perl scripts
1292         https://bugs.webkit.org/show_bug.cgi?id=85545
1293
1294         Reviewed by Daniel Bates.
1295
1296         Chromium still supports Leopard (for a little while yet), but AppleMacWebKit dropped
1297         Leopard support several months ago.  Remove the last bits of isLeopard() branching
1298         in our Perl scripts, since no one is using this (Chromium has never used this Perl code).
1299
1300         * Scripts/build-webkit:
1301         * Scripts/old-run-webkit-tests:
1302         (countAndPrintLeaks):
1303         (captureSavedCrashLog):
1304         * Scripts/webkitdirs.pm:
1305         (determineArchitecture):
1306         (argumentsForRunAndDebugMacWebKitApp):
1307
1308 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1309
1310         [Chromium] Don't show the WebKit Linux ASAN bot on the flakiness dashboard.
1311         https://bugs.webkit.org/show_bug.cgi?id=85551
1312
1313         Reviewed by Dirk Pranke.
1314
1315         It's not uploading results and in the short term we don't plan to have it start.
1316
1317         * TestResultServer/static-dashboards/builders.js:
1318         (isChromiumWebkitTipOfTreeTestRunner):
1319         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1320
1321 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1322
1323         Show the test name in the dashboard if showChrome==false and we're showing multiple tests
1324         https://bugs.webkit.org/show_bug.cgi?id=85243
1325
1326         Reviewed by Dirk Pranke.
1327
1328         This happens for cases where a test name matches multiple tests.
1329         In practice, this only happens for virtual test suites (e.g. the canvas tests).
1330
1331         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1332         (htmlForIndividualTestOnAllBuildersWithChrome):
1333         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1334         (testHtmlForIndividualTestOnAllBuildersWithChromeNonexistant):
1335         (testHtmlForIndividualTestOnAllBuildersWithChrome):
1336         (testHtmlForIndividualTestOnAllBuildersWithChromeWebkitMaster):
1337         (testHtmlForIndividualTests):
1338
1339 2012-05-03  Ojan Vafai  <ojan@chromium.org>
1340
1341         Port the flakiness dashboard unittests to qunit
1342         https://bugs.webkit.org/show_bug.cgi?id=85544
1343
1344         Reviewed by Adam Barth.
1345
1346         * TestResultServer/static-dashboards/dashboard_base.js:
1347         (appendJSONScriptElements):
1348         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1349         (resetGlobals):
1350         (runExpectationsTest):
1351         (test):
1352         * TestResultServer/static-dashboards/run-unittests.html: Added.
1353
1354 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1355
1356         webkitpy: Remove check for readline on Mac.
1357         https://bugs.webkit.org/show_bug.cgi?id=85547
1358
1359         Reviewed by Eric Seidel.
1360
1361         Python 2.5 is not supported by the WebKit project anymore, and
1362         Python's readline documentation does not mention the module not
1363         being available in the supported versions.
1364
1365         * Scripts/webkitpy/common/system/user.py:
1366
1367 2012-05-03  Thiago Marcos P. Santos  <thiago.santos@intel.com>
1368
1369         [NRWT] Run performance tests with lock
1370         https://bugs.webkit.org/show_bug.cgi?id=78628
1371
1372         Reviewed by Tony Chang.
1373
1374         Locking performance tests (like we do for http tests) will force them
1375         to run in serial. This reduces the load of the machine when running perf
1376         tests and minimizes the chances of the tests to fail, specially
1377         timeouts.
1378
1379         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1380         (Manager.__init__):
1381         (Manager._is_perf_test):
1382         (Manager._test_requires_lock):
1383         * Scripts/webkitpy/layout_tests/port/http_lock.py:
1384
1385 2012-05-03  Dirk Pranke  <dpranke@chromium.org>
1386
1387         Attempt to fix a crash on SL ...
1388
1389         Unreviewed, build fix.
1390
1391         It looks like we may be getting an IOError raised during
1392         select() on chromium mac, and that error isn't being caught.
1393         Pushing it into the try/catch block will hopefully catch this
1394         and allow us to see why the subprocess is crashing.
1395
1396         * Scripts/webkitpy/layout_tests/port/server_process.py:
1397         (ServerProcess._wait_for_data_and_update_buffers_using_select):
1398
1399 2012-05-02  Jon Honeycutt  <jhoneycutt@apple.com>
1400
1401         WebFrameLoaderClient::dispatchWillSendSubmitEvent() needs to be
1402         implemented for WebKit2
1403         https://bugs.webkit.org/show_bug.cgi?id=84304
1404
1405         Reviewed by Jessie Berlin.
1406
1407         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1408         Added new files to project.
1409
1410         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent.cpp: Added.
1411         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
1412         Look for the "DidReceiveWillSendSubmitEvent" message. The message body
1413         will be the dictionary of form control names and values passed to the
1414         injected bundle's willSendSubmitEvent handler; look for the pairs
1415         "textField" -> "text field" and "passwordField" -> "password field".
1416         Check that "hiddenField" is not included.
1417         (TestWebKitAPI::setInjectedBundleClient):
1418         (TestWebKitAPI::TEST):
1419
1420         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent_Bundle.cpp: Added.
1421         (TestWebKitAPI::willSendSubmitEvent):
1422         Post a message to the TestWebKitAPI process with the dictionary of form
1423         control names and values that we received.
1424         (TestWebKitAPI::WillSendSubmitEventTest::WillSendSubmitEventTest):
1425         (TestWebKitAPI::WillSendSubmitEventTest::didCreatePage):
1426         Set up the page's form client.
1427
1428         * TestWebKitAPI/Tests/WebKit2/auto-submitting-form.html: Added.
1429
1430 2012-05-03  Eric Seidel  <eric@webkit.org>
1431
1432         Remove dead code from build-webkit for working around old XCode dependency bug
1433         https://bugs.webkit.org/show_bug.cgi?id=85535
1434
1435         Reviewed by Adam Barth.
1436
1437         Xcode used to have a bug whereby it didn't notice if you changed the #defines
1438         passed on the command line to xcodebuild.  I wrote code to work around this
1439         while bringing up SVG support 5-6 years ago.  I think we can finally remove this
1440         code from build-webkit, since SVG is the only feature flag to ever use it
1441         (modern in-progress features like MathML don't seem to be using the support)
1442         an I suspect the XCode bug has long been fixed.
1443
1444         * Scripts/build-webkit:
1445         * Scripts/webkitperl/features.pm:
1446
1447 2012-05-03  Andy Estes  <aestes@apple.com>
1448
1449         InjectedBundleControllerMac.mm needs to include config.h
1450         https://bugs.webkit.org/show_bug.cgi?id=85531
1451
1452         Reviewed by Andreas Kling.
1453
1454         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm:
1455
1456 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1457
1458         webkitpy: Use PlatformInfo whenever possible in server_process.py.
1459         https://bugs.webkit.org/show_bug.cgi?id=85516
1460
1461         Reviewed by Eric Seidel.
1462
1463         Eliminate direct access to sys.platform wherever possible by using
1464         PlatformInfo.is_win() instead.
1465
1466         * Scripts/webkitpy/layout_tests/port/server_process.py:
1467         (ServerProcess.__init__): Remove the executive parameter as it was
1468         not passed by any caller, obtain a SystemHost from port_obj
1469         instead.
1470         (ServerProcess._log): Style, remove extra empty line.
1471         (ServerProcess._start):
1472         (ServerProcess.stop):
1473         (ServerProcess.kill):
1474         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1475         (TrivialMockPort.__init__): Create a MockSystemHost used by
1476         ServerProcess.
1477         (TestServerProcess.test_broken_pipe): Test different platforms.
1478
1479 2012-05-03  Thiago Marcos P. Santos  <thiago.santos@intel.com>
1480
1481         webkitpy: Style checker broken when committing a PNG
1482         https://bugs.webkit.org/show_bug.cgi?id=85504
1483
1484         Reviewed by Dirk Pranke.
1485
1486         * Scripts/webkitpy/style/checkers/png.py:
1487         (PNGChecker.__init__):
1488
1489 2012-05-03  Philippe Normand  <pnormand@igalia.com>
1490
1491         Unreviewed, fix another call to join(), similar to the ones landed
1492         in r115975.
1493
1494         * Scripts/webkitdirs.pm:
1495         (jhbuildConfigurationChanged):
1496
1497 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
1498
1499         webkitpy: Remove sys.platform accesses in CrashLogs.
1500         https://bugs.webkit.org/show_bug.cgi?id=85436
1501
1502         Reviewed by Dirk Pranke.
1503
1504         Instead of directly checking for sys.platform == 'darwin', use a
1505         SystemHost and call PlatformInfo.is_mac().
1506
1507         * Scripts/webkitpy/common/system/crashlogs.py:
1508         (CrashLogs.__init__):
1509         (CrashLogs.find_newest_log):
1510         (CrashLogs._log_directory_darwin):
1511         (CrashLogs._find_newest_log_darwin):
1512         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1513         (CrashLogsTest.test_find_log_darwin):
1514         * Scripts/webkitpy/layout_tests/port/mac.py:
1515         (MacPort._get_crash_log):
1516         * Scripts/webkitpy/layout_tests/port/test.py:
1517         (TestDriver.run_test):
1518         * Scripts/webkitpy/tool/commands/queries.py:
1519         (execute):
1520
1521 2012-05-03  Philippe Normand  <pnormand@igalia.com>
1522
1523         [GTK] join() is wrongly used in webkitdirs.pm since r115532
1524         https://bugs.webkit.org/show_bug.cgi?id=85501
1525
1526         Reviewed by Martin Robinson.
1527
1528         * Scripts/webkitdirs.pm:
1529         (getJhbuildPath):
1530         (buildAutotoolsProject):
1531
1532 2012-05-03  Zan Dobersek  <zandobersek@gmail.com>
1533
1534         [Gtk] ImageDiff crashes when it can't open a display
1535         https://bugs.webkit.org/show_bug.cgi?id=85476
1536
1537         Reviewed by Martin Robinson.
1538
1539         Replace gdk_init() initialization call with the g_type_init()
1540         call. GdkPixbufLoader doesn't actually need the current call to be
1541         made and this change will also remove crashes when ImageDiff
1542         is started in an environment without a DISPLAY env.
1543
1544         * DumpRenderTree/gtk/ImageDiff.cpp:
1545         (main):
1546
1547 2012-05-03  Zan Dobersek  <zandobersek@gmail.com>
1548
1549         ImageDiff should be run inside a properly established environment
1550         https://bugs.webkit.org/show_bug.cgi?id=85292
1551
1552         Reviewed by Martin Robinson.
1553
1554         When creating the ImageDiff server process, pass along a freshly-set-up environment
1555         in which the process should be executed in.
1556
1557         * Scripts/webkitpy/layout_tests/port/webkit.py:
1558         (WebKitPort._start_image_diff_process):
1559
1560 2012-05-03  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1561
1562         [Qt] Allow the web process and WTR to be paused on startup
1563
1564         Makes it easier to debug the web process or run-webkit-tests -2, as you
1565         have ample time to attach gdb to the process.
1566
1567         Reviewed by Simon Hausmann.
1568
1569         * Scripts/webkitpy/layout_tests/port/qt.py:
1570         (QtPort.setup_environ_for_server):
1571         * WebKitTestRunner/qt/main.cpp:
1572         (main):
1573
1574 2012-05-03  Csaba Osztrogonác  <ossy@webkit.org>
1575
1576         Password protect "Stop build" button
1577         https://bugs.webkit.org/show_bug.cgi?id=81982
1578
1579         Reviewed by Ryosuke Niwa.
1580
1581         Only authenticated users should be able to use stopBuild, stopAllBuilds, cancelPendingBuild
1582         features to prevent killing builds on build.webkit.org by SPAM bots.
1583
1584         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1585
1586 2012-05-03  Dongwoo Im  <dw.im@samsung.com>
1587
1588         [DRT] Remove the name of parameter, visibility, from the setPageVisibility stub implementations.
1589         https://bugs.webkit.org/show_bug.cgi?id=85468
1590
1591         Reviewed by Nikolas Zimmermann.
1592
1593         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1594         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
1595         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1596         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
1597         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1598         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
1599         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1600         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
1601         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1602         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
1603
1604 2012-05-03  Stephanie Lewis  <slewis@apple.com>
1605
1606         https://bugs.webkit.org/show_bug.cgi?id=85450 unbounded growth of JSDOMWindowShells loading pages in the same window
1607         <rdar://problem/11320059> REGRESSION (r115083): PLT3 shows linear memory growth and gets slower with each run
1608
1609         Reviewed by Brady Eidson.
1610
1611         The API added for DOMWindowExtension, didCreateGlobalObjectForFrame, would create a global object
1612         for every world, even those that did not need the callback.  This had the side effect of creating a
1613         JSDOMWindowShell that the associated world didn't necessarily know to clean up.  Instead of creating
1614         unnecessary objects change the API to globalObjectIsAvailableForFrame and do not pass the global object
1615         in the API.  The object can be accessed later by those worlds which require it.
1616
1617         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic.cpp:
1618         (TestWebKitAPI):
1619         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
1620         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp:
1621         (TestWebKitAPI):
1622         (DOMWindowExtensionBasic):
1623         (TestWebKitAPI::DOMWindowExtensionBasic::didCreatePage):
1624         (TestWebKitAPI::DOMWindowExtensionBasic::globalObjectIsAvailableForFrame):
1625         (TestWebKitAPI::globalObjectIsAvailableForFrameCallback):
1626
1627 2012-05-03  Nikolas Zimmermann  <nzimmermann@rim.com>
1628
1629         Not reviewed. Fix mac build, it was missing setPageVisibility/resetPageVisibility stub implementations.
1630
1631         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1632         (LayoutTestController::resetPageVisibility):
1633         (LayoutTestController::setPageVisibility):
1634
1635 2012-05-02  Alexander Færøy  <ahf@0x90.dk>
1636
1637         Rename deviceDPI to devicePixelRatio
1638         https://bugs.webkit.org/show_bug.cgi?id=85049
1639
1640         Reviewed by Kenneth Rohde Christiansen.
1641
1642         Use 1.5 as device pixel ratio for the Qt MiniBrowser.
1643
1644         * MiniBrowser/qt/qml/BrowserWindow.qml:
1645
1646 2012-05-03  Dongwoo Im  <dw.im@samsung.com>
1647
1648         [EFL][DRT] Implement the LayoutTestController's methods related to the Page Visibility API.
1649         https://bugs.webkit.org/show_bug.cgi?id=85347
1650
1651         Reviewed by Nikolas Zimmermann.
1652
1653         * DumpRenderTree/LayoutTestController.h:
1654         (LayoutTestController):
1655         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1656         (LayoutTestController::resetPageVisibility): Not implemented.
1657         (LayoutTestController::setPageVisibility): Not implemented.
1658         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1659         (LayoutTestController::resetPageVisibility): Reset the page visibility.
1660         (LayoutTestController::setPageVisibility): Set the page visibility.
1661         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1662         (LayoutTestController::resetPageVisibility): Not implemented.
1663         (LayoutTestController::setPageVisibility): Not implemented.
1664         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1665         (LayoutTestController::resetPageVisibility): Not implemented.
1666         (LayoutTestController::setPageVisibility): Not implemented.
1667         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1668         (LayoutTestController::resetPageVisibility): Not implemented.
1669         (LayoutTestController::setPageVisibility): Not implemented.
1670         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1671         (LayoutTestController::resetPageVisibility): Not implemented.
1672         (LayoutTestController::setPageVisibility): Not implemented.
1673
1674 2012-05-02  Sheriff Bot  <webkit.review.bot@gmail.com>
1675
1676         Unreviewed, rolling out r115918.
1677         http://trac.webkit.org/changeset/115918
1678         https://bugs.webkit.org/show_bug.cgi?id=85452
1679
1680         "Broke test-webkitpy on Apple Lion bots" (Requested by estes
1681         on #webkit).
