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