Tools/ChangeLog

   1 2012-05-11  Dirk Pranke  <dpranke@chromium.org>
   2
   3         webkit-patch rebaseline should fetch stale expectations :)
   4         https://bugs.webkit.org/show_bug.cgi?id=86231
   5
   6         Unreviewed, build fix.
   7
   8         This change reverts r115492 and goes back to fetching individual
   9         baselines from the layout tests directory, rather than fetching
  10         entire zip files. The performance hit over networks for people
  11         far away from the bots has turned out to be severe, so we need a
  12         better solution to this.
  13
  14         * Scripts/webkitpy/tool/commands/rebaseline.py:
  15         (RebaselineTest._results_url):
  16         (RebaselineTest._file_name_for_expected_result):
  17         (RebaselineTest._rebaseline_test):
  18         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
  19         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
  20         (test_rebaseline_updates_expectations_file):
  21         (test_rebaseline_test):
  22         (test_rebaseline_and_copy_test):
  23         (test_rebaseline_and_copy_test_no_existing_result):
  24         (test_rebaseline_and_copy_test_with_lion_result):
  25         (test_rebaseline_and_copy_no_overwrite_test):
  26
  27 2012-05-11  Christophe Dumez  <christophe.dumez@intel.com>
  28
  29         [GTK] LayoutTests/svg/batik/text/verticalText.svg fails to display Chinese characters
  30         https://bugs.webkit.org/show_bug.cgi?id=86198
  31
  32         Reviewed by Martin Robinson.
  33
  34         Update webkitgtk-fonts package to v0.0.3 in order to get new Kochi
  35         Mincho / Gothic fonts to display Japanese and Chinese characters.
  36
  37         * gtk/jhbuild.modules:
  38
  39 2012-05-11  Christophe Dumez  <christophe.dumez@intel.com>
  40
  41         [EFL][DRT] SVG Batik test fails to display Chinese characters
  42         https://bugs.webkit.org/show_bug.cgi?id=84461
  43
  44         Reviewed by Martin Robinson.
  45
  46         Update webkitgtk-fonts package from v0.0.1 to v0.0.3 in order to get
  47         new Kochi Mincho / Gothic fonts to display Japanese and Chinese
  48         characters.
  49
  50         * DumpRenderTree/efl/CMakeLists.txt:
  51         * efl/jhbuild.modules:
  52
  53 2012-05-11  Philippe Normand  <pnormand@igalia.com>
  54
  55         Unreviewed, GTK 64-bit Debug bot fix attempt.
  56
  57         * Scripts/webkitpy/layout_tests/port/gtk.py:
  58         (GtkPort._get_gdb_output): Ignore unicode decoding errors when
  59         processing gdb output.
  60
  61 2012-05-11  Gustavo Noronha Silva  <gns@gnome.org>
  62
  63         REGRESSION(r116205): [GTK]: build no longer supports thin archives
  64         https://bugs.webkit.org/show_bug.cgi?id=86207
  65
  66         * GNUmakefile.am: link libWebCoreModules to DRT.
  67
  68 2012-05-10  Ojan Vafai  <ojan@chromium.org>
  69
  70         Add a perf tab to garden-o-matic
  71         https://bugs.webkit.org/show_bug.cgi?id=86169
  72
  73         Reviewed by Dimitri Glazkov.
  74
  75         This is a first stab at making it easier to view the Chromium WebKit canary
  76         perf graphs. For now, it's just a couple select elements and an iframe.
  77
  78         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
  79         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
  80         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders.js:
  81         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
  82         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
  83         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
  84         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/perf.js: Added.
  85         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/perf_unittests.js: Added.
  86         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
  87         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/perf.css: Added.
  88         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
  89
  90 2012-05-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
  91
  92         Move resumeAnimations to use Internals interface
  93         https://bugs.webkit.org/show_bug.cgi?id=86063
  94
  95         Reviewed by Alexey Proskuryakov.
  96
  97         Remove resumeAnimations functions, because it is able to work in the
  98         cross-port way through the Internals interface.
  99
 100         * DumpRenderTree/LayoutTestController.cpp:
 101         (LayoutTestController::staticFunctions):
 102         * DumpRenderTree/LayoutTestController.h:
 103         (LayoutTestController):
 104         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
 105         * DumpRenderTree/chromium/LayoutTestController.cpp:
 106         (LayoutTestController::LayoutTestController):
 107         * DumpRenderTree/chromium/LayoutTestController.h:
 108         (LayoutTestController):
 109         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 110         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
 111         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
 112         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
 113         * DumpRenderTree/qt/LayoutTestControllerQt.h:
 114         (LayoutTestController):
 115         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
 116         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
 117         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
 118         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
 119         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
 120         (LayoutTestController):
 121
 122 2012-05-10  Alexey Proskuryakov  <ap@apple.com>
 123
 124         Crash in 3rd party WebKit apps that disable cache at a wrong time
 125         https://bugs.webkit.org/show_bug.cgi?id=86027
 126         <rdar://problem/10615880>
 127
 128         Reviewed by Antti Koivisto.
 129
 130         Added a test that's very similar to MemoryCachePruneWithinResourceLoadDelegate,
 131         but for disabling the cache instead of triggering a prune.
 132
 133         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
 134         * TestWebKitAPI/Tests/mac/MemoryCacheDisableWithinResourceLoadDelegate.html: Added.
 135         * TestWebKitAPI/Tests/mac/MemoryCacheDisableWithinResourceLoadDelegate.mm: Added.
 136         (-[MemoryCacheDisableTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
 137         (-[MemoryCacheDisableTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
 138         (-[MemoryCacheDisableTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
 139         (-[MemoryCacheDisableTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
 140         (TestWebKitAPI::TEST):
 141
 142 2012-05-10  Anders Carlsson  <andersca@apple.com>
 143
 144         WebKit2: Add a way to blacklist specific plug-ins/plug-in versions
 145         https://bugs.webkit.org/show_bug.cgi?id=86164
 146         <rdar://problem/9551196>
 147
 148         Reviewed by Sam Weinig.
 149
 150         Update for WK2 API changes.
 151
 152         * MiniBrowser/mac/BrowserWindowController.m:
 153         (-[BrowserWindowController awakeFromNib]):
 154         * WebKitTestRunner/TestController.cpp:
 155         (WTR::TestController::createOtherPage):
 156         (WTR::TestController::initialize):
 157
 158 2012-05-10  Dirk Pranke  <dpranke@chromium.org>
 159
 160         nrwt: [chromium] fix path to asan symbolize script
 161         https://bugs.webkit.org/show_bug.cgi?id=86161
 162
 163         Reviewed by Abhishek Arya.
 164
 165         * Scripts/webkitpy/layout_tests/port/chromium.py:
 166         (ChromiumPort._get_crash_log):
 167
 168 2012-05-10  Lei Zhang  <thestig@chromium.org>
 169
 170         [chromium] Set Valgrind environment variables for layout tests.
 171         https://bugs.webkit.org/show_bug.cgi?id=86156
 172
 173         Reviewed by Dirk Pranke.
 174
 175         * Scripts/webkitpy/layout_tests/port/chromium.py:
 176         (ChromiumPort.setup_environ_for_server):
 177
 178 2012-05-10  Raphael Kubo da Costa  <rakuco@webkit.org>
 179
 180         [webkitpy][EFL] It should be possible to use a wrapper command for ImageDiff.
 181         https://bugs.webkit.org/show_bug.cgi?id=86145
 182
 183         Reviewed by Dirk Pranke.
 184
 185         The GTK+ and EFL ports use jhbuild to build a few base libraries
 186         which are then used when building the port itself. Programs such
 187         as DRT and ImageDiff are then supposed to be run with the
 188         `run-with-jhbuild' wrapper so that the libraries are properly
 189         found and loaded.
 190
 191         This was already the case for DumpRenderTree (EflPort defines the
 192         'wrapper' option for that), but it was not possible to do the same
 193         with ImageDiff so far.
 194
 195         The code which created the ImageDiff command-line call has been
 196         moved to a separate method, _image_diff_command, which is then
 197         overridden by EflPort to add the jhbuild wrapper.
 198
 199         * Scripts/webkitpy/layout_tests/port/efl.py:
 200         (EflPort.__init__):
 201         (EflPort._image_diff_command):
 202         * Scripts/webkitpy/layout_tests/port/webkit.py:
 203         (WebKitPort):
 204         (WebKitPort._image_diff_command):
 205         (WebKitPort._start_image_diff_process):
 206
 207 2012-05-10  Zan Dobersek  <zandobersek@gmail.com>
 208
 209         [GTK] ENABLE_IFRAME_SEAMLESS support
 210         https://bugs.webkit.org/show_bug.cgi?id=85843
 211
 212         Reviewed by Eric Seidel.
 213
 214         Add the iframe-seamless option to the options list. This makes it
 215         possible to enable the iframe seamless support through build-webkit
 216         script. Enabled by default for all the ports.
 217
 218         * Scripts/webkitperl/FeatureList.pm:
 219
 220 2012-05-10  Raphael Kubo da Costa  <rakuco@webkit.org>
 221
 222         [EFL][webkitpy] Do not redefine check_build() in EflPort.
 223         https://bugs.webkit.org/show_bug.cgi?id=86124
 224
 225         Reviewed by Dirk Pranke.
 226
 227         Remove an old FIXME and do not redefine check_build in EflPort;
 228         once build-dumprendertree recognizes --efl, we can use the
 229         check_build implementation in WebKitPort.
 230
 231         * Scripts/build-dumprendertree:
 232         * Scripts/webkitpy/layout_tests/port/efl.py:
 233         (EflPort.check_build):
 234
 235 2012-05-10  Tommy Widenflycht  <tommyw@google.com>
 236
 237         [chromium] MediaStream API: Fix the ExtraData functionality in WebMediaStreamDescriptor
 238         https://bugs.webkit.org/show_bug.cgi?id=86087
 239
 240         Reviewed by Adam Barth.
 241
 242         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
 243         (MockExtraData):
 244         (WebKit):
 245         (WebKit::WebUserMediaClientMock::requestUserMedia):
 246
 247 2012-05-10  Varun Jain  <varunjain@google.com>
 248
 249         [chromium] Trigger context menu for long press gesture
 250         https://bugs.webkit.org/show_bug.cgi?id=85919
 251
 252         Reviewed by Adam Barth.
 253
 254         * DumpRenderTree/chromium/EventSender.cpp:
 255         (EventSender::EventSender):
 256         (EventSender::gestureLongPress):
 257         (EventSender::gestureEvent):
 258         * DumpRenderTree/chromium/EventSender.h:
 259         (EventSender):
 260
 261 2012-05-10  Takashi Toyoshima  <toyoshim@chromium.org>
 262
 263         don't use autoinstall to import pywebsocket but check it in WebKit directly.
 264         https://bugs.webkit.org/show_bug.cgi?id=86107
 265
 266         Reviewed by Adam Barth.
 267
 268         This change removes pywebsocket from webkitpy's autoinstall list and
 269         imports pywebsocket 0.7.4 directly into webkitpy/thirparty.
 270
 271         * Scripts/webkitpy/layout_tests/servers/websocket_server.py:
 272         (PyWebSocket._prepare_config):
 273         * Scripts/webkitpy/thirdparty/__init__.py:
 274         (AutoinstallImportHook.find_module):
 275         (AutoinstallImportHook._install_irc):
 276         * Scripts/webkitpy/thirdparty/__init___unittest.py:
 277         (ThirdpartyTest.test_import_hook.MockImportHook.__init__):
 278         (ThirdpartyTest.test_import_hook.MockImportHook._install_eliza):
 279         (ThirdpartyTest):
 280         (ThirdpartyTest.test_import_hook):
 281         * Scripts/webkitpy/thirdparty/mod_pywebsocket/COPYING: Added.
 282         * Scripts/webkitpy/thirdparty/mod_pywebsocket/__init__.py: Added.
 283         * Scripts/webkitpy/thirdparty/mod_pywebsocket/_stream_base.py: Added.
 284         (ConnectionTerminatedException):
 285         (InvalidFrameException):
 286         (BadOperationException):
 287         (UnsupportedFrameException):
 288         (InvalidUTF8Exception):
 289         (StreamBase):
 290         (StreamBase.__init__):
 291         (StreamBase._read):
 292         (StreamBase._write):
 293         (StreamBase.receive_bytes):
 294         (StreamBase._read_until):
 295         * Scripts/webkitpy/thirdparty/mod_pywebsocket/_stream_hixie75.py: Added.
 296         (for):
 297         (StreamHixie75):
 298         (StreamHixie75.for):
 299         (StreamHixie75.__init__):
 300         (StreamHixie75.send_message):
 301         (StreamHixie75._read_payload_length_hixie75):
 302         (StreamHixie75.receive_message):
 303         (StreamHixie75._send_closing_handshake):
 304         (StreamHixie75.close_connection):
 305         (StreamHixie75.send_ping):
 306         * Scripts/webkitpy/thirdparty/mod_pywebsocket/_stream_hybi.py: Added.
 307         (Frame):
 308         (Frame.__init__):
 309         (create_length_header):
 310         (create_header):
 311         (_build_frame):
 312         (_filter_and_format_frame_object):
 313         (create_binary_frame):
 314         (create_text_frame):
 315         (FragmentedFrameBuilder):
 316         (FragmentedFrameBuilder.to):
 317         (FragmentedFrameBuilder.__init__):
 318         (FragmentedFrameBuilder.build):
 319         (_create_control_frame):
 320         (create_ping_frame):
 321         (create_pong_frame):
 322         (create_close_frame):
 323         (StreamOptions):
 324         (StreamOptions.__init__):
 325         (Stream):
 326         (Stream.for):
 327         (Stream.__init__):
 328         (Stream._receive_frame):
 329         (Stream._receive_frame_as_frame_object):
 330         (Stream.send_message):
 331         (Stream.receive_message):
 332         (Stream._send_closing_handshake):
 333         (Stream.close_connection):
 334         (Stream.send_ping):
 335         (Stream._send_pong):
 336         (Stream._drain_received_data):
 337         * Scripts/webkitpy/thirdparty/mod_pywebsocket/common.py: Added.
 338         (is_control_opcode):
 339         (ExtensionParameter):
 340         (ExtensionParameter.__init__):
 341         (ExtensionParameter.name):
 342         (ExtensionParameter.add_parameter):
 343         (ExtensionParameter.get_parameters):
 344         (ExtensionParameter.get_parameter_names):
 345         (ExtensionParameter.has_parameter):
 346         (ExtensionParameter.get_parameter_value):
 347         * Scripts/webkitpy/thirdparty/mod_pywebsocket/dispatch.py: Added.
 348         (DispatchException):
 349         (DispatchException.__init__):
 350         (_default_passive_closing_handshake_handler):
 351         (_normalize_path):
 352         (_create_path_to_resource_converter):
 353         (_create_path_to_resource_converter.converter):
 354         (_enumerate_handler_file_paths):
 355         (_HandlerSuite):
 356         (_HandlerSuite.__init__):
 357         (_source_handler_file):
 358         (_extract_handler):
 359         (Dispatcher):
 360         (Dispatcher.maintains):
 361         (Dispatcher.__init__):
 362         (Dispatcher.add_resource_path_alias):
 363         (Dispatcher.source_warnings):
 364         (Dispatcher.do_extra_handshake):
 365         (Dispatcher.transfer_data):
 366         (Dispatcher.passive_closing_handshake):
 367         (Dispatcher.get_handler_suite):
 368         (Dispatcher._source_handler_files_in_dir):
 369         * Scripts/webkitpy/thirdparty/mod_pywebsocket/extensions.py: Added.
 370         (ExtensionProcessorInterface):
 371         (ExtensionProcessorInterface.get_extension_response):
 372         (ExtensionProcessorInterface.setup_stream_options):
 373         (DeflateStreamExtensionProcessor):
 374         (DeflateStreamExtensionProcessor.__init__):
 375         (DeflateStreamExtensionProcessor.get_extension_response):
 376         (DeflateStreamExtensionProcessor.setup_stream_options):
 377         (DeflateFrameExtensionProcessor):
 378         (DeflateFrameExtensionProcessor.__init__):
 379         (DeflateFrameExtensionProcessor.get_extension_response):
 380         (DeflateFrameExtensionProcessor.setup_stream_options):
 381         (DeflateFrameExtensionProcessor.setup_stream_options._OutgoingFilter):
 382         (DeflateFrameExtensionProcessor.setup_stream_options._OutgoingFilter.__init__):
 383         (DeflateFrameExtensionProcessor.setup_stream_options._OutgoingFilter.filter):
 384         (DeflateFrameExtensionProcessor.setup_stream_options._IncomingFilter):
 385         (DeflateFrameExtensionProcessor.setup_stream_options._IncomingFilter.__init__):
 386         (DeflateFrameExtensionProcessor.setup_stream_options._IncomingFilter.filter):
 387         (DeflateFrameExtensionProcessor.set_response_window_bits):
 388         (DeflateFrameExtensionProcessor.set_response_no_context_takeover):
 389         (DeflateFrameExtensionProcessor.enable_outgoing_compression):
 390         (DeflateFrameExtensionProcessor.disable_outgoing_compression):
 391         (DeflateFrameExtensionProcessor._outgoing_filter):
 392         (DeflateFrameExtensionProcessor._incoming_filter):
 393         (get_extension_processor):
 394         (get_extension_processor.is):
 395         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/__init__.py: Added.
 396         (try):
 397         (do_handshake):
 398         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/_base.py: Added.
 399         (AbortedByUserException):
 400         (HandshakeException):
 401         (HandshakeException.__init__):
 402         (VersionException):
 403         (VersionException.__init__):
 404         (get_default_port):
 405         (validate_subprotocol):
 406         (parse_host_header):
 407         (format_header):
 408         (build_location):
 409         (get_mandatory_header):
 410         (validate_mandatory_header):
 411         (check_request_line):
 412         (check_header_lines):
 413         (parse_token_list):
 414         (_parse_extension_param):
 415         (_parse_extension):
 416         (parse_extensions):
 417         (format_extensions):
 418         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/draft75.py: Added.
 419         (Handshaker):
 420         (Handshaker.performs):
 421         (Handshaker.__init__):
 422         (Handshaker.do_handshake):
 423         (Handshaker._set_resource):
 424         (Handshaker._set_origin):
 425         (Handshaker._set_location):
 426         (Handshaker._set_subprotocol):
 427         (Handshaker._set_protocol_version):
 428         (Handshaker._sendall):
 429         (Handshaker._send_handshake):
 430         (Handshaker._check_header_lines):
 431         (Handshaker._check_first_lines):
 432         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/hybi.py: Added.
 433         (compute_accept):
 434         (Handshaker):
 435         (Handshaker.__init__):
 436         (Handshaker._validate_connection_header):
 437         (Handshaker.do_handshake):
 438         (Handshaker._get_origin):
 439         (Handshaker._check_version):
 440         (Handshaker._set_protocol):
 441         (Handshaker._parse_extensions):
 442         (Handshaker._validate_key):
 443         (Handshaker._get_key):
 444         (Handshaker._send_handshake):
 445         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/hybi00.py: Added.
 446         (Handshaker):
 447         (Handshaker.__init__):
 448         (Handshaker.do_handshake):
 449         (Handshaker._set_resource):
 450         (Handshaker._set_subprotocol):
 451         (Handshaker._set_location):
 452         (Handshaker._set_origin):
 453         (Handshaker._set_protocol_version):
 454         (Handshaker._set_challenge_response):
 455         (Handshaker._get_key_value):
 456         (Handshaker._get_challenge):
 457         (Handshaker._send_handshake):
 458         * Scripts/webkitpy/thirdparty/mod_pywebsocket/headerparserhandler.py: Added.
 459         (ApacheLogHandler):
 460         (ApacheLogHandler.__init__):
 461         (ApacheLogHandler.emit):
 462         (_configure_logging):
 463         (_parse_option):
 464         (_create_dispatcher):
 465         (headerparserhandler):
 466         * Scripts/webkitpy/thirdparty/mod_pywebsocket/http_header_util.py: Added.
 467         (_is_char):
 468         (_is_ctl):
 469         (ParsingState):
 470         (ParsingState.__init__):
 471         (peek):
 472         (consume):
 473         (consume_string):
 474         (consume_lws):
 475         (consume_lwses):
 476         (consume_token):
 477         (consume_token_or_quoted_string):
 478         (quote_if_necessary):
 479         (parse_uri):
 480         * Scripts/webkitpy/thirdparty/mod_pywebsocket/memorizingfile.py: Added.
 481         (MemorizingFile):
 482         (MemorizingFile.__init__):
 483         (MemorizingFile.__getattribute__):
 484         (MemorizingFile.readline):
 485         (MemorizingFile.get_memorized_lines):
 486         * Scripts/webkitpy/thirdparty/mod_pywebsocket/msgutil.py: Added.
 487         (close_connection):
 488         (send_message):
 489         (receive_message):
 490         (send_ping):
 491         (MessageReceiver):
 492         (MessageReceiver.receives):
 493         (MessageReceiver.provides):
 494         (MessageReceiver.should):
 495         (MessageReceiver.__init__):
 496         (MessageReceiver.run):
 497         (MessageReceiver.receive):
 498         (MessageReceiver.receive_nowait):
 499         (MessageReceiver.stop):
 500         (MessageSender):
 501         (MessageSender.sends):
 502         (MessageSender.provides):
 503         (MessageSender.should):
 504         (MessageSender.__init__):
 505         (MessageSender.run):
 506         (MessageSender.send):
 507         (MessageSender.send_nowait):
 508         * Scripts/webkitpy/thirdparty/mod_pywebsocket/standalone.py: Added.
