[Chromium] ImageDiff should be build for host on Android
[WebKit-https.git] / Tools / ChangeLog
1 2012-05-14  Wei James  <james.wei@intel.com>
2
3         [Chromium] ImageDiff should be build for host on Android
4         https://bugs.webkit.org/show_bug.cgi?id=82039
5
6         Reviewed by Adam Barth.
7
8         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
9
10 2012-05-14  Joone Hur  <joone@webkit.org>
11
12         Unreviewed. Updating email for committers.py script.
13
14         * Scripts/webkitpy/common/config/committers.py:
15
16 2012-05-14  Carlos Garcia Campos  <cgarcia@igalia.com>
17
18         Unreviewed. Fix make distcheck.
19
20         * GNUmakefile.am: Add run-api-tests script to EXTRA_DIST.
21         * gtk/run-api-tests:
22         (TestRunner._lookup_atspi2_binary): Handle the case when
23         pkg_config_file_variable() return None.
24
25 2012-05-14  Sheriff Bot  <webkit.review.bot@gmail.com>
26
27         Unreviewed, rolling out r116935.
28         http://trac.webkit.org/changeset/116935
29         https://bugs.webkit.org/show_bug.cgi?id=86373
30
31         Something is still broken (Requested by Ossy on #webkit).
32
33         * Scripts/webkitpy/common/checksvnconfigfile.py: Removed.
34         * Scripts/webkitpy/style/checkers/png.py:
35         (PNGChecker.check):
36         (PNGChecker):
37         (PNGChecker._config_file_path):
38         * Scripts/webkitpy/tool/commands/download.py:
39         (Land):
40         * Scripts/webkitpy/tool/steps/__init__.py:
41         * Scripts/webkitpy/tool/steps/addsvnmimetypeforpng.py: Removed.
42         * Scripts/webkitpy/tool/steps/addsvnmimetypeforpng_unittest.py: Removed.
43
44 2012-05-14  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
45
46         [Qt] Doesn't build with ENABLE_INSPECTOR=0
47         https://bugs.webkit.org/show_bug.cgi?id=85056
48
49         Reviewed by Tor Arne Vestbø.
50
51         * WebKitTestRunner/TestInvocation.cpp: Ifdef otherwise unused static function.
52         * qmake/mkspecs/features/features.prf:
53         * qmake/mkspecs/features/features.pri:
54
55 2012-05-11  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
56
57         [Qt] Make sure dialogs have their parent and attached properties set before Component.onCompleted
58         https://bugs.webkit.org/show_bug.cgi?id=86222
59
60         The WebView QML api for dialogs is based on delegates for each dialog, which
61         are implemented as components. Since there's no explicit API to show the dialog,
62         the natural place to allow the dialog to choose how to display itself is in
63         Component.onCompleted.
64
65         The issue was that we were using QQmlComponent::create() when instantiating the
66         component, at which point the component did not have a parent or it's attached
67         WebView property set. That meant that we had to do some ugly workaround in our
68         test, using a 1ms timer, to get a callback once the parent was there.
69
70         We now use the QQmlComponent::beginCreate() -> QQmlComponent::completeCreate()
71         pair, which allows us to set the relevant properties before triggering the
72         Component.onCompleted callback. We were already doing this for the context
73         menus (except for setting the parent, which we did afterwards).
74
75         Reviewed by Simon Hausmann.
76
77         * MiniBrowser/qt/qml/BrowserWindow.qml:
78         * MiniBrowser/qt/qml/Dialog.qml:
79
80 2012-05-14  Balazs Ankes  <bank@inf.u-szeged.hu>
81
82         webkit-patch land should automatically add svn:mime-type for .png files
83         https://bugs.webkit.org/show_bug.cgi?id=75825
84         I refactored the png.py to avoid code duplication.
85
86         Reviewed by Dirk Pranke.
87
88         * Scripts/webkitpy/common/checksvnconfigfile.py: Added.
89         (check):
90         (_config_file_path):
91         (errorstr_autoprop):
92         (errorstr_png):
93         * Scripts/webkitpy/style/checkers/png.py:
94         (PNGChecker.check):
95         * Scripts/webkitpy/tool/commands/download.py:
96         (Land):
97         * Scripts/webkitpy/tool/steps/__init__.py:
98         * Scripts/webkitpy/tool/steps/addsvnmimetypeforpng.py: Added.
99         (AddSvnMimetypeForPng):
100         (AddSvnMimetypeForPng.__init__):
101         (AddSvnMimetypeForPng.run):
102         (AddSvnMimetypeForPng._check_pngs):
103         * Scripts/webkitpy/tool/steps/addsvnmimetypeforpng_unittest.py: Added.
104         (MockSCMDetector):
105         (MockSCMDetector.__init__):
106         (MockSCMDetector.display_name):
107         (AddSvnMimetypeForPngTest):
108         (AddSvnMimetypeForPngTest.test_run):
109
110 2012-05-14  David Kilzer  <ddkilzer@apple.com>
111
112         webkit-patch apply-attachment is very slow for big patches
113         https://bugs.webkit.org/show_bug.cgi?id=74929
114
115         Add "excludefield=attachmentdata" to the URL everywhere
116         "ctype=xml" is used to exclude this field from the XML.
117
118         Reviewed by Adam Barth.
119
120         * Scripts/prepare-ChangeLog:
121         (fetchBugDescriptionFromURL):
122         * Scripts/webkitpy/common/config/urls.py:
123         * Scripts/webkitpy/common/config/urls_unittest.py:
124         (URLsTest.test_parse_bug_id):
125         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
126         (Bugzilla.bug_url_for_bug_id):
127
128 2012-05-13  Ryosuke Niwa  <rniwa@webkit.org>
129
130         performance tests should be able to measure runs/sec rather than time
131         https://bugs.webkit.org/show_bug.cgi?id=86021
132
133         Reviewed by Ojan Vafai.
134
135         Allow " runs/s" or " ms" to appear after numerical values in tests.
136
137         * Scripts/webkitpy/performance_tests/perftest.py:
138         (PerfTest):
139
140 2012-05-12  Tim Horton  <timothy_horton@apple.com>
141
142         run-safari and co. should support --guard-malloc command line argument
143         https://bugs.webkit.org/show_bug.cgi?id=86287
144
145         Reviewed by Dan Bernstein.
146
147         Add setUpGuardMallocIfNeeded(), which consults the argument --guard-malloc on OS X,
148         setting DYLD_INSERT_LIBRARIES="/usr/lib/libgmalloc.dylib" if it's enabled.
149
150         Use setUpGuardMallocIfNeeded() before running a WebKit app (whether standalone or
151         in the debugger). This will cause gmalloc to be loaded for gdb as well, but this
152         tends to be harmless.
153
154         Change "malloc guard" to the more common "Guard Malloc" in run-webkit-tests.
155
156         * Scripts/old-run-webkit-tests:
157         * Scripts/webkitdirs.pm:
158         (appendToEnvironmentVariableList):
159         (setUpGuardMallocIfNeeded):
160         (printHelpAndExitForRunAndDebugWebKitAppIfNeeded):
161         (runMacWebKitApp):
162         (execMacWebKitAppForDebugging):
163         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
164         (parse_args):
165
166 2012-05-11  Ojan Vafai  <ojan@chromium.org>
167
168         Add previous/next buttons the garden-o-matics perf panel
169         https://bugs.webkit.org/show_bug.cgi?id=86280
170
171         Reviewed by Dimitri Glazkov.
172
173         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/perf.js:
174         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/perf_unittests.js:
175         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/perf.css:
176
177 2012-05-11  Kevin Ollivier  <kevino@theolliviers.com>
178
179         [wx] Unreviewed build fix. Fix memory exhausted errors from 32-bit ld on *nix.
180
181         * waf/build/settings.py:
182         (common_configure):
183
184 2012-05-11  Dirk Pranke  <dpranke@chromium.org>
185
186         webkit-patch rebaseline should fetch stale expectations :)
187         https://bugs.webkit.org/show_bug.cgi?id=86231
188
189         Unreviewed, build fix.
190
191         This change reverts r115492 and goes back to fetching individual
192         baselines from the layout tests directory, rather than fetching
193         entire zip files. The performance hit over networks for people
194         far away from the bots has turned out to be severe, so we need a
195         better solution to this.
196
197         * Scripts/webkitpy/tool/commands/rebaseline.py:
198         (RebaselineTest._results_url):
199         (RebaselineTest._file_name_for_expected_result):
200         (RebaselineTest._rebaseline_test):
201         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
202         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
203         (test_rebaseline_updates_expectations_file):
204         (test_rebaseline_test):
205         (test_rebaseline_and_copy_test):
206         (test_rebaseline_and_copy_test_no_existing_result):
207         (test_rebaseline_and_copy_test_with_lion_result):
208         (test_rebaseline_and_copy_no_overwrite_test):
209
210 2012-05-11  Christophe Dumez  <christophe.dumez@intel.com>
211
212         [GTK] LayoutTests/svg/batik/text/verticalText.svg fails to display Chinese characters
213         https://bugs.webkit.org/show_bug.cgi?id=86198
214
215         Reviewed by Martin Robinson.
216
217         Update webkitgtk-fonts package to v0.0.3 in order to get new Kochi
218         Mincho / Gothic fonts to display Japanese and Chinese characters.
219
220         * gtk/jhbuild.modules:
221
222 2012-05-11  Christophe Dumez  <christophe.dumez@intel.com>
223
224         [EFL][DRT] SVG Batik test fails to display Chinese characters
225         https://bugs.webkit.org/show_bug.cgi?id=84461
226
227         Reviewed by Martin Robinson.
228
229         Update webkitgtk-fonts package from v0.0.1 to v0.0.3 in order to get
230         new Kochi Mincho / Gothic fonts to display Japanese and Chinese
231         characters.
232
233         * DumpRenderTree/efl/CMakeLists.txt:
234         * efl/jhbuild.modules:
235
236 2012-05-11  Philippe Normand  <pnormand@igalia.com>
237
238         Unreviewed, GTK 64-bit Debug bot fix attempt.
239
240         * Scripts/webkitpy/layout_tests/port/gtk.py:
241         (GtkPort._get_gdb_output): Ignore unicode decoding errors when
242         processing gdb output.
243
244 2012-05-11  Gustavo Noronha Silva  <gns@gnome.org>
245
246         REGRESSION(r116205): [GTK]: build no longer supports thin archives
247         https://bugs.webkit.org/show_bug.cgi?id=86207
248
249         * GNUmakefile.am: link libWebCoreModules to DRT.
250
251 2012-05-10  Ojan Vafai  <ojan@chromium.org>
252
253         Add a perf tab to garden-o-matic
254         https://bugs.webkit.org/show_bug.cgi?id=86169
255
256         Reviewed by Dimitri Glazkov.
257
258         This is a first stab at making it easier to view the Chromium WebKit canary
259         perf graphs. For now, it's just a couple select elements and an iframe.
260
261         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
262         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
263         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders.js:
264         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
265         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
266         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
267         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/perf.js: Added.
268         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/perf_unittests.js: Added.
269         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
270         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/perf.css: Added.
271         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
272
273 2012-05-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
274
275         Move resumeAnimations to use Internals interface
276         https://bugs.webkit.org/show_bug.cgi?id=86063
277
278         Reviewed by Alexey Proskuryakov.
279
280         Remove resumeAnimations functions, because it is able to work in the 
281         cross-port way through the Internals interface.