1682
1683         * Scripts/webkitpy/common/system/crashlogs.py:
1684         (CrashLogs.__init__):
1685         (CrashLogs.find_newest_log):
1686         (CrashLogs._log_directory_darwin):
1687         (CrashLogs._find_newest_log_darwin):
1688         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1689         (CrashLogsTest.test_find_log_darwin):
1690         * Scripts/webkitpy/layout_tests/port/mac.py:
1691         (MacPort._get_crash_log):
1692         * Scripts/webkitpy/layout_tests/port/test.py:
1693         (TestDriver.run_test):
1694         * Scripts/webkitpy/tool/commands/queries.py:
1695         (execute):
1696
1697 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1698
1699         webkitpy: Remove sys.platform accesses in CrashLogs.
1700         https://bugs.webkit.org/show_bug.cgi?id=85436
1701
1702         Reviewed by Dirk Pranke.
1703
1704         Instead of directly checking for sys.platform == 'darwin', use a
1705         SystemHost and call PlatformInfo.is_mac().
1706
1707         * Scripts/webkitpy/common/system/crashlogs.py:
1708         (CrashLogs.__init__):
1709         (CrashLogs.find_newest_log):
1710         (CrashLogs._log_directory_darwin):
1711         (CrashLogs._find_newest_log_darwin):
1712         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
1713         (CrashLogsTest.test_find_log_darwin):
1714         * Scripts/webkitpy/layout_tests/port/mac.py:
1715         (MacPort._get_crash_log):
1716         * Scripts/webkitpy/layout_tests/port/test.py:
1717         (TestDriver.run_test):
1718         * Scripts/webkitpy/tool/commands/queries.py:
1719         (execute):
1720
1721 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1722
1723         webkitpy: Special-case win32 calls in file_lock and executive.
1724         https://bugs.webkit.org/show_bug.cgi?id=85431
1725
1726         Reviewed by Dirk Pranke.
1727
1728         Simplify accesses to sys.platform by only checking if sys.platform
1729         is "win32" and removing explicit checks for the else case.
1730
1731         * Scripts/webkitpy/common/system/executive.py:
1732         (Executive.check_running_pid):
1733         * Scripts/webkitpy/common/system/file_lock.py:
1734         (FileLock._create_lock):
1735         (FileLock._remove_lock):
1736
1737 2012-05-02  Ojan Vafai  <ojan@chromium.org>
1738
1739         Remove pink from flakiness dashboard
1740         https://bugs.webkit.org/show_bug.cgi?id=85422
1741
1742         Reviewed by Dirk Pranke.
1743
1744         The pink indicates that the test doesn't match it's listing in test_expectations.txt.
1745         In practice, this just added noise and confused people.
1746
1747         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1748         (htmlForSingleTestRow):
1749         (showLegend):
1750         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
1751         (#legend-contents *):
1752         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1753         (testHtmlForSingleTestRow):
1754
1755 2012-05-02  Gustavo Noronha Silva  <gns@gnome.org>
1756
1757         Unreviewed. Fix mistake I committed when applying review comments.
1758
1759         * jhbuild/jhbuild-wrapper: dependencies_path was getting 'Root'
1760         appended to its path, so jhbuild itself was being cloned and
1761         installed in the wrong place when using WEBKITOUTPUTDIR.
1762
1763 2012-05-02  Dirk Pranke  <dpranke@chromium.org>
1764
1765         nrwt: make ServerProcess work on chromium win
1766         https://bugs.webkit.org/show_bug.cgi?id=85333
1767
1768         Reviewed by Ojan Vafai.
1769
1770         This change implements a poor man's select() that will
1771         slow-spin doing non-blocking reads on the stdout and stderr
1772         named pipes connecting the worker to the driver. Seems to work
1773         and I have yet to see much overhead or ill effects but it
1774         probably needs more testing.
1775
1776         * Scripts/webkitpy/layout_tests/port/chromium.py:
1777         (ChromiumDriver.__init__):
1778         * Scripts/webkitpy/layout_tests/port/server_process.py:
1779         (ServerProcess._start):
1780         (ServerProcess._wait_for_data_and_update_buffers_using_select):
1781         (ServerProcess._wait_for_data_and_update_buffers_using_win32_apis):
1782         (ServerProcess):
1783         (ServerProcess._non_blocking_read_win32):
1784         (ServerProcess._read):
1785         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1786         (TestServerProcess.test_basic): Added.
1787
1788 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1789
1790         webkitpy: Use a SystemHost instead of PlatformInfo in the PNG checker.
1791         https://bugs.webkit.org/show_bug.cgi?id=85426
1792
1793         Reviewed by Dirk Pranke.
1794
1795         r115880 replaced the direct use of sys.platform with a
1796         PlatformInfo object. Dirk Pranke later told me SystemHost objects
1797         should be preferred, as nothing out of webkitpy.common.system
1798         should create its own PlatformInfo instances.
1799
1800         * Scripts/webkitpy/common/system/systemhost_mock.py:
1801         (MockSystemHost.__init__): Make it possible to specify a custom
1802         filesystem object.
1803         * Scripts/webkitpy/style/checkers/png.py:
1804         (PNGChecker.__init__): Accept a host object instead of a
1805         filesystem one, which is now obtained from host.
1806         (PNGChecker._config_file_path): Retrieve PlatformInfo from
1807         SystemHost.
1808         * Scripts/webkitpy/style/checkers/png_unittest.py:
1809         (PNGCheckerTest.test_init): Adjust calls to PNGChecker to its new
1810         constructor signature.
1811         (PNGCheckerTest.test_check): Ditto.
1812
1813 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1814
1815         webkitpy: Remove direct checks for sys.platform == 'darwin' in the layout test classes.
1816         https://bugs.webkit.org/show_bug.cgi?id=85424
1817
1818         Reviewed by Tony Chang.
1819
1820         We should reduce direct usage of sys.platform in webkitpy and use
1821         PlatformInfo instead.
1822
1823         * Scripts/webkitpy/layout_tests/servers/http_server.py:
1824         (Lighttpd._prepare_config): Call PlatformInfo.is_mac() instead of
1825         checking for 'darwin' directly.
1826         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
1827         (HttpServerBase.__init__): Ditto.
1828
1829 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1830
1831         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
1832         https://bugs.webkit.org/show_bug.cgi?id=85419
1833
1834         Reviewed by Dirk Pranke.
1835
1836         As pointed out by Dirk Pranke, a PlatformInfo is not supposed to
1837         be created directly, we should should a SystemHost instead.
1838
1839         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1840         (MainTest.setUp): Create a SystemHost and obtain the PlatformInfo
1841         object from it instead of creating PlatformInfo directly.
1842
1843 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1844
1845         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
1846         https://bugs.webkit.org/show_bug.cgi?id=85419
1847
1848         Rubber-stamped by Dirk Pranke.
1849
1850         Follow-up patch to r115887 which fixes some capitalization problems
1851         pointed out in comments 2 and 4.
1852
1853         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1854         (MainTest.setUp):
1855         (MainTest.test_child_processes_2):
1856         (MainTest.test_child_processes_min):
1857         (MainTest.test_exception_raised):
1858         (MainTest.test_keyboard_interrupt):
1859
1860 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1861
1862         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
1863         https://bugs.webkit.org/show_bug.cgi?id=85419
1864
1865         Reviewed by Dirk Pranke.
1866
1867         Another part of the work towards replacing direct sys.platform
1868         usage with PlatformInfo.
1869
1870         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1871         (MainTest.setUp): Move the setting of SHOULD_TEST_PROCESSES
1872         here. Even though it is costlier, unittest.setUpClass was
1873         introduced in Python 2.7.
1874         (MainTest.test_child_processes_2): Use self.SHOULD_TEST_PROCESSES.
1875         (MainTest.test_child_processes_min): Ditto.
1876         (MainTest.test_exception_raised): Ditto.
1877         (MainTest.test_keyboard_interrupt): Ditto.
1878         (MainTest.test_crash_log): Check for PlatformInfo.is_mac().
1879         (MainTest.test_web_process_crash_log): Ditto, and remove duplicate
1880         check at the end.
1881         (MainTest.test_no_http_tests): Remove check for (cygwin && Python
1882         2.5.2), as we only support Python 2.6+.
1883
1884 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1885
1886         webkitpy: Remove direct sys.platform usage in the PNG checker.
1887         https://bugs.webkit.org/show_bug.cgi?id=85407
1888
1889         Reviewed by Tony Chang.
1890
1891         Work towards replacing direct sys.platform usage with
1892         PlatformInfo.
1893
1894         * Scripts/webkitpy/common/system/platforminfo_mock.py:
1895         (MockPlatformInfo.__init__): Accept os_name and os_version as
1896         arguments.
1897         * Scripts/webkitpy/style/checkers/png.py:
1898         (PNGChecker.__init__): Create a PlatformInfo object instead of
1899         using sys.platform when needed.
1900         (PNGChecker._config_file_path): Check for the current platform via
1901         PlatformInfo.
1902         * Scripts/webkitpy/style/checkers/png_unittest.py:
1903         (PNGCheckerTest.test_check): Update tests after the changes above.
1904
1905 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
1906
1907         [watchlist] Update watchlist rules for EFL-related patches.
1908         http://bugs.webkit.org/show_bug.cgi?id=85160
1909
1910         Reviewed by Eric Seidel.
1911
1912         * Scripts/webkitpy/common/config/watchlist: Monitor
1913         Source/WTF/wtf/efl and LayoutTests/platform/efl as well.
1914
1915 2012-05-02  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1916
1917         Revert r115191 - "Make the web view's url property follow the active url"
1918
1919         It uncovered/caused issues in the icon-implementation that can't be
1920         easily fixed, so rolling out instead.
1921
1922         https://bugs.webkit.org/show_bug.cgi?id=77554
1923
1924         Rubber-stamped by Simon Hausmann.
1925
1926         * MiniBrowser/qt/qml/BrowserWindow.qml:
1927         * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp:
1928         (TestWebKitAPI::TEST):
1929
1930 2012-05-02  Zan Dobersek  <zandobersek@gmail.com>
1931
1932         Reftest mismatches are providing undefined diff percent in results.html
1933         https://bugs.webkit.org/show_bug.cgi?id=85283
1934
1935         Reviewed by Ojan Vafai.
1936
1937         When writing test results for a reftest mismatch, also take into account the
1938         difference percent that is produced when generating image diff. This value is
1939         then used when creating test results in the form of dictionaries, stored under
1940         the 'image_diff_percent' key. It ultimately ends up in JSON results data,
1941         providing additional information about the reftest difference.
1942
1943         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1944         (interpret_test_failures): Store the diff percent for reftest failure under the
1945         'image_diff_percent' key.
1946         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1947         (write_test_result): Also set the new diff_percent attribute of the reftest
1948         mismatch failure to the value of the difference percent ImageDiff outputs.
1949         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1950         (FailureReftestMismatch.__init__): Add the new diff_percent attribute with the
1951         default value of None.
1952         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1953         (EndToEndTest.test_reftest_with_two_notrefs): Update the test case expected results.
1954
1955 2012-05-02  Terry Anderson  <tdanderson@chromium.org>
1956
1957         Addition of a contributor to committers.py
1958         https://bugs.webkit.org/show_bug.cgi?id=85324
1959
1960         Reviewed by Unreviewed.
1961
1962         * Scripts/webkitpy/common/config/committers.py:
1963
1964 2012-05-02  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1965
1966         [EFL] EFL's LayoutTestController overridePreference implementation
1967         https://bugs.webkit.org/show_bug.cgi?id=83007
1968
1969         Reviewed by Antonio Gomes.
1970
1971         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1972         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1973         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1974         (toBool): aux function converting JSStringRef to bool
1975         (toInt): aux function converting JSStringRef to int
1976         (LayoutTestController::overridePreference): added implementation
1977
1978 2012-05-01  Raphael Kubo da Costa  <rakuco@webkit.org>
1979
1980         webkitpy: Fix sys.platform check in find_files_unittest.
1981         https://bugs.webkit.org/show_bug.cgi?id=85339
1982
1983         Reviewed by Ryosuke Niwa.
1984
1985         Follow-up to r101618 -- 'win' is not a valid sys.platform value,
1986         'win32' should be used instead. The check always returned false
1987         due to this, so assert_normalize() was never called.
1988
1989         * Scripts/webkitpy/common/find_files_unittest.py:
1990         (TestWinNormalize.test_win): Add the missing import for the
1991         FileSystem class and fix the sys.platform check.
1992
1993 2012-05-02  Zan Dobersek  <zandobersek@gmail.com>
1994
1995         [GTK] drag&drop related failures
1996         https://bugs.webkit.org/show_bug.cgi?id=84675
1997
1998         Reviewed by Philippe Normand.
1999
2000         Rather than checking options to see whether we're running pixel tests
2001         the pixel_tests argument passed to GtkDriver._start should be used as
2002         it also covers cases when pixel tests should be run because of testing
2003         reftests. This again creates different Xvfb processes with proper
2004         display IDs for a worker that's performing both ordinary and pixel
2005         tests, as was the case before changes made in r114788.
2006
2007         * Scripts/webkitpy/layout_tests/port/gtk.py:
2008         (GtkDriver._start):
2009
2010 2012-05-02  Simon Hausmann  <simon.hausmann@nokia.com>
2011
2012         [Qt] Enable device orientation/motion with Qt 5
2013
2014         Reviewed by Tor Arne Vestbø.
2015
2016         Enable device orientation/motion features with Qt 5 if Qt Sensors is available.
2017
2018         * qmake/mkspecs/features/features.prf:
2019
2020 2012-05-01  Raphael Kubo da Costa  <rakuco@webkit.org>
2021
2022         webkitpy: Remove unused platform imports.
2023         https://bugs.webkit.org/show_bug.cgi?id=85341
2024
2025         Reviewed by Eric Seidel.
2026
2027         Remove a few uses of `import platform' which were either rendered
2028         unneeded after r105931 or were never needed at all.
2029
2030         * Scripts/webkitpy/common/system/executive.py:
2031         * Scripts/webkitpy/tool/bot/flakytestreporter.py:
2032
2033 2012-05-01  Terry Anderson  <tdanderson@chromium.org>
2034
2035         [chromium] Accept four parameters when dispatching a WebInputEvent::GestureTap in chromium DRT eventSender
2036         https://bugs.webkit.org/show_bug.cgi?id=85289
2037
2038         Reviewed by Eric Seidel.
2039
2040         * DumpRenderTree/chromium/EventSender.cpp:
2041         (EventSender::gestureEvent):
2042             This change will allow for testing of WK85101. Currently
2043             no tests use the extra parameters deltaX and deltaY. Added
2044             a check for the size of |arguments|.
2045
2046 2012-05-01  Eric Seidel  <eric@webkit.org>
2047
2048         Add myself to the watchlist to watch all rendering changes, as well as sort the watchers in the list
2049         https://bugs.webkit.org/show_bug.cgi?id=85305
2050
2051         Reviewed by Adam Barth.
2052
2053         * Scripts/webkitpy/common/config/watchlist:
2054
2055 2012-05-01  Sheriff Bot  <webkit.review.bot@gmail.com>
2056
2057         Unreviewed, rolling out r115735.
2058         http://trac.webkit.org/changeset/115735
2059         https://bugs.webkit.org/show_bug.cgi?id=85314
2060
2061         cause two tests to crash due to an ASSERTION failure
2062         (Requested by zhenyao on #webkit).
2063
2064         * DumpRenderTree/chromium/EventSender.cpp:
2065         (EventSender::gestureEvent):
2066
2067 2012-05-01  Terry Anderson  <tdanderson@chromium.org>
2068
2069         [chromium] Accept four parameters when dispatching a WebInputEvent::GestureTap in chromium DRT eventSender
2070         https://bugs.webkit.org/show_bug.cgi?id=85289
2071
2072         Reviewed by Eric Seidel.
2073
2074         * DumpRenderTree/chromium/EventSender.cpp:
2075         (EventSender::gestureEvent):
2076             This change will allow for testing of WK85101. Currently no tests use the extra
2077             parameters deltaX and deltaY.