 509         (_StandaloneConnection):
 510         (_StandaloneConnection.__init__):
 511         (_StandaloneConnection.get_local_addr):
 512         (_StandaloneConnection.get_remote_addr):
 513         (_StandaloneConnection.write):
 514         (_StandaloneConnection.read):
 515         (_StandaloneConnection.get_memorized_lines):
 516         (_StandaloneRequest):
 517         (_StandaloneRequest.__init__):
 518         (_StandaloneRequest.get_uri):
 519         (_StandaloneRequest.get_method):
 520         (_StandaloneRequest.is_https):
 521         (_StandaloneRequest._drain_received_data):
 522         (_StandaloneSSLConnection):
 523         (_StandaloneSSLConnection.for):
 524         (_StandaloneSSLConnection.__init__):
 525         (_StandaloneSSLConnection.__getattribute__):
 526         (_StandaloneSSLConnection.__setattr__):
 527         (_StandaloneSSLConnection.makefile):
 528         (WebSocketServer):
 529         (WebSocketServer.__init__):
 530         (WebSocketServer._create_sockets):
 531         (WebSocketServer.server_bind):
 532         (WebSocketServer.server_activate):
 533         (WebSocketServer.server_close):
 534         (WebSocketServer.fileno):
 535         (WebSocketServer.handle_error):
 536         (WebSocketServer.get_request):
 537         (WebSocketServer.serve_forever):
 538         (WebSocketServer.shutdown):
 539         (WebSocketRequestHandler):
 540         (WebSocketRequestHandler.setup):
 541         (WebSocketRequestHandler.setup.SocketServer):
 542         (WebSocketRequestHandler.__init__):
 543         (WebSocketRequestHandler.parse_request):
 544         (WebSocketRequestHandler.log_request):
 545         (WebSocketRequestHandler.log_error):
 546         (WebSocketRequestHandler.is_cgi):
 547         (_configure_logging):
 548         (_alias_handlers):
 549         (_build_option_parser):
 550         (ThreadMonitor):
 551         (ThreadMonitor.__init__):
 552         (ThreadMonitor.run):
 553         (_parse_args_and_config):
 554         (_main):
 555         (_main.if):
 556         (_main.if.__check_script):
 557         * Scripts/webkitpy/thirdparty/mod_pywebsocket/stream.py: Copied from Tools/Scripts/webkitpy/thirdparty/__init___unittest.py.
 558         * Scripts/webkitpy/thirdparty/mod_pywebsocket/util.py: Added.
 559         (get_stack_trace):
 560         (prepend_message_to_exception):
 561         (__translate_interp):
 562         (get_script_interp):
 563         (wrap_popen3_for_win):
 564         (wrap_popen3_for_win.__wrap_popen3):
 565         (hexify):
 566         (get_class_logger):
 567         (NoopMasker):
 568         (NoopMasker.__init__):
 569         (NoopMasker.mask):
 570         (RepeatedXorMasker):
 571         (RepeatedXorMasker.__init__):
 572         (RepeatedXorMasker.mask):
 573         (DeflateRequest):
 574         (DeflateRequest.for):
 575         (DeflateRequest.__init__):
 576         (DeflateRequest.__getattribute__):
 577         (DeflateRequest.__setattr__):
 578         (_Deflater):
 579         (_Deflater.__init__):
 580         (_Deflater.compress_and_flush):
 581         (_Inflater):
 582         (_Inflater.__init__):
 583         (_Inflater.decompress):
 584         (_Inflater.append):
 585         (_Inflater.reset):
 586         (_RFC1979Deflater):
 587         (_RFC1979Deflater.that):
 588         (_RFC1979Deflater.__init__):
 589         (_RFC1979Deflater.filter):
 590         (_RFC1979Inflater):
 591         (_RFC1979Inflater.for):
 592         (_RFC1979Inflater.__init__):
 593         (_RFC1979Inflater.filter):
 594         (DeflateSocket):
 595         (DeflateSocket.for):
 596         (DeflateSocket.__init__):
 597         (DeflateSocket.recv):
 598         (DeflateSocket.sendall):
 599         (DeflateSocket.send):
 600         (DeflateConnection):
 601         (DeflateConnection.for):
 602         (DeflateConnection.__init__):
 603         (DeflateConnection.get_remote_addr):
 604         (DeflateConnection.put_bytes):
 605         (DeflateConnection.read):
 606         (DeflateConnection.write):
 607         (_is_ewouldblock_errno):
 608         (drain_received_data):
 609
 610 2012-05-10  Raphael Kubo da Costa  <rakuco@webkit.org>
 611
 612         [EFL][webkitpy] Define the 'wrapper' option in EflPort instead of creating a EflDriver.
 613         https://bugs.webkit.org/show_bug.cgi?id=86117
 614
 615         Reviewed by Adam Barth.
 616
 617         Simplify the code used to wrap driver calls with jhbuild: we can
 618         just define the 'wrapper' option instead of having to create a
 619         Driver class just for that.
 620
 621         * Scripts/webkitpy/layout_tests/port/efl.py:
 622         (EflPort.__init__):
 623         (EflPort._port_flag_for_scripts):
 624
 625 2012-05-10  Sheriff Bot  <webkit.review.bot@gmail.com>
 626
 627         Unreviewed, rolling out r116633.
 628         http://trac.webkit.org/changeset/116633
 629         https://bugs.webkit.org/show_bug.cgi?id=86093
 630
 631         Breaks EFL tests (Requested by kov on #webkit).
 632
 633         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 634         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
 635         * DumpRenderTree/efl/EventSender.cpp:
 636         (keyDownCallback):
 637         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 638         (LayoutTestController::setTabKeyCyclesThroughElements):
 639
 640 2012-05-10  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
 641
 642         [EFL] Memory leaks found in EFL DRT
 643         https://bugs.webkit.org/show_bug.cgi?id=85889
 644
 645         Reviewed by Gustavo Noronha Silva.
 646
 647         Fix two memory leaks in EFL's DRT by using JSRetainPtr for JSString
 648         created with JSStringCreateWithUTF8CString.
 649
 650         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 651         (DumpRenderTreeChrome::onWindowObjectCleared):
 652         * DumpRenderTree/efl/EventSender.cpp:
 653         (modifiersFromJSValue):
 654
 655 2012-05-10  Alexander Shalamov  <alexander.shalamov@intel.com>
 656
 657         [EFL] LayoutTestController needs implementation of setTabKeyCyclesThroughElements
 658         https://bugs.webkit.org/show_bug.cgi?id=82864
 659
 660         Reviewed by Gustavo Noronha Silva.
 661
 662         Add missing implementation for setTabKeyCyclesThroughElements to EFL's LayoutTestController
 663         in order to unskip fast/events/keypress-insert-tab.html
 664
 665         * DumpRenderTree/efl/EventSender.cpp:
 666         (keyDownCallback):
 667         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 668         (LayoutTestController::setTabKeyCyclesThroughElements):
 669
 670 2012-05-10  Alexander Shalamov  <alexander.shalamov@intel.com>
 671
 672         [EFL][DRT] EventSender - set delayed event information if it was not set before
 673         https://bugs.webkit.org/show_bug.cgi?id=85972
 674
 675         Reviewed by Gustavo Noronha Silva.
 676
 677         Reset delayed event information if it is missing.
 678
 679         * DumpRenderTree/efl/EventSender.cpp:
 680         (feedOrQueueMouseEvent):
 681
 682 2012-05-07  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
 683
 684         WebPageProxy::activeURL() should return the pending API request, even when there's no main frame
 685         https://bugs.webkit.org/show_bug.cgi?id=85806
 686
 687         The m_pendingAPIRequestURL member is used (presumably) to mask over the async
 688         nature of WebKit2, so that starting a load of a URL will reflect that URL
 689         immedeatly from activeURL, even if the request has not been passed over to
 690         the web process yet and reflected there.
 691
 692         This works well, except in the case of the initial request, where the main
 693         frame creation happens on the web process side and is notified back to the
 694         UI process. Until we've recived the notification we don't know about the main
 695         frame, and this race condition will potentially give us an empty url instead
 696         of the pending request.
 697
 698         To solve this we always return the pending API request if it's set, even
 699         when there's no mainframe yet (that we known about).
 700
 701         Reviewed by Simon Hausmann.
 702
 703         * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp:
 704
 705 2012-05-10  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
 706
 707         [EFL][DRT] Clear added user style sheets before a new testcase execution.
 708         https://bugs.webkit.org/show_bug.cgi?id=84792
 709
 710         Reviewed Gustavo Noronha Silva.
 711
 712         The added user style sheets from the previous testcases are cleared in order to avoid
 713         flaky results.
 714
 715         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
 716         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
 717
 718 2012-05-09  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
 719
 720         Move suspendAnimations to use Internals interface.
 721         https://bugs.webkit.org/show_bug.cgi?id=85986
 722
 723         Reviewed by Ryosuke Niwa.
 724
 725         Remove suspendAnimations functions, because it is able to work in the
 726         cross-port way through the Internals interface
 727
 728         * DumpRenderTree/LayoutTestController.cpp:
 729         (LayoutTestController::staticFunctions):
 730         * DumpRenderTree/LayoutTestController.h:
 731         (LayoutTestController):
 732         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
 733         * DumpRenderTree/chromium/LayoutTestController.cpp:
 734         (LayoutTestController::LayoutTestController):
 735         * DumpRenderTree/chromium/LayoutTestController.h:
 736         (LayoutTestController):
 737         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
 738         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
 739         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
 740         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
 741         * DumpRenderTree/qt/LayoutTestControllerQt.h:
 742         (LayoutTestController):
 743         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
 744         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
 745
 746 2012-05-09  Kent Tamura  <tkent@chromium.org>
 747
 748         [Chromium] Attempt to fix an Android build error.
 749         https://bugs.webkit.org/show_bug.cgi?id=86045
 750
 751         * DumpRenderTree/DumpRenderTree.gypi:
 752
 753 2012-05-09  Ojan Vafai  <ojan@chromium.org>
 754
 755         Don't show partytime if there are non-layout test failures.
 756         https://bugs.webkit.org/show_bug.cgi?id=86056
 757
 758         Reviewed by Dimitri Glazkov.
 759
 760         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
 761         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers_unittests.js:
 762         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
 763
 764 2012-05-09  Hayato Ito  <hayato@chromium.org>
 765
 766         Improve a warning message for a reftest having an unused expectation file.
 767         https://bugs.webkit.org/show_bug.cgi?id=85975
 768
 769         Reviewed by Adam Barth.
 770
 771         The current warning message is not intuitive. It might be better to update the message
 772         so that it encourage users to remove unused expectation files.
 773
 774         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
 775         (SingleTestRunner.__init__):
 776
 777 2012-05-09  Ojan Vafai  <ojan@chromium.org>
 778
 779         Change "losing test coverage" to cover all non-layout test problems in garden-o-matic
 780         https://bugs.webkit.org/show_bug.cgi?id=86043
 781
 782         Reviewed by Adam Barth.
 783
 784         -Get rid of the gtest iframe.
 785         -Cover all non-layout test problems. Before we wouldn't catch compile failures
 786         or gtest failures since they were not on the webkit testing bots.
 787         -Include the list of failing steps next to the builder name. This is a lot less
 788         cluttered than you'd expect and already helped me find a compile failure early.
 789
 790         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders.js:
 791         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
 792         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
 793         Delete unused map.
 794         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
 795         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
 796         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
 797         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js:
 798         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
 799         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js:
 800         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
 801         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
 802
 803 2012-05-09  Jessie Berlin  <jberlin@apple.com>
 804
 805         Crash using the new WKBundleDOMWindowExtensions APIs.
 806         https://bugs.webkit.org/show_bug.cgi?id=85888
 807
 808         Reviewed by Brady Eidson.
 809
 810         Cached frames can live slightly longer than the page, but most clients unregister themselves
 811         and do other cleanup in the willDestroyPage callback, making them miss the
 812         willDestroyGlobalObjectForDOMWindowExtension callbacks.
 813
 814         The calls to willDestroyGlobalObjectForDOMWindowExtension in the DOMWindowExtensionBasic
 815         test were all being invoked underneath WebPage::close. This is unrealistic. Update that test
 816         to destroy the BundleDOMWindowExtensions in response to the willDestroyPage callback.
 817
 818         Add a test to verify that willDestroyGlobalObjectForDOMWindowExtension is being called for
 819         pages that don't go into the page cache.
 820
 821         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
 822         Add DOMWindowExtensionNoCache.cpp, DOMWindowExtensionNoCache_Bundle.cpp, simple-unload.html
 823         and simple-iframe-unload.html
 824
 825         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic.cpp:
 826         (TestWebKitAPI):
 827         Remove the expected messages for willDestroyGlobalObjectForDOMWindowExtension.
 828         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
 829         Do not bother to keep track of the live extension count - all of them are expected to be
 830         live until the test completes.
 831         (TestWebKitAPI::TEST):
 832         Fix the calls to EXPECT to pass the expected value first, and use EXPECT_WK_STREQ so that
 833         message failures will be clearer.
 834         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp:
 835         (TestWebKitAPI::DOMWindowExtensionBasic::willDestroyPage):
 836         Clean up the BundleDOMWindowExtensions.
 837         (TestWebKitAPI::DOMWindowExtensionBasic::willDestroyGlobalObjectForDOMWindowExtension):
 838         Add an ASSERT_NOT_REACHED.
 839
 840         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionNoCache.cpp: Added.
 841         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
 842         Keep track of the messages received so they can be checked at the end of the test.
 843         (TestWebKitAPI::TEST):
 844         Navigate to uncacheable pages and back.
 845         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionNoCache_Bundle.cpp: Copied from Tools/TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp.
 846         (DOMWindowExtensionNoCache):
 847         (TestWebKitAPI::DOMWindowExtensionNoCache::DOMWindowExtensionNoCache):
 848         Set up all the states for each BundleDOMWindowExtension.
 849         (TestWebKitAPI::DOMWindowExtensionNoCache::frameLoadFinished):
 850         Tell the UI Process about the states of the BundleDOMWindowExtensions.
 851         (TestWebKitAPI::DOMWindowExtensionNoCache::sendExtensionStateMessage):
 852         (TestWebKitAPI::DOMWindowExtensionNoCache::initialize):
 853         (TestWebKitAPI::DOMWindowExtensionNoCache::didCreatePage):
 854         (TestWebKitAPI::DOMWindowExtensionNoCache::willDestroyPage):
 855         Remvoe the remaining BundleDOMWindowExtensions, send the updated state, and finish the test.
 856         (TestWebKitAPI::DOMWindowExtensionNoCache::updateExtensionStateRecord):
 857         (TestWebKitAPI::DOMWindowExtensionNoCache::sendBundleMessage):
 858         (TestWebKitAPI::DOMWindowExtensionNoCache::globalObjectIsAvailableForFrame):
 859         (TestWebKitAPI::DOMWindowExtensionNoCache::willDisconnectDOMWindowExtensionFromGlobalObject):
 860         ASSERT that these pages not going into the page cache are not getting disconnected to go into
 861         the page cache.
 862         (TestWebKitAPI::DOMWindowExtensionNoCache::didReconnectDOMWindowExtensionToGlobalObject):
 863         Ditto about getting reconnected when coming out of the page cache.
 864         (TestWebKitAPI::DOMWindowExtensionNoCache::willDestroyGlobalObjectForDOMWindowExtension):
 865         Tell the UI Process, update the state, and get rid of the BundleDOMWindowExtension.
 866         (TestWebKitAPI::didFinishLoadForFrameCallback):
 867         (TestWebKitAPI::globalObjectIsAvailableForFrameCallback):
 868         (TestWebKitAPI::willDisconnectDOMWindowExtensionFromGlobalObjectCallback):
 869         (TestWebKitAPI::didReconnectDOMWindowExtensionToGlobalObjectCallback):
 870         (TestWebKitAPI::willDestroyGlobalObjectForDOMWindowExtensionCallback):
 871
 872         * TestWebKitAPI/Tests/WebKit2/simple-iframe-unload.html: Added.
 873         * TestWebKitAPI/Tests/WebKit2/simple-unload.html: Added.
 874
 875 2012-05-09  Adam Barth  <abarth@webkit.org>
 876
 877         [Chromium] Android's DRT doesn't want to build WebUserMediaClientMock
 878         https://bugs.webkit.org/show_bug.cgi?id=86045
 879
 880         Reviewed by Eric Seidel.
 881
 882         * DumpRenderTree/DumpRenderTree.gypi:
 883
 884 2012-05-09  Adam Barth  <abarth@webkit.org>
 885
 886         [Chromium] MockWebSpeechInputController.cpp should be guarded by ENABLE(INPUT_SPEECH)
 887         https://bugs.webkit.org/show_bug.cgi?id=86041
 888
 889         Reviewed by Eric Seidel.
 890
 891         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp:
 892
 893 2012-05-09  Adam Barth  <abarth@webkit.org>
 894
 895         [Chromium] Android needs to swizzle pixel dumps in order to get the same hash as other platforms
 896         https://bugs.webkit.org/show_bug.cgi?id=86040
 897
 898         Reviewed by Kent Tamura.
 899
 900         * DumpRenderTree/chromium/TestShell.cpp:
 901         (TestShell::dumpImage):
 902
 903 2012-05-09  Raphael Kubo da Costa  <rakuco@webkit.org>
 904
 905         [EFL] Build fix after r116464.
 906
 907         `ulong' is a type defined by glibc, so it is not very
 908         portable. Use `unsigned long' instead.
 909
 910         * DumpRenderTree/efl/EventSender.cpp:
 911         (DelayedEvent::DelayedEvent):
 912         (DelayedEvent):
 913         (leapForwardCallback):
 914
 915 2012-05-09  Adam Barth  <abarth@webkit.org>
 916
 917         [Chromium] LayoutTestController should use ENABLE guards for some features
 918         https://bugs.webkit.org/show_bug.cgi?id=86030
 919
 920         Reviewed by Eric Seidel.
 921
 922         These features are disabled on OS(ANDROID) and so shouldn't be compiled
 923         when the feature is off.
 924
 925         * DumpRenderTree/chromium/LayoutTestController.cpp:
 926         (LayoutTestController::LayoutTestController):
 927         * DumpRenderTree/chromium/LayoutTestController.h:
 928         (LayoutTestController):
 929         * DumpRenderTree/chromium/TestShell.h:
 930         (TestShell):
 931
 932 2012-05-09  Tom Zakrajsek  <tomz@codeaurora.org>
 933
 934         Fix overzealous re-opened since blocked by... message
 935         https://bugs.webkit.org/show_bug.cgi?id=86020
 936
 937         Reviewed by Adam Barth.
 938
 939         Changed CreateBug behavior to only re-open a blocked bug if it
 940         was RESOLVED.  Also had to update the test infrastructure slightly
 941         to allow for distinct mock commit revisions, so one could point
 942         to a RESOLVED bug for testing while another pointed to an OPEN bug.
 943
 944         Also moved default created mock bug to be 60001 instead of 50004
 945         which already exists as a reference mock bug.
 946
 947         * Scripts/webkitpy/common/checkout/checkout_mock.py:
 948         (MockCheckout.__init__):
 949         (MockCheckout.commit_info_for_revision):
 950         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
 951         (MockBugzilla.create_bug):
 952         * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py:
 953         * Scripts/webkitpy/tool/commands/download_unittest.py:
 954         (test_create_rollout_resolved):
 955         * Scripts/webkitpy/tool/commands/queries_unittest.py:
 956         (FailureReasonTest.test_blame_line_for_revision):
 957         * Scripts/webkitpy/tool/steps/createbug.py:
 958         (CreateBug.run):
 959
 960 2012-05-09  Hugo Parente Lima  <hugo.lima@openbossa.org>
 961
 962         Use suitable viewport values on XHTML-MP pages.
 963         https://bugs.webkit.org/show_bug.cgi?id=85425
 964
 965         Reviewed by Kenneth Rohde Christiansen.
 966
 967         Enable LEGACY_VIEWPORT_ADAPTION on Qt port.
 968
 969         * qmake/mkspecs/features/features.prf:
 970
 971 2012-05-09  Jochen Eisinger  <jochen@chromium.org>
 972
 973         When creating a new page during a navigation, prime the initial document with the correct referrer policy
 974         https://bugs.webkit.org/show_bug.cgi?id=86001
 975
 976         Reviewed by Adam Barth.
 977
 978         Store the referrer policy in the request, so the network stack can comply with it.
 979
 980         * DumpRenderTree/chromium/WebViewHost.cpp:
 981         (WebViewHost::willSendRequest):
 982
 983 2012-05-09  Vivek Galatage  <vivekgalatage@gmail.com>
 984
 985         Remove CYGWIN=tty from environment variable as its no longer supported
 986         https://bugs.webkit.org/show_bug.cgi?id=85791
 987
 988         The CYGWIN=tty environment variable is no longer supported from Cygwin
 989         version 1.7.10. This fix avoids addition of this environment variable
 990         for Cygwin versions greater than 1.7.9. For supporting older versions
 991         i.e. 1.7.9 or older, the script takes care of retaining the environment
 992         variable.
 993
 994         Reviewed by Adam Roben.
 995
 996         * Scripts/webkitdirs.pm:
 997         (setupAppleWinEnv):
 998
 999 2012-05-09  Dirk Pranke  <dpranke@chromium.org>
1000
1001         nrwt: --skip-failing-tests should also skip flaky tests
1002         https://bugs.webkit.org/show_bug.cgi?id=85941
1003
1004         Reviewed by Ojan Vafai.
1005
1006         This changes --skip-failing-tests so that we skip tests that
1007         are believed to pass sometimes as well as tests that are
1008         believed to always fail.
1009
1010         In addition, in testing this I discovered that we had lousy
1011         test coverage of the details of retrying test failures and
1012         --clobber-old-results, so I wrote some more tests.
1013
1014         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1015         (Manager.prepare_lists_and_print_output):
1016         * Scripts/webkitpy/layout_tests/port/test.py:
1017         (unit_test_list):
1018         (TestPort.__init__):
1019         (TestDriver.run_test):
1020         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1021         (MainTest.test_skip_failing_tests):
1022         (MainTest.test_retrying_and_flaky_tests):
1023
1024 2012-05-09  Christophe Dumez  <christophe.dumez@intel.com>
1025
1026         [EFL][DRT] EFL's DRT needs to support LayoutTestController.dumpIconChanges()
1027         https://bugs.webkit.org/show_bug.cgi?id=84435
1028
1029         Reviewed by Antonio Gomes.
1030
1031         Catch the "icon,changed" signal on the frames in EFL's DumpRenderTree
1032         to support LayoutTestController's dumpIconChanges().
1033
1034         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1035         (DumpRenderTreeChrome::createView):
1036         (DumpRenderTreeChrome::onFrameIconChanged):
1037         (DumpRenderTreeChrome::onFrameCreated):
1038         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
1039         (DumpRenderTreeChrome):
1040
1041 2012-05-09  Dominik Röttsches  <dominik.rottsches@intel.com>
1042
1043         webkit-patch upload Fails Behind Load-balancing Proxies
1044         https://bugs.webkit.org/show_bug.cgi?id=85980
1045
1046         Reviewed by Adam Barth.
1047
1048         Unchecking the login restriction to IP to allow the tool
1049         to function behind load-balancing proxies that use multiple
1050         outgoing IPs.