282
283         * DumpRenderTree/LayoutTestController.cpp:
284         (LayoutTestController::staticFunctions):
285         * DumpRenderTree/LayoutTestController.h:
286         (LayoutTestController):
287         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
288         * DumpRenderTree/chromium/LayoutTestController.cpp:
289         (LayoutTestController::LayoutTestController):
290         * DumpRenderTree/chromium/LayoutTestController.h:
291         (LayoutTestController):
292         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
293         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
294         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
295         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
296         * DumpRenderTree/qt/LayoutTestControllerQt.h:
297         (LayoutTestController):
298         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
299         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
300         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
301         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
302         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
303         (LayoutTestController):
304
305 2012-05-10  Alexey Proskuryakov  <ap@apple.com>
306
307         Crash in 3rd party WebKit apps that disable cache at a wrong time
308         https://bugs.webkit.org/show_bug.cgi?id=86027
309         <rdar://problem/10615880>
310
311         Reviewed by Antti Koivisto.
312
313         Added a test that's very similar to MemoryCachePruneWithinResourceLoadDelegate,
314         but for disabling the cache instead of triggering a prune.
315
316         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
317         * TestWebKitAPI/Tests/mac/MemoryCacheDisableWithinResourceLoadDelegate.html: Added.
318         * TestWebKitAPI/Tests/mac/MemoryCacheDisableWithinResourceLoadDelegate.mm: Added.
319         (-[MemoryCacheDisableTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
320         (-[MemoryCacheDisableTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
321         (-[MemoryCacheDisableTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
322         (-[MemoryCacheDisableTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
323         (TestWebKitAPI::TEST):
324
325 2012-05-10  Anders Carlsson  <andersca@apple.com>
326
327         WebKit2: Add a way to blacklist specific plug-ins/plug-in versions
328         https://bugs.webkit.org/show_bug.cgi?id=86164
329         <rdar://problem/9551196>
330
331         Reviewed by Sam Weinig.
332
333         Update for WK2 API changes.
334
335         * MiniBrowser/mac/BrowserWindowController.m:
336         (-[BrowserWindowController awakeFromNib]):
337         * WebKitTestRunner/TestController.cpp:
338         (WTR::TestController::createOtherPage):
339         (WTR::TestController::initialize):
340
341 2012-05-10  Dirk Pranke  <dpranke@chromium.org>
342
343         nrwt: [chromium] fix path to asan symbolize script
344         https://bugs.webkit.org/show_bug.cgi?id=86161
345
346         Reviewed by Abhishek Arya.
347
348         * Scripts/webkitpy/layout_tests/port/chromium.py:
349         (ChromiumPort._get_crash_log):
350
351 2012-05-10  Lei Zhang  <thestig@chromium.org>
352
353         [chromium] Set Valgrind environment variables for layout tests.
354         https://bugs.webkit.org/show_bug.cgi?id=86156
355
356         Reviewed by Dirk Pranke.
357
358         * Scripts/webkitpy/layout_tests/port/chromium.py:
359         (ChromiumPort.setup_environ_for_server):
360
361 2012-05-10  Raphael Kubo da Costa  <rakuco@webkit.org>
362
363         [webkitpy][EFL] It should be possible to use a wrapper command for ImageDiff.
364         https://bugs.webkit.org/show_bug.cgi?id=86145
365
366         Reviewed by Dirk Pranke.
367
368         The GTK+ and EFL ports use jhbuild to build a few base libraries
369         which are then used when building the port itself. Programs such
370         as DRT and ImageDiff are then supposed to be run with the
371         `run-with-jhbuild' wrapper so that the libraries are properly
372         found and loaded.
373
374         This was already the case for DumpRenderTree (EflPort defines the
375         'wrapper' option for that), but it was not possible to do the same
376         with ImageDiff so far.
377
378         The code which created the ImageDiff command-line call has been
379         moved to a separate method, _image_diff_command, which is then
380         overridden by EflPort to add the jhbuild wrapper.
381
382         * Scripts/webkitpy/layout_tests/port/efl.py:
383         (EflPort.__init__):
384         (EflPort._image_diff_command):
385         * Scripts/webkitpy/layout_tests/port/webkit.py:
386         (WebKitPort):
387         (WebKitPort._image_diff_command):
388         (WebKitPort._start_image_diff_process):
389
390 2012-05-10  Zan Dobersek  <zandobersek@gmail.com>
391
392         [GTK] ENABLE_IFRAME_SEAMLESS support
393         https://bugs.webkit.org/show_bug.cgi?id=85843
394
395         Reviewed by Eric Seidel.
396
397         Add the iframe-seamless option to the options list. This makes it
398         possible to enable the iframe seamless support through build-webkit
399         script. Enabled by default for all the ports.
400
401         * Scripts/webkitperl/FeatureList.pm:
402
403 2012-05-10  Raphael Kubo da Costa  <rakuco@webkit.org>
404
405         [EFL][webkitpy] Do not redefine check_build() in EflPort.
406         https://bugs.webkit.org/show_bug.cgi?id=86124
407
408         Reviewed by Dirk Pranke.
409
410         Remove an old FIXME and do not redefine check_build in EflPort;
411         once build-dumprendertree recognizes --efl, we can use the
412         check_build implementation in WebKitPort.
413
414         * Scripts/build-dumprendertree:
415         * Scripts/webkitpy/layout_tests/port/efl.py:
416         (EflPort.check_build):
417
418 2012-05-10  Tommy Widenflycht  <tommyw@google.com>
419
420         [chromium] MediaStream API: Fix the ExtraData functionality in WebMediaStreamDescriptor
421         https://bugs.webkit.org/show_bug.cgi?id=86087
422
423         Reviewed by Adam Barth.
424
425         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
426         (MockExtraData):
427         (WebKit):
428         (WebKit::WebUserMediaClientMock::requestUserMedia):
429
430 2012-05-10  Varun Jain  <varunjain@google.com>
431
432         [chromium] Trigger context menu for long press gesture
433         https://bugs.webkit.org/show_bug.cgi?id=85919
434
435         Reviewed by Adam Barth.
436
437         * DumpRenderTree/chromium/EventSender.cpp:
438         (EventSender::EventSender):
439         (EventSender::gestureLongPress):
440         (EventSender::gestureEvent):
441         * DumpRenderTree/chromium/EventSender.h:
442         (EventSender):
443
444 2012-05-10  Takashi Toyoshima  <toyoshim@chromium.org>
445
446         don't use autoinstall to import pywebsocket but check it in WebKit directly.
447         https://bugs.webkit.org/show_bug.cgi?id=86107
448
449         Reviewed by Adam Barth.
450
451         This change removes pywebsocket from webkitpy's autoinstall list and
452         imports pywebsocket 0.7.4 directly into webkitpy/thirparty.
453
454         * Scripts/webkitpy/layout_tests/servers/websocket_server.py:
455         (PyWebSocket._prepare_config):
456         * Scripts/webkitpy/thirdparty/__init__.py:
457         (AutoinstallImportHook.find_module):
458         (AutoinstallImportHook._install_irc):
459         * Scripts/webkitpy/thirdparty/__init___unittest.py:
460         (ThirdpartyTest.test_import_hook.MockImportHook.__init__):
461         (ThirdpartyTest.test_import_hook.MockImportHook._install_eliza):
462         (ThirdpartyTest):
463         (ThirdpartyTest.test_import_hook):
464         * Scripts/webkitpy/thirdparty/mod_pywebsocket/COPYING: Added.
465         * Scripts/webkitpy/thirdparty/mod_pywebsocket/__init__.py: Added.
466         * Scripts/webkitpy/thirdparty/mod_pywebsocket/_stream_base.py: Added.
467         (ConnectionTerminatedException):
468         (InvalidFrameException):
469         (BadOperationException):
470         (UnsupportedFrameException):
471         (InvalidUTF8Exception):
472         (StreamBase):
473         (StreamBase.__init__):
474         (StreamBase._read):
475         (StreamBase._write):
476         (StreamBase.receive_bytes):
477         (StreamBase._read_until):
478         * Scripts/webkitpy/thirdparty/mod_pywebsocket/_stream_hixie75.py: Added.
479         (for):
480         (StreamHixie75):
481         (StreamHixie75.for):
482         (StreamHixie75.__init__):
483         (StreamHixie75.send_message):
484         (StreamHixie75._read_payload_length_hixie75):
485         (StreamHixie75.receive_message):
486         (StreamHixie75._send_closing_handshake):
487         (StreamHixie75.close_connection):
488         (StreamHixie75.send_ping):
489         * Scripts/webkitpy/thirdparty/mod_pywebsocket/_stream_hybi.py: Added.
490         (Frame):
491         (Frame.__init__):
492         (create_length_header):
493         (create_header):
494         (_build_frame):
495         (_filter_and_format_frame_object):
496         (create_binary_frame):
497         (create_text_frame):
498         (FragmentedFrameBuilder):
499         (FragmentedFrameBuilder.to):
500         (FragmentedFrameBuilder.__init__):
501         (FragmentedFrameBuilder.build):
502         (_create_control_frame):
503         (create_ping_frame):
504         (create_pong_frame):
505         (create_close_frame):
506         (StreamOptions):
507         (StreamOptions.__init__):
508         (Stream):
509         (Stream.for):
510         (Stream.__init__):
511         (Stream._receive_frame):
512         (Stream._receive_frame_as_frame_object):
513         (Stream.send_message):
514         (Stream.receive_message):
515         (Stream._send_closing_handshake):
516         (Stream.close_connection):
517         (Stream.send_ping):
518         (Stream._send_pong):
519         (Stream._drain_received_data):
520         * Scripts/webkitpy/thirdparty/mod_pywebsocket/common.py: Added.
521         (is_control_opcode):
522         (ExtensionParameter):
523         (ExtensionParameter.__init__):
524         (ExtensionParameter.name):
525         (ExtensionParameter.add_parameter):
526         (ExtensionParameter.get_parameters):
527         (ExtensionParameter.get_parameter_names):
528         (ExtensionParameter.has_parameter):
529         (ExtensionParameter.get_parameter_value):
530         * Scripts/webkitpy/thirdparty/mod_pywebsocket/dispatch.py: Added.
531         (DispatchException):
532         (DispatchException.__init__):
533         (_default_passive_closing_handshake_handler):
534         (_normalize_path):
535         (_create_path_to_resource_converter):
536         (_create_path_to_resource_converter.converter):
537         (_enumerate_handler_file_paths):
538         (_HandlerSuite):
539         (_HandlerSuite.__init__):
540         (_source_handler_file):
541         (_extract_handler):
542         (Dispatcher):
543         (Dispatcher.maintains):
544         (Dispatcher.__init__):
545         (Dispatcher.add_resource_path_alias):
546         (Dispatcher.source_warnings):
547         (Dispatcher.do_extra_handshake):
548         (Dispatcher.transfer_data):
549         (Dispatcher.passive_closing_handshake):
550         (Dispatcher.get_handler_suite):
551         (Dispatcher._source_handler_files_in_dir):
552         * Scripts/webkitpy/thirdparty/mod_pywebsocket/extensions.py: Added.