2078
2079 2012-05-01  Zan Dobersek  <zandobersek@gmail.com>
2080
2081         Print out stderr output of ImageDiff if it is present
2082         https://bugs.webkit.org/show_bug.cgi?id=85285
2083
2084         Reviewed by Philippe Normand.
2085
2086         Log the stderr output of ImageDiff as a warning if present. This is
2087         especially helpful when ImageDiff crashes and stderr data might contain
2088         relevant information about what caused the crash.
2089
2090         * Scripts/webkitpy/layout_tests/port/webkit.py:
2091         (WebKitPort._read_image_diff):
2092
2093 2012-05-01  Kenneth Rohde Christiansen  <kenneth@webkit.org>
2094
2095         [Qt] Add an experimental extension to set the min. contents width
2096         https://bugs.webkit.org/show_bug.cgi?id=85281
2097
2098         Reviewed by Antonio Gomes.
2099
2100         * MiniBrowser/qt/qml/BrowserWindow.qml:
2101
2102         Set the value of preferredMinimumContentsWidth to 980.
2103
2104 2012-04-30  Ojan Vafai  <ojan@chromium.org>
2105
2106         Fix tests to not depend on the actual list of builders at build.chromium.org.
2107
2108         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2109         (testHtmlForIndividualTestOnAllBuildersWithChrome):
2110         (testHtmlForIndividualTestOnAllBuildersWithChromeWebkitMaster):
2111
2112 2012-04-29  Sam Weinig  <sam@webkit.org>
2113
2114         Remove BlobBuilder
2115         https://bugs.webkit.org/show_bug.cgi?id=84036
2116
2117         Reviewed by Anders Carlsson.
2118
2119         * Scripts/build-webkit:
2120         * qmake/mkspecs/features/features.prf:
2121         Make exposing the WebKitBlobBuilder JS object conditional on a new
2122         ENABLE_LEGACY_WEBKIT_BLOB_BUILDER flag. Don't enable this for the Mac,
2123         but do for everyone else.
2124
2125 2012-04-30  Amruth Raj Padmanabhuni  <amruthraj@motorola.com>
2126
2127         [GTK] Add glib as a dependency to build glib-networking
2128         https://bugs.webkit.org/show_bug.cgi?id=85181
2129
2130         Reviewed by Martin Robinson.
2131
2132         * gtk/jhbuild.modules:
2133
2134 2012-04-30  Alexis Menard  <alexis.menard@openbossa.org>
2135
2136         Move myself to reviewers.
2137
2138         Reviewed by Andreas Kling.
2139
2140         * Scripts/webkitpy/common/config/committers.py:
2141
2142 2012-04-30  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
2143
2144         Unreviewed, move myself to reviewers
2145
2146         * Scripts/webkitpy/common/config/committers.py:
2147
2148 2012-04-30  Carlos Garcia Campos  <cgarcia@igalia.com>
2149
2150         Unreviewed. Fix make distcheck.
2151
2152         * TestWebKitAPI/GNUmakefile.am: Add missing headers to compilation
2153         and resources to EXTRA_DIST.
2154
2155 2012-04-30  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2156
2157         [GTK] DRT support for layoutTestController.removeOriginAccessWhitelistEntry
2158         https://bugs.webkit.org/show_bug.cgi?id=85156
2159
2160         Reviewed by Martin Robinson.
2161
2162         Add missing implementation removeOriginAccessWhitelistEntry to
2163         GTK's LayoutTestController.
2164
2165         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2166         (LayoutTestController::removeOriginAccessWhitelistEntry):
2167
2168 2012-04-29  Maciej Stachowiak  <mjs@apple.com>
2169
2170         REGRESSION: On Lion, run-webkit-tests changes the display color profile even when not running pixel tests
2171         https://bugs.webkit.org/show_bug.cgi?id=81729
2172
2173         Reviewed by Dan Bernstein.
2174
2175         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2176         (Manager._set_up_run): Only launch LayoutTestHelper when running pixel tests.
2177
2178 2012-04-29  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2179
2180         [GTK] DRT needs an implementation of layoutTestController.setDomainRelaxationForbiddenForURLScheme
2181         https://bugs.webkit.org/show_bug.cgi?id=85131
2182
2183         Reviewed by Martin Robinson.
2184
2185         Add missing implementation setDomainRelaxationForbiddenForURLScheme to
2186         GTK's LayoutTestController.
2187
2188         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2189         (LayoutTestController::setDomainRelaxationForbiddenForURLScheme):
2190
2191 2012-04-28  Sam Weinig  <sam@webkit.org>
2192
2193         Smooth scrolling needs a new key
2194         <rdar://problem/11331632>
2195
2196         Reviewed by Dan Bernstein.
2197
2198         * DumpRenderTree/mac/DumpRenderTree.mm:
2199         (resetDefaultsToConsistentValues):
2200         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm:
2201         (TestWebKitAPI::InjectedBundleController::platformInitialize):
2202         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
2203         (WTR::InjectedBundle::platformInitialize):
2204         Update for new key.
2205
2206 2012-04-28  Zan Dobersek  <zandobersek@gmail.com>
2207
2208         [WK2] http/tests/navigation/anchor-frames-gbk.html fails
2209         https://bugs.webkit.org/show_bug.cgi?id=76896
2210
2211         Reviewed by Darin Adler.
2212
2213         Only dump as text if currently dumping render tree. This ensures that
2214         calling dumpAsText in a test with dumpChildFramesAsText already being
2215         called doesn't override the first decision, possibly making tests fail.
2216
2217         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2218         (WTR::LayoutTestController::dumpAsText):
2219
2220 2012-04-28  Dan Winship  <danw@gnome.org>
2221
2222         Unreviewed; Add me to the SoupNetwork watchlist
2223
2224         * Scripts/webkitpy/common/config/watchlist:
2225
2226 2012-04-27  Gustavo Noronha Silva  <gns@gnome.org>
2227
2228         [GTK] jhbuild cloning is not following WEBKITOUTPUTDIR.
2229         https://bugs.webkit.org/show_bug.cgi?id=76161
2230
2231         Reviewed by Martin Robinson.
2232
2233         * Scripts/webkitdirs.pm:
2234         (getJhbuildPath): New method to obtain the jhbuild base directory,
2235         using the product base directory
2236         (jhbuildConfigurationChanged): Use the new method
2237         (buildAutotoolsProject): Ditto.
2238         * efl/jhbuildrc: Use WEBKITOUTPUTDIR when calculating the path.
2239         * gtk/jhbuildrc: Ditto.
2240         * jhbuild/jhbuild-wrapper: Ditto.
2241
2242 2012-04-27  Scott Graham  <scottmg@chromium.org>
2243
2244         [chromium] make case of npTestNetscapePlugIn.dll match case in copy rule
2245         https://bugs.webkit.org/show_bug.cgi?id=85083
2246
2247         Reviewed by Dirk Pranke.
2248
2249         Make output product_name for npTestNetscapePlugIn.dll match the case
2250         of the copy_TestNetscapePlugIn rule. This is required for ninja, which
2251         is more particular about case matching in rules than the VS build.
2252
2253         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2254
2255 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
2256
2257         [chromium] use "drt-style" output, not "test-shell-style" output, on mac and linux DRT
2258         https://bugs.webkit.org/show_bug.cgi?id=84917
2259
2260         Unreviewed, build fix.
2261
2262         Reland the change in r115453 with a fix for chromium win.
2263         The logic in ChromiumDriver was busted on windows, causing DRT
2264         to get launched without the --test-shell flag.
2265
2266         * Scripts/webkitpy/layout_tests/port/chromium.py:
2267         (ChromiumDriver):
2268         (ChromiumDriver.__init__):
2269         (ChromiumDriver._wrapper_options):
2270         (ChromiumDriver.cmd_line):
2271         (ChromiumDriver._start):
2272         (ChromiumDriver.has_crashed):
2273         (ChromiumDriver.run_test):
2274         (ChromiumDriver.stop):
2275         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2276         (ChromiumDriverTest.setUp):
2277         (ChromiumDriverTest.test_stop):
2278         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
2279         (ChromiumDriverTest.test_two_drivers):
2280
2281 2012-04-27  Peter Beverloo  <peter@chromium.org>
2282
2283         Add John Grabowski as a non-committer to committers.py
2284         https://bugs.webkit.org/show_bug.cgi?id=85080
2285
2286         Reviewed by Dirk Pranke.
2287
2288         He's being CC'ed often enough because of his involvement in the Chromium
2289         side of Chrome for Android upstreaming.
2290
2291         * Scripts/webkitpy/common/config/committers.py:
2292           add jrg@chromium.org
2293
2294 2012-04-27  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2295
2296         [Qt] Fix minimal build.
2297         https://bugs.webkit.org/show_bug.cgi?id=85045
2298
2299         Reviewed by Tor Arne Vestbø.
2300
2301         Do not set conflicting values of ENABLE_XSLT.
2302
2303         * qmake/mkspecs/features/features.prf:
2304
2305 2012-04-27  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2306
2307         [EFL][DRT] EFL should sanitize pulseaudio module configuration too
2308         https://bugs.webkit.org/show_bug.cgi?id=85047
2309
2310         Reviewed by Dirk Pranke.
2311
2312         PulseAudio test setup sanitization shared between EFL and GTK.
2313
2314         * Scripts/webkitpy/layout_tests/port/efl.py:
2315         (EflPort):
2316         (EflPort.setup_test_run):
2317         (EflPort.clean_up_test_run):
2318         * Scripts/webkitpy/layout_tests/port/gtk.py:
2319         (GtkPort): Moving pulseaudio unload/restore out to shared class.
2320         * Scripts/webkitpy/layout_tests/port/pulseaudio_sanitizer.py: Added.
2321         (PulseAudioSanitizer):
2322         (PulseAudioSanitizer._unload_pulseaudio_module):
2323         (PulseAudioSanitizer._restore_pulseaudio_module):
2324
2325 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
2326
2327         Unreviewed, revert r115453 - broke chromium win bots
2328
2329         * Scripts/webkitpy/layout_tests/port/chromium.py:
2330         (ChromiumPort._path_to_image_diff):
2331         (ChromiumDriver):
2332         (ChromiumDriver.__init__):
2333         (ChromiumDriver._wrapper_options):
2334         (ChromiumDriver.cmd_line):
2335         (ChromiumDriver.stop):
2336         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2337         (ChromiumDriverTest.setUp):
2338         (ChromiumDriverTest.test_stop):
2339         (ChromiumDriverTest.test_two_drivers):
2340         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
2341
2342 2012-04-27  Ryosuke Niwa  <rniwa@webkit.org>
2343
2344         Chromium Windows Perf bots timeout due to not output
2345         https://bugs.webkit.org/show_bug.cgi?id=84940
2346
2347         Reviewed by Dirk Pranke.
2348
2349         Dirk and I investigated the issue on the bot but we couldn't figure out what is going wrong.
2350         Since we run-perf-tests don't need any of fancy feature printer provides, just use python's
2351         built-in logging module instead. Printing out to stdout and stderr seem to work so hopefully
2352         this will fix the issue on the bot.
2353
2354         * Scripts/run-perf-tests:
2355         * Scripts/webkitpy/performance_tests/perftest.py:
2356         (PerfTest.run):
2357         (PerfTest.run_failed):
2358         (PerfTest.parse_output):
2359         (PerfTest.output_statistics):
2360         (ChromiumStylePerfTest.parse_output):
2361         (PageLoadingPerfTest.run):
2362         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2363         (MainTest.test_parse_output):
2364         (MainTest.test_parse_output_with_failing_line):
2365         (TestPageLoadingPerfTest):
2366         (TestPageLoadingPerfTest.test_run):
2367         (TestPageLoadingPerfTest.test_run_with_bad_output):
2368         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2369         (PerfTestsRunner.__init__):
2370         (PerfTestsRunner._parse_args):
2371         (PerfTestsRunner.run):
2372         (PerfTestsRunner._upload_json):
2373         (PerfTestsRunner._print_status):
2374         (PerfTestsRunner._run_tests_set):
2375         (PerfTestsRunner._run_single_test):
2376         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2377         (create_runner):
2378         (test_run_test_set):
2379         (test_run_test_set_kills_drt_per_run):
2380         (test_run_test_pause_before_testing):
2381         (test_run_test_set_for_parser_tests):
2382         (test_run_test_set_with_json_output):
2383         (test_run_test_set_with_json_source):
2384         (test_run_test_set_with_multiple_repositories):
2385         (test_upload_json):
2386         (test_parse_args):
2387
2388 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
2389
2390         [chromium] use "drt-style" output, not "test-shell-style" output, on mac and linux DRT
2391         https://bugs.webkit.org/show_bug.cgi?id=84917
2392
2393         Reviewed by Ojan Vafai.
2394
2395         This change changes the chromium new-run-webkit-tests
2396         implementation to use the code in WebKitDriver and ServerProcess
2397         by default on Mac and Linux, instead of using the old-style
2398         Chromium/test_shell style of output.
2399
2400         We will still use the test_shell style on Windows for now
2401         and you can still get the old style of behavior with
2402         new-run-webkit-tests --additional-drt-flag --test-shell
2403
2404         * Scripts/webkitpy/layout_tests/port/chromium.py:
2405         (ChromiumDriver):
2406         (ChromiumDriver.__init__):
2407         (ChromiumDriver._wrapper_options):
2408         (ChromiumDriver.cmd_line):
2409         (ChromiumDriver._start):
2410         (ChromiumDriver.has_crashed):
2411         (ChromiumDriver.run_test):
2412         (ChromiumDriver.stop):
2413         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2414         (ChromiumDriverTest.setUp):
2415         (ChromiumDriverTest.test_stop):
2416         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
2417         (ChromiumDriverTest.test_two_drivers):
2418
2419 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
2420
2421         webkitpy: fix race in unit tests for manager_worker_broker
2422         https://bugs.webkit.org/show_bug.cgi?id=84925
2423
2424         Reviewed by Ojan Vafai.
2425
2426         I have been intermittently seeing hangs when running
2427         test-webkitpy on Mac SL; it appears that there's a bug (?) in
2428         the multiprocessing module where queue references are getting
2429         leaked or not cleaned up somehow. By explicitly closing the
2430         queues, things seem to be happy, so this patch adds a cleanup()
2431         method to the message broker and modifies the manager (and the
2432         unit tests) to call it. This may get rid of the intermittent
2433         error on shutdown on Linux I've seen as well.
2434
2435         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2436         (Manager._run_tests):
2437         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
2438         (_Broker.__del__):
2439         (_Broker):
2440         (_Broker.cleanup):
2441         (_BrokerConnection.cleanup):
2442         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
2443         (_TestsMixin.test_name):
2444         (_TestsMixin.test_cancel):
2445         (_TestsMixin.test_done):
2446         (_TestsMixin.test_unknown_message):
2447
2448 2012-04-27  Gavin Peters  <gavinp@chromium.org>
2449
2450         Add new ENABLE_LINK_PRERENDER define to control the Prerendering API
2451         https://bugs.webkit.org/show_bug.cgi?id=84871
2452
2453         Reviewed by Adam Barth.
2454
2455         Prerendering is currently covered by the ENABLE_LINK_PREFETCH macro, but the new Prerendering
2456         API separates it from prefetching.  Having separate include guards lets ports enable prefetching,
2457         a relatively easy change, without needing to build the infrastructure for prerendering, which
2458         is considerably more complicated.
2459
2460 2012-04-27  Philippe Normand  <pnormand@igalia.com>
2461
2462         Unreviewed, GTK gardening.
2463
2464         * Scripts/run-gtk-tests:
2465         (TestRunner): Skip flaky WebKit2 API test.