1051
1052         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1053         (Bugzilla.authenticate):
1054
1055 2012-05-09  Hao Zheng  <zhenghao@chromium.org>
1056
1057         [Chromium-android] Add unit test for Android layout test script (chromium_android.py)
1058         https://bugs.webkit.org/show_bug.cgi?id=80852
1059
1060         Reviewed by Adam Barth.
1061
1062         Add some tests to ensure that chromium android port works to some
1063         extent. Some methods is hard to test, as they rely on the actual
1064         behavior on device, e.g. pushing files to device.
1065
1066         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1067         (ChromiumAndroidPort.get_last_stacktrace): Check if dir exists.
1068         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
1069         (ChromiumAndroidPortTest):
1070         (ChromiumAndroidPortTest.test_attributes):
1071         (ChromiumAndroidPortTest.test_get_last_stacktrace):
1072         (ChromiumAndroidPortTest.test_get_last_stacktrace.mock_run_command_no_dir):
1073         (ChromiumAndroidPortTest.test_get_last_stacktrace.mock_run_command_no_file):
1074         (ChromiumAndroidPortTest.test_get_last_stacktrace.mock_run_command_non_empty):
1075         (ChromiumAndroidDriverTest):
1076         (ChromiumAndroidDriverTest.setUp):
1077         (ChromiumAndroidDriverTest.test_get_drt_return_value):
1078         (ChromiumAndroidDriverTest.test_has_crash_hint):
1079         (ChromiumAndroidDriverTest.test_read_prompt):
1080         (ChromiumAndroidDriverTest.test_test_shell_command):
1081         (ChromiumAndroidDriverTest.test_write_command_and_read_line):
1082
1083 2012-05-09  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1084
1085         [Qt] ORWT fails with Qt 5 and WK2.
1086         https://bugs.webkit.org/show_bug.cgi?id=85968
1087
1088         Reviewed by Csaba Osztrogonác.
1089
1090         The Qt5 binary does not have a -style argument, so avoid using it.
1091
1092         * Scripts/old-run-webkit-tests:
1093
1094 2012-05-09  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1095
1096         [EFL][DRT] LayoutTestController addUserScript implementation
1097         https://bugs.webkit.org/show_bug.cgi?id=84718
1098
1099         Reviewed by Antonio Gomes.
1100
1101         Added implementation of addUserScript() to LayoutTestController.
1102
1103         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1104         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): clear added user scripts
1105         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1106         (LayoutTestController::addUserScript): added implementation
1107
1108 2012-05-09  Dan Bernstein  <mitz@apple.com>
1109
1110         Made build-webkit copy WebKitLibraries to the right location again after r116257
1111
1112         Reviewed by Mark Rowe.
1113
1114         * Scripts/build-webkit: Pass the built product directory to copy-webkitlibraries-to-product-directory.
1115         * Scripts/copy-webkitlibraries-to-product-directory: Use the first argument, if present, as the
1116         product directory.
1117
1118 2012-05-08  Ojan Vafai  <ojan@chromium.org>
1119
1120         Show the chromium non-layout test bots in garden-o-matic
1121         https://bugs.webkit.org/show_bug.cgi?id=85903
1122
1123         Reviewed by Eric Seidel.
1124
1125         This is just a single-line high-level overview of the chromium non-layout test bots.
1126         This means that Chromium WebKit gardeners only need to keep garden-o-matic open in
1127         order to do their job.
1128
1129         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1130         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1131         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1132         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
1133         (#chromium-gtests iframe):
1134         (#chromium-gtests):
1135         (#gtest-label):
1136
1137 2012-05-08  Kangil Han  <kangil.han@samsung.com>
1138
1139         [EFL][DRT] Implement TextInputController.
1140         https://bugs.webkit.org/show_bug.cgi?id=78559
1141
1142         Reviewed by Martin Robinson.
1143
1144         TextInputController has been created for an input emulation purpose in DRT.
1145         However, not like other ports, EFL doesn't have it.
1146         Therefore, this patch will enable EFL port to have an input emulation in DRT.
1147
1148         * DumpRenderTree/efl/CMakeLists.txt:
1149         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1150         (DumpRenderTreeChrome::onWindowObjectCleared):
1151         * DumpRenderTree/efl/TextInputController.cpp: Added.
1152         (setMarkedTextCallback):
1153         (hasMarkedTextCallback):
1154         (markedRangeCallback):
1155         (insertTextCallback):
1156         (unmarkTextCallback):
1157         (firstRectForCharacterRangeCallback):
1158         (selectedRangeCallback):
1159         (makeTextInputController):
1160         * DumpRenderTree/efl/TextInputController.h: Added.
1161
1162 2012-04-23  Raphael Kubo da Costa  <rakuco@webkit.org>
1163
1164         [EFL] Pass a WTF::String directly from DRT to ewebkit when possible
1165         https://bugs.webkit.org/show_bug.cgi?id=84634
1166
1167         Reviewed by Gustavo Noronha Silva.
1168
1169         Drop unnecessary const char* -> WTF::String conversions by
1170         creating String objects directly in LayoutTestController out of
1171         JSStringRef's and passing them to DumpRenderTreeSupportEfl.
1172
1173         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1174         (LayoutTestController::queueLoad):
1175         (LayoutTestController::setValueForUser):
1176         (LayoutTestController::findString):
1177         (LayoutTestController::addUserStyleSheet):
1178
1179 2012-05-08  Alexander Shalamov  <alexander.shalamov@intel.com>
1180
1181         [EFL][DRT] EventSender needs implementation of leapForward function
1182         https://bugs.webkit.org/show_bug.cgi?id=83486
1183
1184         Reviewed by Eric Seidel.
1185
1186         Implemented leapForward function in Efl's DRT. This function is required
1187         to delay and queue event execution and simulate real user behavior.
1188
1189         * DumpRenderTree/efl/EventSender.cpp:
1190         (DelayedEvent):
1191         (MouseEventInfo::MouseEventInfo):
1192         (MouseEventInfo):
1193         (sendClick):
1194         (mouseDownCallback):
1195         (mouseUpCallback):
1196         (mouseMoveToCallback):
1197         (leapForwardCallback):
1198         (mouseScrollByCallback):
1199         (feedOrQueueMouseEvent):
1200         (feedMouseEvent):
1201         (feedQueuedMouseEvents):
1202
1203 2012-05-08  Ryuan Choi  <ryuan.choi@samsung.com>
1204
1205         [EFL][DRT] WebKit/DRT should not paint webview for the pixel tests.
1206         https://bugs.webkit.org/show_bug.cgi?id=79853
1207
1208         Reviewed by Eric Seidel.
1209
1210         Screen buffer can be different from rendered results.
1211         For example, WebKit/Efl just moves tiles for ewk_view_tiled when scrolling
1212         contents.
1213         It means that pixel tests can't check regression of ewk_view_tiled.
1214
1215         This patch changes a way to provide bitmap from painting webview to capturing
1216         evas.
1217
1218         * DumpRenderTree/efl/PixelDumpSupportEfl.cpp:
1219         (createBitmapContextFromWebView):
1220
1221 2012-05-08  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1222
1223         [GTK] Memory leak in DRT EventSender.cpp
1224         https://bugs.webkit.org/show_bug.cgi?id=85886
1225
1226         Reviewed by Martin Robinson.
1227
1228         Fix a memory leak in gdkModifersFromJSValue() by using JSRetainPtr for
1229         JSString created with JSStringCreateWithUTF8CString.
1230
1231         * DumpRenderTree/gtk/EventSender.cpp:
1232         (gdkModifersFromJSValue):
1233
1234 2012-05-08  Christophe Dumez  <christophe.dumez@intel.com>
1235
1236         [EFL] glib dependency needs to be updated in jhbuild
1237         https://bugs.webkit.org/show_bug.cgi?id=85878
1238
1239         Reviewed by Martin Robinson.
1240
1241         Bump glib version from 2.31.8 to 2.32.2 in jhbuild.modules. This
1242         addresses flakiness in media tests due to possible deadlock with the
1243         following output on stderr:
1244         GStreamer-WARNING **: wrong STREAM_LOCK count 0
1245
1246         The bug was fixed upstream by:
1247         http://git.gnome.org/browse/glib/commit/?id=265f265c677bc5399dbb1cfa449cb4b915a6459f
1248
1249         * efl/jhbuild.modules:
1250
1251 2012-05-07  Alexis Menard  <alexis.menard@openbossa.org>
1252
1253         [Qt] Unbreak debugging of WebKit2.
1254         https://bugs.webkit.org/show_bug.cgi?id=85839
1255
1256         Reviewed by Simon Hausmann.
1257
1258         When you attach GDB to a running process, it stops it.
1259         http://trac.webkit.org/changeset/115958 introduced a pause()
1260         call to wait the debugger to be attached to then continue
1261         the execution of the WebProcess. Unfortunately the pause()
1262         function does not return unless a signal handler is called.
1263         This patch introduce an event handler to exit from the paused
1264         state when the debugger send the signal SIGCONT. The old code
1265         works with older version of GDB (<7.0) but not with newer
1266         versions where the behavior of pause() is correct.
1267
1268         * WebKitTestRunner/qt/main.cpp:
1269         (sigcontHandler):
1270         (main):
1271
1272 2012-05-08  Christophe Dumez  <christophe.dumez@intel.com>
1273
1274         [EFL][DRT] EFL's DRT does not fully support page visibility
1275         https://bugs.webkit.org/show_bug.cgi?id=85032
1276
1277         Reviewed by Antonio Gomes.
1278
1279         * DumpRenderTree/LayoutTestController.cpp:
1280         (resetPageVisibilityCallback): Fix incorrect argumentCount check.
1281         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1282         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): Properly
1283         reset page visibility setting between tests.
1284         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1285         (LayoutTestController::setPageVisibility): Add support for "preview"
1286         visibility state.
1287
1288 2012-05-08  Gavin Peters  <gavinp@chromium.org>
1289
1290         [webkit-patch] Return meaningful results from single-result queries.
1291         https://bugs.webkit.org/show_bug.cgi?id=85367
1292
1293         Reviewed by Dirk Pranke.
1294
1295         Sometimes a quicksearch returns a single result by just redirecting to the page for that single
1296         bug. This patch hacks around that by noticing the redirect target is a single page, and constructing
1297         a result list consisting of that one bug.
1298
1299         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1300         (BugzillaQueries._fetch_bugs_from_advanced_query):
1301
1302 2012-05-08  Gavin Peters  <gavinp@chromium.org>
1303
1304         [webkit-patch] Add new chrome-channels command to track down current chrome release channels for a committed bug.
1305         https://bugs.webkit.org/show_bug.cgi?id=85368
1306
1307         Reviewed by Eric Seidel.
1308
1309         * Scripts/webkitpy/common/config/urls.py:
1310         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
1311         (MockBugzillaQueries.fetch_bugs_matching_quicksearch):
1312         * Scripts/webkitpy/common/net/omahaproxy.py: Added.
1313         (OmahaProxy):
1314         (OmahaProxy.__init__):
1315         (OmahaProxy.set_url):
1316         (OmahaProxy._json_url):
1317         (OmahaProxy._get_json):
1318         (OmahaProxy.get_revisions):
1319         * Scripts/webkitpy/common/net/omahaproxy_unittest.py: Added.
1320         (MockOmahaProxy):
1321         (MockOmahaProxy.__init__):
1322         (OmahaProxyTest):
1323         (OmahaProxyTest.test_get_revisions):
1324         * Scripts/webkitpy/tool/commands/__init__.py:
1325         * Scripts/webkitpy/tool/commands/bugsearch.py:
1326         (BugSearch):
1327         * Scripts/webkitpy/tool/commands/chromechannels.py: Added.
1328         (ChromeChannels):
1329         (__init__):
1330         (execute):
1331         * Scripts/webkitpy/tool/commands/chromechannels_unittest.py: Added.
1332         (MockOmahaProxy):
1333         (MockOmahaProxy.get_revisions):
1334         (TestableChromeChannels):
1335         (TestableChromeChannels.__init__):
1336         (ChromeChannelsTest):
1337         (ChromeChannelsTest.test_single_bug):
1338         (ChromeChannelsTest.test_with_query):
1339
1340 2012-05-08  Carlos Garcia Campos  <cgarcia@igalia.com>
1341
1342         Unreviewed. Fix GTK+ unit tests.
1343
1344         * gtk/common.py:
1345         (pkg_config_file_variable):
1346
1347 2012-05-08  Carlos Garcia Campos  <cgarcia@igalia.com>
1348
1349         [GTK] Google tests that time out are leaked
1350         https://bugs.webkit.org/show_bug.cgi?id=84973
1351
1352         Reviewed by Philippe Normand.
1353
1354         Move Tools/Scripts/run-gtk-tests to Tools/gtk/run-api-tests
1355         leaving run-gtk-tests as simple wrapper that runs run-api-tests in
1356         a jhbuild environment when jhbuild is available.
1357
1358         * Scripts/run-gtk-tests: Run Tools/gtk/run-api-tests directly or
1359         with jhbuild if present.
1360         * gtk/common.py:
1361         (pkg_config_file_variable): Generic function to get a variable
1362         from a pkg-config file.
1363         (prefix_of_pkg_config_file): Use pkg_config_file_variable with 'prefix'.
1364         * gtk/run-api-tests: Copied from Tools/Scripts/run-gtk-tests.
1365         (TestRunner.__init__): Use common.build_path() to get the programs path.
1366         (TestRunner._lookup_atspi2_binary): Use
1367         common.pkg_config_file_variable() to get the 'exec_prefix'.
1368         (TestRunner._start_accessibility_daemons): Use subprocess.Popen()
1369         directly instead of helper _create_process that has been removed.
1370         (TestRunner._setup_testing_environment): Use subprocess.Popen()
1371         directly instead of helper _create_process that has been
1372         removed. Use common.top_level_path() and common.build_path()
1373         instead of custom methods that have been removed too.
1374         (TestRunner._run_test_command): Use subprocess.Popen() directly
1375         instead of helper _create_process that has been removed.
1376
1377 2012-05-08  Thiago Marcos P. Santos  <thiago.santos@intel.com>
1378
1379         [EFL] [DRT] Clear extra views before setting a blank page
1380         https://bugs.webkit.org/show_bug.cgi?id=85608
1381
1382         Reviewed by Kenneth Rohde Christiansen.
1383
1384         We need to clear the extra views before setting a blank page,
1385         otherwise it will cause more events interactions and sometimes
1386         trigger a unload(). This was causing frame-crash-with-page-cache.html
1387         to leak events to the next test causing any subsequent test to
1388         fail.
1389
1390         * DumpRenderTree/efl/DumpRenderTree.cpp:
1391         (runTest):
1392         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1393         (DumpRenderTreeChrome::clearExtraViews):
1394         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1395         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
1396         (DumpRenderTreeChrome):
1397
1398 2012-05-08  Balazs Kelemen  <kbalazs@webkit.org>
1399
1400         [Qt] X11 plugins need to be reworked for Qt5+WK1
1401         https://bugs.webkit.org/show_bug.cgi?id=80691
1402
1403         Reviewed by Simon Hausmann.
1404
1405         Implement basic windowless plugin support with Qt5.
1406
1407         * qmake/mkspecs/features/features.prf:
1408         Enable NPAPI plugins if X11 libraries are available.
1409         * qmake/mkspecs/features/functions.prf:
1410         Added a convenience function to determine availability
1411         of X11 libraries.
1412
1413 2012-05-07  Raphael Kubo da Costa  <rakuco@webkit.org>
1414
1415         [webkitpy] Remove Python 2.5-specific workaround from http_server_base.py.
1416         https://bugs.webkit.org/show_bug.cgi?id=85852
1417
1418         Reviewed by Dirk Pranke.
1419
1420         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
1421         (HttpServerBase._is_server_running_on_all_ports): Revert the part
1422         of r89412 which was a workaround for an issue with Python 2.5 on
1423         Mac 10.5.
1424
1425 2012-05-07  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1426
1427         webkitdirs.pm: Make determineNumberOfCPUs work on FreeBSD.
1428         https://bugs.webkit.org/show_bug.cgi?id=81481
1429
1430         Reviewed by Dirk Pranke.
1431
1432         FreeBSD has the same sysctl interface present in Darwin to query
1433         the number of CPUs available, so use that too.
1434
1435         * Scripts/webkitdirs.pm:
1436         (determineNumberOfCPUs): Share the sysctl call with Darwin.
1437         (isFreeBSD): Added.
1438
1439 2012-05-07  Alexis Menard  <alexis.menard@openbossa.org>
1440
1441         Unreviewed Qt build fix in Debug.
1442
1443         We need some system includes here for getpid() to be recognized.
1444
1445         * WebKitTestRunner/qt/main.cpp:
1446
1447 2012-05-07  Ojan Vafai  <ojan@chromium.org>
1448
1449         When embedding the flakiness dashboard, hide popups when the frame is blurred.
1450         https://bugs.webkit.org/show_bug.cgi?id=85684
1451
1452         Reviewed by Dirk Pranke.
1453
1454         If you show a popup, then click elsewhere in garden-o-matic, it will now
1455         properly hide it.
1456
1457         * TestResultServer/static-dashboards/dashboard_base.js:
1458         (showPopup):
1459         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1460         (showPopupForBuild):
1461         (generatePageForIndividualTests.if):
1462         (generatePageForIndividualTests):
1463         * TestResultServer/static-dashboards/flakiness_dashboard_embedded_unittests.js: Added.
1464         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1465         * TestResultServer/static-dashboards/run-embedded-unittests.html: Added.
1466
1467 2012-05-07  Ian Vollick  <vollick@chromium.org>
1468
1469         [chromium] WebViewHost should setCompositorSurfaceReady
1470         https://bugs.webkit.org/show_bug.cgi?id=85418
1471
1472         Reviewed by James Robinson.
1473
1474         * DumpRenderTree/chromium/WebViewHost.cpp:
1475         (WebViewHost::setWebWidget):
1476
1477 2012-05-07  Jon Lee  <jonlee@apple.com>
1478
1479         WebKitTestRunner needs layoutTestController.dumpFrameLoadCallbacks
1480         https://bugs.webkit.org/show_bug.cgi?id=42331
1481         <rdar://problem/8193641>
1482
1483         Reviewed by Darin Adler.
1484
1485         This patch sets up the framework for supporting dumpFrameLoadCallbacks, as well as
1486         outputting some of the required strings. As tests become unskipped the other callbacks
1487         can be supported.
1488
1489         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: Add dumpFrameLoadCallbacks()
1490         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1491         (WTR::InjectedBundle::didReceiveMessage): pass in the dictionary containing keys needed to initialize
1492         dump callbacks for the layoutTestController.
1493         (WTR::InjectedBundle::booleanForKey): Helper function to extract boolean value from key in WKDictionaryRef.
1494         Output some warning if we are trying to extract a value that is not a boolean.
1495         (WTR::InjectedBundle::beginTesting): Update to take in the dictionary passed in by the TestController.
1496         We cannot set the bits on the layoutTestController until it has been created in beginTesting().
1497         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1498         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1499         (WTR::InjectedBundle::didReceiveMessage):
1500         (WTR):
1501         (WTR::InjectedBundle::beginTesting):
1502         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1503         (InjectedBundle):
1504         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1505         (WTR):
1506         (WTR::dumpFrameDescriptionSuitableForTestResult): Static function similar to
1507         -[WebFrame _drt_descriptionSuitableForTestResult] in DumpRenderTree.
1508         (WTR::InjectedBundlePage::didStartProvisionalLoadForFrame): Output string.
1509         (WTR::InjectedBundlePage::didCommitLoadForFrame): Output string.
1510         (WTR::InjectedBundlePage::didFinishLoadForFrame): Output string.
1511         (WTR::InjectedBundlePage::didReceiveTitleForFrame): Output string.
1512         (WTR::InjectedBundlePage::didCancelClientRedirectForFrame): Output string.
1513         (WTR::InjectedBundlePage::willPerformClientRedirectForFrame): Output string. In WK1 it's
1514         "willPerformClientRedirectToURL" so we will have to use that value here.
1515         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame): Output string.
1516         (WTR::InjectedBundlePage::didHandleOnloadEventsForFrame): Output string.
1517         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1518         (WTR::LayoutTestController::LayoutTestController): Initialize to not dump frame load callbacks.
1519         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1520         (WTR::LayoutTestController::dumpFrameLoadCallbacks): Change to use setShouldDumpFrameLoadCallbacks.
1521         (WTR::LayoutTestController::setShouldDumpFrameLoadCallbacks): Set bit.
1522         (WTR::LayoutTestController::shouldDumpFrameLoadCallbacks): Return bit.
1523         (LayoutTestController): Added m_dumpFrameLoadCallbacks to track whether to dump the output.
1524         * WebKitTestRunner/TestInvocation.cpp:
1525         (WTR::shouldLogFrameLoadDelegates): Similar to DRT.
1526         (WTR::TestInvocation::invoke): Based on the URL loaded, set the bit in the dictionary that gets passed
1527         to the injected bundle.
1528
1529 2012-05-07  Tommy Widenflycht  <tommyw@google.com>
1530
1531         [chromium] MediaStream API: Enhance WebUserMediaClientMock and add a test for it
1532         https://bugs.webkit.org/show_bug.cgi?id=85618
1533
1534         Reviewed by Adam Barth.
1535
1536         Adding the capability to control which MediaStreamTracks actually get created,
1537         and adding a test that exercises the JS api using that capability.
1538
1539         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
1540         (WebKit::WebUserMediaClientMock::requestUserMedia):
1541         * DumpRenderTree/chromium/WebUserMediaClientMock.h:
1542         * DumpRenderTree/chromium/WebViewHost.cpp:
1543         (WebViewHost::testMediaStreamClient):
1544         * DumpRenderTree/chromium/WebViewHost.h:
1545         (WebViewHost):
1546
1547 2012-05-07  Zan Dobersek  <zandobersek@gmail.com>
1548
1549         [Gtk] WebGL feature is not built anymore through build-webkit after r116251
1550         https://bugs.webkit.org/show_bug.cgi?id=85788
1551
1552         Reviewed by Martin Robinson.
1553
1554         Enable WebGL by default for the Gtk port. The feature is already enabled by
1555         default in configure.ac, so there's no reason it shouldn't be enabled when
1556         building through build-webkit.