553         (ExtensionProcessorInterface):
554         (ExtensionProcessorInterface.get_extension_response):
555         (ExtensionProcessorInterface.setup_stream_options):
556         (DeflateStreamExtensionProcessor):
557         (DeflateStreamExtensionProcessor.__init__):
558         (DeflateStreamExtensionProcessor.get_extension_response):
559         (DeflateStreamExtensionProcessor.setup_stream_options):
560         (DeflateFrameExtensionProcessor):
561         (DeflateFrameExtensionProcessor.__init__):
562         (DeflateFrameExtensionProcessor.get_extension_response):
563         (DeflateFrameExtensionProcessor.setup_stream_options):
564         (DeflateFrameExtensionProcessor.setup_stream_options._OutgoingFilter):
565         (DeflateFrameExtensionProcessor.setup_stream_options._OutgoingFilter.__init__):
566         (DeflateFrameExtensionProcessor.setup_stream_options._OutgoingFilter.filter):
567         (DeflateFrameExtensionProcessor.setup_stream_options._IncomingFilter):
568         (DeflateFrameExtensionProcessor.setup_stream_options._IncomingFilter.__init__):
569         (DeflateFrameExtensionProcessor.setup_stream_options._IncomingFilter.filter):
570         (DeflateFrameExtensionProcessor.set_response_window_bits):
571         (DeflateFrameExtensionProcessor.set_response_no_context_takeover):
572         (DeflateFrameExtensionProcessor.enable_outgoing_compression):
573         (DeflateFrameExtensionProcessor.disable_outgoing_compression):
574         (DeflateFrameExtensionProcessor._outgoing_filter):
575         (DeflateFrameExtensionProcessor._incoming_filter):
576         (get_extension_processor):
577         (get_extension_processor.is):
578         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/__init__.py: Added.
579         (try):
580         (do_handshake):
581         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/_base.py: Added.
582         (AbortedByUserException):
583         (HandshakeException):
584         (HandshakeException.__init__):
585         (VersionException):
586         (VersionException.__init__):
587         (get_default_port):
588         (validate_subprotocol):
589         (parse_host_header):
590         (format_header):
591         (build_location):
592         (get_mandatory_header):
593         (validate_mandatory_header):
594         (check_request_line):
595         (check_header_lines):
596         (parse_token_list):
597         (_parse_extension_param):
598         (_parse_extension):
599         (parse_extensions):
600         (format_extensions):
601         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/draft75.py: Added.
602         (Handshaker):
603         (Handshaker.performs):
604         (Handshaker.__init__):
605         (Handshaker.do_handshake):
606         (Handshaker._set_resource):
607         (Handshaker._set_origin):
608         (Handshaker._set_location):
609         (Handshaker._set_subprotocol):
610         (Handshaker._set_protocol_version):
611         (Handshaker._sendall):
612         (Handshaker._send_handshake):
613         (Handshaker._check_header_lines):
614         (Handshaker._check_first_lines):
615         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/hybi.py: Added.
616         (compute_accept):
617         (Handshaker):
618         (Handshaker.__init__):
619         (Handshaker._validate_connection_header):
620         (Handshaker.do_handshake):
621         (Handshaker._get_origin):
622         (Handshaker._check_version):
623         (Handshaker._set_protocol):
624         (Handshaker._parse_extensions):
625         (Handshaker._validate_key):
626         (Handshaker._get_key):
627         (Handshaker._send_handshake):
628         * Scripts/webkitpy/thirdparty/mod_pywebsocket/handshake/hybi00.py: Added.
629         (Handshaker):
630         (Handshaker.__init__):
631         (Handshaker.do_handshake):
632         (Handshaker._set_resource):
633         (Handshaker._set_subprotocol):
634         (Handshaker._set_location):
635         (Handshaker._set_origin):
636         (Handshaker._set_protocol_version):
637         (Handshaker._set_challenge_response):
638         (Handshaker._get_key_value):
639         (Handshaker._get_challenge):
640         (Handshaker._send_handshake):
641         * Scripts/webkitpy/thirdparty/mod_pywebsocket/headerparserhandler.py: Added.
642         (ApacheLogHandler):
643         (ApacheLogHandler.__init__):
644         (ApacheLogHandler.emit):
645         (_configure_logging):
646         (_parse_option):
647         (_create_dispatcher):
648         (headerparserhandler):
649         * Scripts/webkitpy/thirdparty/mod_pywebsocket/http_header_util.py: Added.
650         (_is_char):
651         (_is_ctl):
652         (ParsingState):
653         (ParsingState.__init__):
654         (peek):
655         (consume):
656         (consume_string):
657         (consume_lws):
658         (consume_lwses):
659         (consume_token):
660         (consume_token_or_quoted_string):
661         (quote_if_necessary):
662         (parse_uri):
663         * Scripts/webkitpy/thirdparty/mod_pywebsocket/memorizingfile.py: Added.
664         (MemorizingFile):
665         (MemorizingFile.__init__):
666         (MemorizingFile.__getattribute__):
667         (MemorizingFile.readline):
668         (MemorizingFile.get_memorized_lines):
669         * Scripts/webkitpy/thirdparty/mod_pywebsocket/msgutil.py: Added.
670         (close_connection):
671         (send_message):
672         (receive_message):
673         (send_ping):
674         (MessageReceiver):
675         (MessageReceiver.receives):
676         (MessageReceiver.provides):
677         (MessageReceiver.should):
678         (MessageReceiver.__init__):
679         (MessageReceiver.run):
680         (MessageReceiver.receive):
681         (MessageReceiver.receive_nowait):
682         (MessageReceiver.stop):
683         (MessageSender):
684         (MessageSender.sends):
685         (MessageSender.provides):
686         (MessageSender.should):
687         (MessageSender.__init__):
688         (MessageSender.run):
689         (MessageSender.send):
690         (MessageSender.send_nowait):
691         * Scripts/webkitpy/thirdparty/mod_pywebsocket/standalone.py: Added.
692         (_StandaloneConnection):
693         (_StandaloneConnection.__init__):
694         (_StandaloneConnection.get_local_addr):
695         (_StandaloneConnection.get_remote_addr):
696         (_StandaloneConnection.write):
697         (_StandaloneConnection.read):
698         (_StandaloneConnection.get_memorized_lines):
699         (_StandaloneRequest):
700         (_StandaloneRequest.__init__):
701         (_StandaloneRequest.get_uri):
702         (_StandaloneRequest.get_method):
703         (_StandaloneRequest.is_https):
704         (_StandaloneRequest._drain_received_data):
705         (_StandaloneSSLConnection):
706         (_StandaloneSSLConnection.for):
707         (_StandaloneSSLConnection.__init__):
708         (_StandaloneSSLConnection.__getattribute__):
709         (_StandaloneSSLConnection.__setattr__):
710         (_StandaloneSSLConnection.makefile):
711         (WebSocketServer):
712         (WebSocketServer.__init__):
713         (WebSocketServer._create_sockets):
714         (WebSocketServer.server_bind):
715         (WebSocketServer.server_activate):
716         (WebSocketServer.server_close):
717         (WebSocketServer.fileno):
718         (WebSocketServer.handle_error):
719         (WebSocketServer.get_request):
720         (WebSocketServer.serve_forever):
721         (WebSocketServer.shutdown):
722         (WebSocketRequestHandler):
723         (WebSocketRequestHandler.setup):
724         (WebSocketRequestHandler.setup.SocketServer):
725         (WebSocketRequestHandler.__init__):
726         (WebSocketRequestHandler.parse_request):
727         (WebSocketRequestHandler.log_request):
728         (WebSocketRequestHandler.log_error):
729         (WebSocketRequestHandler.is_cgi):
730         (_configure_logging):
731         (_alias_handlers):
732         (_build_option_parser):
733         (ThreadMonitor):
734         (ThreadMonitor.__init__):
735         (ThreadMonitor.run):
736         (_parse_args_and_config):
737         (_main):
738         (_main.if):
739         (_main.if.__check_script):
740         * Scripts/webkitpy/thirdparty/mod_pywebsocket/stream.py: Copied from Tools/Scripts/webkitpy/thirdparty/__init___unittest.py.
741         * Scripts/webkitpy/thirdparty/mod_pywebsocket/util.py: Added.
742         (get_stack_trace):
743         (prepend_message_to_exception):
744         (__translate_interp):
745         (get_script_interp):
746         (wrap_popen3_for_win):
747         (wrap_popen3_for_win.__wrap_popen3):
748         (hexify):
749         (get_class_logger):
750         (NoopMasker):
751         (NoopMasker.__init__):
752         (NoopMasker.mask):
753         (RepeatedXorMasker):
754         (RepeatedXorMasker.__init__):
755         (RepeatedXorMasker.mask):
756         (DeflateRequest):
757         (DeflateRequest.for):
758         (DeflateRequest.__init__):
759         (DeflateRequest.__getattribute__):
760         (DeflateRequest.__setattr__):
761         (_Deflater):
762         (_Deflater.__init__):
763         (_Deflater.compress_and_flush):
764         (_Inflater):
765         (_Inflater.__init__):
766         (_Inflater.decompress):
767         (_Inflater.append):
768         (_Inflater.reset):
769         (_RFC1979Deflater):
770         (_RFC1979Deflater.that):
771         (_RFC1979Deflater.__init__):
772         (_RFC1979Deflater.filter):
773         (_RFC1979Inflater):
774         (_RFC1979Inflater.for):
775         (_RFC1979Inflater.__init__):
776         (_RFC1979Inflater.filter):
777         (DeflateSocket):
778         (DeflateSocket.for):
779         (DeflateSocket.__init__):
780         (DeflateSocket.recv):
781         (DeflateSocket.sendall):
782         (DeflateSocket.send):
783         (DeflateConnection):
784         (DeflateConnection.for):
785         (DeflateConnection.__init__):
786         (DeflateConnection.get_remote_addr):
787         (DeflateConnection.put_bytes):
788         (DeflateConnection.read):
789         (DeflateConnection.write):
790         (_is_ewouldblock_errno):
791         (drain_received_data):
792
793 2012-05-10  Raphael Kubo da Costa  <rakuco@webkit.org>
794
795         [EFL][webkitpy] Define the 'wrapper' option in EflPort instead of creating a EflDriver.
796         https://bugs.webkit.org/show_bug.cgi?id=86117
797
798         Reviewed by Adam Barth.
799
800         Simplify the code used to wrap driver calls with jhbuild: we can
801         just define the 'wrapper' option instead of having to create a
802         Driver class just for that.
803
804         * Scripts/webkitpy/layout_tests/port/efl.py:
805         (EflPort.__init__):
806         (EflPort._port_flag_for_scripts):
807
808 2012-05-10  Sheriff Bot  <webkit.review.bot@gmail.com>
809
810         Unreviewed, rolling out r116633.
811         http://trac.webkit.org/changeset/116633
812         https://bugs.webkit.org/show_bug.cgi?id=86093
813
814         Breaks EFL tests (Requested by kov on #webkit).
815
816         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
817         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
818         * DumpRenderTree/efl/EventSender.cpp:
819         (keyDownCallback):
820         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
821         (LayoutTestController::setTabKeyCyclesThroughElements):
822
823 2012-05-10  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
824
825         [EFL] Memory leaks found in EFL DRT
826         https://bugs.webkit.org/show_bug.cgi?id=85889
827
828         Reviewed by Gustavo Noronha Silva.
829
830         Fix two memory leaks in EFL's DRT by using JSRetainPtr for JSString
831         created with JSStringCreateWithUTF8CString.