2466
2467 2012-04-27  Philippe Normand  <pnormand@igalia.com>
2468
2469         Unreviewed, GTK gardening.
2470
2471         * Scripts/run-gtk-tests:
2472         (TestRunner): Fix another typo. So easy to get confused by a Test
2473         name and the actual name of the executable...
2474
2475 2012-04-27  Philippe Normand  <pnormand@igalia.com>
2476
2477         Unreviewed, GTK gardening.
2478
2479         * Scripts/run-gtk-tests:
2480         (TestRunner): Skip failing WebKit2 API test.
2481
2482 2012-04-27  Christophe Dumez  <christophe.dumez@intel.com>
2483
2484         [EFL] DRT should support LayoutTestController's dumpResourceLoadCallbacks()
2485         https://bugs.webkit.org/show_bug.cgi?id=81891
2486
2487         Reviewed by Philippe Normand.
2488
2489         Implement support for LayoutTestController's
2490         dumpResourceLoadCallbacks() in EFL's DumpRenderTree by catching new
2491         ewk_view signals about resource loading and print out the expected
2492         information. This allows for several test cases to be removed from the
2493         skip list.
2494
2495         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2496         (DumpRenderTreeChrome::createView):
2497         (descriptionSuitableForTestResult):
2498         (DumpRenderTreeChrome::onResponseReceived):
2499         (DumpRenderTreeChrome::onResourceLoadFinished):
2500         (DumpRenderTreeChrome::onResourceLoadFailed):
2501         (DumpRenderTreeChrome::onNewResourceRequest):
2502         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
2503         (DumpRenderTreeChrome):
2504
2505 2012-04-27  Philippe Normand  <pnormand@igalia.com>
2506
2507         Unreviewed, GTK typo fix.
2508
2509         * Scripts/run-gtk-tests:
2510         (TestRunner): Fix-up skipped WebKit2 API test name TestWKConnection.
2511
2512 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2513
2514         [SOUP] Add a way to register custom uri schemes in WebKit2
2515         https://bugs.webkit.org/show_bug.cgi?id=84130
2516
2517         Reviewed by Martin Robinson.
2518
2519         * Scripts/webkitpy/style/checkers/cpp.py:
2520         (check_identifier_name_in_declaration): Add webkit_soup prefix to
2521         the list of exceptions to the underscores in identifiers rule.
2522
2523 2012-04-26  Ryosuke Niwa  <rniwa@webkit.org>
2524
2525         Extract PerfTestFactory
2526         https://bugs.webkit.org/show_bug.cgi?id=83944
2527
2528         Reviewed by Dirk Pranke.
2529
2530         Extracted PerfTestFactory out of PerfTest. This class creates appropriate PerfTest object given
2531         a test name and a path. Also removed dirname from the constructors of PerfTest and its subclasses.
2532
2533         Also added simple test cases to sanity check PerfTestFactory.
2534
2535         * Scripts/webkitpy/performance_tests/perftest.py:
2536         (PerfTest.__init__):
2537         (PerfTest.test_name):
2538         (ChromiumStylePerfTest.__init__):
2539         (PageLoadingPerfTest.__init__):
2540         (PageLoadingPerfTest.run):
2541         (PerfTestFactory):
2542         (PerfTestFactory.create_perf_test):
2543         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
2544         (MainTest.test_parse_output):
2545         (MainTest.test_parse_output_with_failing_line):
2546         (TestPageLoadingPerfTest.test_run):
2547         (TestPageLoadingPerfTest.test_run_with_bad_output):
2548         (TestPerfTestFactory):
2549         (TestPerfTestFactory.test_regular_test):
2550         (TestPerfTestFactory.test_inspector_test):
2551         (TestPerfTestFactory.test_page_loading_test):
2552         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2553         (PerfTestsRunner):
2554         (PerfTestsRunner._collect_tests):
2555         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2556         (run_test):
2557         (_tests_for_runner):
2558
2559 2012-04-26  Dirk Pranke  <dpranke@chromium.org>
2560
2561         nrwt: clean up server process, webkit driver so chromium can use it
2562         https://bugs.webkit.org/show_bug.cgi?id=84910
2563
2564         Reviewed by Ojan Vafai.
2565
2566         This change moves the "sample a process" logic out of
2567         server_process.py and into a port-specific class (where really
2568         only the mac has an implementation), and also preemptively kills
2569         DRT when a test times out in WebKitDriver (rather than waiting through the
2570         additional delays caused by calling stop() when we would want to
2571         restart the driver generically in worker.py).
2572
2573         These changes will make it possible for the chromium port to
2574         switch over to the stock WebKitDriver implementation, at least
2575         on mac and linux.
2576
2577         * Scripts/webkitpy/layout_tests/port/base.py:
2578         (Port.sample_process):
2579         * Scripts/webkitpy/layout_tests/port/mac.py:
2580         (MacPort.sample_process):
2581         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2582         (test_helper_fails_to_stop):
2583         (test_sample_process):
2584         (test_sample_process.logging_run_command):
2585         (test_sample_process_throws_exception):
2586         (test_sample_process_throws_exception.throwing_run_command):
2587         * Scripts/webkitpy/layout_tests/port/server_process.py:
2588         (ServerProcess._log):
2589         (ServerProcess._handle_timeout):
2590         (ServerProcess.stop):
2591         (ServerProcess):
2592         (ServerProcess.kill): Here we add a method to immediately stop
2593         the process rather than trying to shut it down cleanly.
2594         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2595         (TestServerProcess.test_broken_pipe):
2596         * Scripts/webkitpy/layout_tests/port/webkit.py:
2597         (WebKitDriver.run_test): Fix an issue where we weren't passing
2598         along any per-test args (only needed for Chromium, but still).
2599         Also, kill the driver immediately if we time out a test.
2600
2601 2012-04-26  Benjamin Poulain  <bpoulain@apple.com>
2602
2603         ObjcClass::methodsNamed() can leak if buffer is dynamically allocated
2604         https://bugs.webkit.org/show_bug.cgi?id=84668
2605
2606         Reviewed by Alexey Proskuryakov.
2607
2608         Extend ObjCPlugin to support the new layout tests of the Objective-C bridge.
2609
2610         * DumpRenderTree/mac/ObjCPlugin.m:
2611         (+[ObjCPlugin isSelectorExcludedFromWebScript:]):
2612         (+[ObjCPlugin webScriptNameForSelector:]):
2613         (-[ObjCPlugin methodMappedToLongName]):
2614         (-[ObjCPlugin testConversionColon:]):
2615         (-[ObjCPlugin _:]):
2616
2617 2012-04-26  Dimitri Glazkov  <dglazkov@chromium.org>
2618
2619         Unreviewed, rolling out r115340.
2620         http://trac.webkit.org/changeset/115340
2621         https://bugs.webkit.org/show_bug.cgi?id=83048
2622
2623         Does not work with Windows.
2624
2625         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2626         (SingleTestRunner._run_reftest):
2627         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2628         (copy_file):
2629         * Scripts/webkitpy/layout_tests/port/base.py:
2630         (Port._parse_reftest_list):
2631         (Port.reference_files):
2632         (Port._real_tests):
2633         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2634         (PortTest.test_reference_files):
2635         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2636         (ChromiumDriverTest.test_write_command_and_read_line):
2637
2638 2012-04-26  Christophe Dumez  <christophe.dumez@intel.com>
2639
2640         [EFL] Enable VIDEO_TRACK feature
2641         https://bugs.webkit.org/show_bug.cgi?id=84830
2642
2643         Reviewed by Gustavo Noronha Silva.
2644
2645         Enable VIDEO_TRACK by default on EFL port whenever build-webkit is
2646         used.
2647
2648         * Scripts/build-webkit:
2649
2650 2012-04-08  Robert Hogan  <robert@webkit.org>
2651
2652         new-webkit-run-tests: handle ref tests from the CSS test suite
2653         https://bugs.webkit.org/show_bug.cgi?id=83048
2654
2655         Reviewed by Ryosuke Niwa.
2656
2657         This allows new-run-webkit-tests to run reference tests created for the CSS test suite. It removes a big swathe
2658         of pixel and text results that are no longer required now that the reference results are used and adds in the support
2659         files from the CSS test suite that allow the ref tests to run.
2660
2661         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2662         (SingleTestRunner._run_reftest): Assert that the reference result exists. Previously the test would
2663          run without it and report a pass!
2664         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2665         (copy_file): ensure the output directory exists, it may not if the reference test is
2666           contained in a folder that contains nothing but reference tests.
2667         * Scripts/webkitpy/layout_tests/port/base.py:
2668         (Port.available_reference_files): Add as a helper. Handle embedded relative paths, e.g: some/../path
2669         (Port.reference_files): Only return reference results in the reftest list that actually exist on
2670           disk. Also, don't ignore potential reference results on disk that are not in the reftest list, and
2671           don't ignore them just because we have a reftest list for the test's directory.
2672         (Port._real_tests): ignore anything that looks like a test in a 'support' directory, this semantic
2673           is used by the CSS 2.1 test suite.
2674         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2675         (PortTest.test_reference_files_when_reftestlist_and_reference_file_present): Ensure that a test with a reference result
2676           on disk but no reference result listed in the directory's reftest list picks up its reference result.
2677         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2678         (ChromiumDriverTest.test_strip_uri): Test that some/test.html and some/other/../test.html are treated
2679          as the same file.
2680
2681 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2682
2683         Unreviewed. Fix GTK+ build after r115314.
2684
2685         For some reason it fails while bulding GtkVersioning for WTR, but
2686         it didn't fail neither in EWS nor in my laptop. We don't really
2687         need GtkVersioning in WTR since it already depends on GTK+3
2688         unconditionally.
2689
2690         * WebKitTestRunner/GNUmakefile.am: Remove GtkVersioning.c from
2691         compilation.
2692         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
2693         (WTR::EventSenderProxy::createMouseButtonEvent): Use GTK+ 3 API
2694         directly instead of getDefaultGDKPointerDevice() helper.
2695         (WTR::EventSenderProxy::keyDown): Ditto.
2696         (WTR::EventSenderProxy::mouseMoveTo): Ditto.
2697
2698 2012-04-24  Ojan Vafai  <ojan@chromium.org>
2699
2700         Show flakiness dashboard data in garden-o-matic
2701         https://bugs.webkit.org/show_bug.cgi?id=83716
2702
2703         Reviewed by Dimitri Glazkov.
2704
2705         Put an iframe below the expected/actual results in the Results view.
2706         Size the iframe to it's height.
2707
2708         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2709         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2710         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2711         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2712         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
2713         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2714         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2715
2716 2012-04-26  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2717
2718         [EFL] [DRT] LayoutTestController needs implementation of setJavaScriptProfilingEnabled
2719         https://bugs.webkit.org/show_bug.cgi?id=84576
2720
2721         Reviewed by Martin Robinson.
2722
2723         Add missing implementation setJavaScriptProfilingEnabled to EFL's
2724         LayoutTestController in order to unskip tests rely on this setting.
2725
2726         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2727         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
2728         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2729         (LayoutTestController::setJavaScriptProfilingEnabled):
2730
2731 2012-04-26  Xianzhu Wang  <wangxianzhu@chromium.org>
2732
2733         Add Jing Zhao, Min Qin, Tien-Ren Chen and Hao Zheng into contributors list.
2734
2735         Not reviewed.
2736
2737         * Scripts/webkitpy/common/config/committers.py:
2738
2739 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2740
2741         [GTK] Test TestWebKitAPI/WebKit2/TestNewFirstVisuallyNonEmptyLayout times out
2742         https://bugs.webkit.org/show_bug.cgi?id=84957
2743
2744         Reviewed by Martin Robinson.
2745
2746         * Scripts/run-gtk-tests:
2747         (TestRunner): Unskip TestNewFirstVisuallyNonEmptyLayout and
2748         TestNewFirstVisuallyNonEmptyLayoutForImages.
2749         * TestWebKitAPI/gtk/PlatformWebViewGtk.cpp:
2750         (TestWebKitAPI::PlatformWebView::PlatformWebView): Show the
2751         toplevel window containing the web view to make sure events are
2752         received.
2753
2754 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2755
2756         Unreviewed. Add bug numbers to skipped tests.
2757
2758         * Scripts/run-gtk-tests:
2759         (TestRunner):
2760
2761 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2762
2763         [GTK] Build and run TestWebKitAPI WebKit2 unit tests
2764         https://bugs.webkit.org/show_bug.cgi?id=84446
2765
2766         Reviewed by Philippe Normand.
2767
2768         * Scripts/run-gtk-tests:
2769         (TestRunner): Add TestWebKitAPI/WebKit2 to the list of directories
2770         containing tests. Unskip FunctionalTest.RefCountedStorage, since
2771         we are now using a custom main that initializes WTF threading.
2772         (TestRunner._setup_testing_environment): Set environment variables
2773         required to run WebKit2 tests.
2774         * TestWebKitAPI/GNUmakefile.am:
2775         * TestWebKitAPI/JavaScriptTest.cpp: Use
2776         JavaScriptCore/JSContextRef.h instead of
2777         JavaScriptCore/JavaScriptCore.h since it includes JSStringRefCF.h
2778         unconditionally.
2779         * TestWebKitAPI/PlatformWebView.h: Define PlatformWKView and
2780         PlatformWindow for GTK.
2781         * TestWebKitAPI/Tests/WebKit2/EvaluateJavaScript.cpp: Remove
2782         unneded header that breaks the build of GTK port.
2783         * TestWebKitAPI/Tests/WebKit2/PreventEmptyUserAgent.cpp: Use
2784         JavaScriptCore/JSContextRef.h instead of
2785         JavaScriptCore/JavaScriptCore.h since it includes JSStringRefCF.h
2786         unconditionally.
2787         * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp:
2788         (TestWebKitAPI::TEST): Add expected values for GTK+ port.
2789         * TestWebKitAPI/gtk/InjectedBundleControllerGtk.cpp:
2790         (TestWebKitAPI):
2791         (TestWebKitAPI::InjectedBundleController::platformInitialize):
2792         * TestWebKitAPI/gtk/PlatformUtilitiesGtk.cpp: Added.
2793         (TestWebKitAPI::Util::checkTestFinished):
2794         (TestWebKitAPI::Util::run):
2795         (TestWebKitAPI::Util::sleep):
2796         (TestWebKitAPI::Util::getFilenameFromEnvironmentVariableAsUTF8):
2797         (TestWebKitAPI::Util::createInjectedBundlePath):
2798         (TestWebKitAPI::Util::createURLForResource):
2799         (TestWebKitAPI::Util::URLForNonExistentResource):
2800         (TestWebKitAPI::Util::isKeyDown):
2801         * TestWebKitAPI/gtk/PlatformWebViewGtk.cpp: Added.
2802         (TestWebKitAPI::PlatformWebView::PlatformWebView):
2803         (TestWebKitAPI::PlatformWebView::~PlatformWebView):
2804         (TestWebKitAPI::PlatformWebView::page):
2805         (TestWebKitAPI::PlatformWebView::resizeTo):
2806         (TestWebKitAPI::doKeyStroke):
2807         (TestWebKitAPI::PlatformWebView::simulateSpacebarKeyPress):
2808         (TestWebKitAPI::PlatformWebView::simulateAltKeyPress):
2809         (TestWebKitAPI::doMouseButtonEvent):
2810         (TestWebKitAPI::PlatformWebView::simulateRightClick):
2811         (TestWebKitAPI::PlatformWebView::simulateMouseMove):
2812         * TestWebKitAPI/gtk/main.cpp:
2813         (main):
2814
2815 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2816
2817         [GTK] run-gtk-tests: Use a timeout per test instead of a global timeout
2818         https://bugs.webkit.org/show_bug.cgi?id=84695
2819
2820         Reviewed by Philippe Normand.
2821
2822         It also adds a command line option to be able to pass a custom
2823         timeout value, instead of hard-coding it. The default timeout is 10
2824         seconds if no other value is passed to the script.