1557
1558         * Scripts/webkitperl/FeatureList.pm:
1559
1560 2012-05-07  Csaba Osztrogonác  <ossy@webkit.org>
1561
1562         Unittest fix after r116288, because now there are more than 5 hit for the 'Eric' word.
1563         We have only 2 'Gavin', so it is a better choice than 'Eric' for unit testing.
1564
1565         Reviewed by Tor Arne Vestbø.
1566
1567         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
1568         (IRCCommandTest.test_whois):
1569
1570 2012-05-07  Sheriff Bot  <webkit.review.bot@gmail.com>
1571
1572         Unreviewed, rolling out r116299, r116301, and r116303.
1573         http://trac.webkit.org/changeset/116299
1574         http://trac.webkit.org/changeset/116301
1575         http://trac.webkit.org/changeset/116303
1576         https://bugs.webkit.org/show_bug.cgi?id=85795
1577
1578         Build is still broken (Requested by Ossy on #webkit).
1579
1580         * DumpRenderTree/qt/DumpRenderTree.pro:
1581         * DumpRenderTree/qt/main.cpp:
1582         (main):
1583         * QtTestPlatformPlugin/QtTestPlatformPlugin.pro: Removed.
1584         * QtTestPlatformPlugin/TestIntegration.cpp: Removed.
1585         * QtTestPlatformPlugin/TestIntegration.h: Removed.
1586         * QtTestPlatformPlugin/mac/TestFontDatabase.h: Removed.
1587         * QtTestPlatformPlugin/mac/TestFontDatabase.mm: Removed.
1588         * QtTestPlatformPlugin/mac/TestIntegrationMac.mm: Removed.
1589         * QtTestPlatformPlugin/main.cpp: Removed.
1590         * QtTestPlatformPlugin/testplatform.json: Removed.
1591         * Tools.pro:
1592         * WebKitTestRunner/Target.pri:
1593         * WebKitTestRunner/qt/main.cpp:
1594         (main):
1595
1596 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
1597
1598         Unreviewed Qt build fix.
1599
1600         One more try to learn how to use
1601         const with a pointer array.
1602
1603         * DumpRenderTree/qt/main.cpp:
1604         (initializeTestPlatformPlugin):
1605
1606 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
1607
1608         Unreviewed build fix after r116299.
1609
1610         * DumpRenderTree/qt/main.cpp:
1611         (initializeTestPlatformPlugin):
1612
1613 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
1614
1615         [Qt] Add test specific platform plugin to achieve unified layout test results
1616         https://bugs.webkit.org/show_bug.cgi?id=80996
1617
1618         Reviewed by Simon Hausmann.
1619
1620         Added QtTestPlatformPlugin as a new project under Tools.
1621         This is a Qt5-ish platform plugin that can be used to tweak the
1622         platform support interfaces in order to unify layout test results.
1623         For now it only overrides the font database on Mac and redirects
1624         everything else to the real platform plugin. The font database it
1625         provides mimics the way how we set up test fonts with fontconfig on Linux.
1626         Make DumpRenderTree and WebKitTestRunner use this platform plugin.
1627
1628         * DumpRenderTree/qt/DumpRenderTree.pro:
1629         * DumpRenderTree/qt/main.cpp:
1630         (initializeTestPlatformPlugin):
1631         (main):
1632         * QtTestPlatformPlugin/QtTestPlatformPlugin.pro: Added.
1633         * QtTestPlatformPlugin/TestIntegration.cpp: Added.
1634         (TestIntegration::TestIntegration):
1635         (TestIntegration::fontDatabase):
1636         * QtTestPlatformPlugin/TestIntegration.h: Added.
1637         (TestIntegration):
1638         (TestIntegration::hasCapability):
1639         (TestIntegration::createPlatformPixmap):
1640         (TestIntegration::createPlatformWindow):
1641         (TestIntegration::createPlatformBackingStore):
1642         (TestIntegration::createPlatformOpenGLContext):
1643         (TestIntegration::createPlatformSharedGraphicsCache):
1644         (TestIntegration::guiThreadEventDispatcher):
1645         (TestIntegration::clipboard):
1646         (TestIntegration::drag):
1647         (TestIntegration::inputContext):
1648         (TestIntegration::accessibility):
1649         (TestIntegration::nativeInterface):
1650         (TestIntegration::services):
1651         (TestIntegration::styleHint):
1652         (TestIntegration::platformTheme):
1653         * QtTestPlatformPlugin/mac/TestFontDatabase.h: Added.
1654         (TestFontDatabase):
1655         * QtTestPlatformPlugin/mac/TestFontDatabase.mm: Added.
1656         (TestFontDatabase::populateFontDatabase):
1657         * QtTestPlatformPlugin/mac/TestIntegrationMac.mm: Added.
1658         (TestIntegration::fontDatabase):
1659         * QtTestPlatformPlugin/main.cpp: Added.
1660         (TestIntegrationPlugin::keys):
1661         (TestIntegrationPlugin::create):
1662         (TestIntegrationPlugin::initialize):
1663         * QtTestPlatformPlugin/testplatform.json: Added.
1664         * Tools.pro:
1665         * WebKitTestRunner/Target.pri:
1666         * WebKitTestRunner/qt/main.cpp:
1667         (main):
1668
1669 2012-05-07  Christophe Dumez  <christophe.dumez@intel.com>
1670
1671         [EFL] media/video-poster-blocked-by-willsendrequest.html is flaky
1672         https://bugs.webkit.org/show_bug.cgi?id=85609
1673
1674         Reviewed by Simon Hausmann.
1675
1676         Flakiness was caused by the fact that
1677         media/crash-closing-page-with-media-as-plugin-fallback.html and
1678         media/video-poster-blocked-by-willsendrequest.html use the same PNG
1679         file as poster attribute in the video tag. The PNG file gets cached
1680         when the first test is executed and this somehow causes the second
1681         test to fail.
1682
1683         The flakiness was addressed by clearing the memory and page caches
1684         between tests.
1685
1686         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1687         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1688
1689 2012-05-07  Eric Seidel  <eric@webkit.org>
1690
1691         Add values for all features to Qt's features.pri
1692         https://bugs.webkit.org/show_bug.cgi?id=85746
1693
1694         Reviewed by Tor Arne Vestbø.
1695
1696         When generate-feature-files generates features.pri it provides a value
1697         for every possible ENABLE_.  In prepration for landing generate-feature-files
1698         it seems appropriate to add a value for all possible ENABLEs to the existing
1699         features.pri.  This lets me make sure that I'm getting them all right!
1700
1701         * qmake/mkspecs/features/features.pri:
1702
1703 2012-05-07  Adam Bergkvist  <adam.bergkvist@ericsson.com>
1704
1705         Added myself as a committer.
1706
1707         Not reviewed.
1708
1709         * Scripts/webkitpy/common/config/committers.py:
1710
1711 2012-05-07  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1712
1713         [EFL][DRT] Implement LayoutTestController::execCommand
1714         https://bugs.webkit.org/show_bug.cgi?id=82286
1715
1716         Reviewed by Antonio Gomes.
1717
1718         Add missing implementation execCommand and isCommandEnabled to EFL's
1719         LayoutTestController.
1720
1721         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1722         (LayoutTestController::execCommand):
1723         (LayoutTestController::isCommandEnabled):
1724
1725 2012-05-06  Hayato Ito  <hayato@chromium.org>
1726
1727         [prepare-ChangeLog] Excludes a reference file used by reftests from test files.
1728         https://bugs.webkit.org/show_bug.cgi?id=74268
1729
1730         Reviewed by Ryosuke Niwa.
1731
1732         * Scripts/prepare-ChangeLog:
1733         (generateFileList):
1734
1735 2012-05-06  Mary Wu  <mary.wu@torchmobile.com.cn>
1736
1737         Add a contributor to committers.py
1738         https://bugs.webkit.org/show_bug.cgi?id=85761
1739
1740         Reviewed by Unreviewed.
1741
1742         Add myself to committers.py contributor.
1743
1744         * Scripts/webkitpy/common/config/committers.py:
1745
1746 2012-05-06  Dan Bernstein  <mitz@apple.com>
1747
1748         Tools part of: Building and debugging WebKit in the Xcode IDE requires a lot of setup
1749         https://bugs.webkit.org/show_bug.cgi?id=85739
1750
1751         Reviewed by Daniel Bates.
1752
1753         * Scripts/build-webkit: Moved code that copies from WebKitLibraries to the product directory
1754         from here...
1755         * Scripts/copy-webkitlibraries-to-product-directory: ...to this new script.
1756
1757 2012-05-06  Andy Estes  <aestes@apple.com>
1758
1759         Fix the build after r116246.
1760
1761         * DumpRenderTree/LayoutTestController.cpp: Remove an unused function.
1762
1763 2012-05-06  Eric Seidel  <eric@webkit.org>
1764
1765         Update several build-webkit options to match the #defines they toggle
1766         https://bugs.webkit.org/show_bug.cgi?id=85744
1767
1768         Reviewed by Adam Barth.
1769
1770         Most notable here is fixing --3d-canvas to be --webgl these days --
1771         the define was renamed 15 months ago!
1772
1773         * Scripts/webkitperl/FeatureList.pm:
1774
1775 2012-05-06  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1776
1777         [EFL] EFL's LayoutTestController setJavaScriptCanAccessClipboard implementation
1778         https://bugs.webkit.org/show_bug.cgi?id=83687
1779
1780         Reviewed by Antonio Gomes.
1781
1782         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1783         (LayoutTestController::setJavaScriptCanAccessClipboard):
1784
1785 2012-05-06  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1786
1787         Convert isPageBoxVisible to use Internals interface.
1788         https://bugs.webkit.org/show_bug.cgi?id=85692
1789
1790         Reviewed by Darin Adler.
1791
1792         Remove isPageBoxVisible functions, because it is able to work in the
1793         cross-port way through the Internals interface.
1794
1795         * DumpRenderTree/LayoutTestController.cpp:
1796         (LayoutTestController::staticFunctions):
1797         * DumpRenderTree/LayoutTestController.h:
1798         (LayoutTestController):
1799         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1800         * DumpRenderTree/chromium/LayoutTestController.cpp:
1801         (LayoutTestController::LayoutTestController):
1802         * DumpRenderTree/chromium/LayoutTestController.h:
1803         (LayoutTestController):
1804         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1805         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1806         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1807         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1808         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1809         (LayoutTestController):
1810         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1811         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1812
1813 2012-05-06  Tom Zakrajsek  <tomz@codeaurora.org>
1814
1815         sheriffbot isn't reopening patches after it lands rollouts
1816         https://bugs.webkit.org/show_bug.cgi?id=64418
1817
1818         Reviewed by Adam Barth.
1819
1820         * Scripts/webkitpy/tool/commands/download_unittest.py:
1821         * Scripts/webkitpy/tool/steps/createbug.py:
1822         (CreateBug.run):
1823
1824 2012-05-04  Raphael Kubo da Costa  <rakuco@webkit.org>
1825
1826         webkitpy: Use os.pathsep instead of manually finding out the path separator in NRWT.
1827         https://bugs.webkit.org/show_bug.cgi?id=85697
1828
1829         Reviewed by Eric Seidel.
1830
1831         * Scripts/new-run-webkit-tests: Instead of manually detecting whether
1832         the path separator should be ':' or ';' by checking sys.platform, rely
1833         on os.pathsep which has exactly the same information (os.pathsep is
1834         ':' for Cygwin, from what I could check in Python's sources).
1835
1836 2012-05-05  Sheriff Bot  <webkit.review.bot@gmail.com>
1837
1838         Unreviewed, rolling out r116223.
1839         http://trac.webkit.org/changeset/116223
1840         https://bugs.webkit.org/show_bug.cgi?id=85718
1841
1842         The bots need to set XDG_DATA_DIRS (Requested by rakuco on
1843         #webkit).
1844
1845         * efl/jhbuildrc:
1846         * gtk/jhbuildrc:
1847
1848 2012-05-05  Raphael Kubo da Costa  <rakuco@webkit.org>
1849
1850         [jhbuild] Do not set XDG_DATA_DIRS in jhbuildrc.
1851         https://bugs.webkit.org/show_bug.cgi?id=85717
1852
1853         Reviewed by Martin Robinson.
1854
1855         After r116209, the XDG_DATA_DIRS environment variable is now
1856         passed by webkitpy's layout_tests code to run-with-jhbuild, so
1857         instead of setting a few defaults in jhbuildrc we can now rely on
1858         the values set and really used by each system.
1859
1860         * efl/jhbuildrc:
1861         * gtk/jhbuildrc:
1862
1863 2012-05-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1864
1865         [Qt] Buildfix for --minimal build after r116107.
1866
1867         Reviewed by Csaba Osztrogonác.
1868
1869         * qmake/mkspecs/features/features.prf:
1870
1871 2012-05-05  Zan Dobersek  <zandobersek@gmail.com>
1872
1873         [Gtk] GtkDriver is not required anymore after r116134
1874         https://bugs.webkit.org/show_bug.cgi?id=85699
1875
1876         Reviewed by Csaba Osztrogonác.
1877
1878         Remove GtkDriver as it was replaced by the XvfbDriver and is
1879         no longer needed.
1880
1881         * Scripts/webkitpy/layout_tests/port/gtk.py:
1882
1883 2012-05-05  Zan Dobersek  <zandobersek@gmail.com>
1884
1885         Increase the display ID for the Xvfb process to use if pixel_tests argument
1886         is true rather than if pixel_tests option is present. This fixes the
1887         XvfbDriver in the same way r115825 fixed the now-redundant GtkDriver.
1888
1889         Rubber-stamped by Csaba Osztrogonác.
1890
1891         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py:
1892         (XvfbDriver._start):
1893
1894 2012-05-04  Raphael Kubo da Costa  <rakuco@webkit.org>
1895
1896         webkitpy: Preserve XDG_DATA_DIRS in the environment passed to ServerProcess.
1897         https://bugs.webkit.org/show_bug.cgi?id=85694
1898
1899         Reviewed by Martin Robinson.
1900
1901         This is sort of a follow-up to r100674: $XDG_DATA_DIRS should be
1902         preserved, as it is particularly useful for Linux/Unix
1903         environments.  As a real use case, if shared-mime-info is
1904         installed into a non-trivial prefix, Glib and other libraries will
1905         have trouble to properly identify mimetypes (in ResourceResponses,
1906         for example) as the base freedesktop.org mimetype file will not be
1907         found.
1908
1909         * Scripts/webkitpy/layout_tests/port/base.py:
1910         (Port.to.setup_environ_for_server):
1911
1912 2012-05-04  Gustavo Noronha Silva  <gns@gnome.org>
1913
1914         [GTK] Simplify how libWebCoreModules is linked in, and fix WebKit2 build
1915         https://bugs.webkit.org/show_bug.cgi?id=85691
1916
1917         * GNUmakefile.am: no longer link libWebCoreModules to DumpRenderTree.
1918
1919 2012-05-04  Eric Seidel  <eric@webkit.org>
1920
1921         Make capitalization and descriptions in FeatureList.pm more consistent to make easier to autogenerate
1922         https://bugs.webkit.org/show_bug.cgi?id=85583
1923
1924         Reviewed by Daniel Bates.
1925
1926         * Scripts/webkitperl/FeatureList.pm:
1927
1928 2012-05-04  Eric Seidel  <eric@webkit.org>
1929
1930         Sort entries in FeaturesList.pm to make them easier to autogenerate
1931         https://bugs.webkit.org/show_bug.cgi?id=85584
1932
1933         Reviewed by Daniel Bates.
1934
1935         No functional changes, only sorting.
1936
1937         * Scripts/webkitperl/FeatureList.pm:
1938
1939 2012-05-04  Andy Estes  <aestes@apple.com>
1940
1941         REGRESSION (r115752): WebKit2.NewFirstVisuallyNonEmptyLayout API test times out
1942         https://bugs.webkit.org/show_bug.cgi?id=85675
1943
1944         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout.cpp:
1945         (TestWebKitAPI::TEST): Disable the test.
1946
1947 2012-04-30  Jer Noble  <jer.noble@apple.com>
1948
1949         Full screen will exit during a provisional load of a non-ancestor iframe.
1950         https://bugs.webkit.org/show_bug.cgi?id=85230
1951
1952         Reviewed by Maciej Stachowiak .
1953
1954         Add support for the new UIDelegate webView:closeFullScreenWithListener: method:
1955         * DumpRenderTree/mac/UIDelegate.mm:
1956         (-[UIDelegate webView:closeFullScreenWithListener:]):
1957
1958         Add support for the new closeFullScreen injected bundle method:
1959         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1960         (WTR::InjectedBundlePage::InjectedBundlePage):
1961         (WTR::InjectedBundlePage::closeFullScreen):
1962         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
1963
1964 2012-05-04  Ojan Vafai  <ojan@chromium.org>
1965
1966         Fix test naming to be camel-case like the methods they are testing.
1967
1968         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1969         (test):
1970
1971 2012-05-04  Ojan Vafai  <ojan@chromium.org>
1972
1973         Show the results links in the embedded flakiness dashboard
1974         https://bugs.webkit.org/show_bug.cgi?id=85660
1975
1976         Reviewed by Adam Barth.
1977
1978         Turns out these links don't add much visual noise and it's annoying
1979         to have to go to the dashboard from garden-o-matic just to view the
1980         results for a test.
1981
1982         This is useful for cases where garden-o-matic doesn't properly show the
1983         results (e.g. missing tests). If we address all those cases, then
1984         we could easily hide the results links again if we wanted to.
1985
1986         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1987         (showUpdateInfoForTest):
1988         (htmlForIndividualTestOnAllBuildersWithResultsLinks):
1989         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1990         (test):
1991
1992 2012-05-04  Dirk Pranke  <dpranke@chromium.org>
1993
1994         Revert to using chromium --test-shell mode on SL in NRWT
1995
1996         https://bugs.webkit.org/show_bug.cgi?id=83076
1997
1998         Unreviewed, build fix.
1999
2000         We seem to be seeing a lot more flakiness on the bot since
2001         I switched NRWT to "DRT" mode. We'll try reverting back and
2002         see if this helps things.
2003
2004         * Scripts/webkitpy/layout_tests/port/chromium.py:
2005         (ChromiumDriver.__init__):
2006
2007 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
2008
2009         [EFL] Implement layoutTestController.dumpResourceResponseMIMETypes
2010         https://bugs.webkit.org/show_bug.cgi?id=84941
2011
2012         Reviewed by Martin Robinson.
2013
2014         Print MIME type detection messages in EFL's DumpRenderTree if
2015         LayoutTestController's dumpResourceResponseMIMETypes() returns true.
2016
2017         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2018         (DumpRenderTreeChrome::onResponseReceived):
2019
2020 2012-05-04  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2021
2022         [GTK] DRT needs an implementation of LayoutTestController::setDefersLoading and ::goBack
2023         https://bugs.webkit.org/show_bug.cgi?id=85134
2024
2025         Reviewed by Martin Robinson.
2026
2027         Add missing implementation setDefersLoading and goBack to GTK's
2028         LayoutTestController so that the test case rely on these methods
2029         can be enabled.
2030
2031         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2032         (resetDefaultsToConsistentValues):
2033         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2034         (LayoutTestController::goBack):
2035         (LayoutTestController::setDefersLoading):
2036
2037 2012-05-04  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
2038
2039         [Qt][NRWT] Run each DRT in it's own xvfb
2040         https://bugs.webkit.org/show_bug.cgi?id=77335
2041
2042         Reviewed by Dirk Pranke.
2043
2044         * Scripts/webkitpy/layout_tests/port/gtk.py:
2045         (GtkPort._driver_class):
2046         * Scripts/webkitpy/layout_tests/port/qt.py:
2047         (QtPort._driver_class):
2048         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Added.
2049         (XvfbDriver):
2050         (XvfbDriver._start):
2051         (XvfbDriver._start.x_filter):
2052         (XvfbDriver.stop):
2053
2054 2012-05-04  Zan Dobersek  <zandobersek@gmail.com>
2055
2056         Use destructuring assingment for the return values of the
2057         port.diff_image() method call.
2058
2059         Rubber-stamped by Dirk Pranke.
2060
2061         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2062         (write_test_result):
2063
2064 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2065
2066         Store fewer runs by default in the flakiness dashboard
2067         https://bugs.webkit.org/show_bug.cgi?id=85568
2068
2069         Reviewed by Dirk Pranke.
2070
2071         This will greatly shrink the size of the json files we load,
2072         resulting in a much faster initial load. People rarely care
2073         about more than 100 runs. For the cases they do, they can still click
2074         the show all runs checkbox to show 500 runs.
2075
2076         * TestResultServer/model/jsonresults.py:
2077
2078 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
2079
2080         [EFL] Frame load callbacks output is missing for redirections
2081         https://bugs.webkit.org/show_bug.cgi?id=85173
2082
2083         Reviewed by Antonio Gomes.
2084
2085         Catch the new ewk_frame signals to print the expected output in case
2086         of redirections, when LayoutTestController's dumpFrameLoadCallbacks()
2087         returns true.
2088
2089         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2090         (DumpRenderTreeChrome::createView):
2091         (DumpRenderTreeChrome::onWillSendRequest):
2092         (DumpRenderTreeChrome::onFrameCreated):
2093         (DumpRenderTreeChrome::onFrameRedirectForProvisionalLoad):
2094         (DumpRenderTreeChrome::onFrameRedirectRequested):
2095         (DumpRenderTreeChrome::onResponseReceived):
2096         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
2097         (DumpRenderTreeChrome):
2098
2099 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2100
2101         [Qt] Fix up warning about missing GLib/Gio/GStreamer
2102
2103         We don't actually fall back.
2104
2105         Rubber-stamped by Csaba Osztrogonác.
2106
2107         * qmake/mkspecs/features/features.prf:
2108
2109 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2110
2111         [Qt] Clarify warning about missing GLib/Gio/GStreamer for media support
2112
2113         And only print it once.
2114
2115         Reviewed by Csaba Osztrogonác.
2116
2117         * qmake/mkspecs/features/features.prf:
2118
2119 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2120
2121         [Qt] Clean up and split features.prf into a static list of defaults
2122
2123         The static list of feature defaults is used as a fallback for any
2124         feature that's not dynamically detected or overriden on the command
2125         line (though build-webkit or passing DEFINES+= to qmake).