832
833         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
834         (DumpRenderTreeChrome::onWindowObjectCleared):
835         * DumpRenderTree/efl/EventSender.cpp:
836         (modifiersFromJSValue):
837
838 2012-05-10  Alexander Shalamov  <alexander.shalamov@intel.com>
839
840         [EFL] LayoutTestController needs implementation of setTabKeyCyclesThroughElements
841         https://bugs.webkit.org/show_bug.cgi?id=82864
842
843         Reviewed by Gustavo Noronha Silva.
844
845         Add missing implementation for setTabKeyCyclesThroughElements to EFL's LayoutTestController
846         in order to unskip fast/events/keypress-insert-tab.html
847
848         * DumpRenderTree/efl/EventSender.cpp:
849         (keyDownCallback):
850         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
851         (LayoutTestController::setTabKeyCyclesThroughElements):
852
853 2012-05-10  Alexander Shalamov  <alexander.shalamov@intel.com>
854
855         [EFL][DRT] EventSender - set delayed event information if it was not set before
856         https://bugs.webkit.org/show_bug.cgi?id=85972
857
858         Reviewed by Gustavo Noronha Silva.
859
860         Reset delayed event information if it is missing.
861
862         * DumpRenderTree/efl/EventSender.cpp:
863         (feedOrQueueMouseEvent):
864
865 2012-05-07  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
866
867         WebPageProxy::activeURL() should return the pending API request, even when there's no main frame
868         https://bugs.webkit.org/show_bug.cgi?id=85806
869
870         The m_pendingAPIRequestURL member is used (presumably) to mask over the async
871         nature of WebKit2, so that starting a load of a URL will reflect that URL
872         immedeatly from activeURL, even if the request has not been passed over to
873         the web process yet and reflected there.
874
875         This works well, except in the case of the initial request, where the main
876         frame creation happens on the web process side and is notified back to the
877         UI process. Until we've recived the notification we don't know about the main
878         frame, and this race condition will potentially give us an empty url instead
879         of the pending request.
880
881         To solve this we always return the pending API request if it's set, even
882         when there's no mainframe yet (that we known about).
883
884         Reviewed by Simon Hausmann.
885
886         * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp:
887
888 2012-05-10  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
889
890         [EFL][DRT] Clear added user style sheets before a new testcase execution.
891         https://bugs.webkit.org/show_bug.cgi?id=84792
892
893         Reviewed Gustavo Noronha Silva.
894
895         The added user style sheets from the previous testcases are cleared in order to avoid 
896         flaky results.
897
898         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
899         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
900
901 2012-05-09  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
902
903         Move suspendAnimations to use Internals interface.
904         https://bugs.webkit.org/show_bug.cgi?id=85986
905
906         Reviewed by Ryosuke Niwa.
907
908         Remove suspendAnimations functions, because it is able to work in the 
909         cross-port way through the Internals interface
910
911         * DumpRenderTree/LayoutTestController.cpp:
912         (LayoutTestController::staticFunctions):
913         * DumpRenderTree/LayoutTestController.h:
914         (LayoutTestController):
915         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
916         * DumpRenderTree/chromium/LayoutTestController.cpp:
917         (LayoutTestController::LayoutTestController):
918         * DumpRenderTree/chromium/LayoutTestController.h:
919         (LayoutTestController):
920         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
921         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
922         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
923         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
924         * DumpRenderTree/qt/LayoutTestControllerQt.h:
925         (LayoutTestController):
926         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
927         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
928
929 2012-05-09  Kent Tamura  <tkent@chromium.org>
930
931         [Chromium] Attempt to fix an Android build error.
932         https://bugs.webkit.org/show_bug.cgi?id=86045
933
934         * DumpRenderTree/DumpRenderTree.gypi:
935
936 2012-05-09  Ojan Vafai  <ojan@chromium.org>
937
938         Don't show partytime if there are non-layout test failures.
939         https://bugs.webkit.org/show_bug.cgi?id=86056
940
941         Reviewed by Dimitri Glazkov.
942
943         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
944         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers_unittests.js:
945         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
946
947 2012-05-09  Hayato Ito  <hayato@chromium.org>
948
949         Improve a warning message for a reftest having an unused expectation file.
950         https://bugs.webkit.org/show_bug.cgi?id=85975
951
952         Reviewed by Adam Barth.
953
954         The current warning message is not intuitive. It might be better to update the message
955         so that it encourage users to remove unused expectation files.
956
957         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
958         (SingleTestRunner.__init__):
959
960 2012-05-09  Ojan Vafai  <ojan@chromium.org>
961
962         Change "losing test coverage" to cover all non-layout test problems in garden-o-matic
963         https://bugs.webkit.org/show_bug.cgi?id=86043
964
965         Reviewed by Adam Barth.
966
967         -Get rid of the gtest iframe.
968         -Cover all non-layout test problems. Before we wouldn't catch compile failures
969         or gtest failures since they were not on the webkit testing bots.
970         -Include the list of failing steps next to the builder name. This is a lot less
971         cluttered than you'd expect and already helped me find a compile failure early.
972
973         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders.js:
974         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
975         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
976         Delete unused map.
977         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
978         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
979         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
980         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js:
981         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
982         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js:
983         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
984         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
985
986 2012-05-09  Jessie Berlin  <jberlin@apple.com>
987
988         Crash using the new WKBundleDOMWindowExtensions APIs.
989         https://bugs.webkit.org/show_bug.cgi?id=85888
990
991         Reviewed by Brady Eidson.
992
993         Cached frames can live slightly longer than the page, but most clients unregister themselves
994         and do other cleanup in the willDestroyPage callback, making them miss the
995         willDestroyGlobalObjectForDOMWindowExtension callbacks.
996
997         The calls to willDestroyGlobalObjectForDOMWindowExtension in the DOMWindowExtensionBasic
998         test were all being invoked underneath WebPage::close. This is unrealistic. Update that test
999         to destroy the BundleDOMWindowExtensions in response to the willDestroyPage callback.
1000
1001         Add a test to verify that willDestroyGlobalObjectForDOMWindowExtension is being called for
1002         pages that don't go into the page cache.
1003
1004         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1005         Add DOMWindowExtensionNoCache.cpp, DOMWindowExtensionNoCache_Bundle.cpp, simple-unload.html
1006         and simple-iframe-unload.html
1007
1008         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic.cpp:
1009         (TestWebKitAPI):
1010         Remove the expected messages for willDestroyGlobalObjectForDOMWindowExtension.
1011         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
1012         Do not bother to keep track of the live extension count - all of them are expected to be
1013         live until the test completes.
1014         (TestWebKitAPI::TEST):
1015         Fix the calls to EXPECT to pass the expected value first, and use EXPECT_WK_STREQ so that
1016         message failures will be clearer.
1017         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp:
1018         (TestWebKitAPI::DOMWindowExtensionBasic::willDestroyPage):
1019         Clean up the BundleDOMWindowExtensions.
1020         (TestWebKitAPI::DOMWindowExtensionBasic::willDestroyGlobalObjectForDOMWindowExtension):
1021         Add an ASSERT_NOT_REACHED.
1022
1023         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionNoCache.cpp: Added.
1024         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
1025         Keep track of the messages received so they can be checked at the end of the test.
1026         (TestWebKitAPI::TEST):
1027         Navigate to uncacheable pages and back.
1028         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionNoCache_Bundle.cpp: Copied from Tools/TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp.
1029         (DOMWindowExtensionNoCache):
1030         (TestWebKitAPI::DOMWindowExtensionNoCache::DOMWindowExtensionNoCache):
1031         Set up all the states for each BundleDOMWindowExtension.
1032         (TestWebKitAPI::DOMWindowExtensionNoCache::frameLoadFinished):
1033         Tell the UI Process about the states of the BundleDOMWindowExtensions.
1034         (TestWebKitAPI::DOMWindowExtensionNoCache::sendExtensionStateMessage):
1035         (TestWebKitAPI::DOMWindowExtensionNoCache::initialize):
1036         (TestWebKitAPI::DOMWindowExtensionNoCache::didCreatePage):
1037         (TestWebKitAPI::DOMWindowExtensionNoCache::willDestroyPage):
1038         Remvoe the remaining BundleDOMWindowExtensions, send the updated state, and finish the test.
1039         (TestWebKitAPI::DOMWindowExtensionNoCache::updateExtensionStateRecord):
1040         (TestWebKitAPI::DOMWindowExtensionNoCache::sendBundleMessage):
1041         (TestWebKitAPI::DOMWindowExtensionNoCache::globalObjectIsAvailableForFrame):
1042         (TestWebKitAPI::DOMWindowExtensionNoCache::willDisconnectDOMWindowExtensionFromGlobalObject):
1043         ASSERT that these pages not going into the page cache are not getting disconnected to go into
1044         the page cache.
1045         (TestWebKitAPI::DOMWindowExtensionNoCache::didReconnectDOMWindowExtensionToGlobalObject):
1046         Ditto about getting reconnected when coming out of the page cache.
1047         (TestWebKitAPI::DOMWindowExtensionNoCache::willDestroyGlobalObjectForDOMWindowExtension):
1048         Tell the UI Process, update the state, and get rid of the BundleDOMWindowExtension.
1049         (TestWebKitAPI::didFinishLoadForFrameCallback):
1050         (TestWebKitAPI::globalObjectIsAvailableForFrameCallback):
1051         (TestWebKitAPI::willDisconnectDOMWindowExtensionFromGlobalObjectCallback):
1052         (TestWebKitAPI::didReconnectDOMWindowExtensionToGlobalObjectCallback):
1053         (TestWebKitAPI::willDestroyGlobalObjectForDOMWindowExtensionCallback):
1054
1055         * TestWebKitAPI/Tests/WebKit2/simple-iframe-unload.html: Added.
1056         * TestWebKitAPI/Tests/WebKit2/simple-unload.html: Added.
1057
1058 2012-05-09  Adam Barth  <abarth@webkit.org>
1059
1060         [Chromium] Android's DRT doesn't want to build WebUserMediaClientMock
1061         https://bugs.webkit.org/show_bug.cgi?id=86045
1062
1063         Reviewed by Eric Seidel.
1064
1065         * DumpRenderTree/DumpRenderTree.gypi:
1066
1067 2012-05-09  Adam Barth  <abarth@webkit.org>
1068
1069         [Chromium] MockWebSpeechInputController.cpp should be guarded by ENABLE(INPUT_SPEECH)
1070         https://bugs.webkit.org/show_bug.cgi?id=86041
1071
1072         Reviewed by Eric Seidel.
1073
1074         * DumpRenderTree/chromium/MockWebSpeechInputController.cpp:
1075
1076 2012-05-09  Adam Barth  <abarth@webkit.org>
1077
1078         [Chromium] Android needs to swizzle pixel dumps in order to get the same hash as other platforms
1079         https://bugs.webkit.org/show_bug.cgi?id=86040
1080
1081         Reviewed by Kent Tamura.
1082
1083         * DumpRenderTree/chromium/TestShell.cpp:
1084         (TestShell::dumpImage):
1085
1086 2012-05-09  Raphael Kubo da Costa  <rakuco@webkit.org>
1087
1088         [EFL] Build fix after r116464.
1089
1090         `ulong' is a type defined by glibc, so it is not very
1091         portable. Use `unsigned long' instead.