2825
2826         * Scripts/run-gtk-tests:
2827         (TestTimeout): Exception raised when a test times out.
2828         (TestRunner._get_child_pid_from_test_output): Helper function to
2829         get the pid of the running test from gtester output.
2830         (TestRunner._kill_process): Helper funtion ot kill a process
2831         ignoring exceptions if the process is already died.
2832         (TestRunner._run_test_command): Run the test command raising
2833         TestTimeout exception if the test doesn't finish before the given
2834         timeout in seconds.
2835         (TestRunner._run_test_command.alarm_handler): Alarm handler that
2836         raises TestTimeout exception.
2837         (TestRunner._run_test_glib): Use _run_test_command() to run the test.
2838         (TestRunner._run_test_google): Ditto.
2839         (TestRunner.run_tests): Add current test to the list of timed out
2840         test if TestTimeout exception is raised. Show the list of timed
2841         out test at the end.
2842
2843 2012-04-26  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2844
2845         [EFL] [DRT] LayoutTestController needs implementation of addOriginAccessWhitelistEntry and removeOriginAccessWhitelistEntry
2846         https://bugs.webkit.org/show_bug.cgi?id=83880
2847
2848         Reviewed by Antonio Gomes.
2849
2850         Add missing implementation addOriginAccessWhitelistEntry and
2851         removeOriginAccessWhitelistEntry to EFL's LayoutTestController
2852         in order to unskip tests which rely on this.
2853
2854         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2855         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
2856         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2857         (LayoutTestController::addOriginAccessWhitelistEntry):
2858         (LayoutTestController::removeOriginAccessWhitelistEntry):
2859
2860 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2861
2862         [GTK] Add downloads support to MiniBrowser
2863         https://bugs.webkit.org/show_bug.cgi?id=73662
2864
2865         Reviewed by Philippe Normand.
2866
2867         * MiniBrowser/gtk/BrowserDownloadsBar.c: Added.
2868         (browserDownloadsBarChildRemoved): Close the downloads bar if
2869         there aren't active downloads.
2870         (browserDownloadsBarResponse): Destroy the downloads bar when
2871         closed.
2872         (browser_downloads_bar_init):
2873         (browser_downloads_bar_class_init):
2874         (browser_downloads_bar_new): Create a new downloads bar.
2875         (actionButtonClicked): Cancel the download if cancel button was
2876         clicked, or open the downloaded file if the download finished.
2877         (browserDownloadFinalize):
2878         (browser_download_init):
2879         (browser_download_class_init):
2880         (downloadReceivedResponse): Update download message using the
2881         response URI.
2882         (remainingTime): Helper function to get human readable remaining
2883         time.
2884         (downloadProgress): Update downloads progress.
2885         (downloadReceivedData):
2886         (downloadFinished): Update download message and turn download
2887         button into open.
2888         (downloadFailed): Show error message or destroy the download UI if
2889         it was cancelled.
2890         (browserDownloadNew): Create a new download widget for the given
2891         WebKitDownload.
2892         (browser_downloads_bar_add_download): Add download widget to the
2893         downloads bar.
2894         * MiniBrowser/gtk/BrowserDownloadsBar.h: Added.
2895         * MiniBrowser/gtk/BrowserWindow.c:
2896         (downloadStarted): Create a download widget and add it to the
2897         downloads bar when a new download is started.
2898         (browserWindowConstructed): Connect to
2899         WebKitWebContext::download-started signal.
2900         * MiniBrowser/gtk/GNUmakefile.am: Add new files to compilation.
2901
2902 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2903
2904         Unreviewed. Skip unit test failing in debug bot.
2905
2906         * Scripts/run-gtk-tests:
2907         (TestRunner):
2908
2909 2012-04-26  Christophe Dumez  <christophe.dumez@intel.com>
2910
2911         [EFL][DRT] "resource,request,willsend" signal needs to provide more information
2912         https://bugs.webkit.org/show_bug.cgi?id=84670
2913
2914         Reviewed by Andreas Kling.
2915
2916         Handle in EFL's DumpRenderTree the new Ewk_Frame_Resource_Messages now
2917         passed with the "resource,request,willsend" signal and dump
2918         information about the network request and the possible network
2919         response if LayoutTestController's dumpResourceLoadCallbacks() returns
2920         true.
2921
2922         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2923         (pathSuitableForTestResult):
2924         (urlSuitableForTestResult):
2925         (descriptionSuitableForTestResult):
2926         (DumpRenderTreeChrome::onWillSendRequest):
2927
2928 2012-04-25  Csaba Osztrogonác  <ossy@webkit.org>
2929
2930         [Qt][WK2] Unreviewed, rolling out r113703 because of crashing tests.
2931         http://trac.webkit.org/changeset/113703
2932         https://bugs.webkit.org/show_bug.cgi?id=83024
2933
2934         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
2935         * Tools.pro:
2936         * qmake/mkspecs/features/features.prf:
2937         * qmake/mkspecs/features/functions.prf:
2938
2939 2012-04-25  Maciej Stachowiak  <mjs@apple.com>
2940
2941         Teach LayoutTestHelper to correctly restore the color profile when it's a factory profile
2942         https://bugs.webkit.org/show_bug.cgi?id=84931
2943
2944         Reviewed by Filip Pizlo.
2945
2946         * DumpRenderTree/mac/LayoutTestHelper.m:
2947         (installLayoutTestColorProfile): Account for the possibility of the default
2948         profile being a factory profile.
2949
2950 2012-04-25  Sheriff Bot  <webkit.review.bot@gmail.com>
2951
2952         Unreviewed, rolling out r115240.
2953         http://trac.webkit.org/changeset/115240
2954         https://bugs.webkit.org/show_bug.cgi?id=84928
2955
2956         It broke everything (Requested by Ossy_HOME on #webkit).
2957
2958         * Scripts/webkitpy/layout_tests/port/gtk.py:
2959         (GtkDriver):
2960         (GtkDriver._start):
2961         (GtkDriver._start.x_filter):
2962         (GtkDriver.stop):
2963         (GtkPort._driver_class):
2964         * Scripts/webkitpy/layout_tests/port/qt.py:
2965         (QtPort._build_driver):
2966         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Removed.
2967
2968 2012-04-25  Dirk Pranke  <dpranke@chromium.org>
2969
2970         [chromium] change DRT's "DRT" output to match the other ports
2971         https://bugs.webkit.org/show_bug.cgi?id=84904
2972
2973         Reviewed by Kent Tamura.
2974
2975         Since we run Chromium's DRT in --test-shell mode, the
2976         non-test-shell output has never been really beaten on to make
2977         sure it's right. This fixes a few issues and will be tested
2978         with upcoming NRWT changes that'll land in a separate patch.
2979         In particular we should never output framing text like "\n"
2980         or "#EOF" outside of something outside of TestEventPrinter,
2981         and we weren't handling audio output properly at all.
2982
2983         * DumpRenderTree/chromium/TestEventPrinter.cpp:
2984         (DRTPrinter):
2985         (TestShellPrinter):
2986         (DRTPrinter::handleTextFooter):
2987         (DRTPrinter::handleAudioFooter):
2988         (DRTPrinter::handleTestFooter):
2989         (TestShellPrinter::handleAudioFooter):
2990         * DumpRenderTree/chromium/TestEventPrinter.h:
2991         (TestEventPrinter):
2992         * DumpRenderTree/chromium/TestShell.cpp:
2993         (TestShell::TestShell):
2994         (TestShell::dump):
2995
2996 2012-04-25  Joshua Bell  <jsbell@chromium.org>
2997
2998         Unreviewed: Adding Alec Flett as a contributor (non-committer).
2999
3000         * Scripts/webkitpy/common/config/committers.py:
3001
3002 2012-04-25  Abhishek Arya  <inferno@chromium.org>
3003
3004         Mark myself as a reviewer.
3005         https://bugs.webkit.org/show_bug.cgi?id=84902
3006
3007         Reviewed by Dimitri Glazkov.
3008
3009         * Scripts/webkitpy/common/config/committers.py:
3010
3011 2012-04-25  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
3012
3013         [Qt][NRWT] Run each DRT in it's own xvfb
3014         https://bugs.webkit.org/show_bug.cgi?id=77335
3015
3016         Reviewed by Dirk Pranke.
3017
3018         * Scripts/webkitpy/layout_tests/port/gtk.py:
3019         (GtkPort._driver_class):
3020         * Scripts/webkitpy/layout_tests/port/qt.py:
3021         (QtPort._driver_class):
3022         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Added.
3023         (XvfbDriver):
3024         (XvfbDriver._start):
3025         (XvfbDriver._start.x_filter):
3026         (XvfbDriver.stop):
3027
3028 2012-04-25  Sheriff Bot  <webkit.review.bot@gmail.com>
3029
3030         Unreviewed, rolling out r115222.
3031         http://trac.webkit.org/changeset/115222
3032         https://bugs.webkit.org/show_bug.cgi?id=84874
3033
3034         Caused test regressions on the EFL build (Requested by rakuco
3035         on #webkit).
3036
3037         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3038         (LayoutTestController::addOriginAccessWhitelistEntry):
3039         (LayoutTestController::removeOriginAccessWhitelistEntry):
3040
3041 2012-04-25  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3042
3043         [EFL] [DRT] LayoutTestController needs implementation of addOriginAccessWhitelistEntry and removeOriginAccessWhitelistEntry
3044         https://bugs.webkit.org/show_bug.cgi?id=83880
3045
3046         Reviewed by Antonio Gomes.
3047
3048         Add missing implementation addOriginAccessWhitelistEntry and
3049         removeOriginAccessWhitelistEntry to EFL's LayoutTestController
3050         in order to unskip tests which rely on this.
3051
3052         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3053         (LayoutTestController::addOriginAccessWhitelistEntry):
3054         (LayoutTestController::removeOriginAccessWhitelistEntry):
3055
3056 2012-04-25  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3057
3058         [EFL] [DRT] Reset text zoom factor before running a test
3059         https://bugs.webkit.org/show_bug.cgi?id=84861
3060
3061         Reviewed by Antonio Gomes.
3062
3063         This fix eliminates most of the current flakiness on EFL build
3064         bots, although some CSS tests will now need rebasiling.
3065
3066         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3067         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
3068
3069 2012-04-25  Milian Wolff  <milian.wolff@kdab.com>
3070
3071         [Qt] add LayoutTestController::setPrinting support to Qt unit tests
3072         https://bugs.webkit.org/show_bug.cgi?id=84246
3073
3074         Reviewed by Simon Hausmann.
3075
3076         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3077         (WebCore::DumpRenderTree::dump):
3078
3079 2012-04-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3080
3081         [Qt] Make the web view's url property follow the active url
3082
3083         https://bugs.webkit.org/show_bug.cgi?id=77554
3084
3085         Update  the location bar in the minibrowser to behave
3086         a bit more like normal browsers in terms of when the url will change
3087         and how active focus is handled.
3088
3089         Reviewed by Simon Hausmann.
3090
3091         * MiniBrowser/qt/qml/BrowserWindow.qml:
3092
3093 2012-04-25  Philippe Normand  <pnormand@igalia.com>
3094
3095         Webkit build fails due to missing gstreamer include file on Kubuntu 8.04
3096         https://bugs.webkit.org/show_bug.cgi?id=81913
3097
3098         Reviewed by Tor Arne Vestbø.
3099
3100         * qmake/mkspecs/features/features.prf: Check for gstreamer-0.10
3101         and gstreamer-plugins-base >= 0.10.30 just like the GTK port
3102         does. 0.10.27 is unfortunately not recent enough to get the media
3103         player working reliably.
3104
3105
3106 2012-04-25  Xiaobo Wang  <xbwang@torchmobile.com.cn>
3107
3108         [BlackBerry] Run tests in '/http/tests/local' with local URI
3109         https://bugs.webkit.org/show_bug.cgi?id=84820
3110
3111         Reviewed by Nikolas Zimmermann.
3112
3113         Currently we run all tests under folder "/http/tests/" as HTTP
3114         tests (http://...). However in WebKit ORWT and NRWT scripts, tests in
3115         'http/tests/local' are run with local path (like '/developer/LayoutTests
3116         /http/tests/local/absolute-url-strip-whitespace.html').
3117         With this patch we can keep the same behaviour as other portings.
3118
3119         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
3120         (BlackBerry::WebKit::DumpRenderTree::isHTTPTest):
3121
3122 2012-04-25  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
3123
3124         [EFL][DRT] Reset DRT settings to their default values after a testcase was executed.
3125         https://bugs.webkit.org/show_bug.cgi?id=84709
3126
3127         Reviewed by Eric Seidel.
3128
3129         Not all the settings that can be changed by layout testcases through a JS API were
3130         reset in DumpRenderTreeChrome::resetDefaultsToConsistentValues().
3131
3132         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3133         (defaultEditingBehavior): returns editing behavior depending on current platform.
3134         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): reset some missed DRT settings
3135
3136 2012-04-24  Dirk Pranke  <dpranke@chromium.org>
3137
3138         nrwt: add ASAN handling to chromium crash logs
3139         https://bugs.webkit.org/show_bug.cgi?id=84664
3140
3141         Reviewed by Ojan Vafai.
3142
3143         * Scripts/webkitpy/layout_tests/port/chromium.py:
3144         (ChromiumPort._get_crash_log):
3145
3146 2012-04-24  Dirk Pranke  <dpranke@chromium.org>
3147
3148         webkit-patch rebaseline-test is retrieving stale expectations
3149         https://bugs.webkit.org/show_bug.cgi?id=84762
3150
3151         Reviewed by Ojan Vafai.
3152
3153         webkit-patch rebaseline-expectations is fetching files directly
3154         from the layout-test-results directory on the buildbots, rather
3155         than fetching them from the archive. This is problematic because
3156         we don't clobber the directory after each run and so you can end
3157         up fetching stale failing results.
3158
3159         This change temporarily changes the code to fetch the zip file
3160         instead -- making things much slower -- until I can confer w/
3161         abarth and ojan over the best way to fix this for the long-term.
3162
3163         * Scripts/webkitpy/tool/commands/rebaseline.py:
3164         (RebaselineTest.__init__):
3165         (RebaselineTest._zip_file_set):
3166         (RebaselineTest):
3167         (RebaselineTest._fetch_baseline):
3168         (RebaselineTest._rebaseline_test):
3169         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3170         (TestRebaseline.stub_rebaseline_test_command_and_tool):
3171         (TestRebaseline.stub_rebaseline_test_command_and_tool.FakeZipFileSet):
3172         (TestRebaseline.stub_rebaseline_test_command_and_tool.FakeZipFileSet.read):
3173         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3174         (test_rebaseline_updates_expectations_file):
3175         (test_rebaseline_test):
3176         (test_rebaseline_and_copy_test):
3177         (test_rebaseline_and_copy_test_no_existing_result):
3178         (test_rebaseline_and_copy_test_with_lion_result):
3179         (test_rebaseline_and_copy_no_overwrite_test):
3180
3181 2012-04-24  Ojan Vafai  <ojan@chromium.org>
3182
3183         Improve code for embedding flakiness dashboard as an iframe
3184         https://bugs.webkit.org/show_bug.cgi?id=84782
3185
3186         Reviewed by Dirk Pranke.
3187
3188         -Include the horizontal scrollbar height in the measured height of the page.
3189         -Post a message everytime the height changes instead of in response to a queryContentHeight message.
3190         This way, the embedder doesn't need to poll.
3191
3192         No new tests. This is hard to test because it involves postMessage and iframes.
3193         Once we switch the dashboard tests over to qunit, we should be able to write an
3194         asyncTest for this.
3195
3196         * TestResultServer/static-dashboards/flakiness_dashboard.js:
3197         (generatePage):
3198         (appendHTML):
3199         (appendExpectations):
3200         (postHeightChangedMessage):
3201         (generatePageForIndividualTests.if):
3202         (generatePageForIndividualTests):
3203         * TestResultServer/static-dashboards/flakiness_dashboard_embedded.html: Added.