2126
2127         The static list is complete, which allows for auto-generation based
2128         on Features.py (see bug https://bugs.webkit.org/show_bug.cgi?id=85456)
2129
2130         https://bugs.webkit.org/show_bug.cgi?id=85611
2131
2132         Reviewed by Simon Hausmann.
2133
2134         * Scripts/build-webkit:
2135         * Tools.pro:
2136         * qmake/mkspecs/features/features.prf:
2137         * qmake/mkspecs/features/features.pri: Added.
2138
2139 2012-05-04  Sheriff Bot  <webkit.review.bot@gmail.com>
2140
2141         Unreviewed, rolling out r116085, r116091, and r116095.
2142         http://trac.webkit.org/changeset/116085
2143         http://trac.webkit.org/changeset/116091
2144         http://trac.webkit.org/changeset/116095
2145         https://bugs.webkit.org/show_bug.cgi?id=85628
2146
2147         We are not ready with dependencies on all platform yet (mac) +
2148         problems with debug builds. (Requested by Zoltan on #webkit).
2149
2150         * qmake/mkspecs/features/features.prf:
2151
2152 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
2153
2154         [EFL] LayoutTestController does not support overriding WebKitLoadSiteIconsKey
2155         https://bugs.webkit.org/show_bug.cgi?id=85171
2156
2157         Reviewed by Andreas Kling.
2158
2159         Implement support for overriding "WebKitLoadSiteIconsKey" setting in
2160         EFL's LayoutTestController.
2161
2162         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2163         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
2164         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2165         (LayoutTestController::overridePreference):
2166
2167 2012-05-03  Alexis Menard  <alexis.menard@openbossa.org>
2168
2169         [Qt] Enable fullscreen API for WebKit2.
2170         https://bugs.webkit.org/show_bug.cgi?id=85498
2171
2172         Reviewed by Simon Hausmann.
2173
2174         Enable the fullscreen API on the MiniBrowser and turn
2175         it default on WebKit2, disable it on WK1.
2176
2177         * MiniBrowser/qt/BrowserWindow.cpp:
2178         (BrowserWindow::BrowserWindow):
2179         * MiniBrowser/qt/qml/BrowserWindow.qml:
2180         * Scripts/webkitperl/FeatureList.pm:
2181         * qmake/mkspecs/features/features.prf:
2182
2183 2012-05-04  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2184
2185         REGRESSION(r116008) old-run-webkit-tests broken
2186         https://bugs.webkit.org/show_bug.cgi?id=85597
2187
2188         Reviewed by Simon Hausmann.
2189
2190         Reinstate libraryContainsSymbol.
2191
2192         * Scripts/webkitperl/features.pm:
2193         (libraryContainsSymbol):
2194
2195 2012-05-04  Zoltan Horvath  <zoltan@webkit.org>
2196
2197         [Qt] Set WebCore imagedecoders as default and add fallback to QImageDecoder
2198         https://bugs.webkit.org/show_bug.cgi?id=80400
2199
2200         Remove WTF_USE_QT_IMAGE_DECODER macro.
2201
2202         Reviewed by Simon Hausmann.
2203
2204         * qmake/mkspecs/features/features.prf:
2205
2206 2012-05-03  Christophe Dumez  <christophe.dumez@intel.com>
2207
2208         [EFL] DRT should obey layoutTestController's addURLToRedirect()
2209         https://bugs.webkit.org/show_bug.cgi?id=82722
2210
2211         Reviewed by Maciej Stachowiak.
2212
2213         Implement support for layoutTestController's addURLToRedirect()
2214         in EFL's DRT so that it redirects properly when indicated.
2215
2216         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2217         (DumpRenderTreeChrome::onWillSendRequest):
2218
2219 2012-05-03  Christophe Dumez  <christophe.dumez@intel.com>
2220
2221         [EFL] DRT should support LayoutTestController's willSendRequestReturnsNullOnRedirect()
2222         https://bugs.webkit.org/show_bug.cgi?id=82704
2223
2224         Reviewed by Antonio Gomes.
2225
2226         EFL's DRT needs to obey LayoutTestController's
2227         willSendRequestReturnsNullOnRedirect().
2228
2229         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2230         (DumpRenderTreeChrome::onWillSendRequest):
2231
2232 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2233
2234         git-add-reviewer: Do not put the bug title and its URL in the same line in the commit message.
2235         https://bugs.webkit.org/show_bug.cgi?id=85570
2236
2237         Reviewed by Adam Roben.
2238
2239         The standard format of commit messages and ChangeLog entries has a
2240         short summary in the first line, followed by the bug URL in
2241         Bugzilla. Passing "%s" in the pretty format line to git-rev-list
2242         merges these two lines into a single one separated by a space
2243         character, which is not what we want.
2244
2245         Since "%s%n%n%b" is in practice "%B" (raw message), use it
2246         instead, as the first two lines do not end up being merged
2247         together.
2248
2249         * Scripts/git-add-reviewer:
2250         (writeCommitMessageToFile):
2251
2252 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2253
2254         Improve UI for garden-o-matic examine view when there are fewer than 5 tests.
2255         https://bugs.webkit.org/show_bug.cgi?id=85566
2256
2257         Reviewed by Adam Barth.
2258
2259         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2260         (.):
2261         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
2262         (.):
2263         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2264
2265 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2266
2267         webkitpy: Recognize FreeBSD as a valid platform.
2268         https://bugs.webkit.org/show_bug.cgi?id=81467
2269
2270         Reviewed by Adam Barth.
2271
2272         Add PlatformInfo.is_freebsd() so that trying to use webkit-patch
2273         on FreeBSD does not raise an "unrecognized platform" exception.
2274
2275         Most of this patch contains some unit tests for the mentioned
2276         method, as currently platform-specific code is needed only on
2277         Windows.
2278
2279         * Scripts/webkitpy/common/system/platforminfo.py:
2280         (PlatformInfo.__init__):
2281         (PlatformInfo.is_freebsd):
2282         (PlatformInfo._determine_os_name):
2283         * Scripts/webkitpy/common/system/platforminfo_mock.py:
2284         (MockPlatformInfo.is_freebsd):
2285         * Scripts/webkitpy/common/system/platforminfo_unittest.py:
2286         (fake_platform):
2287         (fake_platform.FakePlatformModule.release):
2288         (TestPlatformInfo.test_real_code):
2289         (TestPlatformInfo.test_os_name_and_wrappers):
2290         (TestPlatformInfo.test_os_version):
2291         (TestPlatformInfo.test_display_name):
2292         (TestPlatformInfo.test_total_bytes_memory):
2293         (TestPlatformInfo.test_free_bytes_memory):
2294         * Scripts/webkitpy/layout_tests/port/factory.py:
2295         (PortFactory._default_port):
2296
2297 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2298
2299         Make the table headers white-space:nowrap. This decreases the vertical size,
2300         which makes for more minimal embedding in garden-o-matic.
2301
2302         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
2303         (.table-header-content *):
2304
2305 2012-05-03  Eric Seidel  <eric@webkit.org>
2306
2307         Split build-webkit's feature option list off into its own module in preparation for autogeneration
2308         https://bugs.webkit.org/show_bug.cgi?id=85548
2309
2310         Reviewed by Daniel Bates.
2311
2312         My Perl-fu is very weak.  This moves the feature option list off into a separate module
2313         file so that I can easily autogenerate just that file with the fancy new generate-feature-files command.
2314         It's a bit odd to have an array which we return out to build-webkit to modify, but it turns out
2315         not to matter.  Also, build-webkit doesn't actually use any of the variables we're using to
2316         store the results from the option parse, so those end up inaccessible from build-webkit (except by reference)
2317         which is actually cleaner.  Callers can't assume that they would get a new array every time, but someone
2318         with more Perl-fu than I could create something fancier than this if we need in the future.
2319
2320         * Scripts/build-webkit:
2321         * Scripts/webkitperl/FeatureList.pm: Added.
2322         (getFeatureOptionList):
2323
2324 2012-05-03  Dirk Pranke  <dpranke@chromium.org>
2325
2326         REGRESSION: run-webkit-tests failing on Chromium Mac
2327         https://bugs.webkit.org/show_bug.cgi?id=85459
2328
2329         Unreviewed, build fix.
2330
2331         Turns out select doesn't throw IOErrors, it has it's own kind of
2332         error :(. Trap that instead.
2333
2334         * Scripts/webkitpy/layout_tests/port/server_process.py:
2335         (ServerProcess._wait_for_data_and_update_buffers_using_select):
2336
2337 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2338
2339         [CMake] Rewrite FindCairo.cmake.
2340         https://bugs.webkit.org/show_bug.cgi?id=84895
2341
2342         Reviewed by Daniel Bates.
2343
2344         The old approach relied on pkg-config for finding Cairo (which
2345         introduced a dependency on pkg-config that could be avoided), used
2346         the LibFindMacros code that we should probably remove in the
2347         future and did not use the FindPackageHandleStandardArguments
2348         module.
2349
2350         Change all that by rewriting the module.
2351         - Use the pkg-config output optionally instead of requiring it
2352         like LibFindMacros did.
2353         - Remove the implicit dependency on FreeType which often found it
2354         the wrong way via pkg-config and without considering
2355         CMAKE_PREFIX_PATH.
2356         - Retrieve the Cairo version by looking at cairo-version.h instead
2357         of relying on pkg-config. It requires some additional code for
2358         checking if the desired version has been found, but that will not
2359         be needed once we start depending on CMake 2.8.3 or later.
2360
2361         The only downside is that FPHSA sets <UPPERCASED_NAME>_FOUND
2362         instead of <Name>_FOUND, and to keep things consistent
2363         Cairo_LIBRARIES and Cairo_INCLUDE_DIRS have become CAIRO_LIBRARIES
2364         and CAIRO_INCLUDE_DIRS.
2365
2366         * DumpRenderTree/efl/CMakeLists.txt: Use CAIRO_FOO instead of
2367         Cairo_FOO.
2368         * EWebLauncher/CMakeLists.txt: Ditto.
2369
2370 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2371
2372         Fix modifier and bug sorting on the flakiness dashboard
2373         https://bugs.webkit.org/show_bug.cgi?id=85554
2374
2375         Reviewed by Dirk Pranke.
2376
2377         The sorting assumes that the column header name matches a field
2378         in the testResults object and uses that field for sorting.
2379         Filter the bugs out of the modifiers list and put them in a bugs field
2380         and only put the non-bug modifiers in the modifiers list.
2381
2382         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2383         (populateExpectationsData):
2384         (htmlForSingleTestRow):
2385         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2386
2387 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2388
2389         Fix some state resetting when running unittests. HtmlForSingleTestRow
2390         was flaky.
2391
2392         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2393         (htmlForSingleTestRow):
2394         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2395         (resetGlobals):
2396
2397 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2398
2399         webkitpy: Use PlatformInfo wherever possible in the User class.
2400         https://bugs.webkit.org/show_bug.cgi?id=85549
2401
2402         Reviewed by Dirk Pranke.
2403
2404         Instead of relying on sys.platform to detect whether the current
2405         platform is a Mac, use the PlatformInfo class instead.
2406
2407         * Scripts/webkitpy/common/system/user.py:
2408         (User.__init__):
2409         (User.edit_changelog):
2410
2411 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2412
2413         Remove the percent failed column from the flakiness dashboard
2414         https://bugs.webkit.org/show_bug.cgi?id=85552
2415
2416         Reviewed by Dirk Pranke.
2417
2418         This column isn't useful enough to justify the visual noise.
2419
2420         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2421         (processMissingAndExtraExpectations):
2422         (tableHeaders):
2423         (htmlForSingleTestRow):
2424         (sortTests):
2425         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2426         (test):
2427
2428 2012-05-03  Eric Seidel  <eric@webkit.org>
2429
2430         --coverage should not not use the "feature" options system in build-webkit, similarly remove special-casing for WTF_URL
2431         https://bugs.webkit.org/show_bug.cgi?id=85542
2432
2433         Reviewed by Daniel Bates.
2434
2435         Remove these last two odd-men-out, so that we can generate the default feature options
2436         for build-webkit using the new generate-feature-files command.
2437         The Mac build system has some special casing for ENABLE_ defines which gets them
2438         defined as -D on the build command, even though initially they start out as
2439         environment variables ENABLE_FOO = ENABLE_FOO.  The Feature options in build-webkit
2440         depend on this support, but it doesn't work for WTF_USE_ defines like WTF_USE_WTF_URL.
2441         I'm told that WTF_URL is not actively being worked on, so just removing it for now.
2442         It should probably come back to life as an ENABLE_, or build-webkit or the Xcode projects
2443         should be made generically smarter about these USE_WTF defines.
2444         (There is still WTF_USE_TILED_BACKING_STORE in the feature list, but it only seems
2445         used by Qt.  If someone tried build-webkit --tiled-backing-store on Mac today it
2446         would not work!)
2447
2448         * Scripts/build-webkit:
2449
2450 2012-05-03  Eric Seidel  <eric@webkit.org>
2451
2452         Remove the last bits of Leopard support from our perl scripts
2453         https://bugs.webkit.org/show_bug.cgi?id=85545
2454
2455         Reviewed by Daniel Bates.
2456
2457         Chromium still supports Leopard (for a little while yet), but AppleMacWebKit dropped
2458         Leopard support several months ago.  Remove the last bits of isLeopard() branching
2459         in our Perl scripts, since no one is using this (Chromium has never used this Perl code).
2460
2461         * Scripts/build-webkit:
2462         * Scripts/old-run-webkit-tests:
2463         (countAndPrintLeaks):
2464         (captureSavedCrashLog):
2465         * Scripts/webkitdirs.pm:
2466         (determineArchitecture):
2467         (argumentsForRunAndDebugMacWebKitApp):
2468
2469 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2470
2471         [Chromium] Don't show the WebKit Linux ASAN bot on the flakiness dashboard.
2472         https://bugs.webkit.org/show_bug.cgi?id=85551
2473
2474         Reviewed by Dirk Pranke.
2475
2476         It's not uploading results and in the short term we don't plan to have it start.
2477
2478         * TestResultServer/static-dashboards/builders.js:
2479         (isChromiumWebkitTipOfTreeTestRunner):
2480         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2481
2482 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2483
2484         Show the test name in the dashboard if showChrome==false and we're showing multiple tests
2485         https://bugs.webkit.org/show_bug.cgi?id=85243
2486
2487         Reviewed by Dirk Pranke.
2488
2489         This happens for cases where a test name matches multiple tests.
2490         In practice, this only happens for virtual test suites (e.g. the canvas tests).
2491
2492         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2493         (htmlForIndividualTestOnAllBuildersWithChrome):
2494         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2495         (testHtmlForIndividualTestOnAllBuildersWithChromeNonexistant):
2496         (testHtmlForIndividualTestOnAllBuildersWithChrome):
2497         (testHtmlForIndividualTestOnAllBuildersWithChromeWebkitMaster):
2498         (testHtmlForIndividualTests):
2499
2500 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2501
2502         Port the flakiness dashboard unittests to qunit
2503         https://bugs.webkit.org/show_bug.cgi?id=85544
2504
2505         Reviewed by Adam Barth.
2506
2507         * TestResultServer/static-dashboards/dashboard_base.js:
2508         (appendJSONScriptElements):
2509         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2510         (resetGlobals):
2511         (runExpectationsTest):
2512         (test):
2513         * TestResultServer/static-dashboards/run-unittests.html: Added.
2514
2515 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2516
2517         webkitpy: Remove check for readline on Mac.
2518         https://bugs.webkit.org/show_bug.cgi?id=85547
2519
2520         Reviewed by Eric Seidel.
2521
2522         Python 2.5 is not supported by the WebKit project anymore, and
2523         Python's readline documentation does not mention the module not
2524         being available in the supported versions.
2525
2526         * Scripts/webkitpy/common/system/user.py:
2527
2528 2012-05-03  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2529
2530         [NRWT] Run performance tests with lock
2531         https://bugs.webkit.org/show_bug.cgi?id=78628
2532
2533         Reviewed by Tony Chang.
2534
2535         Locking performance tests (like we do for http tests) will force them
2536         to run in serial. This reduces the load of the machine when running perf
2537         tests and minimizes the chances of the tests to fail, specially
2538         timeouts.
2539
2540         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2541         (Manager.__init__):
2542         (Manager._is_perf_test):
2543         (Manager._test_requires_lock):
2544         * Scripts/webkitpy/layout_tests/port/http_lock.py:
2545
2546 2012-05-03  Dirk Pranke  <dpranke@chromium.org>
2547
2548         Attempt to fix a crash on SL ...
2549
2550         Unreviewed, build fix.
2551
2552         It looks like we may be getting an IOError raised during
2553         select() on chromium mac, and that error isn't being caught.
2554         Pushing it into the try/catch block will hopefully catch this
2555         and allow us to see why the subprocess is crashing.
2556
2557         * Scripts/webkitpy/layout_tests/port/server_process.py:
2558         (ServerProcess._wait_for_data_and_update_buffers_using_select):
2559
2560 2012-05-02  Jon Honeycutt  <jhoneycutt@apple.com>
2561
2562         WebFrameLoaderClient::dispatchWillSendSubmitEvent() needs to be
2563         implemented for WebKit2
2564         https://bugs.webkit.org/show_bug.cgi?id=84304
2565
2566         Reviewed by Jessie Berlin.
2567
2568         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2569         Added new files to project.
2570
2571         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent.cpp: Added.
2572         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
2573         Look for the "DidReceiveWillSendSubmitEvent" message. The message body
2574         will be the dictionary of form control names and values passed to the
2575         injected bundle's willSendSubmitEvent handler; look for the pairs
2576         "textField" -> "text field" and "passwordField" -> "password field".
2577         Check that "hiddenField" is not included.
2578         (TestWebKitAPI::setInjectedBundleClient):
2579         (TestWebKitAPI::TEST):
2580
2581         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent_Bundle.cpp: Added.
2582         (TestWebKitAPI::willSendSubmitEvent):
2583         Post a message to the TestWebKitAPI process with the dictionary of form
2584         control names and values that we received.
2585         (TestWebKitAPI::WillSendSubmitEventTest::WillSendSubmitEventTest):
2586         (TestWebKitAPI::WillSendSubmitEventTest::didCreatePage):
2587         Set up the page's form client.
2588
2589         * TestWebKitAPI/Tests/WebKit2/auto-submitting-form.html: Added.
2590
2591 2012-05-03  Eric Seidel  <eric@webkit.org>
2592
2593         Remove dead code from build-webkit for working around old XCode dependency bug
2594         https://bugs.webkit.org/show_bug.cgi?id=85535
2595
2596         Reviewed by Adam Barth.
2597
2598         Xcode used to have a bug whereby it didn't notice if you changed the #defines
2599         passed on the command line to xcodebuild.  I wrote code to work around this
2600         while bringing up SVG support 5-6 years ago.  I think we can finally remove this
2601         code from build-webkit, since SVG is the only feature flag to ever use it
2602         (modern in-progress features like MathML don't seem to be using the support)
2603         an I suspect the XCode bug has long been fixed.
2604
2605         * Scripts/build-webkit:
2606         * Scripts/webkitperl/features.pm:
2607
2608 2012-05-03  Andy Estes  <aestes@apple.com>
2609
2610         InjectedBundleControllerMac.mm needs to include config.h
2611         https://bugs.webkit.org/show_bug.cgi?id=85531
2612
2613         Reviewed by Andreas Kling.
2614
2615         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm:
2616
2617 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2618
2619         webkitpy: Use PlatformInfo whenever possible in server_process.py.
2620         https://bugs.webkit.org/show_bug.cgi?id=85516
2621
2622         Reviewed by Eric Seidel.
2623
2624         Eliminate direct access to sys.platform wherever possible by using
2625         PlatformInfo.is_win() instead.
2626
2627         * Scripts/webkitpy/layout_tests/port/server_process.py:
2628         (ServerProcess.__init__): Remove the executive parameter as it was
2629         not passed by any caller, obtain a SystemHost from port_obj
2630         instead.
2631         (ServerProcess._log): Style, remove extra empty line.
2632         (ServerProcess._start):
2633         (ServerProcess.stop):
2634         (ServerProcess.kill):
2635         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2636         (TrivialMockPort.__init__): Create a MockSystemHost used by
2637         ServerProcess.
2638         (TestServerProcess.test_broken_pipe): Test different platforms.
2639
2640 2012-05-03  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2641
2642         webkitpy: Style checker broken when committing a PNG
2643         https://bugs.webkit.org/show_bug.cgi?id=85504
2644
2645         Reviewed by Dirk Pranke.
2646
2647         * Scripts/webkitpy/style/checkers/png.py:
2648         (PNGChecker.__init__):
2649
2650 2012-05-03  Philippe Normand  <pnormand@igalia.com>
2651
2652         Unreviewed, fix another call to join(), similar to the ones landed
2653         in r115975.
2654
2655         * Scripts/webkitdirs.pm:
2656         (jhbuildConfigurationChanged):
2657
2658 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2659
2660         webkitpy: Remove sys.platform accesses in CrashLogs.
2661         https://bugs.webkit.org/show_bug.cgi?id=85436
2662
2663         Reviewed by Dirk Pranke.
2664
2665         Instead of directly checking for sys.platform == 'darwin', use a
2666         SystemHost and call PlatformInfo.is_mac().
2667
2668         * Scripts/webkitpy/common/system/crashlogs.py:
2669         (CrashLogs.__init__):
2670         (CrashLogs.find_newest_log):
2671         (CrashLogs._log_directory_darwin):
2672         (CrashLogs._find_newest_log_darwin):
2673         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2674         (CrashLogsTest.test_find_log_darwin):
2675         * Scripts/webkitpy/layout_tests/port/mac.py:
2676         (MacPort._get_crash_log):
2677         * Scripts/webkitpy/layout_tests/port/test.py:
2678         (TestDriver.run_test):
2679         * Scripts/webkitpy/tool/commands/queries.py:
2680         (execute):
2681
2682 2012-05-03  Philippe Normand  <pnormand@igalia.com>
2683
2684         [GTK] join() is wrongly used in webkitdirs.pm since r115532
2685         https://bugs.webkit.org/show_bug.cgi?id=85501
2686
2687         Reviewed by Martin Robinson.
2688
2689         * Scripts/webkitdirs.pm:
2690         (getJhbuildPath):
2691         (buildAutotoolsProject):
2692
2693 2012-05-03  Zan Dobersek  <zandobersek@gmail.com>
2694
2695         [Gtk] ImageDiff crashes when it can't open a display
2696         https://bugs.webkit.org/show_bug.cgi?id=85476
2697
2698         Reviewed by Martin Robinson.
2699
2700         Replace gdk_init() initialization call with the g_type_init()
2701         call. GdkPixbufLoader doesn't actually need the current call to be
2702         made and this change will also remove crashes when ImageDiff
2703         is started in an environment without a DISPLAY env.