1092
1093         * DumpRenderTree/efl/EventSender.cpp:
1094         (DelayedEvent::DelayedEvent):
1095         (DelayedEvent):
1096         (leapForwardCallback):
1097
1098 2012-05-09  Adam Barth  <abarth@webkit.org>
1099
1100         [Chromium] LayoutTestController should use ENABLE guards for some features
1101         https://bugs.webkit.org/show_bug.cgi?id=86030
1102
1103         Reviewed by Eric Seidel.
1104
1105         These features are disabled on OS(ANDROID) and so shouldn't be compiled
1106         when the feature is off.
1107
1108         * DumpRenderTree/chromium/LayoutTestController.cpp:
1109         (LayoutTestController::LayoutTestController):
1110         * DumpRenderTree/chromium/LayoutTestController.h:
1111         (LayoutTestController):
1112         * DumpRenderTree/chromium/TestShell.h:
1113         (TestShell):
1114
1115 2012-05-09  Tom Zakrajsek  <tomz@codeaurora.org>
1116
1117         Fix overzealous re-opened since blocked by... message
1118         https://bugs.webkit.org/show_bug.cgi?id=86020
1119
1120         Reviewed by Adam Barth.
1121
1122         Changed CreateBug behavior to only re-open a blocked bug if it
1123         was RESOLVED.  Also had to update the test infrastructure slightly
1124         to allow for distinct mock commit revisions, so one could point
1125         to a RESOLVED bug for testing while another pointed to an OPEN bug.
1126
1127         Also moved default created mock bug to be 60001 instead of 50004
1128         which already exists as a reference mock bug.
1129
1130         * Scripts/webkitpy/common/checkout/checkout_mock.py:
1131         (MockCheckout.__init__):
1132         (MockCheckout.commit_info_for_revision):
1133         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
1134         (MockBugzilla.create_bug):
1135         * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py:
1136         * Scripts/webkitpy/tool/commands/download_unittest.py:
1137         (test_create_rollout_resolved):
1138         * Scripts/webkitpy/tool/commands/queries_unittest.py:
1139         (FailureReasonTest.test_blame_line_for_revision):
1140         * Scripts/webkitpy/tool/steps/createbug.py:
1141         (CreateBug.run):
1142
1143 2012-05-09  Hugo Parente Lima  <hugo.lima@openbossa.org>
1144
1145         Use suitable viewport values on XHTML-MP pages.
1146         https://bugs.webkit.org/show_bug.cgi?id=85425
1147
1148         Reviewed by Kenneth Rohde Christiansen.
1149
1150         Enable LEGACY_VIEWPORT_ADAPTION on Qt port.
1151
1152         * qmake/mkspecs/features/features.prf:
1153
1154 2012-05-09  Jochen Eisinger  <jochen@chromium.org>
1155
1156         When creating a new page during a navigation, prime the initial document with the correct referrer policy
1157         https://bugs.webkit.org/show_bug.cgi?id=86001
1158
1159         Reviewed by Adam Barth.
1160
1161         Store the referrer policy in the request, so the network stack can comply with it.
1162
1163         * DumpRenderTree/chromium/WebViewHost.cpp:
1164         (WebViewHost::willSendRequest):
1165
1166 2012-05-09  Vivek Galatage  <vivekgalatage@gmail.com>
1167
1168         Remove CYGWIN=tty from environment variable as its no longer supported
1169         https://bugs.webkit.org/show_bug.cgi?id=85791
1170
1171         The CYGWIN=tty environment variable is no longer supported from Cygwin
1172         version 1.7.10. This fix avoids addition of this environment variable
1173         for Cygwin versions greater than 1.7.9. For supporting older versions
1174         i.e. 1.7.9 or older, the script takes care of retaining the environment
1175         variable.
1176
1177         Reviewed by Adam Roben.
1178
1179         * Scripts/webkitdirs.pm:
1180         (setupAppleWinEnv):
1181
1182 2012-05-09  Dirk Pranke  <dpranke@chromium.org>
1183
1184         nrwt: --skip-failing-tests should also skip flaky tests
1185         https://bugs.webkit.org/show_bug.cgi?id=85941
1186
1187         Reviewed by Ojan Vafai.
1188
1189         This changes --skip-failing-tests so that we skip tests that
1190         are believed to pass sometimes as well as tests that are
1191         believed to always fail.
1192
1193         In addition, in testing this I discovered that we had lousy
1194         test coverage of the details of retrying test failures and
1195         --clobber-old-results, so I wrote some more tests.
1196
1197         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1198         (Manager.prepare_lists_and_print_output):
1199         * Scripts/webkitpy/layout_tests/port/test.py:
1200         (unit_test_list):
1201         (TestPort.__init__):
1202         (TestDriver.run_test):
1203         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1204         (MainTest.test_skip_failing_tests):
1205         (MainTest.test_retrying_and_flaky_tests):
1206
1207 2012-05-09  Christophe Dumez  <christophe.dumez@intel.com>
1208
1209         [EFL][DRT] EFL's DRT needs to support LayoutTestController.dumpIconChanges()
1210         https://bugs.webkit.org/show_bug.cgi?id=84435
1211
1212         Reviewed by Antonio Gomes.
1213
1214         Catch the "icon,changed" signal on the frames in EFL's DumpRenderTree
1215         to support LayoutTestController's dumpIconChanges().
1216
1217         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1218         (DumpRenderTreeChrome::createView):
1219         (DumpRenderTreeChrome::onFrameIconChanged):
1220         (DumpRenderTreeChrome::onFrameCreated):
1221         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
1222         (DumpRenderTreeChrome):
1223
1224 2012-05-09  Dominik Röttsches  <dominik.rottsches@intel.com>
1225
1226         webkit-patch upload Fails Behind Load-balancing Proxies
1227         https://bugs.webkit.org/show_bug.cgi?id=85980
1228
1229         Reviewed by Adam Barth.
1230
1231         Unchecking the login restriction to IP to allow the tool
1232         to function behind load-balancing proxies that use multiple
1233         outgoing IPs.
1234
1235         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1236         (Bugzilla.authenticate):
1237
1238 2012-05-09  Hao Zheng  <zhenghao@chromium.org>
1239
1240         [Chromium-android] Add unit test for Android layout test script (chromium_android.py)
1241         https://bugs.webkit.org/show_bug.cgi?id=80852
1242
1243         Reviewed by Adam Barth.
1244
1245         Add some tests to ensure that chromium android port works to some
1246         extent. Some methods is hard to test, as they rely on the actual
1247         behavior on device, e.g. pushing files to device.
1248
1249         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1250         (ChromiumAndroidPort.get_last_stacktrace): Check if dir exists.
1251         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
1252         (ChromiumAndroidPortTest):
1253         (ChromiumAndroidPortTest.test_attributes):
1254         (ChromiumAndroidPortTest.test_get_last_stacktrace):
1255         (ChromiumAndroidPortTest.test_get_last_stacktrace.mock_run_command_no_dir):
1256         (ChromiumAndroidPortTest.test_get_last_stacktrace.mock_run_command_no_file):
1257         (ChromiumAndroidPortTest.test_get_last_stacktrace.mock_run_command_non_empty):
1258         (ChromiumAndroidDriverTest):
1259         (ChromiumAndroidDriverTest.setUp):
1260         (ChromiumAndroidDriverTest.test_get_drt_return_value):
1261         (ChromiumAndroidDriverTest.test_has_crash_hint):
1262         (ChromiumAndroidDriverTest.test_read_prompt):
1263         (ChromiumAndroidDriverTest.test_test_shell_command):
1264         (ChromiumAndroidDriverTest.test_write_command_and_read_line):
1265
1266 2012-05-09  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1267
1268         [Qt] ORWT fails with Qt 5 and WK2.
1269         https://bugs.webkit.org/show_bug.cgi?id=85968
1270
1271         Reviewed by Csaba Osztrogonác.
1272
1273         The Qt5 binary does not have a -style argument, so avoid using it.
1274
1275         * Scripts/old-run-webkit-tests:
1276
1277 2012-05-09  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1278
1279         [EFL][DRT] LayoutTestController addUserScript implementation
1280         https://bugs.webkit.org/show_bug.cgi?id=84718
1281
1282         Reviewed by Antonio Gomes.
1283
1284         Added implementation of addUserScript() to LayoutTestController.
1285
1286         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1287         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): clear added user scripts
1288         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1289         (LayoutTestController::addUserScript): added implementation
1290
1291 2012-05-09  Dan Bernstein  <mitz@apple.com>
1292
1293         Made build-webkit copy WebKitLibraries to the right location again after r116257
1294
1295         Reviewed by Mark Rowe.
1296
1297         * Scripts/build-webkit: Pass the built product directory to copy-webkitlibraries-to-product-directory.
1298         * Scripts/copy-webkitlibraries-to-product-directory: Use the first argument, if present, as the
1299         product directory.
1300
1301 2012-05-08  Ojan Vafai  <ojan@chromium.org>
1302
1303         Show the chromium non-layout test bots in garden-o-matic
1304         https://bugs.webkit.org/show_bug.cgi?id=85903
1305
1306         Reviewed by Eric Seidel.
1307
1308         This is just a single-line high-level overview of the chromium non-layout test bots.
1309         This means that Chromium WebKit gardeners only need to keep garden-o-matic open in
1310         order to do their job.
1311
1312         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
1313         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1314         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1315         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
1316         (#chromium-gtests iframe):
1317         (#chromium-gtests):
1318         (#gtest-label):
1319
1320 2012-05-08  Kangil Han  <kangil.han@samsung.com>
1321
1322         [EFL][DRT] Implement TextInputController.
1323         https://bugs.webkit.org/show_bug.cgi?id=78559
1324
1325         Reviewed by Martin Robinson.
1326
1327         TextInputController has been created for an input emulation purpose in DRT.
1328         However, not like other ports, EFL doesn't have it.
1329         Therefore, this patch will enable EFL port to have an input emulation in DRT.
1330
1331         * DumpRenderTree/efl/CMakeLists.txt:
1332         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1333         (DumpRenderTreeChrome::onWindowObjectCleared):
1334         * DumpRenderTree/efl/TextInputController.cpp: Added.
1335         (setMarkedTextCallback):
1336         (hasMarkedTextCallback):
1337         (markedRangeCallback):
1338         (insertTextCallback):
1339         (unmarkTextCallback):
1340         (firstRectForCharacterRangeCallback):
1341         (selectedRangeCallback):
1342         (makeTextInputController):
1343         * DumpRenderTree/efl/TextInputController.h: Added.
1344
1345 2012-04-23  Raphael Kubo da Costa  <rakuco@webkit.org>
1346
1347         [EFL] Pass a WTF::String directly from DRT to ewebkit when possible
1348         https://bugs.webkit.org/show_bug.cgi?id=84634
1349
1350         Reviewed by Gustavo Noronha Silva.
1351
1352         Drop unnecessary const char* -> WTF::String conversions by
1353         creating String objects directly in LayoutTestController out of
1354         JSStringRef's and passing them to DumpRenderTreeSupportEfl.
1355
1356         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1357         (LayoutTestController::queueLoad):
1358         (LayoutTestController::setValueForUser):
1359         (LayoutTestController::findString):
1360         (LayoutTestController::addUserStyleSheet):
1361
1362 2012-05-08  Alexander Shalamov  <alexander.shalamov@intel.com>
1363
1364         [EFL][DRT] EventSender needs implementation of leapForward function
1365         https://bugs.webkit.org/show_bug.cgi?id=83486
1366
1367         Reviewed by Eric Seidel.
1368
1369         Implemented leapForward function in Efl's DRT. This function is required
1370         to delay and queue event execution and simulate real user behavior.