3204
3205 2012-04-16  Ojan Vafai  <ojan@chromium.org>
3206
3207         Replace garden-o-matic accordion UI with a panel UI
3208         https://bugs.webkit.org/show_bug.cgi?id=84081
3209
3210         Reviewed by Dimitri Glazkov.
3211
3212         This is a nicer UI to work with and makes it easier to embed the flakiness dashboard,
3213         since we can reuse the same iframe without moving it in the DOM.
3214
3215         Also, excise unnecessary jquery-isms.
3216
3217         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3218         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
3219         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
3220
3221 2012-04-24  Peter Beverloo  <peter@chromium.org>
3222
3223         [Chromium] DumpRenderTree should depend on base.gyp:test_support_base
3224         https://bugs.webkit.org/show_bug.cgi?id=84729
3225
3226         Reviewed by Dirk Pranke.
3227
3228         This dependency is necessary for Android since it'll soon be
3229         calling InitAndroidOSPathStub() in it's platform support. The
3230         TestWebKitAPI and webkit_unit_tests already have this dependency.
3231
3232         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
3233
3234 2012-04-24  Ojan Vafai  <ojan@chromium.org>
3235
3236         Replace vague informative statement with the list of builders where the test isn't listed
3237         https://bugs.webkit.org/show_bug.cgi?id=84779
3238
3239         Reviewed by Dirk Pranke.
3240
3241         This is more informative and more likely to be noticed. With the old UI, people never
3242         read the line and if they did, they didn't really understand what it meant. Hopefully
3243         have the list of builders will be more clear.
3244
3245         * TestResultServer/static-dashboards/flakiness_dashboard.js:
3246         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
3247         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
3248
3249 2012-04-24  Tommy Widenflycht  <tommyw@google.com>
3250
3251         MediaStream API: Create a new flag for PeerConnection
3252         https://bugs.webkit.org/show_bug.cgi?id=84723
3253
3254         Reviewed by Dimitri Glazkov.
3255
3256         To allow Chrome to make the GetUserMedia functionality available by default we
3257         need another flag for the PeerConnection related parts. This new flag is in
3258         addition to the general media stream flag, and both need to be enabled for
3259         PeerConnection object creation.
3260
3261         * DumpRenderTree/chromium/TestShell.cpp:
3262         (TestShell::TestShell):
3263
3264 2012-04-24  Joe Mason  <jmason@rim.com>
3265
3266         [BlackBerry] Revert broken changes to authentication dialog
3267         https://bugs.webkit.org/show_bug.cgi?id=80135
3268
3269         Reviewed by Antonio Gomes.
3270
3271         The previous patches from this bug caused an infinite loop when using digest auth;
3272         apparently they were only tested with basic.
3273
3274         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
3275         * DumpRenderTree/blackberry/DumpRenderTreeBlackBerry.h:
3276         (WebCore):
3277
3278 2012-04-24  Zan Dobersek  <zandobersek@gmail.com>
3279
3280         [TestResultsServer] Add a unit test for changes in r114868
3281         https://bugs.webkit.org/show_bug.cgi?id=84713
3282
3283         Reviewed by Ojan Vafai.
3284
3285         Add a new unit test to test the isToTWebKit attribute of a BuilderGroup object
3286         that is created for either DEPS or ToT builder groups. The unit test covers changes
3287         made in r114868.
3288
3289         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
3290         (testBuilderGroupIsToTWebKitAttribute.g_handleBuildersListLoaded):
3291         (testBuilderGroupIsToTWebKitAttribute.assertEquals):
3292
3293 2012-04-24  Brady Eidson  <beidson@apple.com>
3294
3295         <rdar://problem/10120155> and https://bugs.webkit.org/show_bug.cgi?id=82664
3296          Need DOMWindow mechanism to supplement UserScripts for page cache notifications
3297
3298         Reviewed by Sam Weinig.
3299
3300         Add a new bundle test that stresses the basics of the WKBundleDOMWindowExtension API.
3301
3302         It makes sure new extensions can be added at the right times, and that all of the expected
3303         callbacks are made.
3304
3305         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic.cpp: Added.
3306         (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Record all messages sent from the bundle.
3307         (TestWebKitAPI::TEST): Run a page through some basic exercises, then compared messages sent from the
3308            bundle to a set of known expected messages.
3309
3310         Listen to all the new client callbacks, create some extensions, and monitor their lifetimes:
3311         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp: Added.
3312         (TestWebKitAPI::DOMWindowExtensionBasic::DOMWindowExtensionBasic):
3313         (TestWebKitAPI::DOMWindowExtensionBasic::frameLoadFinished):
3314         (TestWebKitAPI::DOMWindowExtensionBasic::sendExtensionStateMessage):
3315         (TestWebKitAPI::DOMWindowExtensionBasic::initialize):
3316         (TestWebKitAPI::DOMWindowExtensionBasic::didCreatePage):
3317         (TestWebKitAPI::DOMWindowExtensionBasic::updateExtensionStateRecord):
3318         (TestWebKitAPI::DOMWindowExtensionBasic::sendBundleMessage):
3319         (TestWebKitAPI::DOMWindowExtensionBasic::didCreateGlobalObjectForFrame):
3320         (TestWebKitAPI::DOMWindowExtensionBasic::willDisconnectDOMWindowExtensionFromGlobalObject):
3321         (TestWebKitAPI::DOMWindowExtensionBasic::didReconnectDOMWindowExtensionToGlobalObject):
3322         (TestWebKitAPI::DOMWindowExtensionBasic::willDestroyGlobalObjectForDOMWindowExtension):
3323         (TestWebKitAPI::didFinishLoadForFrameCallback):
3324         (TestWebKitAPI::didCreateGlobalObjectForFrameCallback):
3325         (TestWebKitAPI::willDisconnectDOMWindowExtensionFromGlobalObjectCallback):
3326         (TestWebKitAPI::didReconnectDOMWindowExtensionToGlobalObjectCallback):
3327         (TestWebKitAPI::willDestroyGlobalObjectForDOMWindowExtensionCallback):
3328
3329         * TestWebKitAPI/Tests/WebKit2/InjectedBundleBasic.cpp:
3330         (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Make this method static, as it always should have been.
3331
3332         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3333
3334 2012-04-24  Carlos Garcia Campos  <cgarcia@igalia.com>
3335
3336         [GTK] Build and run TestWebKitAPI unit tests
3337         https://bugs.webkit.org/show_bug.cgi?id=84325
3338
3339         Reviewed by Philippe Normand.
3340
3341         We only build WTF unit tests for now. WebKit2 tests will be
3342         enabled in a follow up patch.
3343
3344         * Scripts/run-gtk-tests:
3345         (TestRunner): Add TestWebKitAPI/WTF to the list of directories
3346         containing unit tests.
3347         (TestRunner._run_test_glib): Run glib tests using gtester.
3348         (TestRunner._run_test_google): Run google tests.
3349         (TestRunner._run_test): Use _run_test_glib or _run_test_google
3350         depending on whether given test is glib or google test
3351         * TestWebKitAPI/GNUmakefile.am: Added.
3352         * TestWebKitAPI/config.h: Use #ifdef instead of # to fix a compile
3353         warning with -Wundef. Include WebKit2 headers only when building
3354         WebKit2.
3355
3356 2012-04-24  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
3357
3358         [Qt][WK2] Convert touch-point area
3359         https://bugs.webkit.org/show_bug.cgi?id=84434
3360
3361         Reviewed by Kenneth Rohde Christiansen.
3362
3363         Fix failing test after r114917 by also adding the new EventSender API
3364         to WebKitTestRunner.
3365
3366         * WebKitTestRunner/EventSenderProxy.h:
3367         (EventSenderProxy):
3368         * WebKitTestRunner/InjectedBundle/Bindings/EventSendingController.idl:
3369         * WebKitTestRunner/InjectedBundle/EventSendingController.cpp:
3370         (WTR::EventSendingController::setTouchPointRadius):
3371         * WebKitTestRunner/InjectedBundle/EventSendingController.h:
3372         (EventSendingController):
3373         * WebKitTestRunner/TestController.cpp:
3374         (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
3375         * WebKitTestRunner/qt/EventSenderProxyQt.cpp:
3376         (WTR::EventSenderProxy::addTouchPoint):
3377         (WTR::EventSenderProxy::updateTouchPoint):
3378         (WTR::EventSenderProxy::setTouchPointRadius):
3379         (WTR::EventSenderProxy::clearTouchPoints):
3380
3381 2012-04-24  Mikhail Naganov  <mnaganov@chromium.org>
3382
3383         [Chromium] Unreviewed test expectations update.
3384         https://bugs.webkit.org/show_bug.cgi?id=84726
3385
3386         Mark webkitpy.layout_tests.servers.http_server_unittest.TestHttpServer.test_start_cmd as failing on Windows.
3387
3388         * Scripts/webkitpy/test/test_finder.py:
3389         (TestFinder._default_names):
3390
3391 2012-04-24  Christophe Dumez  <christophe.dumez@intel.com>
3392
3393         [EFL] Enable Web Timing
3394         https://bugs.webkit.org/show_bug.cgi?id=84705
3395
3396         Reviewed by Tony Gentilcore.
3397
3398         Enable web timing by default whenever build-webkit is used.
3399
3400         * Scripts/build-webkit:
3401
3402 2012-04-23  Eriq Augustine  <eaugusti@chromium.org>
3403
3404         Return value from executed script in Chromium.
3405         https://bugs.webkit.org/show_bug.cgi?id=79851
3406
3407         Reviewed by Adam Barth.
3408
3409         Adding a function to LayoutTestController that evaluates a script in
3410         an isolated wold an returns the result of the evaluation.
3411
3412         * DumpRenderTree/LayoutTestController.cpp:
3413         (evaluateScriptInIsolatedWorldAndReturnValue):
3414         (LayoutTestController::staticFunctions):
3415         * DumpRenderTree/LayoutTestController.h:
3416         (LayoutTestController):
3417         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
3418         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3419         * DumpRenderTree/chromium/LayoutTestController.cpp:
3420         (LayoutTestController::LayoutTestController):
3421         (LayoutTestController::setXSSAuditorEnabled):
3422         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3423         * DumpRenderTree/chromium/LayoutTestController.h:
3424         (LayoutTestController):
3425         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3426         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3427         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3428         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3429         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
3430         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3431         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3432         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3433         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3434         (LayoutTestController):
3435         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3436         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3437         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
3438         (LayoutTestController::evaluateScriptInIsolatedWorldAndReturnValue):
3439
3440 2012-04-23  Dirk Pranke  <dpranke@chromium.org>
3441
3442         run-perf-tests fail intermittently with an exception
3443         https://bugs.webkit.org/show_bug.cgi?id=83532
3444
3445         Reviewed by Ryosuke Niwa.
3446
3447         This bulletproofs port.get_option('child_processes')
3448         against the case where the option wasn't set.
3449
3450         * Scripts/webkitpy/layout_tests/port/mac.py:
3451         (MacPort._get_crash_log):
3452         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3453         (test_get_crash_log):
3454         (test_get_crash_log.fake_time_cb):
3455
3456 2012-04-23  Balazs Kelemen  <kbalazs@webkit.org>
3457
3458         [Qt] Add desktop zooming support for QQuickWebView
3459         https://bugs.webkit.org/show_bug.cgi?id=82337
3460
3461         Unreviewed typo fix after r114900.
3462
3463         * MiniBrowser/qt/BrowserWindow.cpp:
3464         (BrowserWindow::BrowserWindow): Fix the mistyped
3465         value of a zoom level.
3466
3467 2012-04-23  Hao Zheng  <zhenghao@chromium.org>
3468
3469         [chromium-android] Get rid of the trick that replaces system fallback font with DejaVuSans.
3470         https://bugs.webkit.org/show_bug.cgi?id=84580
3471
3472         Reviewed by Ojan Vafai.
3473
3474         Android's first fallback font is changed to DroidSansArabic.ttf in JB.
3475         Take the opportunity to get rid of the trick.
3476
3477         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3478         (ChromiumAndroidPort.start_helper):
3479         (ChromiumAndroidPort._push_fonts):
3480
3481 2012-04-23  Dirk Pranke  <dpranke@chromium.org>
3482
3483         test-webkitpy is failing
3484         https://bugs.webkit.org/show_bug.cgi?id=84617
3485
3486         Reviewed by Eric Seidel.
3487
3488         * Scripts/webkitpy/layout_tests/port/http_lock_unittest.py:
3489         (HttpLockTestWithRealFileSystem.clean_all_lockfile):
3490
3491 2012-04-23  Ojan Vafai  <ojan@chromium.org>
3492
3493         Split up flakiness_dashboard.html into multiple files. This is a step in the direction
3494         of making the code more maintainable (e.g. being able to use qunit for testing).
3495
3496         * TestResultServer/static-dashboards/dashboard_base.js:
3497         * TestResultServer/static-dashboards/flakiness_dashboard.css: Added.
3498         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3499         * TestResultServer/static-dashboards/flakiness_dashboard.js: Copied from Tools/TestResultServer/static-dashboards/flakiness_dashboard.html.
3500         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css: Added.
3501         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js: Renamed from Tools/TestResultServer/static-dashboards/flakiness_dashboard_tests.js.
3502
3503 2012-04-23  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
3504
3505         [Qt][WK2] Convert touch-point area.
3506         https://bugs.webkit.org/show_bug.cgi?id=84434
3507
3508         Reviewed by Kenneth Rohde Christiansen.
3509
3510         Add feature to set touch-point radius for emitted touch-events.
3511
3512         * DumpRenderTree/qt/EventSenderQt.cpp:
3513         (EventSender::addTouchPoint):
3514         (EventSender::updateTouchPoint):
3515         (EventSender::setTouchPointRadius):
3516         (EventSender::clearTouchPoints):
3517         * DumpRenderTree/qt/EventSenderQt.h:
3518         (EventSender):
3519
3520 2012-04-23  Christophe Dumez  <christophe.dumez@intel.com>
3521
3522         [EFL] EFL's LayoutTestController does not implement elementDoesAutoCompleteForElementWithId
3523         https://bugs.webkit.org/show_bug.cgi?id=84425
3524
3525         Reviewed by Antonio Gomes.
3526
3527         Use the new helper in DumpRenderTreeSupportEfl to properly
3528         test whether an element performs autocompletion.
3529
3530         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3531         (LayoutTestController::elementDoesAutoCompleteForElementWithId):
3532
3533 2012-04-23  Csaba Osztrogonác  <ossy@webkit.org>
3534
3535         master.cfg cleanup, remove dead codes after r92792.
3536         https://bugs.webkit.org/show_bug.cgi?id=84433
3537
3538         Relanding r114881 after typo fix in UploadBuiltProduct.
3539
3540         Reviewed by Ryosuke Niwa.
3541
3542         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3543
3544 2012-04-23  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3545
3546         [EFL][DRT] Need to support requestAnimationFrame
3547         https://bugs.webkit.org/show_bug.cgi?id=84585
3548
3549         Reviewed by Martin Robinson.
3550
3551         Making requestAnimationFrame() support a default on the EFL build.
3552
3553         * Scripts/build-webkit:
3554
3555 2012-04-23  Balazs Kelemen  <kbalazs@webkit.org>
3556
3557         [Qt] Add desktop zooming support for QQuickWebView
3558         https://bugs.webkit.org/show_bug.cgi?id=82337
3559
3560         Reviewed by Simon Hausmann.
3561
3562         Add zooming support for MiniBrowser to
3563         test the new API.