2704
2705         * DumpRenderTree/gtk/ImageDiff.cpp:
2706         (main):
2707
2708 2012-05-03  Zan Dobersek  <zandobersek@gmail.com>
2709
2710         ImageDiff should be run inside a properly established environment
2711         https://bugs.webkit.org/show_bug.cgi?id=85292
2712
2713         Reviewed by Martin Robinson.
2714
2715         When creating the ImageDiff server process, pass along a freshly-set-up environment
2716         in which the process should be executed in.
2717
2718         * Scripts/webkitpy/layout_tests/port/webkit.py:
2719         (WebKitPort._start_image_diff_process):
2720
2721 2012-05-03  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2722
2723         [Qt] Allow the web process and WTR to be paused on startup
2724
2725         Makes it easier to debug the web process or run-webkit-tests -2, as you
2726         have ample time to attach gdb to the process.
2727
2728         Reviewed by Simon Hausmann.
2729
2730         * Scripts/webkitpy/layout_tests/port/qt.py:
2731         (QtPort.setup_environ_for_server):
2732         * WebKitTestRunner/qt/main.cpp:
2733         (main):
2734
2735 2012-05-03  Csaba Osztrogonác  <ossy@webkit.org>
2736
2737         Password protect "Stop build" button
2738         https://bugs.webkit.org/show_bug.cgi?id=81982
2739
2740         Reviewed by Ryosuke Niwa.
2741
2742         Only authenticated users should be able to use stopBuild, stopAllBuilds, cancelPendingBuild
2743         features to prevent killing builds on build.webkit.org by SPAM bots.
2744
2745         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2746
2747 2012-05-03  Dongwoo Im  <dw.im@samsung.com>
2748
2749         [DRT] Remove the name of parameter, visibility, from the setPageVisibility stub implementations.
2750         https://bugs.webkit.org/show_bug.cgi?id=85468
2751
2752         Reviewed by Nikolas Zimmermann.
2753
2754         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2755         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2756         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2757         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2758         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2759         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2760         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2761         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2762         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2763         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2764
2765 2012-05-03  Stephanie Lewis  <slewis@apple.com>
2766
2767         https://bugs.webkit.org/show_bug.cgi?id=85450 unbounded growth of JSDOMWindowShells loading pages in the same window
2768         <rdar://problem/11320059> REGRESSION (r115083): PLT3 shows linear memory growth and gets slower with each run
2769
2770         Reviewed by Brady Eidson.
2771
2772         The API added for DOMWindowExtension, didCreateGlobalObjectForFrame, would create a global object
2773         for every world, even those that did not need the callback.  This had the side effect of creating a
2774         JSDOMWindowShell that the associated world didn't necessarily know to clean up.  Instead of creating
2775         unnecessary objects change the API to globalObjectIsAvailableForFrame and do not pass the global object
2776         in the API.  The object can be accessed later by those worlds which require it.
2777
2778         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic.cpp:
2779         (TestWebKitAPI):
2780         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
2781         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp:
2782         (TestWebKitAPI):
2783         (DOMWindowExtensionBasic):
2784         (TestWebKitAPI::DOMWindowExtensionBasic::didCreatePage):
2785         (TestWebKitAPI::DOMWindowExtensionBasic::globalObjectIsAvailableForFrame):
2786         (TestWebKitAPI::globalObjectIsAvailableForFrameCallback):
2787
2788 2012-05-03  Nikolas Zimmermann  <nzimmermann@rim.com>
2789
2790         Not reviewed. Fix mac build, it was missing setPageVisibility/resetPageVisibility stub implementations.
2791
2792         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2793         (LayoutTestController::resetPageVisibility):
2794         (LayoutTestController::setPageVisibility):
2795
2796 2012-05-02  Alexander Færøy  <ahf@0x90.dk>
2797
2798         Rename deviceDPI to devicePixelRatio
2799         https://bugs.webkit.org/show_bug.cgi?id=85049
2800
2801         Reviewed by Kenneth Rohde Christiansen.
2802
2803         Use 1.5 as device pixel ratio for the Qt MiniBrowser.
2804
2805         * MiniBrowser/qt/qml/BrowserWindow.qml:
2806
2807 2012-05-03  Dongwoo Im  <dw.im@samsung.com>
2808
2809         [EFL][DRT] Implement the LayoutTestController's methods related to the Page Visibility API.
2810         https://bugs.webkit.org/show_bug.cgi?id=85347
2811
2812         Reviewed by Nikolas Zimmermann.
2813
2814         * DumpRenderTree/LayoutTestController.h:
2815         (LayoutTestController):
2816         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2817         (LayoutTestController::resetPageVisibility): Not implemented.
2818         (LayoutTestController::setPageVisibility): Not implemented.
2819         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2820         (LayoutTestController::resetPageVisibility): Reset the page visibility.
2821         (LayoutTestController::setPageVisibility): Set the page visibility.
2822         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2823         (LayoutTestController::resetPageVisibility): Not implemented.
2824         (LayoutTestController::setPageVisibility): Not implemented.
2825         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2826         (LayoutTestController::resetPageVisibility): Not implemented.
2827         (LayoutTestController::setPageVisibility): Not implemented.
2828         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2829         (LayoutTestController::resetPageVisibility): Not implemented.
2830         (LayoutTestController::setPageVisibility): Not implemented.
2831         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2832         (LayoutTestController::resetPageVisibility): Not implemented.
2833         (LayoutTestController::setPageVisibility): Not implemented.
2834
2835 2012-05-02  Sheriff Bot  <webkit.review.bot@gmail.com>
2836
2837         Unreviewed, rolling out r115918.
2838         http://trac.webkit.org/changeset/115918
2839         https://bugs.webkit.org/show_bug.cgi?id=85452
2840
2841         "Broke test-webkitpy on Apple Lion bots" (Requested by estes
2842         on #webkit).
2843
2844         * Scripts/webkitpy/common/system/crashlogs.py:
2845         (CrashLogs.__init__):
2846         (CrashLogs.find_newest_log):
2847         (CrashLogs._log_directory_darwin):
2848         (CrashLogs._find_newest_log_darwin):
2849         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2850         (CrashLogsTest.test_find_log_darwin):
2851         * Scripts/webkitpy/layout_tests/port/mac.py:
2852         (MacPort._get_crash_log):
2853         * Scripts/webkitpy/layout_tests/port/test.py:
2854         (TestDriver.run_test):
2855         * Scripts/webkitpy/tool/commands/queries.py:
2856         (execute):
2857
2858 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
2859
2860         webkitpy: Remove sys.platform accesses in CrashLogs.
2861         https://bugs.webkit.org/show_bug.cgi?id=85436
2862
2863         Reviewed by Dirk Pranke.
2864
2865         Instead of directly checking for sys.platform == 'darwin', use a
2866         SystemHost and call PlatformInfo.is_mac().
2867
2868         * Scripts/webkitpy/common/system/crashlogs.py:
2869         (CrashLogs.__init__):
2870         (CrashLogs.find_newest_log):
2871         (CrashLogs._log_directory_darwin):
2872         (CrashLogs._find_newest_log_darwin):
2873         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2874         (CrashLogsTest.test_find_log_darwin):
2875         * Scripts/webkitpy/layout_tests/port/mac.py:
2876         (MacPort._get_crash_log):
2877         * Scripts/webkitpy/layout_tests/port/test.py:
2878         (TestDriver.run_test):
2879         * Scripts/webkitpy/tool/commands/queries.py:
2880         (execute):
2881
2882 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
2883
2884         webkitpy: Special-case win32 calls in file_lock and executive.
2885         https://bugs.webkit.org/show_bug.cgi?id=85431
2886
2887         Reviewed by Dirk Pranke.
2888
2889         Simplify accesses to sys.platform by only checking if sys.platform
2890         is "win32" and removing explicit checks for the else case.
2891
2892         * Scripts/webkitpy/common/system/executive.py:
2893         (Executive.check_running_pid):
2894         * Scripts/webkitpy/common/system/file_lock.py:
2895         (FileLock._create_lock):
2896         (FileLock._remove_lock):
2897
2898 2012-05-02  Ojan Vafai  <ojan@chromium.org>
2899
2900         Remove pink from flakiness dashboard
2901         https://bugs.webkit.org/show_bug.cgi?id=85422
2902
2903         Reviewed by Dirk Pranke.
2904
2905         The pink indicates that the test doesn't match it's listing in test_expectations.txt.
2906         In practice, this just added noise and confused people.
2907
2908         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2909         (htmlForSingleTestRow):
2910         (showLegend):
2911         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
2912         (#legend-contents *):
2913         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2914         (testHtmlForSingleTestRow):
2915
2916 2012-05-02  Gustavo Noronha Silva  <gns@gnome.org>
2917
2918         Unreviewed. Fix mistake I committed when applying review comments.
2919
2920         * jhbuild/jhbuild-wrapper: dependencies_path was getting 'Root'
2921         appended to its path, so jhbuild itself was being cloned and
2922         installed in the wrong place when using WEBKITOUTPUTDIR.
2923
2924 2012-05-02  Dirk Pranke  <dpranke@chromium.org>
2925
2926         nrwt: make ServerProcess work on chromium win
2927         https://bugs.webkit.org/show_bug.cgi?id=85333
2928
2929         Reviewed by Ojan Vafai.
2930
2931         This change implements a poor man's select() that will
2932         slow-spin doing non-blocking reads on the stdout and stderr
2933         named pipes connecting the worker to the driver. Seems to work
2934         and I have yet to see much overhead or ill effects but it
2935         probably needs more testing.
2936
2937         * Scripts/webkitpy/layout_tests/port/chromium.py:
2938         (ChromiumDriver.__init__):
2939         * Scripts/webkitpy/layout_tests/port/server_process.py:
2940         (ServerProcess._start):
2941         (ServerProcess._wait_for_data_and_update_buffers_using_select):
2942         (ServerProcess._wait_for_data_and_update_buffers_using_win32_apis):
2943         (ServerProcess):
2944         (ServerProcess._non_blocking_read_win32):
2945         (ServerProcess._read):
2946         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2947         (TestServerProcess.test_basic): Added.
2948
2949 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
2950
2951         webkitpy: Use a SystemHost instead of PlatformInfo in the PNG checker.
2952         https://bugs.webkit.org/show_bug.cgi?id=85426
2953
2954         Reviewed by Dirk Pranke.
2955
2956         r115880 replaced the direct use of sys.platform with a
2957         PlatformInfo object. Dirk Pranke later told me SystemHost objects
2958         should be preferred, as nothing out of webkitpy.common.system
2959         should create its own PlatformInfo instances.
2960
2961         * Scripts/webkitpy/common/system/systemhost_mock.py:
2962         (MockSystemHost.__init__): Make it possible to specify a custom
2963         filesystem object.
2964         * Scripts/webkitpy/style/checkers/png.py:
2965         (PNGChecker.__init__): Accept a host object instead of a
2966         filesystem one, which is now obtained from host.
2967         (PNGChecker._config_file_path): Retrieve PlatformInfo from
2968         SystemHost.
2969         * Scripts/webkitpy/style/checkers/png_unittest.py:
2970         (PNGCheckerTest.test_init): Adjust calls to PNGChecker to its new
2971         constructor signature.
2972         (PNGCheckerTest.test_check): Ditto.
2973
2974 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
2975
2976         webkitpy: Remove direct checks for sys.platform == 'darwin' in the layout test classes.
2977         https://bugs.webkit.org/show_bug.cgi?id=85424
2978
2979         Reviewed by Tony Chang.
2980
2981         We should reduce direct usage of sys.platform in webkitpy and use
2982         PlatformInfo instead.
2983
2984         * Scripts/webkitpy/layout_tests/servers/http_server.py:
2985         (Lighttpd._prepare_config): Call PlatformInfo.is_mac() instead of
2986         checking for 'darwin' directly.
2987         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
2988         (HttpServerBase.__init__): Ditto.
2989
2990 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
2991
2992         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
2993         https://bugs.webkit.org/show_bug.cgi?id=85419
2994
2995         Reviewed by Dirk Pranke.
2996
2997         As pointed out by Dirk Pranke, a PlatformInfo is not supposed to
2998         be created directly, we should should a SystemHost instead.
2999
3000         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3001         (MainTest.setUp): Create a SystemHost and obtain the PlatformInfo
3002         object from it instead of creating PlatformInfo directly.
3003
3004 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3005
3006         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
3007         https://bugs.webkit.org/show_bug.cgi?id=85419
3008
3009         Rubber-stamped by Dirk Pranke.
3010
3011         Follow-up patch to r115887 which fixes some capitalization problems
3012         pointed out in comments 2 and 4.
3013
3014         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3015         (MainTest.setUp):
3016         (MainTest.test_child_processes_2):
3017         (MainTest.test_child_processes_min):
3018         (MainTest.test_exception_raised):
3019         (MainTest.test_keyboard_interrupt):
3020
3021 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3022
3023         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
3024         https://bugs.webkit.org/show_bug.cgi?id=85419
3025
3026         Reviewed by Dirk Pranke.
3027
3028         Another part of the work towards replacing direct sys.platform
3029         usage with PlatformInfo.
3030
3031         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3032         (MainTest.setUp): Move the setting of SHOULD_TEST_PROCESSES
3033         here. Even though it is costlier, unittest.setUpClass was
3034         introduced in Python 2.7.
3035         (MainTest.test_child_processes_2): Use self.SHOULD_TEST_PROCESSES.
3036         (MainTest.test_child_processes_min): Ditto.
3037         (MainTest.test_exception_raised): Ditto.
3038         (MainTest.test_keyboard_interrupt): Ditto.
3039         (MainTest.test_crash_log): Check for PlatformInfo.is_mac().
3040         (MainTest.test_web_process_crash_log): Ditto, and remove duplicate
3041         check at the end.
3042         (MainTest.test_no_http_tests): Remove check for (cygwin && Python
3043         2.5.2), as we only support Python 2.6+.
3044
3045 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3046
3047         webkitpy: Remove direct sys.platform usage in the PNG checker.
3048         https://bugs.webkit.org/show_bug.cgi?id=85407
3049
3050         Reviewed by Tony Chang.
3051
3052         Work towards replacing direct sys.platform usage with
3053         PlatformInfo.
3054
3055         * Scripts/webkitpy/common/system/platforminfo_mock.py:
3056         (MockPlatformInfo.__init__): Accept os_name and os_version as
3057         arguments.
3058         * Scripts/webkitpy/style/checkers/png.py:
3059         (PNGChecker.__init__): Create a PlatformInfo object instead of
3060         using sys.platform when needed.
3061         (PNGChecker._config_file_path): Check for the current platform via
3062         PlatformInfo.
3063         * Scripts/webkitpy/style/checkers/png_unittest.py:
3064         (PNGCheckerTest.test_check): Update tests after the changes above.
3065
3066 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3067
3068         [watchlist] Update watchlist rules for EFL-related patches.
3069         http://bugs.webkit.org/show_bug.cgi?id=85160
3070
3071         Reviewed by Eric Seidel.
3072
3073         * Scripts/webkitpy/common/config/watchlist: Monitor
3074         Source/WTF/wtf/efl and LayoutTests/platform/efl as well.
3075
3076 2012-05-02  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3077
3078         Revert r115191 - "Make the web view's url property follow the active url"
3079
3080         It uncovered/caused issues in the icon-implementation that can't be
3081         easily fixed, so rolling out instead.
3082
3083         https://bugs.webkit.org/show_bug.cgi?id=77554
3084
3085         Rubber-stamped by Simon Hausmann.
3086
3087         * MiniBrowser/qt/qml/BrowserWindow.qml:
3088         * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp:
3089         (TestWebKitAPI::TEST):
3090
3091 2012-05-02  Zan Dobersek  <zandobersek@gmail.com>
3092
3093         Reftest mismatches are providing undefined diff percent in results.html
3094         https://bugs.webkit.org/show_bug.cgi?id=85283
3095
3096         Reviewed by Ojan Vafai.
3097
3098         When writing test results for a reftest mismatch, also take into account the
3099         difference percent that is produced when generating image diff. This value is
3100         then used when creating test results in the form of dictionaries, stored under
3101         the 'image_diff_percent' key. It ultimately ends up in JSON results data,
3102         providing additional information about the reftest difference.
3103
3104         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3105         (interpret_test_failures): Store the diff percent for reftest failure under the
3106         'image_diff_percent' key.
3107         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
3108         (write_test_result): Also set the new diff_percent attribute of the reftest
3109         mismatch failure to the value of the difference percent ImageDiff outputs.
3110         * Scripts/webkitpy/layout_tests/models/test_failures.py:
3111         (FailureReftestMismatch.__init__): Add the new diff_percent attribute with the
3112         default value of None.
3113         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3114         (EndToEndTest.test_reftest_with_two_notrefs): Update the test case expected results.
3115
3116 2012-05-02  Terry Anderson  <tdanderson@chromium.org>
3117
3118         Addition of a contributor to committers.py
3119         https://bugs.webkit.org/show_bug.cgi?id=85324
3120
3121         Reviewed by Unreviewed.
3122
3123         * Scripts/webkitpy/common/config/committers.py:
3124
3125 2012-05-02  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
3126
3127         [EFL] EFL's LayoutTestController overridePreference implementation
3128         https://bugs.webkit.org/show_bug.cgi?id=83007
3129
3130         Reviewed by Antonio Gomes.
3131
3132         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3133         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
3134         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3135         (toBool): aux function converting JSStringRef to bool
3136         (toInt): aux function converting JSStringRef to int
3137         (LayoutTestController::overridePreference): added implementation
3138
3139 2012-05-01  Raphael Kubo da Costa  <rakuco@webkit.org>
3140
3141         webkitpy: Fix sys.platform check in find_files_unittest.
3142         https://bugs.webkit.org/show_bug.cgi?id=85339
3143
3144         Reviewed by Ryosuke Niwa.
3145
3146         Follow-up to r101618 -- 'win' is not a valid sys.platform value,
3147         'win32' should be used instead. The check always returned false
3148         due to this, so assert_normalize() was never called.
3149
3150         * Scripts/webkitpy/common/find_files_unittest.py:
3151         (TestWinNormalize.test_win): Add the missing import for the
3152         FileSystem class and fix the sys.platform check.
3153
3154 2012-05-02  Zan Dobersek  <zandobersek@gmail.com>
3155
3156         [GTK] drag&drop related failures
3157         https://bugs.webkit.org/show_bug.cgi?id=84675
3158
3159         Reviewed by Philippe Normand.
3160
3161         Rather than checking options to see whether we're running pixel tests
3162         the pixel_tests argument passed to GtkDriver._start should be used as
3163         it also covers cases when pixel tests should be run because of testing
3164         reftests. This again creates different Xvfb processes with proper
3165         display IDs for a worker that's performing both ordinary and pixel
3166         tests, as was the case before changes made in r114788.
3167
3168         * Scripts/webkitpy/layout_tests/port/gtk.py:
3169         (GtkDriver._start):
3170
3171 2012-05-02  Simon Hausmann  <simon.hausmann@nokia.com>
3172
3173         [Qt] Enable device orientation/motion with Qt 5
3174
3175         Reviewed by Tor Arne Vestbø.
3176
3177         Enable device orientation/motion features with Qt 5 if Qt Sensors is available.
3178
3179         * qmake/mkspecs/features/features.prf:
3180
3181 2012-05-01  Raphael Kubo da Costa  <rakuco@webkit.org>
3182
3183         webkitpy: Remove unused platform imports.
3184         https://bugs.webkit.org/show_bug.cgi?id=85341
3185
3186         Reviewed by Eric Seidel.
3187
3188         Remove a few uses of `import platform' which were either rendered
3189         unneeded after r105931 or were never needed at all.
3190
3191         * Scripts/webkitpy/common/system/executive.py:
3192         * Scripts/webkitpy/tool/bot/flakytestreporter.py:
3193
3194 2012-05-01  Terry Anderson  <tdanderson@chromium.org>
3195
3196         [chromium] Accept four parameters when dispatching a WebInputEvent::GestureTap in chromium DRT eventSender
3197         https://bugs.webkit.org/show_bug.cgi?id=85289
3198
3199         Reviewed by Eric Seidel.
3200
3201         * DumpRenderTree/chromium/EventSender.cpp:
3202         (EventSender::gestureEvent):
3203             This change will allow for testing of WK85101. Currently
3204             no tests use the extra parameters deltaX and deltaY. Added
3205             a check for the size of |arguments|.
3206
3207 2012-05-01  Eric Seidel  <eric@webkit.org>
3208
3209         Add myself to the watchlist to watch all rendering changes, as well as sort the watchers in the list
3210         https://bugs.webkit.org/show_bug.cgi?id=85305
3211
3212         Reviewed by Adam Barth.
3213
3214         * Scripts/webkitpy/common/config/watchlist:
3215
3216 2012-05-01  Sheriff Bot  <webkit.review.bot@gmail.com>
3217
3218         Unreviewed, rolling out r115735.
3219         http://trac.webkit.org/changeset/115735
3220         https://bugs.webkit.org/show_bug.cgi?id=85314
3221
3222         cause two tests to crash due to an ASSERTION failure
3223         (Requested by zhenyao on #webkit).
3224
3225         * DumpRenderTree/chromium/EventSender.cpp:
3226         (EventSender::gestureEvent):
3227
3228 2012-05-01  Terry Anderson  <tdanderson@chromium.org>
3229
3230         [chromium] Accept four parameters when dispatching a WebInputEvent::GestureTap in chromium DRT eventSender
3231         https://bugs.webkit.org/show_bug.cgi?id=85289
3232
3233         Reviewed by Eric Seidel.
3234
3235         * DumpRenderTree/chromium/EventSender.cpp:
3236         (EventSender::gestureEvent):
3237             This change will allow for testing of WK85101. Currently no tests use the extra
3238             parameters deltaX and deltaY.
3239
3240 2012-05-01  Zan Dobersek  <zandobersek@gmail.com>
3241
3242         Print out stderr output of ImageDiff if it is present
3243         https://bugs.webkit.org/show_bug.cgi?id=85285
3244
3245         Reviewed by Philippe Normand.
3246
3247         Log the stderr output of ImageDiff as a warning if present. This is
3248         especially helpful when ImageDiff crashes and stderr data might contain
3249         relevant information about what caused the crash.