1371
1372         * DumpRenderTree/efl/EventSender.cpp:
1373         (DelayedEvent):
1374         (MouseEventInfo::MouseEventInfo):
1375         (MouseEventInfo):
1376         (sendClick):
1377         (mouseDownCallback):
1378         (mouseUpCallback):
1379         (mouseMoveToCallback):
1380         (leapForwardCallback):
1381         (mouseScrollByCallback):
1382         (feedOrQueueMouseEvent):
1383         (feedMouseEvent):
1384         (feedQueuedMouseEvents):
1385
1386 2012-05-08  Ryuan Choi  <ryuan.choi@samsung.com>
1387
1388         [EFL][DRT] WebKit/DRT should not paint webview for the pixel tests.
1389         https://bugs.webkit.org/show_bug.cgi?id=79853
1390
1391         Reviewed by Eric Seidel.
1392
1393         Screen buffer can be different from rendered results.
1394         For example, WebKit/Efl just moves tiles for ewk_view_tiled when scrolling
1395         contents.
1396         It means that pixel tests can't check regression of ewk_view_tiled.
1397
1398         This patch changes a way to provide bitmap from painting webview to capturing
1399         evas.
1400
1401         * DumpRenderTree/efl/PixelDumpSupportEfl.cpp:
1402         (createBitmapContextFromWebView):
1403
1404 2012-05-08  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1405
1406         [GTK] Memory leak in DRT EventSender.cpp
1407         https://bugs.webkit.org/show_bug.cgi?id=85886
1408
1409         Reviewed by Martin Robinson.
1410
1411         Fix a memory leak in gdkModifersFromJSValue() by using JSRetainPtr for
1412         JSString created with JSStringCreateWithUTF8CString.
1413
1414         * DumpRenderTree/gtk/EventSender.cpp:
1415         (gdkModifersFromJSValue):
1416
1417 2012-05-08  Christophe Dumez  <christophe.dumez@intel.com>
1418
1419         [EFL] glib dependency needs to be updated in jhbuild
1420         https://bugs.webkit.org/show_bug.cgi?id=85878
1421
1422         Reviewed by Martin Robinson.
1423
1424         Bump glib version from 2.31.8 to 2.32.2 in jhbuild.modules. This
1425         addresses flakiness in media tests due to possible deadlock with the
1426         following output on stderr:
1427         GStreamer-WARNING **: wrong STREAM_LOCK count 0
1428
1429         The bug was fixed upstream by:
1430         http://git.gnome.org/browse/glib/commit/?id=265f265c677bc5399dbb1cfa449cb4b915a6459f
1431
1432         * efl/jhbuild.modules:
1433
1434 2012-05-07  Alexis Menard  <alexis.menard@openbossa.org>
1435
1436         [Qt] Unbreak debugging of WebKit2.
1437         https://bugs.webkit.org/show_bug.cgi?id=85839
1438
1439         Reviewed by Simon Hausmann.
1440
1441         When you attach GDB to a running process, it stops it.
1442         http://trac.webkit.org/changeset/115958 introduced a pause() 
1443         call to wait the debugger to be attached to then continue 
1444         the execution of the WebProcess. Unfortunately the pause()
1445         function does not return unless a signal handler is called.
1446         This patch introduce an event handler to exit from the paused
1447         state when the debugger send the signal SIGCONT. The old code
1448         works with older version of GDB (<7.0) but not with newer
1449         versions where the behavior of pause() is correct.
1450
1451         * WebKitTestRunner/qt/main.cpp:
1452         (sigcontHandler):
1453         (main):
1454
1455 2012-05-08  Christophe Dumez  <christophe.dumez@intel.com>
1456
1457         [EFL][DRT] EFL's DRT does not fully support page visibility
1458         https://bugs.webkit.org/show_bug.cgi?id=85032
1459
1460         Reviewed by Antonio Gomes.
1461
1462         * DumpRenderTree/LayoutTestController.cpp:
1463         (resetPageVisibilityCallback): Fix incorrect argumentCount check.
1464         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1465         (DumpRenderTreeChrome::resetDefaultsToConsistentValues): Properly
1466         reset page visibility setting between tests.
1467         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1468         (LayoutTestController::setPageVisibility): Add support for "preview"
1469         visibility state.
1470
1471 2012-05-08  Gavin Peters  <gavinp@chromium.org>
1472
1473         [webkit-patch] Return meaningful results from single-result queries.
1474         https://bugs.webkit.org/show_bug.cgi?id=85367
1475
1476         Reviewed by Dirk Pranke.
1477
1478         Sometimes a quicksearch returns a single result by just redirecting to the page for that single
1479         bug. This patch hacks around that by noticing the redirect target is a single page, and constructing
1480         a result list consisting of that one bug.
1481
1482         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1483         (BugzillaQueries._fetch_bugs_from_advanced_query):
1484
1485 2012-05-08  Gavin Peters  <gavinp@chromium.org>
1486
1487         [webkit-patch] Add new chrome-channels command to track down current chrome release channels for a committed bug.
1488         https://bugs.webkit.org/show_bug.cgi?id=85368
1489
1490         Reviewed by Eric Seidel.
1491
1492         * Scripts/webkitpy/common/config/urls.py:
1493         * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
1494         (MockBugzillaQueries.fetch_bugs_matching_quicksearch):
1495         * Scripts/webkitpy/common/net/omahaproxy.py: Added.
1496         (OmahaProxy):
1497         (OmahaProxy.__init__):
1498         (OmahaProxy.set_url):
1499         (OmahaProxy._json_url):
1500         (OmahaProxy._get_json):
1501         (OmahaProxy.get_revisions):
1502         * Scripts/webkitpy/common/net/omahaproxy_unittest.py: Added.
1503         (MockOmahaProxy):
1504         (MockOmahaProxy.__init__):
1505         (OmahaProxyTest):
1506         (OmahaProxyTest.test_get_revisions):
1507         * Scripts/webkitpy/tool/commands/__init__.py:
1508         * Scripts/webkitpy/tool/commands/bugsearch.py:
1509         (BugSearch):
1510         * Scripts/webkitpy/tool/commands/chromechannels.py: Added.
1511         (ChromeChannels):
1512         (__init__):
1513         (execute):
1514         * Scripts/webkitpy/tool/commands/chromechannels_unittest.py: Added.
1515         (MockOmahaProxy):
1516         (MockOmahaProxy.get_revisions):
1517         (TestableChromeChannels):
1518         (TestableChromeChannels.__init__):
1519         (ChromeChannelsTest):
1520         (ChromeChannelsTest.test_single_bug):
1521         (ChromeChannelsTest.test_with_query):
1522
1523 2012-05-08  Carlos Garcia Campos  <cgarcia@igalia.com>
1524
1525         Unreviewed. Fix GTK+ unit tests.
1526
1527         * gtk/common.py:
1528         (pkg_config_file_variable):
1529
1530 2012-05-08  Carlos Garcia Campos  <cgarcia@igalia.com>
1531
1532         [GTK] Google tests that time out are leaked
1533         https://bugs.webkit.org/show_bug.cgi?id=84973
1534
1535         Reviewed by Philippe Normand.
1536
1537         Move Tools/Scripts/run-gtk-tests to Tools/gtk/run-api-tests
1538         leaving run-gtk-tests as simple wrapper that runs run-api-tests in
1539         a jhbuild environment when jhbuild is available.
1540
1541         * Scripts/run-gtk-tests: Run Tools/gtk/run-api-tests directly or
1542         with jhbuild if present.
1543         * gtk/common.py:
1544         (pkg_config_file_variable): Generic function to get a variable
1545         from a pkg-config file.
1546         (prefix_of_pkg_config_file): Use pkg_config_file_variable with 'prefix'.
1547         * gtk/run-api-tests: Copied from Tools/Scripts/run-gtk-tests.
1548         (TestRunner.__init__): Use common.build_path() to get the programs path.
1549         (TestRunner._lookup_atspi2_binary): Use
1550         common.pkg_config_file_variable() to get the 'exec_prefix'.
1551         (TestRunner._start_accessibility_daemons): Use subprocess.Popen()
1552         directly instead of helper _create_process that has been removed.
1553         (TestRunner._setup_testing_environment): Use subprocess.Popen()
1554         directly instead of helper _create_process that has been
1555         removed. Use common.top_level_path() and common.build_path()
1556         instead of custom methods that have been removed too.
1557         (TestRunner._run_test_command): Use subprocess.Popen() directly
1558         instead of helper _create_process that has been removed.
1559
1560 2012-05-08  Thiago Marcos P. Santos  <thiago.santos@intel.com>
1561
1562         [EFL] [DRT] Clear extra views before setting a blank page
1563         https://bugs.webkit.org/show_bug.cgi?id=85608
1564
1565         Reviewed by Kenneth Rohde Christiansen.
1566
1567         We need to clear the extra views before setting a blank page,
1568         otherwise it will cause more events interactions and sometimes
1569         trigger a unload(). This was causing frame-crash-with-page-cache.html
1570         to leak events to the next test causing any subsequent test to
1571         fail.
1572
1573         * DumpRenderTree/efl/DumpRenderTree.cpp:
1574         (runTest):
1575         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1576         (DumpRenderTreeChrome::clearExtraViews):
1577         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1578         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
1579         (DumpRenderTreeChrome):
1580
1581 2012-05-08  Balazs Kelemen  <kbalazs@webkit.org>
1582
1583         [Qt] X11 plugins need to be reworked for Qt5+WK1
1584         https://bugs.webkit.org/show_bug.cgi?id=80691
1585
1586         Reviewed by Simon Hausmann.
1587
1588         Implement basic windowless plugin support with Qt5.
1589
1590         * qmake/mkspecs/features/features.prf:
1591         Enable NPAPI plugins if X11 libraries are available.
1592         * qmake/mkspecs/features/functions.prf:
1593         Added a convenience function to determine availability
1594         of X11 libraries.
1595
1596 2012-05-07  Raphael Kubo da Costa  <rakuco@webkit.org>
1597
1598         [webkitpy] Remove Python 2.5-specific workaround from http_server_base.py.
1599         https://bugs.webkit.org/show_bug.cgi?id=85852
1600
1601         Reviewed by Dirk Pranke.
1602
1603         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
1604         (HttpServerBase._is_server_running_on_all_ports): Revert the part
1605         of r89412 which was a workaround for an issue with Python 2.5 on
1606         Mac 10.5.
1607
1608 2012-05-07  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1609
1610         webkitdirs.pm: Make determineNumberOfCPUs work on FreeBSD.
1611         https://bugs.webkit.org/show_bug.cgi?id=81481
1612
1613         Reviewed by Dirk Pranke.
1614
1615         FreeBSD has the same sysctl interface present in Darwin to query
1616         the number of CPUs available, so use that too.
1617
1618         * Scripts/webkitdirs.pm:
1619         (determineNumberOfCPUs): Share the sysctl call with Darwin.
1620         (isFreeBSD): Added.