3564
3565         * MiniBrowser/qt/BrowserWindow.cpp:
3566         (BrowserWindow::BrowserWindow):
3567         (BrowserWindow::zoomIn):
3568         (BrowserWindow::zoomOut):
3569         (BrowserWindow::keyPressEvent):
3570         (BrowserWindow::wheelEvent):
3571         * MiniBrowser/qt/BrowserWindow.h:
3572         (BrowserWindow):
3573
3574 2012-04-23  Csaba Osztrogonác  <ossy@webkit.org>
3575
3576         Unreviewed rolling out r114881, because it broke upload buildstep.
3577
3578         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3579         (ConfigureBuild.__init__):
3580         (ConfigureBuild.start):
3581         (CompileWebKit.start):
3582         (UploadBuiltProduct):
3583         (UploadBuiltProduct.determineExtraFeatures):
3584         (Factory.__init__):
3585         (BuildFactory.__init__):
3586         (TestFactory.__init__):
3587         (loadBuilderConfig):
3588
3589 2012-04-23  Csaba Osztrogonác  <ossy@webkit.org>
3590
3591         master.cfg cleanup, remove dead codes after r92792.
3592         https://bugs.webkit.org/show_bug.cgi?id=84433
3593
3594         Reviewed by Ryosuke Niwa.
3595
3596         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3597
3598 2012-04-23  Milian Wolff  <mail@milianw.de>
3599
3600         [Qt] Make DRT not dump pixel results if test is calling layoutTestController.dumpAsText()
3601         https://bugs.webkit.org/show_bug.cgi?id=84327
3602
3603         Reviewed by Csaba Osztrogonác.
3604
3605         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3606         (WebCore::DumpRenderTree::dump):
3607
3608 2012-04-22  Sriram Neelakandan  <sriram.neelakandan@gmail.com>
3609
3610         [Gtk] Added MOZ_X11 build flag for TARGET_X11
3611         https://bugs.webkit.org/show_bug.cgi?id=40785
3612
3613         Reviewed by Anders Carlsson.
3614
3615         * GNUmakefile.am:
3616
3617 2012-04-22  Zan Dobersek  <zandobersek@gmail.com>
3618
3619         [TestResultsServer] Impossible to inspect builds on non-Chromium builders
3620         https://bugs.webkit.org/show_bug.cgi?id=84544
3621
3622         Reviewed by Ryosuke Niwa.
3623
3624         Use the groupEnum parameter that is passed through to the onBuilderListLoad
3625         function rather than an undefined attribute of the BuilderGroup object. This
3626         makes it again possible to inspect builds on non-Chromium builders in the
3627         test results server by not trying to always add Chromium revision links.
3628
3629         * TestResultServer/static-dashboards/builders.js:
3630
3631 2012-04-20  Martin Robinson  <mrobinson@igalia.com>
3632
3633         [GTK] Revert changes that disabled Cairo fonts metrics hinting
3634         https://bugs.webkit.org/show_bug.cgi?id=84502
3635
3636         Reviewed by Daniel Bates.
3637
3638         Re-enable Cairo metrics hinting, which seems to ensure consistent results in
3639         reference and pixel test results. This makes our font metrics a bit different
3640         than the Mac port's, but it's a decent price to pay for consistent results.
3641
3642         * DumpRenderTree/gtk/DumpRenderTree.cpp:
3643         (initializeGtkFontSettings):
3644         * WebKitTestRunner/InjectedBundle/gtk/ActivateFontsGtk.cpp:
3645         (WTR::initializeGtkSettings):
3646
3647 2012-04-04  Jer Noble  <jer.noble@apple.com>
3648
3649         apple.com top navigation bar appears inside video during full screen exit animation
3650         https://bugs.webkit.org/show_bug.cgi?id=83095
3651
3652         Reviewed by Eric Carlson.
3653
3654         Add a new LayoutTestController function, setHasCustomFullScreenBehavior(), which allows scripts to
3655         disable the standard full screen behavior, and replace it with explicit callbacks to the relevant
3656         Document methods from within the script.  This allows subtle timing bugs to be consistently reproduced
3657         from within LayoutTests.
3658
3659         Add the new methods, hasCustomFullScreenBehavior() and setHasCustomFullScreenBehavior():
3660         * DumpRenderTree/LayoutTestController.cpp:
3661         (setHasCustomFullScreenBehaviorCallback):
3662         (LayoutTestController::staticFunctions):
3663         * DumpRenderTree/LayoutTestController.h:
3664         (LayoutTestController::setHasCustomFullScreenBehavior):
3665         (LayoutTestController::hasCustomFullScreenBehavior):
3666         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
3667         (WTR::InjectedBundlePage::enterFullScreenForElement):
3668         (WTR::InjectedBundlePage::exitFullScreenForElement):
3669         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
3670         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3671         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3672         (WTR::LayoutTestController::setHasCustomFullScreenBehavior):
3673         (WTR::LayoutTestController::hasCustomFullScreenBehavior):
3674         * DumpRenderTree/chromium/LayoutTestController.cpp:
3675         (LayoutTestController::LayoutTestController):
3676         (LayoutTestController::reset):
3677         * DumpRenderTree/chromium/LayoutTestController.h:
3678         (LayoutTestController::setHasCustomFullScreenBehavior):
3679         (LayoutTestController::hasCustomFullScreenBehavior):
3680
3681         Query the new methods before calling will/did/Enter/ExitFullScreen:
3682         * DumpRenderTree/chromium/WebViewHost.cpp:
3683         (WebViewHost::enterFullScreenNow):
3684         (WebViewHost::exitFullScreenNow):
3685         * DumpRenderTree/mac/UIDelegate.mm:
3686         (-[UIDelegate webView:enterFullScreenForElement:listener:]):
3687         (-[UIDelegate webView:exitFullScreenForElement:listener:]):
3688
3689 2012-04-20  Nandor Huszka  <hnandor@inf.u-szeged.hu>
3690
3691         NRWT: option --skip-pixel-test-if-no-baseline support on DRT
3692         https://bugs.webkit.org/show_bug.cgi?id=72841
3693
3694         Reviewed by Dirk Pranke
3695
3696         This patch implements support for --skip-pixel-test-if-no-baseline generically in NRWT,
3697         instead of in every DRT and WTR implementation. We add 'should_run_pixel_test'
3698         and 'reference_files' fields to the TestInput objects, and use those flags
3699         to select the right driver to use in the DriverProxy.
3700
3701         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
3702         (SingleTestRunner.__init__):
3703         (SingleTestRunner._driver_input):
3704         (SingleTestRunner._run_reftest):
3705         * Scripts/webkitpy/layout_tests/controllers/worker.py:
3706         (Worker.handle_test_list):
3707         (Worker._update_test_input):
3708         * Scripts/webkitpy/layout_tests/models/test_input.py:
3709         (TestInput):
3710         (TestInput.__init__):
3711         (TestInput.__repr__):
3712         * Scripts/webkitpy/layout_tests/port/chromium.py:
3713         (ChromiumDriver._wrapper_options):
3714         (ChromiumDriver.run_test):
3715         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3716         (ChromiumDriverTest.test_crash_log):
3717         * Scripts/webkitpy/layout_tests/port/driver.py:
3718         (DriverInput.__init__):
3719         (Driver.__init__):
3720         (DriverProxy.__init__):
3721         (DriverProxy.run_test):
3722         (DriverProxy.start):
3723         (DriverProxy.cmd_line):
3724         * Scripts/webkitpy/layout_tests/port/gtk.py:
3725         (GtkDriver._start):
3726         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
3727         (MockDRT.run):
3728         (MockDRT.input_from_line):
3729         (MockDRT.output_for_test):
3730         (MockDRT.write_test_output):
3731         (MockChromiumDRT.input_from_line):
3732         (MockChromiumDRT.output_for_test):
3733         (MockChromiumDRT.write_test_output):
3734         * Scripts/webkitpy/layout_tests/port/webkit.py:
3735         (WebKitDriver.cmd_line):
3736         (WebKitDriver.run_test):
3737         * WebKitTestRunner/TestController.cpp:
3738         (WTR::TestController::TestController):
3739         (WTR::TestController::initialize):
3740         (WTR::TestController::runTest):
3741         * WebKitTestRunner/TestController.h:
3742         (TestController):
3743         * WebKitTestRunner/TestInvocation.cpp:
3744         (WTR::TestInvocation::TestInvocation):
3745         (WTR::TestInvocation::setIsPixelTest):
3746         * WebKitTestRunner/TestInvocation.h:
3747         (TestInvocation):
3748
3749 2012-04-20  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3750
3751         [EFL] Make it possible to define offline storage quota per origin
3752         https://bugs.webkit.org/show_bug.cgi?id=84170
3753
3754         Improved LayoutTestController::setDatabaseQuota to set quota based on
3755         origin which was causing some tests to fail. Also align the default
3756         quota for testing (5MB) with others ports.
3757
3758         When dumpDatabaseCallbacks is set to true by, also dump some database
3759         information when quota is exceed, which is also required by some tests.
3760
3761         Reviewed by Antonio Gomes.
3762
3763         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3764         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
3765         * DumpRenderTree/efl/DumpRenderTreeView.cpp:
3766         (onExceededDatabaseQuota):
3767         (drtViewAdd):
3768         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3769         (LayoutTestController::setDatabaseQuota):
3770
3771 2012-04-20  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3772
3773         [EFL] [DRT] Send double click event from EventSender
3774         https://bugs.webkit.org/show_bug.cgi?id=82867
3775
3776         Reviewed by Daniel Bates.
3777
3778         Add support for EFL DumpRenderTree EventSender to send
3779         double click events.
3780
3781         * DumpRenderTree/efl/EventSender.cpp:
3782         (sendMouseEvent):
3783
3784 2012-04-20  Sergio Villar Senin  <svillar@igalia.com>
3785
3786         [GTK] Enable Web Timing
3787         https://bugs.webkit.org/show_bug.cgi?id=42432
3788
3789         Reviewed by Gustavo Noronha Silva.
3790
3791         Build web timing by default whenever build-webkit is used.
3792
3793         * Scripts/build-webkit:
3794
3795 2012-04-20  Kamil Blank  <k.blank@samsung.com>
3796
3797         [EFL] Add missing reset for setAuthorAndUserStylesEnabled inside DRT.
3798         https://bugs.webkit.org/show_bug.cgi?id=84430
3799
3800         Reviewed by Daniel Bates.
3801
3802         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3803         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
3804
3805 2012-04-20  Ryosuke Niwa  <rniwa@webkit.org>
3806
3807         REGRESSION(r112781): Windows build slaves are no longer including .pdb files in archives handed to testers
3808         https://bugs.webkit.org/show_bug.cgi?id=84424
3809
3810         Reviewed by Mark Rowe.
3811
3812         Let Windows and Chromium have their own list of files to copy.
3813
3814         * BuildSlaveSupport/built-product-archive:
3815         (removeDirectoryIfExists):
3816         (copyBuildFiles):
3817         (archiveBuiltProduct):
3818
3819 2012-04-20  Kamil Blank  <k.blank@samsung.com>
3820
3821         [EFL] Add setting API for author and user styles.
3822         https://bugs.webkit.org/show_bug.cgi?id=82860
3823
3824         Reviewed by Daniel Bates.
3825
3826         Implementation of setAuthorAndUserStylesEnabled.
3827
3828         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3829         (LayoutTestController::setAuthorAndUserStylesEnabled):
3830
3831 2012-04-19  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3832
3833         [EFL] Missing keycode translation for space key
3834         https://bugs.webkit.org/show_bug.cgi?id=84153
3835
3836         Reviewed by Antonio Gomes.
3837
3838         Set keyname for space key before dispatching event.
3839
3840         * DumpRenderTree/efl/EventSender.cpp:
3841         (keyNameFromJSValue):
3842
3843 2012-04-19  Carlos Garcia Campos  <cgarcia@igalia.com>
3844
3845         [GTK] Allow to use WebInspector in GtkLauncher
3846         https://bugs.webkit.org/show_bug.cgi?id=84143
3847
3848         Reviewed by Martin Robinson.
3849
3850         * GNUmakefile.am: Add new files to compilation.
3851         * GtkLauncher/LauncherInspectorWindow.c: Added.
3852         (launcherInspectorWindowFinalize):
3853         (launcher_inspector_window_init):
3854         (launcher_inspector_window_class_init):
3855         (inspectedURIChanged): Update the window title based on current
3856         inspected URI.
3857         (showInspectorWindow): Show inspector window.
3858         (closeInspectorWindow): Hide inspector window.
3859         (launcherInspectorWindowNew): Create an inspector window.
3860         (launcherInspectorWindowGetWebView): Return the inspected web view.
3861         * GtkLauncher/LauncherInspectorWindow.h: Added.
3862         * GtkLauncher/main.c:
3863         (inspectorInspectWebViewCb): Create an inspector window.
3864         (createBrowser): Connect to inspect-web-view signal of the view
3865         inspector.
3866
3867 2012-04-19  Jeremy Mao  <yujie.mao@intel.com>
3868
3869         Media Stream API: Piece of code cleanup for the chromium port
3870         https://bugs.webkit.org/show_bug.cgi?id=84328
3871
3872         Reviewed by Kent Tamura.
3873
3874         WebUserMediaClientMock use an old requestUserMedia interface and
3875         Replace it with the new interface.
3876
3877         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
3878         (WebKit::WebUserMediaClientMock::requestUserMedia):
3879         Remove the old requestUserMedia and replace it with the new.
3880         * DumpRenderTree/chromium/WebUserMediaClientMock.h:
3881         (WebUserMediaClientMock):
3882
3883 2012-04-19  Christophe Dumez  <christophe.dumez@intel.com>
3884
3885         [EFL] DRT console messages should contain only the file name part of local URIs
3886         https://bugs.webkit.org/show_bug.cgi?id=84236
3887
3888         Reviewed by Kentaro Hara.
3889
3890         Process console messages before printing them in EFL's DumpRenderTree
3891         them so that they contain only the file name part of local URIs. This
3892         is what is expected by test cases.
3893
3894         * DumpRenderTree/efl/DumpRenderTreeView.cpp:
3895         (urlSuitableForTestResult):
3896         (onConsoleMessage):
3897
3898 2012-04-19  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3899
3900         [EFL] [DRT] Should create offline storage inside DUMPRENDERTREE_TEMP
3901         https://bugs.webkit.org/show_bug.cgi?id=84169
3902
3903         All databases were being created inside ~/.webkit which was causing
3904         flaky tests when running multiple tests in parallel, specially timeouts
3905         due to concurrency.
3906
3907         Reviewed by Eric Seidel.
3908
3909         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3910         (DumpRenderTreeChrome::initialize):
3911
3912 2012-04-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3913
3914         [GTK] 3 fast/ layout tests failing after upgrading libsoup to 2.37.92
3915         https://bugs.webkit.org/show_bug.cgi?id=82484
3916
3917         Reviewed by Martin Robinson.
3918
3919         Bumping libsoup to version 2.38.1 which fixes the above WebKit bug.
3920
3921         * efl/jhbuild.modules:
3922         * gtk/jhbuild.modules:
3923
3924 2012-04-19  Lucas Forschler  <lforschler@apple.com>
3925
3926         Move apple-xserve-10 from Lion Test to SnowLeopard Debug Build
3927
3928         * BuildSlaveSupport/build.webkit.org-config/config.json:
3929
3930 2012-04-19  Christophe Dumez  <christophe.dumez@intel.com>
3931
3932         [EFL] dispatchDidCancelClientRedirect needs implementation
3933         https://bugs.webkit.org/show_bug.cgi?id=83669
3934
3935         Reviewed by Antonio Gomes.
3936
3937         Catch the new ewk_frame "redirect,cancelled" signal in EFL's DRT and
3938         and print out information that is expected by several test cases if
3939         LayoutTestController's dumpFrameLoadCallback() returns true.
3940
3941         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3942         (DumpRenderTreeChrome::createView):
3943         (DumpRenderTreeChrome::onFrameCreated):
3944         (DumpRenderTreeChrome::onFrameRedirectCancelled):
3945         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
3946         (DumpRenderTreeChrome):
3947
3948 2012-04-18  Hironori Bono  <hbono@chromium.org>
3949
3950         Split SpellChecker::didCheck() to SpellChecker::didCheckSucceeded() and SpellChecker::didCheckCanceled()
3951         https://bugs.webkit.org/show_bug.cgi?id=83748
3952
3953         Reviewed by Ryosuke Niwa.
3954
3955         This change replaces a call for SpellChecker::didCheck() with one for
3956         SpellChecker::didCheckCenceled() because didCheck() has been split into
3957         didCheckSucceeded() and didCheckCanceled().