3250
3251         * Scripts/webkitpy/layout_tests/port/webkit.py:
3252         (WebKitPort._read_image_diff):
3253
3254 2012-05-01  Kenneth Rohde Christiansen  <kenneth@webkit.org>
3255
3256         [Qt] Add an experimental extension to set the min. contents width
3257         https://bugs.webkit.org/show_bug.cgi?id=85281
3258
3259         Reviewed by Antonio Gomes.
3260
3261         * MiniBrowser/qt/qml/BrowserWindow.qml:
3262
3263         Set the value of preferredMinimumContentsWidth to 980.
3264
3265 2012-04-30  Ojan Vafai  <ojan@chromium.org>
3266
3267         Fix tests to not depend on the actual list of builders at build.chromium.org.
3268
3269         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
3270         (testHtmlForIndividualTestOnAllBuildersWithChrome):
3271         (testHtmlForIndividualTestOnAllBuildersWithChromeWebkitMaster):
3272
3273 2012-04-29  Sam Weinig  <sam@webkit.org>
3274
3275         Remove BlobBuilder
3276         https://bugs.webkit.org/show_bug.cgi?id=84036
3277
3278         Reviewed by Anders Carlsson.
3279
3280         * Scripts/build-webkit:
3281         * qmake/mkspecs/features/features.prf:
3282         Make exposing the WebKitBlobBuilder JS object conditional on a new
3283         ENABLE_LEGACY_WEBKIT_BLOB_BUILDER flag. Don't enable this for the Mac,
3284         but do for everyone else.
3285
3286 2012-04-30  Amruth Raj Padmanabhuni  <amruthraj@motorola.com>
3287
3288         [GTK] Add glib as a dependency to build glib-networking
3289         https://bugs.webkit.org/show_bug.cgi?id=85181
3290
3291         Reviewed by Martin Robinson.
3292
3293         * gtk/jhbuild.modules:
3294
3295 2012-04-30  Alexis Menard  <alexis.menard@openbossa.org>
3296
3297         Move myself to reviewers.
3298
3299         Reviewed by Andreas Kling.
3300
3301         * Scripts/webkitpy/common/config/committers.py:
3302
3303 2012-04-30  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
3304
3305         Unreviewed, move myself to reviewers
3306
3307         * Scripts/webkitpy/common/config/committers.py:
3308
3309 2012-04-30  Carlos Garcia Campos  <cgarcia@igalia.com>
3310
3311         Unreviewed. Fix make distcheck.
3312
3313         * TestWebKitAPI/GNUmakefile.am: Add missing headers to compilation
3314         and resources to EXTRA_DIST.
3315
3316 2012-04-30  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3317
3318         [GTK] DRT support for layoutTestController.removeOriginAccessWhitelistEntry
3319         https://bugs.webkit.org/show_bug.cgi?id=85156
3320
3321         Reviewed by Martin Robinson.
3322
3323         Add missing implementation removeOriginAccessWhitelistEntry to
3324         GTK's LayoutTestController.
3325
3326         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3327         (LayoutTestController::removeOriginAccessWhitelistEntry):
3328
3329 2012-04-29  Maciej Stachowiak  <mjs@apple.com>
3330
3331         REGRESSION: On Lion, run-webkit-tests changes the display color profile even when not running pixel tests
3332         https://bugs.webkit.org/show_bug.cgi?id=81729
3333
3334         Reviewed by Dan Bernstein.
3335
3336         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3337         (Manager._set_up_run): Only launch LayoutTestHelper when running pixel tests.
3338
3339 2012-04-29  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3340
3341         [GTK] DRT needs an implementation of layoutTestController.setDomainRelaxationForbiddenForURLScheme
3342         https://bugs.webkit.org/show_bug.cgi?id=85131
3343
3344         Reviewed by Martin Robinson.
3345
3346         Add missing implementation setDomainRelaxationForbiddenForURLScheme to
3347         GTK's LayoutTestController.
3348
3349         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3350         (LayoutTestController::setDomainRelaxationForbiddenForURLScheme):
3351
3352 2012-04-28  Sam Weinig  <sam@webkit.org>
3353
3354         Smooth scrolling needs a new key
3355         <rdar://problem/11331632>
3356
3357         Reviewed by Dan Bernstein.
3358
3359         * DumpRenderTree/mac/DumpRenderTree.mm:
3360         (resetDefaultsToConsistentValues):
3361         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm:
3362         (TestWebKitAPI::InjectedBundleController::platformInitialize):
3363         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
3364         (WTR::InjectedBundle::platformInitialize):
3365         Update for new key.
3366
3367 2012-04-28  Zan Dobersek  <zandobersek@gmail.com>
3368
3369         [WK2] http/tests/navigation/anchor-frames-gbk.html fails
3370         https://bugs.webkit.org/show_bug.cgi?id=76896
3371
3372         Reviewed by Darin Adler.
3373
3374         Only dump as text if currently dumping render tree. This ensures that
3375         calling dumpAsText in a test with dumpChildFramesAsText already being
3376         called doesn't override the first decision, possibly making tests fail.
3377
3378         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3379         (WTR::LayoutTestController::dumpAsText):
3380
3381 2012-04-28  Dan Winship  <danw@gnome.org>
3382
3383         Unreviewed; Add me to the SoupNetwork watchlist
3384
3385         * Scripts/webkitpy/common/config/watchlist:
3386
3387 2012-04-27  Gustavo Noronha Silva  <gns@gnome.org>
3388
3389         [GTK] jhbuild cloning is not following WEBKITOUTPUTDIR.
3390         https://bugs.webkit.org/show_bug.cgi?id=76161
3391
3392         Reviewed by Martin Robinson.
3393
3394         * Scripts/webkitdirs.pm:
3395         (getJhbuildPath): New method to obtain the jhbuild base directory,
3396         using the product base directory
3397         (jhbuildConfigurationChanged): Use the new method
3398         (buildAutotoolsProject): Ditto.
3399         * efl/jhbuildrc: Use WEBKITOUTPUTDIR when calculating the path.
3400         * gtk/jhbuildrc: Ditto.
3401         * jhbuild/jhbuild-wrapper: Ditto.
3402
3403 2012-04-27  Scott Graham  <scottmg@chromium.org>
3404
3405         [chromium] make case of npTestNetscapePlugIn.dll match case in copy rule
3406         https://bugs.webkit.org/show_bug.cgi?id=85083
3407
3408         Reviewed by Dirk Pranke.
3409
3410         Make output product_name for npTestNetscapePlugIn.dll match the case
3411         of the copy_TestNetscapePlugIn rule. This is required for ninja, which
3412         is more particular about case matching in rules than the VS build.
3413
3414         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
3415
3416 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
3417
3418         [chromium] use "drt-style" output, not "test-shell-style" output, on mac and linux DRT
3419         https://bugs.webkit.org/show_bug.cgi?id=84917
3420
3421         Unreviewed, build fix.
3422
3423         Reland the change in r115453 with a fix for chromium win.
3424         The logic in ChromiumDriver was busted on windows, causing DRT
3425         to get launched without the --test-shell flag.
3426
3427         * Scripts/webkitpy/layout_tests/port/chromium.py:
3428         (ChromiumDriver):
3429         (ChromiumDriver.__init__):
3430         (ChromiumDriver._wrapper_options):
3431         (ChromiumDriver.cmd_line):
3432         (ChromiumDriver._start):
3433         (ChromiumDriver.has_crashed):
3434         (ChromiumDriver.run_test):
3435         (ChromiumDriver.stop):
3436         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3437         (ChromiumDriverTest.setUp):
3438         (ChromiumDriverTest.test_stop):
3439         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
3440         (ChromiumDriverTest.test_two_drivers):
3441
3442 2012-04-27  Peter Beverloo  <peter@chromium.org>
3443
3444         Add John Grabowski as a non-committer to committers.py
3445         https://bugs.webkit.org/show_bug.cgi?id=85080
3446
3447         Reviewed by Dirk Pranke.
3448
3449         He's being CC'ed often enough because of his involvement in the Chromium
3450         side of Chrome for Android upstreaming.
3451
3452         * Scripts/webkitpy/common/config/committers.py:
3453           add jrg@chromium.org
3454
3455 2012-04-27  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
3456
3457         [Qt] Fix minimal build.
3458         https://bugs.webkit.org/show_bug.cgi?id=85045
3459
3460         Reviewed by Tor Arne Vestbø.
3461
3462         Do not set conflicting values of ENABLE_XSLT.
3463
3464         * qmake/mkspecs/features/features.prf:
3465
3466 2012-04-27  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3467
3468         [EFL][DRT] EFL should sanitize pulseaudio module configuration too
3469         https://bugs.webkit.org/show_bug.cgi?id=85047
3470
3471         Reviewed by Dirk Pranke.
3472
3473         PulseAudio test setup sanitization shared between EFL and GTK.
3474
3475         * Scripts/webkitpy/layout_tests/port/efl.py:
3476         (EflPort):
3477         (EflPort.setup_test_run):
3478         (EflPort.clean_up_test_run):
3479         * Scripts/webkitpy/layout_tests/port/gtk.py:
3480         (GtkPort): Moving pulseaudio unload/restore out to shared class.
3481         * Scripts/webkitpy/layout_tests/port/pulseaudio_sanitizer.py: Added.
3482         (PulseAudioSanitizer):
3483         (PulseAudioSanitizer._unload_pulseaudio_module):
3484         (PulseAudioSanitizer._restore_pulseaudio_module):
3485
3486 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
3487
3488         Unreviewed, revert r115453 - broke chromium win bots
3489
3490         * Scripts/webkitpy/layout_tests/port/chromium.py:
3491         (ChromiumPort._path_to_image_diff):
3492         (ChromiumDriver):
3493         (ChromiumDriver.__init__):
3494         (ChromiumDriver._wrapper_options):
3495         (ChromiumDriver.cmd_line):
3496         (ChromiumDriver.stop):
3497         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3498         (ChromiumDriverTest.setUp):
3499         (ChromiumDriverTest.test_stop):
3500         (ChromiumDriverTest.test_two_drivers):
3501         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
3502
3503 2012-04-27  Ryosuke Niwa  <rniwa@webkit.org>
3504
3505         Chromium Windows Perf bots timeout due to not output
3506         https://bugs.webkit.org/show_bug.cgi?id=84940
3507
3508         Reviewed by Dirk Pranke.
3509
3510         Dirk and I investigated the issue on the bot but we couldn't figure out what is going wrong.
3511         Since we run-perf-tests don't need any of fancy feature printer provides, just use python's
3512         built-in logging module instead. Printing out to stdout and stderr seem to work so hopefully
3513         this will fix the issue on the bot.
3514
3515         * Scripts/run-perf-tests:
3516         * Scripts/webkitpy/performance_tests/perftest.py:
3517         (PerfTest.run):
3518         (PerfTest.run_failed):
3519         (PerfTest.parse_output):
3520         (PerfTest.output_statistics):
3521         (ChromiumStylePerfTest.parse_output):
3522         (PageLoadingPerfTest.run):
3523         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
3524         (MainTest.test_parse_output):
3525         (MainTest.test_parse_output_with_failing_line):
3526         (TestPageLoadingPerfTest):
3527         (TestPageLoadingPerfTest.test_run):
3528         (TestPageLoadingPerfTest.test_run_with_bad_output):
3529         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3530         (PerfTestsRunner.__init__):
3531         (PerfTestsRunner._parse_args):
3532         (PerfTestsRunner.run):
3533         (PerfTestsRunner._upload_json):
3534         (PerfTestsRunner._print_status):
3535         (PerfTestsRunner._run_tests_set):
3536         (PerfTestsRunner._run_single_test):
3537         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
3538         (create_runner):
3539         (test_run_test_set):
3540         (test_run_test_set_kills_drt_per_run):
3541         (test_run_test_pause_before_testing):
3542         (test_run_test_set_for_parser_tests):
3543         (test_run_test_set_with_json_output):
3544         (test_run_test_set_with_json_source):
3545         (test_run_test_set_with_multiple_repositories):
3546         (test_upload_json):
3547         (test_parse_args):
3548
3549 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
3550
3551         [chromium] use "drt-style" output, not "test-shell-style" output, on mac and linux DRT
3552         https://bugs.webkit.org/show_bug.cgi?id=84917
3553
3554         Reviewed by Ojan Vafai.
3555
3556         This change changes the chromium new-run-webkit-tests
3557         implementation to use the code in WebKitDriver and ServerProcess
3558         by default on Mac and Linux, instead of using the old-style
3559         Chromium/test_shell style of output.
3560
3561         We will still use the test_shell style on Windows for now
3562         and you can still get the old style of behavior with
3563         new-run-webkit-tests --additional-drt-flag --test-shell
3564
3565         * Scripts/webkitpy/layout_tests/port/chromium.py:
3566         (ChromiumDriver):
3567         (ChromiumDriver.__init__):
3568         (ChromiumDriver._wrapper_options):
3569         (ChromiumDriver.cmd_line):
3570         (ChromiumDriver._start):
3571         (ChromiumDriver.has_crashed):
3572         (ChromiumDriver.run_test):
3573         (ChromiumDriver.stop):
3574         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3575         (ChromiumDriverTest.setUp):
3576         (ChromiumDriverTest.test_stop):
3577         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
3578         (ChromiumDriverTest.test_two_drivers):
3579
3580 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
3581
3582         webkitpy: fix race in unit tests for manager_worker_broker
3583         https://bugs.webkit.org/show_bug.cgi?id=84925
3584
3585         Reviewed by Ojan Vafai.
3586
3587         I have been intermittently seeing hangs when running
3588         test-webkitpy on Mac SL; it appears that there's a bug (?) in
3589         the multiprocessing module where queue references are getting
3590         leaked or not cleaned up somehow. By explicitly closing the
3591         queues, things seem to be happy, so this patch adds a cleanup()
3592         method to the message broker and modifies the manager (and the
3593         unit tests) to call it. This may get rid of the intermittent
3594         error on shutdown on Linux I've seen as well.
3595
3596         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3597         (Manager._run_tests):
3598         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
3599         (_Broker.__del__):
3600         (_Broker):
3601         (_Broker.cleanup):
3602         (_BrokerConnection.cleanup):
3603         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
3604         (_TestsMixin.test_name):
3605         (_TestsMixin.test_cancel):
3606         (_TestsMixin.test_done):
3607         (_TestsMixin.test_unknown_message):
3608
3609 2012-04-27  Gavin Peters  <gavinp@chromium.org>
3610
3611         Add new ENABLE_LINK_PRERENDER define to control the Prerendering API
3612         https://bugs.webkit.org/show_bug.cgi?id=84871
3613
3614         Reviewed by Adam Barth.
3615
3616         Prerendering is currently covered by the ENABLE_LINK_PREFETCH macro, but the new Prerendering
3617         API separates it from prefetching.  Having separate include guards lets ports enable prefetching,
3618         a relatively easy change, without needing to build the infrastructure for prerendering, which
3619         is considerably more complicated.
3620
3621 2012-04-27  Philippe Normand  <pnormand@igalia.com>
3622
3623         Unreviewed, GTK gardening.
3624
3625         * Scripts/run-gtk-tests:
3626         (TestRunner): Skip flaky WebKit2 API test.
3627
3628 2012-04-27  Philippe Normand  <pnormand@igalia.com>
3629
3630         Unreviewed, GTK gardening.
3631
3632         * Scripts/run-gtk-tests:
3633         (TestRunner): Fix another typo. So easy to get confused by a Test
3634         name and the actual name of the executable...
3635
3636 2012-04-27  Philippe Normand  <pnormand@igalia.com>
3637
3638         Unreviewed, GTK gardening.
3639
3640         * Scripts/run-gtk-tests:
3641         (TestRunner): Skip failing WebKit2 API test.
3642
3643 2012-04-27  Christophe Dumez  <christophe.dumez@intel.com>
3644
3645         [EFL] DRT should support LayoutTestController's dumpResourceLoadCallbacks()
3646         https://bugs.webkit.org/show_bug.cgi?id=81891
3647
3648         Reviewed by Philippe Normand.
3649
3650         Implement support for LayoutTestController's
3651         dumpResourceLoadCallbacks() in EFL's DumpRenderTree by catching new
3652         ewk_view signals about resource loading and print out the expected
3653         information. This allows for several test cases to be removed from the
3654         skip list.
3655
3656         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3657         (DumpRenderTreeChrome::createView):
3658         (descriptionSuitableForTestResult):
3659         (DumpRenderTreeChrome::onResponseReceived):
3660         (DumpRenderTreeChrome::onResourceLoadFinished):
3661         (DumpRenderTreeChrome::onResourceLoadFailed):
3662         (DumpRenderTreeChrome::onNewResourceRequest):
3663         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
3664         (DumpRenderTreeChrome):
3665
3666 2012-04-27  Philippe Normand  <pnormand@igalia.com>
3667
3668         Unreviewed, GTK typo fix.
3669
3670         * Scripts/run-gtk-tests:
3671         (TestRunner): Fix-up skipped WebKit2 API test name TestWKConnection.
3672
3673 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
3674
3675         [SOUP] Add a way to register custom uri schemes in WebKit2
3676         https://bugs.webkit.org/show_bug.cgi?id=84130
3677
3678         Reviewed by Martin Robinson.
3679
3680         * Scripts/webkitpy/style/checkers/cpp.py:
3681         (check_identifier_name_in_declaration): Add webkit_soup prefix to
3682         the list of exceptions to the underscores in identifiers rule.
3683
3684 2012-04-26  Ryosuke Niwa  <rniwa@webkit.org>
3685
3686         Extract PerfTestFactory
3687         https://bugs.webkit.org/show_bug.cgi?id=83944
3688
3689         Reviewed by Dirk Pranke.
3690
3691         Extracted PerfTestFactory out of PerfTest. This class creates appropriate PerfTest object given
3692         a test name and a path. Also removed dirname from the constructors of PerfTest and its subclasses.
3693
3694         Also added simple test cases to sanity check PerfTestFactory.
3695
3696         * Scripts/webkitpy/performance_tests/perftest.py:
3697         (PerfTest.__init__):
3698         (PerfTest.test_name):
3699         (ChromiumStylePerfTest.__init__):
3700         (PageLoadingPerfTest.__init__):
3701         (PageLoadingPerfTest.run):
3702         (PerfTestFactory):
3703         (PerfTestFactory.create_perf_test):
3704         * Scripts/webkitpy/performance_tests/perftest_unittest.py:
3705         (MainTest.test_parse_output):
3706         (MainTest.test_parse_output_with_failing_line):
3707         (TestPageLoadingPerfTest.test_run):
3708         (TestPageLoadingPerfTest.test_run_with_bad_output):
3709         (TestPerfTestFactory):
3710         (TestPerfTestFactory.test_regular_test):
3711         (TestPerfTestFactory.test_inspector_test):
3712         (TestPerfTestFactory.test_page_loading_test):
3713         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3714         (PerfTestsRunner):
3715         (PerfTestsRunner._collect_tests):
3716         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
3717         (run_test):
3718         (_tests_for_runner):
3719
3720 2012-04-26  Dirk Pranke  <dpranke@chromium.org>
3721
3722         nrwt: clean up server process, webkit driver so chromium can use it
3723         https://bugs.webkit.org/show_bug.cgi?id=84910
3724
3725         Reviewed by Ojan Vafai.
3726
3727         This change moves the "sample a process" logic out of
3728         server_process.py and into a port-specific class (where really
3729         only the mac has an implementation), and also preemptively kills
3730         DRT when a test times out in WebKitDriver (rather than waiting through the
3731         additional delays caused by calling stop() when we would want to
3732         restart the driver generically in worker.py).
3733
3734         These changes will make it possible for the chromium port to
3735         switch over to the stock WebKitDriver implementation, at least
3736         on mac and linux.
3737
3738         * Scripts/webkitpy/layout_tests/port/base.py:
3739         (Port.sample_process):
3740         * Scripts/webkitpy/layout_tests/port/mac.py:
3741         (MacPort.sample_process):
3742         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3743         (test_helper_fails_to_stop):
3744         (test_sample_process):
3745         (test_sample_process.logging_run_command):
3746         (test_sample_process_throws_exception):
3747         (test_sample_process_throws_exception.throwing_run_command):
3748         * Scripts/webkitpy/layout_tests/port/server_process.py:
3749         (ServerProcess._log):
3750         (ServerProcess._handle_timeout):
3751         (ServerProcess.stop):
3752         (ServerProcess):
3753         (ServerProcess.kill): Here we add a method to immediately stop
3754         the process rather than trying to shut it down cleanly.
3755         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
3756         (TestServerProcess.test_broken_pipe):
3757         * Scripts/webkitpy/layout_tests/port/webkit.py:
3758         (WebKitDriver.run_test): Fix an issue where we weren't passing
3759         along any per-test args (only needed for Chromium, but still).
3760         Also, kill the driver immediately if we time out a test.
3761
3762 2012-04-26  Benjamin Poulain  <bpoulain@apple.com>
3763
3764         ObjcClass::methodsNamed() can leak if buffer is dynamically allocated
3765         https://bugs.webkit.org/show_bug.cgi?id=84668
3766
3767         Reviewed by Alexey Proskuryakov.
3768
3769         Extend ObjCPlugin to support the new layout tests of the Objective-C bridge.
3770
3771         * DumpRenderTree/mac/ObjCPlugin.m:
3772         (+[ObjCPlugin isSelectorExcludedFromWebScript:]):
3773         (+[ObjCPlugin webScriptNameForSelector:]):
3774         (-[ObjCPlugin methodMappedToLongName]):
3775         (-[ObjCPlugin testConversionColon:]):
3776         (-[ObjCPlugin _:]):
3777
3778 2012-04-26  Dimitri Glazkov  <dglazkov@chromium.org>
3779
3780         Unreviewed, rolling out r115340.
3781         http://trac.webkit.org/changeset/115340
3782         https://bugs.webkit.org/show_bug.cgi?id=83048
3783
3784         Does not work with Windows.
3785
3786         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
3787         (SingleTestRunner._run_reftest):
3788         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
3789         (copy_file):
3790         * Scripts/webkitpy/layout_tests/port/base.py:
3791         (Port._parse_reftest_list):
3792         (Port.reference_files):
3793         (Port._real_tests):
3794         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3795         (PortTest.test_reference_files):
3796         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3797         (ChromiumDriverTest.test_write_command_and_read_line):
3798
3799 2012-04-26  Christophe Dumez  <christophe.dumez@intel.com>
3800
3801         [EFL] Enable VIDEO_TRACK feature
3802         https://bugs.webkit.org/show_bug.cgi?id=84830
3803
3804         Reviewed by Gustavo Noronha Silva.
3805
3806         Enable VIDEO_TRACK by default on EFL port whenever build-webkit is
3807         used.