1621
1622 2012-05-07  Alexis Menard  <alexis.menard@openbossa.org>
1623
1624         Unreviewed Qt build fix in Debug.
1625
1626         We need some system includes here for getpid() to be recognized.
1627
1628         * WebKitTestRunner/qt/main.cpp:
1629
1630 2012-05-07  Ojan Vafai  <ojan@chromium.org>
1631
1632         When embedding the flakiness dashboard, hide popups when the frame is blurred.
1633         https://bugs.webkit.org/show_bug.cgi?id=85684
1634
1635         Reviewed by Dirk Pranke.
1636
1637         If you show a popup, then click elsewhere in garden-o-matic, it will now
1638         properly hide it.
1639
1640         * TestResultServer/static-dashboards/dashboard_base.js:
1641         (showPopup):
1642         * TestResultServer/static-dashboards/flakiness_dashboard.js:
1643         (showPopupForBuild):
1644         (generatePageForIndividualTests.if):
1645         (generatePageForIndividualTests):
1646         * TestResultServer/static-dashboards/flakiness_dashboard_embedded_unittests.js: Added.
1647         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
1648         * TestResultServer/static-dashboards/run-embedded-unittests.html: Added.
1649
1650 2012-05-07  Ian Vollick  <vollick@chromium.org>
1651
1652         [chromium] WebViewHost should setCompositorSurfaceReady
1653         https://bugs.webkit.org/show_bug.cgi?id=85418
1654
1655         Reviewed by James Robinson.
1656
1657         * DumpRenderTree/chromium/WebViewHost.cpp:
1658         (WebViewHost::setWebWidget):
1659
1660 2012-05-07  Jon Lee  <jonlee@apple.com>
1661
1662         WebKitTestRunner needs layoutTestController.dumpFrameLoadCallbacks
1663         https://bugs.webkit.org/show_bug.cgi?id=42331
1664         <rdar://problem/8193641>
1665
1666         Reviewed by Darin Adler.
1667
1668         This patch sets up the framework for supporting dumpFrameLoadCallbacks, as well as
1669         outputting some of the required strings. As tests become unskipped the other callbacks
1670         can be supported.
1671
1672         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl: Add dumpFrameLoadCallbacks()
1673         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1674         (WTR::InjectedBundle::didReceiveMessage): pass in the dictionary containing keys needed to initialize
1675         dump callbacks for the layoutTestController.
1676         (WTR::InjectedBundle::booleanForKey): Helper function to extract boolean value from key in WKDictionaryRef.
1677         Output some warning if we are trying to extract a value that is not a boolean.
1678         (WTR::InjectedBundle::beginTesting): Update to take in the dictionary passed in by the TestController.
1679         We cannot set the bits on the layoutTestController until it has been created in beginTesting().
1680         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1681         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
1682         (WTR::InjectedBundle::didReceiveMessage):
1683         (WTR):
1684         (WTR::InjectedBundle::beginTesting):
1685         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
1686         (InjectedBundle):
1687         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
1688         (WTR):
1689         (WTR::dumpFrameDescriptionSuitableForTestResult): Static function similar to
1690         -[WebFrame _drt_descriptionSuitableForTestResult] in DumpRenderTree.
1691         (WTR::InjectedBundlePage::didStartProvisionalLoadForFrame): Output string.
1692         (WTR::InjectedBundlePage::didCommitLoadForFrame): Output string.
1693         (WTR::InjectedBundlePage::didFinishLoadForFrame): Output string.
1694         (WTR::InjectedBundlePage::didReceiveTitleForFrame): Output string.
1695         (WTR::InjectedBundlePage::didCancelClientRedirectForFrame): Output string.
1696         (WTR::InjectedBundlePage::willPerformClientRedirectForFrame): Output string. In WK1 it's
1697         "willPerformClientRedirectToURL" so we will have to use that value here.
1698         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame): Output string.
1699         (WTR::InjectedBundlePage::didHandleOnloadEventsForFrame): Output string.
1700         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
1701         (WTR::LayoutTestController::LayoutTestController): Initialize to not dump frame load callbacks.
1702         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
1703         (WTR::LayoutTestController::dumpFrameLoadCallbacks): Change to use setShouldDumpFrameLoadCallbacks.
1704         (WTR::LayoutTestController::setShouldDumpFrameLoadCallbacks): Set bit.
1705         (WTR::LayoutTestController::shouldDumpFrameLoadCallbacks): Return bit.
1706         (LayoutTestController): Added m_dumpFrameLoadCallbacks to track whether to dump the output.
1707         * WebKitTestRunner/TestInvocation.cpp:
1708         (WTR::shouldLogFrameLoadDelegates): Similar to DRT.
1709         (WTR::TestInvocation::invoke): Based on the URL loaded, set the bit in the dictionary that gets passed
1710         to the injected bundle.
1711
1712 2012-05-07  Tommy Widenflycht  <tommyw@google.com>
1713
1714         [chromium] MediaStream API: Enhance WebUserMediaClientMock and add a test for it
1715         https://bugs.webkit.org/show_bug.cgi?id=85618
1716
1717         Reviewed by Adam Barth.
1718
1719         Adding the capability to control which MediaStreamTracks actually get created,
1720         and adding a test that exercises the JS api using that capability.
1721
1722         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
1723         (WebKit::WebUserMediaClientMock::requestUserMedia):
1724         * DumpRenderTree/chromium/WebUserMediaClientMock.h:
1725         * DumpRenderTree/chromium/WebViewHost.cpp:
1726         (WebViewHost::testMediaStreamClient):
1727         * DumpRenderTree/chromium/WebViewHost.h:
1728         (WebViewHost):
1729
1730 2012-05-07  Zan Dobersek  <zandobersek@gmail.com>
1731
1732         [Gtk] WebGL feature is not built anymore through build-webkit after r116251
1733         https://bugs.webkit.org/show_bug.cgi?id=85788
1734
1735         Reviewed by Martin Robinson.
1736
1737         Enable WebGL by default for the Gtk port. The feature is already enabled by
1738         default in configure.ac, so there's no reason it shouldn't be enabled when
1739         building through build-webkit.
1740
1741         * Scripts/webkitperl/FeatureList.pm:
1742
1743 2012-05-07  Csaba Osztrogonác  <ossy@webkit.org>
1744
1745         Unittest fix after r116288, because now there are more than 5 hit for the 'Eric' word.
1746         We have only 2 'Gavin', so it is a better choice than 'Eric' for unit testing.
1747
1748         Reviewed by Tor Arne Vestbø.
1749
1750         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
1751         (IRCCommandTest.test_whois):
1752
1753 2012-05-07  Sheriff Bot  <webkit.review.bot@gmail.com>
1754
1755         Unreviewed, rolling out r116299, r116301, and r116303.
1756         http://trac.webkit.org/changeset/116299
1757         http://trac.webkit.org/changeset/116301
1758         http://trac.webkit.org/changeset/116303
1759         https://bugs.webkit.org/show_bug.cgi?id=85795
1760
1761         Build is still broken (Requested by Ossy on #webkit).
1762
1763         * DumpRenderTree/qt/DumpRenderTree.pro:
1764         * DumpRenderTree/qt/main.cpp:
1765         (main):
1766         * QtTestPlatformPlugin/QtTestPlatformPlugin.pro: Removed.
1767         * QtTestPlatformPlugin/TestIntegration.cpp: Removed.
1768         * QtTestPlatformPlugin/TestIntegration.h: Removed.
1769         * QtTestPlatformPlugin/mac/TestFontDatabase.h: Removed.
1770         * QtTestPlatformPlugin/mac/TestFontDatabase.mm: Removed.
1771         * QtTestPlatformPlugin/mac/TestIntegrationMac.mm: Removed.
1772         * QtTestPlatformPlugin/main.cpp: Removed.
1773         * QtTestPlatformPlugin/testplatform.json: Removed.
1774         * Tools.pro:
1775         * WebKitTestRunner/Target.pri:
1776         * WebKitTestRunner/qt/main.cpp:
1777         (main):
1778
1779 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
1780
1781         Unreviewed Qt build fix.
1782
1783         One more try to learn how to use
1784         const with a pointer array.
1785
1786         * DumpRenderTree/qt/main.cpp:
1787         (initializeTestPlatformPlugin):
1788
1789 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
1790
1791         Unreviewed build fix after r116299.
1792
1793         * DumpRenderTree/qt/main.cpp:
1794         (initializeTestPlatformPlugin):
1795
1796 2012-05-07  Balazs Kelemen  <kbalazs@webkit.org>
1797
1798         [Qt] Add test specific platform plugin to achieve unified layout test results
1799         https://bugs.webkit.org/show_bug.cgi?id=80996
1800
1801         Reviewed by Simon Hausmann.
1802
1803         Added QtTestPlatformPlugin as a new project under Tools.
1804         This is a Qt5-ish platform plugin that can be used to tweak the
1805         platform support interfaces in order to unify layout test results.
1806         For now it only overrides the font database on Mac and redirects
1807         everything else to the real platform plugin. The font database it
1808         provides mimics the way how we set up test fonts with fontconfig on Linux.
1809         Make DumpRenderTree and WebKitTestRunner use this platform plugin.
1810
1811         * DumpRenderTree/qt/DumpRenderTree.pro:
1812         * DumpRenderTree/qt/main.cpp:
1813         (initializeTestPlatformPlugin):
1814         (main):
1815         * QtTestPlatformPlugin/QtTestPlatformPlugin.pro: Added.
1816         * QtTestPlatformPlugin/TestIntegration.cpp: Added.
1817         (TestIntegration::TestIntegration):
1818         (TestIntegration::fontDatabase):
1819         * QtTestPlatformPlugin/TestIntegration.h: Added.
1820         (TestIntegration):
1821         (TestIntegration::hasCapability):
1822         (TestIntegration::createPlatformPixmap):
1823         (TestIntegration::createPlatformWindow):
1824         (TestIntegration::createPlatformBackingStore):
1825         (TestIntegration::createPlatformOpenGLContext):
1826         (TestIntegration::createPlatformSharedGraphicsCache):
1827         (TestIntegration::guiThreadEventDispatcher):
1828         (TestIntegration::clipboard):
1829         (TestIntegration::drag):
1830         (TestIntegration::inputContext):
1831         (TestIntegration::accessibility):
1832         (TestIntegration::nativeInterface):
1833         (TestIntegration::services):
1834         (TestIntegration::styleHint):
1835         (TestIntegration::platformTheme):
1836         * QtTestPlatformPlugin/mac/TestFontDatabase.h: Added.
1837         (TestFontDatabase):
1838         * QtTestPlatformPlugin/mac/TestFontDatabase.mm: Added.
1839         (TestFontDatabase::populateFontDatabase):
1840         * QtTestPlatformPlugin/mac/TestIntegrationMac.mm: Added.
1841         (TestIntegration::fontDatabase):
1842         * QtTestPlatformPlugin/main.cpp: Added.
1843         (TestIntegrationPlugin::keys):
1844         (TestIntegrationPlugin::create):
1845         (TestIntegrationPlugin::initialize):
1846         * QtTestPlatformPlugin/testplatform.json: Added.