3958
3959         * DumpRenderTree/chromium/WebViewHost.cpp:
3960         (WebViewHost::requestCheckingOfText): Replaced didCheck with didCheckCanceled.
3961
3962 2012-04-18  Dirk Pranke  <dpranke@chromium.org>
3963
3964         Fix failure in test-webkitpy after r114583.
3965
3966         Unreviewed, build fix.
3967
3968         Updating the call site in r114583 w/o updating the test caused
3969         the test to fail. Somehow I forgot to run the unit tests :(.
3970
3971         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3972         (ChromiumDriverTest.test_crash_log):
3973
3974 2012-04-18  Filip Pizlo  <fpizlo@apple.com>
3975
3976         JSC testing should have complete coverage over typed array types
3977         https://bugs.webkit.org/show_bug.cgi?id=84302
3978
3979         Reviewed by Geoff Garen.
3980
3981         Made run-fast-jsc print something and a newline if jsc threw an exception or
3982         otherwise returned with an error code, but did not crash.
3983
3984         * Scripts/run-fast-jsc:
3985
3986 2012-04-18  Dirk Pranke  <dpranke@chromium.org>
3987
3988         Fix crash in chromium.py after r114877.
3989
3990         Unreviewed, build fix.
3991
3992         Missed updating one call site for get_crash_log() :(.
3993
3994         * Scripts/webkitpy/layout_tests/port/chromium.py:
3995         (ChromiumDriver.run_test):
3996
3997 2012-04-18  Philippe Normand  <pnormand@igalia.com>
3998
3999         [GTK] NRWT bails out if more than one X server runs on the machine
4000         https://bugs.webkit.org/show_bug.cgi?id=84276
4001
4002         Reviewed by Dirk Pranke.
4003
4004         Set up GTK driver's Xvfb display number based on Xorg servers
4005         currently running.
4006
4007         * Scripts/webkitpy/layout_tests/port/gtk.py:
4008         (GtkDriver._start):
4009         (GtkDriver._start.x_filter):
4010
4011 2012-04-18  Dirk Pranke  <dpranke@chromium.org>
4012
4013         run-webkit-tests picked up an old crash log
4014         https://bugs.webkit.org/show_bug.cgi?id=84196
4015
4016         Reviewed by Ryosuke Niwa.
4017
4018         Modify the crash log-finding code so that we can
4019         limit our search to logs newer than X - this prevents
4020         us from finding old crash logs for a given process/pid
4021         combination even when the PIDs might've wrapped.
4022
4023         Also fix a bug in filesystem_mock where files_under() would
4024         return paths that didn't exist.
4025
4026         * Scripts/webkitpy/common/system/crashlogs.py:
4027         (CrashLogs.find_newest_log):
4028         (CrashLogs._find_newest_log_darwin):
4029         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
4030         (CrashLogsTest.test_find_log_darwin):
4031         (CrashLogsTest.test_find_log_darwin.bad_read):
4032         (CrashLogsTest):
4033         * Scripts/webkitpy/common/system/filesystem_mock.py:
4034         (MockFileSystem.files_under):
4035         * Scripts/webkitpy/layout_tests/port/base.py:
4036         (Port._get_crash_log):
4037         * Scripts/webkitpy/layout_tests/port/gtk.py:
4038         (GtkPort._get_crash_log):
4039         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
4040         * Scripts/webkitpy/layout_tests/port/mac.py:
4041         (MacPort._get_crash_log):
4042         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
4043         (PortTestCase.test_get_crash_log):
4044         * Scripts/webkitpy/layout_tests/port/webkit.py:
4045         (WebKitDriver.run_test):
4046
4047 2012-04-18  Dirk Pranke  <dpranke@chromium.org>
4048
4049         nrwt: support additional test expectations files
4050         https://bugs.webkit.org/show_bug.cgi?id=84222
4051
4052         Reviewed by Ojan Vafai.
4053
4054         This patch adds an --additional-expectations flag that can be
4055         used to point to more files that will override the default set
4056         of expectations. This will allow us to remove the
4057         'google_chrome' ports and eventually help to clean up override
4058         handling in the chromium ports and unify expectations files and
4059         Skipped files; this can also be used to specify overrides for
4060         tests that run differently on a given machine.
4061
4062         * Scripts/webkitpy/layout_tests/port/base.py:
4063         (Port.test_expectations_overrides):
4064         * Scripts/webkitpy/layout_tests/port/chromium.py:
4065         (ChromiumPort.test_expectations_overrides):
4066         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
4067         (parse_args):
4068         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
4069         (MainTest.test_additional_expectations):
4070
4071 2012-04-18  Jer Noble  <jer.noble@apple.com>
4072
4073         Lion Production Test failing with error: "Failed to stop httpd: pid file still exists"
4074         https://bugs.webkit.org/show_bug.cgi?id=84177
4075
4076         Reviewed by Dirk Pranke.
4077
4078         Try to delete the pid file left by Apache in the case it was killed forcefully.
4079
4080         * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
4081         (LayoutTestApacheHttpd._stop_running_server):
4082
4083 2012-04-18  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
4084
4085         Clean-up WheelEvent Conversion.
4086         https://bugs.webkit.org/show_bug.cgi?id=84243
4087
4088         Reviewed by Simon Hausmann.
4089
4090         Calculate proper wheel-delta from pixel-delta given.
4091
4092         * DumpRenderTree/qt/EventSenderQt.cpp:
4093         (EventSender::EventSender):
4094         (EventSender::mouseScrollBy):
4095         (EventSender::continuousMouseScrollBy):
4096
4097 2012-04-18  Sheriff Bot  <webkit.review.bot@gmail.com>
4098
4099         Unreviewed, rolling out r114506.
4100         http://trac.webkit.org/changeset/114506
4101         https://bugs.webkit.org/show_bug.cgi?id=84254
4102
4103         Seems to be making gtk's DumpRenderTree crash occasionally
4104         (Requested by tomz on #webkit).
4105
4106         * DumpRenderTree/gtk/DumpRenderTree.cpp:
4107         (resetDefaultsToConsistentValues):
4108
4109 2012-04-18  Thiago Marcos P. Santos  <thiago.santos@intel.com>
4110
4111         [CMake] Use jsc target instead of ONLY_BUILD_JAVASCRIPTCORE
4112         https://bugs.webkit.org/show_bug.cgi?id=84229
4113
4114         CMake caches ONLY_BUILD_JAVASCRIPTCORE=1 and will reuse the setting for
4115         subsequent builds, stopping WebCore from being built ever again.
4116
4117         Instead of trying to emulate a build target using command line options,
4118         just build the existing jsc target directly.
4119
4120         Reviewed by Rob Buis.
4121
4122         * Scripts/build-jsc:
4123
4124 2012-04-18  Simon Pena  <spena@igalia.com>
4125
4126         [GTK] r114021 triggered media flakyness
4127         https://bugs.webkit.org/show_bug.cgi?id=83874
4128
4129         Reviewed by Philippe Normand.
4130
4131         Now that there is gtk API to update the
4132         MediaPlaybackRequiresUserGesture property, set it to false from the
4133         DumpRenderTree tool when resetting defaults to consistent values.
4134
4135         * DumpRenderTree/gtk/DumpRenderTree.cpp:
4136         (resetDefaultsToConsistentValues):
4137
4138 2012-04-18  Vivek Galatage  <vivekgalatage@gmail.com>
4139
4140         Web Inspector: Disabling Inspector causes build failure on Windows
4141         https://bugs.webkit.org/show_bug.cgi?id=83557
4142
4143         Reviewed by Pavel Feldman.
4144
4145         The calls to inspector APIs are not put under ENABLE(INSPECTOR) guard.
4146         This change brings all such API calls under the guard.
4147
4148         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
4149         (WTR::LayoutTestController::showWebInspector):
4150         (WTR::LayoutTestController::closeWebInspector):
4151         (WTR::LayoutTestController::evaluateInWebInspector):
4152         (WTR::LayoutTestController::setJavaScriptProfilingEnabled):
4153         * WebKitTestRunner/TestInvocation.cpp:
4154         (WTR::TestInvocation::invoke):
4155
4156 2012-04-18  Mario Sanchez Prada  <msanchez@igalia.com>
4157
4158         [GTK] atk/textChangedNotifications API test fails
4159         https://bugs.webkit.org/show_bug.cgi?id=84010
4160
4161         Unreviewed, unskipping test after bug being fixed by r114311.
4162
4163         * Scripts/run-gtk-tests:
4164         (TestRunner): Unskip now passing test.
4165
4166 2012-04-17  Jacky Jiang  <zhajiang@rim.com>
4167
4168         [BlackBerry] Viewport metatag doesn't disable double-tap zoom
4169         https://bugs.webkit.org/show_bug.cgi?id=84199
4170
4171         Reviewed by George Staikos.
4172
4173         Revert a part of the change master_32/SHA:fa0f8ee9 and use
4174         WebPage::setUserScalable instead.
4175
4176         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
4177         (BlackBerry::WebKit::DumpRenderTree::resetToConsistentStateBeforeTesting):
4178
4179 2012-04-17  Dirk Pranke  <dpranke@chromium.org>
4180
4181         build-webkit: make ninja work with chromium linux
4182         https://bugs.webkit.org/show_bug.cgi?id=84213
4183
4184         Reviewed by Eric Seidel.
4185
4186         Fix typos so that build-webkit will use ninja to
4187         build chromium if a build.ninja file exists and is up-to-date.
4188
4189         * Scripts/webkitdirs.pm:
4190         (determineIsChromiumNinja):
4191
4192 2012-04-17  Ojan Vafai  <ojan@chromium.org>
4193
4194         Add caching to the flakiness dashboard json files
4195         https://bugs.webkit.org/show_bug.cgi?id=84198
4196
4197         Reviewed by Dirk Pranke.
4198
4199         * TestResultServer/handlers/testfilehandler.py:
4200         (GetFile._get_file_list):
4201         (GetFile._get_file_content):
4202         (GetFile._get_file_content_from_key):
4203         (GetFile._get_test_list_json):
4204         (GetFile._serve_json):
4205         Add a Last-Modified header and handle If-Modified-Since headers.
4206         (GetFile.get):
4207
4208 2012-04-17  Anders Carlsson  <andersca@apple.com>
4209
4210         Need a client callback for when the user tries to interact with an already unresponsive page
4211         https://bugs.webkit.org/show_bug.cgi?id=84201
4212         <rdar://problem/11140862>
4213
4214         Reviewed by Andreas Kling.
4215
4216         Add new client functions.
4217
4218         * MiniBrowser/mac/BrowserWindowController.m:
4219         (-[BrowserWindowController awakeFromNib]):
4220         * WebKitTestRunner/TestController.cpp:
4221         (WTR::TestController::initialize):
4222
4223 2012-04-17  Dirk Pranke  <dpranke@chromium.org>
4224
4225         nrwt: handle stopping layout test helper after a ctrl-c cleanly on the chromium port
4226         https://bugs.webkit.org/show_bug.cgi?id=84190
4227
4228         Reviewed by Ojan Vafai.
4229
4230         LayoutTestHelper would exit when the user typed ctrl-c, but we
4231         would attempt to write to it to shut it down, generating an
4232         IOError.
4233
4234         * Scripts/webkitpy/layout_tests/port/chromium.py:
4235         (ChromiumPort.stop_helper):
4236
4237 2012-04-17  Dirk Pranke  <dpranke@chromium.org>
4238
4239         rename --experimental-fully-parallel
4240         https://bugs.webkit.org/show_bug.cgi?id=84188
4241
4242         Reviewed by Ojan Vafai.
4243
4244         I thought this was done long ago; rename
4245         --experimental-fully-parallel to just --fully-parallel and
4246         remove the --no- switch that AFAIK no one uses.
4247
4248         * Scripts/webkitpy/layout_tests/controllers/manager.py:
4249         (Manager._run_tests):
4250         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
4251         (parse_args):
4252
4253 2012-04-17  Christophe Dumez  <christophe.dumez@intel.com>
4254
4255         EFL's LayoutTestController does not implement callShouldCloseOnWebView
4256         https://bugs.webkit.org/show_bug.cgi?id=82301
4257
4258         Reviewed by Martin Robinson.
4259
4260         Implement callShouldCloseOnWebView in EFL's LayoutTestController by
4261         calling shouldClose() on the FrameLoader, allowing the related test
4262         to be removed from the skip list.
4263
4264         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
4265         (LayoutTestController::callShouldCloseOnWebView):
4266
4267 2012-04-17  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
4268
4269         [EFL] Add setting API to enable/disable XSSAuditor
4270         https://bugs.webkit.org/show_bug.cgi?id=83281
4271
4272         Add missing implementation setXSSAuditorEnabled to EFL's LayoutTestController
4273         in order to unskip tests in http/tests/security/xssAuditor. Also, catch the signal
4274         'xss,detected' in DRT to enable a test, which is expecting a line containing
4275         'didDetectXSS' in the output when reflected XSS is encountered in the page.
4276
4277         Reviewed by Antonio Gomes.
4278
4279         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
4280         (DumpRenderTreeChrome::createView):
4281         (DumpRenderTreeChrome::onFrameCreated):
4282         (DumpRenderTreeChrome::onDidDetectXSS):
4283         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
4284         (DumpRenderTreeChrome):
4285         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
4286         (LayoutTestController::setXSSAuditorEnabled):
4287
4288 2012-04-17  Nandor Huszka  <hnandor@inf.u-szeged.hu>
4289
4290         [Qt] JSC build should handle --no-webkit2 option to avoid unwanted clean-builds
4291         https://bugs.webkit.org/show_bug.cgi?id=77533
4292
4293         Reviewed by Simon Hausmann.
4294
4295         Pick up the --no-webkit2 option from BUILD_WEBKIT_ARGS and pass on
4296         the corresponding parameter to the buildQMakeProjects function.
4297
4298         * Scripts/build-jsc:
4299
4300 2012-04-17  Christophe Dumez  <christophe.dumez@intel.com>
4301
4302         [EFL] DRT should support LayoutTestController's dumpFrameLoadCallbacks()
4303         https://bugs.webkit.org/show_bug.cgi?id=83030
4304
4305         Reviewed by Antonio Gomes.
4306
4307         Implement support for LayoutTestController's dumpFrameLoadCallbacks()
4308         in EFL's DumpRenderTree. Make use of the new frame argument passed to
4309         the "load,started" and "load,finished" signals.
4310
4311         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
4312         (DumpRenderTreeChrome::createView):
4313         (DumpRenderTreeChrome::onLoadStarted):
4314         (DumpRenderTreeChrome::processWork):
4315         (DumpRenderTreeChrome::topLoadingFrameLoadFinished):
4316         (DumpRenderTreeChrome::onWebViewOnloadEvent):
4317         (DumpRenderTreeChrome::onInsecureContentRun):
4318         (DumpRenderTreeChrome::onInsecureContentDisplayed):
4319         (DumpRenderTreeChrome::onFrameCreated):
4320         (DumpRenderTreeChrome::onFrameProvisionalLoad):
4321         (DumpRenderTreeChrome::onFrameLoadCommitted):
4322         (DumpRenderTreeChrome::onFrameLoadFinished):
4323         (DumpRenderTreeChrome::onFrameLoadError):
4324         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
4325         (DumpRenderTreeChrome):
4326
4327 2012-04-17  Sheriff Bot  <webkit.review.bot@gmail.com>
4328
4329         Unreviewed, rolling out r114376.