3808
3809         * Scripts/build-webkit:
3810
3811 2012-04-08  Robert Hogan  <robert@webkit.org>
3812
3813         new-webkit-run-tests: handle ref tests from the CSS test suite
3814         https://bugs.webkit.org/show_bug.cgi?id=83048
3815
3816         Reviewed by Ryosuke Niwa.
3817
3818         This allows new-run-webkit-tests to run reference tests created for the CSS test suite. It removes a big swathe
3819         of pixel and text results that are no longer required now that the reference results are used and adds in the support
3820         files from the CSS test suite that allow the ref tests to run.
3821
3822         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
3823         (SingleTestRunner._run_reftest): Assert that the reference result exists. Previously the test would
3824          run without it and report a pass!
3825         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
3826         (copy_file): ensure the output directory exists, it may not if the reference test is
3827           contained in a folder that contains nothing but reference tests.
3828         * Scripts/webkitpy/layout_tests/port/base.py:
3829         (Port.available_reference_files): Add as a helper. Handle embedded relative paths, e.g: some/../path
3830         (Port.reference_files): Only return reference results in the reftest list that actually exist on
3831           disk. Also, don't ignore potential reference results on disk that are not in the reftest list, and
3832           don't ignore them just because we have a reftest list for the test's directory.
3833         (Port._real_tests): ignore anything that looks like a test in a 'support' directory, this semantic
3834           is used by the CSS 2.1 test suite.
3835         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3836         (PortTest.test_reference_files_when_reftestlist_and_reference_file_present): Ensure that a test with a reference result
3837           on disk but no reference result listed in the directory's reftest list picks up its reference result.
3838         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3839         (ChromiumDriverTest.test_strip_uri): Test that some/test.html and some/other/../test.html are treated
3840          as the same file.
3841
3842 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
3843
3844         Unreviewed. Fix GTK+ build after r115314.
3845
3846         For some reason it fails while bulding GtkVersioning for WTR, but
3847         it didn't fail neither in EWS nor in my laptop. We don't really
3848         need GtkVersioning in WTR since it already depends on GTK+3
3849         unconditionally.
3850
3851         * WebKitTestRunner/GNUmakefile.am: Remove GtkVersioning.c from
3852         compilation.
3853         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
3854         (WTR::EventSenderProxy::createMouseButtonEvent): Use GTK+ 3 API
3855         directly instead of getDefaultGDKPointerDevice() helper.
3856         (WTR::EventSenderProxy::keyDown): Ditto.
3857         (WTR::EventSenderProxy::mouseMoveTo): Ditto.
3858
3859 2012-04-24  Ojan Vafai  <ojan@chromium.org>
3860
3861         Show flakiness dashboard data in garden-o-matic
3862         https://bugs.webkit.org/show_bug.cgi?id=83716
3863
3864         Reviewed by Dimitri Glazkov.
3865
3866         Put an iframe below the expected/actual results in the Results view.
3867         Size the iframe to it's height.
3868
3869         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
3870         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
3871         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
3872         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
3873         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
3874         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
3875         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
3876
3877 2012-04-26  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3878
3879         [EFL] [DRT] LayoutTestController needs implementation of setJavaScriptProfilingEnabled
3880         https://bugs.webkit.org/show_bug.cgi?id=84576
3881
3882         Reviewed by Martin Robinson.
3883
3884         Add missing implementation setJavaScriptProfilingEnabled to EFL's
3885         LayoutTestController in order to unskip tests rely on this setting.
3886
3887         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3888         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
3889         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3890         (LayoutTestController::setJavaScriptProfilingEnabled):
3891
3892 2012-04-26  Xianzhu Wang  <wangxianzhu@chromium.org>
3893
3894         Add Jing Zhao, Min Qin, Tien-Ren Chen and Hao Zheng into contributors list.
3895
3896         Not reviewed.
3897
3898         * Scripts/webkitpy/common/config/committers.py:
3899
3900 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
3901
3902         [GTK] Test TestWebKitAPI/WebKit2/TestNewFirstVisuallyNonEmptyLayout times out
3903         https://bugs.webkit.org/show_bug.cgi?id=84957
3904
3905         Reviewed by Martin Robinson.
3906
3907         * Scripts/run-gtk-tests:
3908         (TestRunner): Unskip TestNewFirstVisuallyNonEmptyLayout and
3909         TestNewFirstVisuallyNonEmptyLayoutForImages.
3910         * TestWebKitAPI/gtk/PlatformWebViewGtk.cpp:
3911         (TestWebKitAPI::PlatformWebView::PlatformWebView): Show the
3912         toplevel window containing the web view to make sure events are
3913         received.
3914
3915 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
3916
3917         Unreviewed. Add bug numbers to skipped tests.
3918
3919         * Scripts/run-gtk-tests:
3920         (TestRunner):
3921
3922 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
3923
3924         [GTK] Build and run TestWebKitAPI WebKit2 unit tests
3925         https://bugs.webkit.org/show_bug.cgi?id=84446
3926
3927         Reviewed by Philippe Normand.
3928
3929         * Scripts/run-gtk-tests:
3930         (TestRunner): Add TestWebKitAPI/WebKit2 to the list of directories
3931         containing tests. Unskip FunctionalTest.RefCountedStorage, since
3932         we are now using a custom main that initializes WTF threading.
3933         (TestRunner._setup_testing_environment): Set environment variables
3934         required to run WebKit2 tests.
3935         * TestWebKitAPI/GNUmakefile.am:
3936         * TestWebKitAPI/JavaScriptTest.cpp: Use
3937         JavaScriptCore/JSContextRef.h instead of
3938         JavaScriptCore/JavaScriptCore.h since it includes JSStringRefCF.h
3939         unconditionally.
3940         * TestWebKitAPI/PlatformWebView.h: Define PlatformWKView and
3941         PlatformWindow for GTK.
3942         * TestWebKitAPI/Tests/WebKit2/EvaluateJavaScript.cpp: Remove
3943         unneded header that breaks the build of GTK port.
3944         * TestWebKitAPI/Tests/WebKit2/PreventEmptyUserAgent.cpp: Use
3945         JavaScriptCore/JSContextRef.h instead of
3946         JavaScriptCore/JavaScriptCore.h since it includes JSStringRefCF.h
3947         unconditionally.
3948         * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp:
3949         (TestWebKitAPI::TEST): Add expected values for GTK+ port.
3950         * TestWebKitAPI/gtk/InjectedBundleControllerGtk.cpp:
3951         (TestWebKitAPI):
3952         (TestWebKitAPI::InjectedBundleController::platformInitialize):
3953         * TestWebKitAPI/gtk/PlatformUtilitiesGtk.cpp: Added.
3954         (TestWebKitAPI::Util::checkTestFinished):
3955         (TestWebKitAPI::Util::run):
3956         (TestWebKitAPI::Util::sleep):
3957         (TestWebKitAPI::Util::getFilenameFromEnvironmentVariableAsUTF8):
3958         (TestWebKitAPI::Util::createInjectedBundlePath):
3959         (TestWebKitAPI::Util::createURLForResource):
3960         (TestWebKitAPI::Util::URLForNonExistentResource):
3961         (TestWebKitAPI::Util::isKeyDown):
3962         * TestWebKitAPI/gtk/PlatformWebViewGtk.cpp: Added.
3963         (TestWebKitAPI::PlatformWebView::PlatformWebView):
3964         (TestWebKitAPI::PlatformWebView::~PlatformWebView):
3965         (TestWebKitAPI::PlatformWebView::page):
3966         (TestWebKitAPI::PlatformWebView::resizeTo):
3967         (TestWebKitAPI::doKeyStroke):
3968         (TestWebKitAPI::PlatformWebView::simulateSpacebarKeyPress):
3969         (TestWebKitAPI::PlatformWebView::simulateAltKeyPress):
3970         (TestWebKitAPI::doMouseButtonEvent):
3971         (TestWebKitAPI::PlatformWebView::simulateRightClick):
3972         (TestWebKitAPI::PlatformWebView::simulateMouseMove):
3973         * TestWebKitAPI/gtk/main.cpp:
3974         (main):
3975
3976 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
3977
3978         [GTK] run-gtk-tests: Use a timeout per test instead of a global timeout
3979         https://bugs.webkit.org/show_bug.cgi?id=84695
3980
3981         Reviewed by Philippe Normand.
3982
3983         It also adds a command line option to be able to pass a custom
3984         timeout value, instead of hard-coding it. The default timeout is 10
3985         seconds if no other value is passed to the script.
3986
3987         * Scripts/run-gtk-tests:
3988         (TestTimeout): Exception raised when a test times out.
3989         (TestRunner._get_child_pid_from_test_output): Helper function to
3990         get the pid of the running test from gtester output.
3991         (TestRunner._kill_process): Helper funtion ot kill a process
3992         ignoring exceptions if the process is already died.
3993         (TestRunner._run_test_command): Run the test command raising
3994         TestTimeout exception if the test doesn't finish before the given
3995         timeout in seconds.
3996         (TestRunner._run_test_command.alarm_handler): Alarm handler that
3997         raises TestTimeout exception.
3998         (TestRunner._run_test_glib): Use _run_test_command() to run the test.
3999         (TestRunner._run_test_google): Ditto.
4000         (TestRunner.run_tests): Add current test to the list of timed out
4001         test if TestTimeout exception is raised. Show the list of timed
4002         out test at the end.
4003
4004 2012-04-26  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
4005
4006         [EFL] [DRT] LayoutTestController needs implementation of addOriginAccessWhitelistEntry and removeOriginAccessWhitelistEntry
4007         https://bugs.webkit.org/show_bug.cgi?id=83880
4008
4009         Reviewed by Antonio Gomes.
4010
4011         Add missing implementation addOriginAccessWhitelistEntry and
4012         removeOriginAccessWhitelistEntry to EFL's LayoutTestController
4013         in order to unskip tests which rely on this.
4014
4015         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
4016         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
4017         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
4018         (LayoutTestController::addOriginAccessWhitelistEntry):
4019         (LayoutTestController::removeOriginAccessWhitelistEntry):
4020
4021 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
4022
4023         [GTK] Add downloads support to MiniBrowser
4024         https://bugs.webkit.org/show_bug.cgi?id=73662
4025
4026         Reviewed by Philippe Normand.
4027
4028         * MiniBrowser/gtk/BrowserDownloadsBar.c: Added.
4029         (browserDownloadsBarChildRemoved): Close the downloads bar if
4030         there aren't active downloads.
4031         (browserDownloadsBarResponse): Destroy the downloads bar when
4032         closed.
4033         (browser_downloads_bar_init):
4034         (browser_downloads_bar_class_init):
4035         (browser_downloads_bar_new): Create a new downloads bar.
4036         (actionButtonClicked): Cancel the download if cancel button was
4037         clicked, or open the downloaded file if the download finished.
4038         (browserDownloadFinalize):
4039         (browser_download_init):
4040         (browser_download_class_init):
4041         (downloadReceivedResponse): Update download message using the
4042         response URI.
4043         (remainingTime): Helper function to get human readable remaining
4044         time.
4045         (downloadProgress): Update downloads progress.
4046         (downloadReceivedData):
4047         (downloadFinished): Update download message and turn download
4048         button into open.
4049         (downloadFailed): Show error message or destroy the download UI if
4050         it was cancelled.
4051         (browserDownloadNew): Create a new download widget for the given
4052         WebKitDownload.
4053         (browser_downloads_bar_add_download): Add download widget to the
4054         downloads bar.
4055         * MiniBrowser/gtk/BrowserDownloadsBar.h: Added.
4056         * MiniBrowser/gtk/BrowserWindow.c:
4057         (downloadStarted): Create a download widget and add it to the
4058         downloads bar when a new download is started.
4059         (browserWindowConstructed): Connect to
4060         WebKitWebContext::download-started signal.
4061         * MiniBrowser/gtk/GNUmakefile.am: Add new files to compilation.
4062
4063 2012-04-26  Carlos Garcia Campos  <cgarcia@igalia.com>
4064
4065         Unreviewed. Skip unit test failing in debug bot.
4066
4067         * Scripts/run-gtk-tests:
4068         (TestRunner):
4069
4070 2012-04-26  Christophe Dumez  <christophe.dumez@intel.com>
4071
4072         [EFL][DRT] "resource,request,willsend" signal needs to provide more information
4073         https://bugs.webkit.org/show_bug.cgi?id=84670
4074
4075         Reviewed by Andreas Kling.
4076
4077         Handle in EFL's DumpRenderTree the new Ewk_Frame_Resource_Messages now
4078         passed with the "resource,request,willsend" signal and dump
4079         information about the network request and the possible network
4080         response if LayoutTestController's dumpResourceLoadCallbacks() returns
4081         true.
4082
4083         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
4084         (pathSuitableForTestResult):
4085         (urlSuitableForTestResult):
4086         (descriptionSuitableForTestResult):
4087         (DumpRenderTreeChrome::onWillSendRequest):
4088
4089 2012-04-25  Csaba Osztrogonác  <ossy@webkit.org>
4090
4091         [Qt][WK2] Unreviewed, rolling out r113703 because of crashing tests.
4092         http://trac.webkit.org/changeset/113703
4093         https://bugs.webkit.org/show_bug.cgi?id=83024
4094
4095         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
4096         * Tools.pro:
4097         * qmake/mkspecs/features/features.prf:
4098         * qmake/mkspecs/features/functions.prf:
4099
4100 2012-04-25  Maciej Stachowiak  <mjs@apple.com>
4101
4102         Teach LayoutTestHelper to correctly restore the color profile when it's a factory profile
4103         https://bugs.webkit.org/show_bug.cgi?id=84931
4104
4105         Reviewed by Filip Pizlo.
4106
4107         * DumpRenderTree/mac/LayoutTestHelper.m:
4108         (installLayoutTestColorProfile): Account for the possibility of the default
4109         profile being a factory profile.
4110
4111 2012-04-25  Sheriff Bot  <webkit.review.bot@gmail.com>
4112
4113         Unreviewed, rolling out r115240.
4114         http://trac.webkit.org/changeset/115240
4115         https://bugs.webkit.org/show_bug.cgi?id=84928
4116
4117         It broke everything (Requested by Ossy_HOME on #webkit).
4118
4119         * Scripts/webkitpy/layout_tests/port/gtk.py:
4120         (GtkDriver):
4121         (GtkDriver._start):
4122         (GtkDriver._start.x_filter):
4123         (GtkDriver.stop):
4124         (GtkPort._driver_class):
4125         * Scripts/webkitpy/layout_tests/port/qt.py:
4126         (QtPort._build_driver):
4127         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Removed.
4128
4129 2012-04-25  Dirk Pranke  <dpranke@chromium.org>
4130
4131         [chromium] change DRT's "DRT" output to match the other ports
4132         https://bugs.webkit.org/show_bug.cgi?id=84904
4133
4134         Reviewed by Kent Tamura.
4135
4136         Since we run Chromium's DRT in --test-shell mode, the
4137         non-test-shell output has never been really beaten on to make
4138         sure it's right. This fixes a few issues and will be tested
4139         with upcoming NRWT changes that'll land in a separate patch.
4140         In particular we should never output framing text like "\n"
4141         or "#EOF" outside of something outside of TestEventPrinter,
4142         and we weren't handling audio output properly at all.
4143
4144         * DumpRenderTree/chromium/TestEventPrinter.cpp:
4145         (DRTPrinter):
4146         (TestShellPrinter):
4147         (DRTPrinter::handleTextFooter):
4148         (DRTPrinter::handleAudioFooter):
4149         (DRTPrinter::handleTestFooter):
4150         (TestShellPrinter::handleAudioFooter):
4151         * DumpRenderTree/chromium/TestEventPrinter.h:
4152         (TestEventPrinter):
4153         * DumpRenderTree/chromium/TestShell.cpp:
4154         (TestShell::TestShell):
4155         (TestShell::dump):
4156
4157 2012-04-25  Joshua Bell  <jsbell@chromium.org>
4158
4159         Unreviewed: Adding Alec Flett as a contributor (non-committer).
4160
4161         * Scripts/webkitpy/common/config/committers.py:
4162
4163 2012-04-25  Abhishek Arya  <inferno@chromium.org>
4164
4165         Mark myself as a reviewer.
4166         https://bugs.webkit.org/show_bug.cgi?id=84902
4167
4168         Reviewed by Dimitri Glazkov.
4169
4170         * Scripts/webkitpy/common/config/committers.py:
4171
4172 2012-04-25  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
4173
4174         [Qt][NRWT] Run each DRT in it's own xvfb
4175         https://bugs.webkit.org/show_bug.cgi?id=77335
4176
4177         Reviewed by Dirk Pranke.
4178
4179         * Scripts/webkitpy/layout_tests/port/gtk.py:
4180         (GtkPort._driver_class):
4181         * Scripts/webkitpy/layout_tests/port/qt.py:
4182         (QtPort._driver_class):
4183         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Added.
4184         (XvfbDriver):
4185         (XvfbDriver._start):
4186         (XvfbDriver._start.x_filter):
4187         (XvfbDriver.stop):
4188
4189 2012-04-25  Sheriff Bot  <webkit.review.bot@gmail.com>
4190
4191         Unreviewed, rolling out r115222.
4192         http://trac.webkit.org/changeset/115222
4193         https://bugs.webkit.org/show_bug.cgi?id=84874
4194
4195         Caused test regressions on the EFL build (Requested by rakuco
4196         on #webkit).
4197
4198         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
4199         (LayoutTestController::addOriginAccessWhitelistEntry):
4200         (LayoutTestController::removeOriginAccessWhitelistEntry):
4201
4202 2012-04-25  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
4203
4204         [EFL] [DRT] LayoutTestController needs implementation of addOriginAccessWhitelistEntry and removeOriginAccessWhitelistEntry
4205         https://bugs.webkit.org/show_bug.cgi?id=83880
4206
4207         Reviewed by Antonio Gomes.
4208
4209         Add missing implementation addOriginAccessWhitelistEntry and
4210         removeOriginAccessWhitelistEntry to EFL's LayoutTestController
4211         in order to unskip tests which rely on this.
4212
4213         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
4214         (LayoutTestController::addOriginAccessWhitelistEntry):
4215         (LayoutTestController::removeOriginAccessWhitelistEntry):
4216
4217 2012-04-25  Thiago Marcos P. Santos  <thiago.santos@intel.com>
4218
4219         [EFL] [DRT] Reset text zoom factor before running a test
4220         https://bugs.webkit.org/show_bug.cgi?id=84861
4221
4222         Reviewed by Antonio Gomes.
4223
4224         This fix eliminates most of the current flakiness on EFL build
4225         bots, although some CSS tests will now need rebasiling.
4226
4227         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
4228         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
4229
4230 2012-04-25  Milian Wolff  <milian.wolff@kdab.com>
4231
4232         [Qt] add LayoutTestController::setPrinting support to Qt unit tests
4233         https://bugs.webkit.org/show_bug.cgi?id=84246
4234
4235         Reviewed by Simon Hausmann.
4236
4237         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
4238         (WebCore::DumpRenderTree::dump):
4239
4240 2012-04-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
4241
4242         [Qt] Make the web view's url property follow the active url
4243
4244         https://bugs.webkit.org/show_bug.cgi?id=77554
4245
4246         Update  the location bar in the minibrowser to behave
4247         a bit more like normal browsers in terms of when the url will change
4248         and how active focus is handled.
4249
4250         Reviewed by Simon Hausmann.
4251
4252         * MiniBrowser/qt/qml/BrowserWindow.qml:
4253
4254 2012-04-25  Philippe Normand  <pnormand@igalia.com>
4255
4256         Webkit build fails due to missing gstreamer include file on Kubuntu 8.04
4257         https://bugs.webkit.org/show_bug.cgi?id=81913
4258
4259         Reviewed by Tor Arne Vestbø.
4260
4261         * qmake/mkspecs/features/features.prf: Check for gstreamer-0.10
4262         and gstreamer-plugins-base >= 0.10.30 just like the GTK port
4263         does. 0.10.27 is unfortunately not recent enough to get the media
4264         player working reliably.
4265
4266
4267 2012-04-25  Xiaobo Wang  <xbwang@torchmobile.com.cn>
4268
4269         [BlackBerry] Run tests in '/http/tests/local' with local URI
4270         https://bugs.webkit.org/show_bug.cgi?id=84820
4271
4272         Reviewed by Nikolas Zimmermann.
4273
4274         Currently we run all tests under folder "/http/tests/" as HTTP
4275         tests (http://...). However in WebKit ORWT and NRWT scripts, tests in
4276         'http/tests/local' are run with local path (like '/developer/LayoutTests
4277         /http/tests/local/absolute-url-strip-whitespace.html').
4278         With this patch we can keep the same behaviour as other portings.
4279
4280         * DumpRenderTree/blackberry/DumpRenderTree.cpp:
4281         (BlackBerry::WebKit::DumpRenderTree::isHTTPTest):
4282
4283 2012-04-25  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
4284
4285         [EFL][DRT] Reset DRT settings to their default values after a testcase was executed.
4286         https://bugs.webkit.org/show_bug.cgi?id=84709
4287
4288         Reviewed by Eric Seidel.
4289
4290         Not all the settings that can be changed by layout testcases through a JS API were
4291         reset in DumpRenderTreeChrome::resetDefaultsToConsistentValues().
4292
4293         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
4294         (defaultEditingBehavior): returns editing behavior depending on current platform.
4295         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): reset some missed DRT settings
4296
4297 2012-04-24  Dirk Pranke  <dpranke@chromium.org>
4298
4299         nrwt: add ASAN handling to chromium crash logs
4300         https://bugs.webkit.org/show_bug.cgi?id=84664
4301
4302         Reviewed by Ojan Vafai.
4303
4304         * Scripts/webkitpy/layout_tests/port/chromium.py:
4305         (ChromiumPort._get_crash_log):
4306
4307 2012-04-24  Dirk Pranke  <dpranke@chromium.org>
4308
4309         webkit-patch rebaseline-test is retrieving stale expectations
4310    &n