1847         * Tools.pro:
1848         * WebKitTestRunner/Target.pri:
1849         * WebKitTestRunner/qt/main.cpp:
1850         (main):
1851
1852 2012-05-07  Christophe Dumez  <christophe.dumez@intel.com>
1853
1854         [EFL] media/video-poster-blocked-by-willsendrequest.html is flaky
1855         https://bugs.webkit.org/show_bug.cgi?id=85609
1856
1857         Reviewed by Simon Hausmann.
1858
1859         Flakiness was caused by the fact that
1860         media/crash-closing-page-with-media-as-plugin-fallback.html and
1861         media/video-poster-blocked-by-willsendrequest.html use the same PNG
1862         file as poster attribute in the video tag. The PNG file gets cached
1863         when the first test is executed and this somehow causes the second
1864         test to fail. 
1865
1866         The flakiness was addressed by clearing the memory and page caches
1867         between tests.
1868
1869         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
1870         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
1871
1872 2012-05-07  Eric Seidel  <eric@webkit.org>
1873
1874         Add values for all features to Qt's features.pri
1875         https://bugs.webkit.org/show_bug.cgi?id=85746
1876
1877         Reviewed by Tor Arne Vestbø.
1878
1879         When generate-feature-files generates features.pri it provides a value
1880         for every possible ENABLE_.  In prepration for landing generate-feature-files
1881         it seems appropriate to add a value for all possible ENABLEs to the existing
1882         features.pri.  This lets me make sure that I'm getting them all right!
1883
1884         * qmake/mkspecs/features/features.pri:
1885
1886 2012-05-07  Adam Bergkvist  <adam.bergkvist@ericsson.com>
1887
1888         Added myself as a committer.
1889
1890         Not reviewed.
1891
1892         * Scripts/webkitpy/common/config/committers.py:
1893
1894 2012-05-07  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1895
1896         [EFL][DRT] Implement LayoutTestController::execCommand
1897         https://bugs.webkit.org/show_bug.cgi?id=82286
1898
1899         Reviewed by Antonio Gomes.
1900
1901         Add missing implementation execCommand and isCommandEnabled to EFL's
1902         LayoutTestController. 
1903
1904         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1905         (LayoutTestController::execCommand):
1906         (LayoutTestController::isCommandEnabled):
1907
1908 2012-05-06  Hayato Ito  <hayato@chromium.org>
1909
1910         [prepare-ChangeLog] Excludes a reference file used by reftests from test files.
1911         https://bugs.webkit.org/show_bug.cgi?id=74268
1912
1913         Reviewed by Ryosuke Niwa.
1914
1915         * Scripts/prepare-ChangeLog:
1916         (generateFileList):
1917
1918 2012-05-06  Mary Wu  <mary.wu@torchmobile.com.cn>
1919
1920         Add a contributor to committers.py
1921         https://bugs.webkit.org/show_bug.cgi?id=85761
1922
1923         Reviewed by Unreviewed.
1924
1925         Add myself to committers.py contributor.
1926
1927         * Scripts/webkitpy/common/config/committers.py:
1928
1929 2012-05-06  Dan Bernstein  <mitz@apple.com>
1930
1931         Tools part of: Building and debugging WebKit in the Xcode IDE requires a lot of setup
1932         https://bugs.webkit.org/show_bug.cgi?id=85739
1933
1934         Reviewed by Daniel Bates.
1935
1936         * Scripts/build-webkit: Moved code that copies from WebKitLibraries to the product directory
1937         from here...
1938         * Scripts/copy-webkitlibraries-to-product-directory: ...to this new script.
1939
1940 2012-05-06  Andy Estes  <aestes@apple.com>
1941
1942         Fix the build after r116246.
1943
1944         * DumpRenderTree/LayoutTestController.cpp: Remove an unused function.
1945
1946 2012-05-06  Eric Seidel  <eric@webkit.org>
1947
1948         Update several build-webkit options to match the #defines they toggle
1949         https://bugs.webkit.org/show_bug.cgi?id=85744
1950
1951         Reviewed by Adam Barth.
1952
1953         Most notable here is fixing --3d-canvas to be --webgl these days --
1954         the define was renamed 15 months ago!
1955
1956         * Scripts/webkitperl/FeatureList.pm:
1957
1958 2012-05-06  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
1959
1960         [EFL] EFL's LayoutTestController setJavaScriptCanAccessClipboard implementation
1961         https://bugs.webkit.org/show_bug.cgi?id=83687
1962
1963         Reviewed by Antonio Gomes.
1964
1965         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1966         (LayoutTestController::setJavaScriptCanAccessClipboard):
1967
1968 2012-05-06  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1969
1970         Convert isPageBoxVisible to use Internals interface.
1971         https://bugs.webkit.org/show_bug.cgi?id=85692
1972
1973         Reviewed by Darin Adler.
1974
1975         Remove isPageBoxVisible functions, because it is able to work in the
1976         cross-port way through the Internals interface.
1977
1978         * DumpRenderTree/LayoutTestController.cpp:
1979         (LayoutTestController::staticFunctions):
1980         * DumpRenderTree/LayoutTestController.h:
1981         (LayoutTestController):
1982         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1983         * DumpRenderTree/chromium/LayoutTestController.cpp:
1984         (LayoutTestController::LayoutTestController):
1985         * DumpRenderTree/chromium/LayoutTestController.h:
1986         (LayoutTestController):
1987         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1988         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1989         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1990         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
1991         * DumpRenderTree/qt/LayoutTestControllerQt.h:
1992         (LayoutTestController):
1993         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1994         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1995
1996 2012-05-06  Tom Zakrajsek  <tomz@codeaurora.org>
1997
1998         sheriffbot isn't reopening patches after it lands rollouts
1999         https://bugs.webkit.org/show_bug.cgi?id=64418
2000
2001         Reviewed by Adam Barth.
2002
2003         * Scripts/webkitpy/tool/commands/download_unittest.py:
2004         * Scripts/webkitpy/tool/steps/createbug.py:
2005         (CreateBug.run):
2006
2007 2012-05-04  Raphael Kubo da Costa  <rakuco@webkit.org>
2008
2009         webkitpy: Use os.pathsep instead of manually finding out the path separator in NRWT.
2010         https://bugs.webkit.org/show_bug.cgi?id=85697
2011
2012         Reviewed by Eric Seidel.
2013
2014         * Scripts/new-run-webkit-tests: Instead of manually detecting whether
2015         the path separator should be ':' or ';' by checking sys.platform, rely
2016         on os.pathsep which has exactly the same information (os.pathsep is
2017         ':' for Cygwin, from what I could check in Python's sources).
2018
2019 2012-05-05  Sheriff Bot  <webkit.review.bot@gmail.com>
2020
2021         Unreviewed, rolling out r116223.
2022         http://trac.webkit.org/changeset/116223
2023         https://bugs.webkit.org/show_bug.cgi?id=85718
2024
2025         The bots need to set XDG_DATA_DIRS (Requested by rakuco on
2026         #webkit).
2027
2028         * efl/jhbuildrc:
2029         * gtk/jhbuildrc:
2030
2031 2012-05-05  Raphael Kubo da Costa  <rakuco@webkit.org>
2032
2033         [jhbuild] Do not set XDG_DATA_DIRS in jhbuildrc.
2034         https://bugs.webkit.org/show_bug.cgi?id=85717
2035
2036         Reviewed by Martin Robinson.
2037
2038         After r116209, the XDG_DATA_DIRS environment variable is now
2039         passed by webkitpy's layout_tests code to run-with-jhbuild, so
2040         instead of setting a few defaults in jhbuildrc we can now rely on
2041         the values set and really used by each system.
2042
2043         * efl/jhbuildrc:
2044         * gtk/jhbuildrc:
2045
2046 2012-05-05  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2047
2048         [Qt] Buildfix for --minimal build after r116107.
2049
2050         Reviewed by Csaba Osztrogonác.
2051
2052         * qmake/mkspecs/features/features.prf:
2053
2054 2012-05-05  Zan Dobersek  <zandobersek@gmail.com>
2055
2056         [Gtk] GtkDriver is not required anymore after r116134
2057         https://bugs.webkit.org/show_bug.cgi?id=85699
2058
2059         Reviewed by Csaba Osztrogonác.
2060
2061         Remove GtkDriver as it was replaced by the XvfbDriver and is
2062         no longer needed.
2063
2064         * Scripts/webkitpy/layout_tests/port/gtk.py:
2065
2066 2012-05-05  Zan Dobersek  <zandobersek@gmail.com>
2067
2068         Increase the display ID for the Xvfb process to use if pixel_tests argument
2069         is true rather than if pixel_tests option is present. This fixes the
2070         XvfbDriver in the same way r115825 fixed the now-redundant GtkDriver.
2071
2072         Rubber-stamped by Csaba Osztrogonác.
2073
2074         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py:
2075         (XvfbDriver._start):
2076
2077 2012-05-04  Raphael Kubo da Costa  <rakuco@webkit.org>
2078
2079         webkitpy: Preserve XDG_DATA_DIRS in the environment passed to ServerProcess.
2080         https://bugs.webkit.org/show_bug.cgi?id=85694
2081
2082         Reviewed by Martin Robinson.
2083
2084         This is sort of a follow-up to r100674: $XDG_DATA_DIRS should be
2085         preserved, as it is particularly useful for Linux/Unix
2086         environments.  As a real use case, if shared-mime-info is
2087         installed into a non-trivial prefix, Glib and other libraries will
2088         have trouble to properly identify mimetypes (in ResourceResponses,
2089         for example) as the base freedesktop.org mimetype file will not be
2090         found.
2091
2092         * Scripts/webkitpy/layout_tests/port/base.py:
2093         (Port.to.setup_environ_for_server):
2094
2095 2012-05-04  Gustavo Noronha Silva  <gns@gnome.org>
2096
2097         [GTK] Simplify how libWebCoreModules is linked in, and fix WebKit2 build
2098         https://bugs.webkit.org/show_bug.cgi?id=85691
2099
2100         * GNUmakefile.am: no longer link libWebCoreModules to DumpRenderTree.
2101
2102 2012-05-04  Eric Seidel  <eric@webkit.org>
2103
2104         Make capitalization and descriptions in FeatureList.pm more consistent to make easier to autogenerate
2105         https://bugs.webkit.org/show_bug.cgi?id=85583
2106
2107         Reviewed by Daniel Bates.
2108
2109         * Scripts/webkitperl/FeatureList.pm:
2110
2111 2012-05-04  Eric Seidel  <eric@webkit.org>
2112
2113         Sort entries in FeaturesList.pm to make them easier to autogenerate
2114         https://bugs.webkit.org/show_bug.cgi?id=85584
2115
2116         Reviewed by Daniel Bates.
2117
2118         No functional changes, only sorting.
2119
2120         * Scripts/webkitperl/FeatureList.pm:
2121
2122 2012-05-04  Andy Estes  <aestes@apple.com>
2123
2124         REGRESSION (r115752): WebKit2.NewFirstVisuallyNonEmptyLayout API test times out
2125         https://bugs.webkit.org/show_bug.cgi?id=85675
2126
2127         * TestWebKitAPI/Tests/WebKit2/NewFirstVisuallyNonEmptyLayout.cpp:
2128         (TestWebKitAPI::TEST): Disable the test.
2129
2130 2012-04-30  Jer Noble  <jer.noble@apple.com>
2131
2132         Full screen will exit during a provisional load of a non-ancestor iframe.
2133         https://bugs.webkit.org/show_bug.cgi?id=85230
2134
2135         Reviewed by Maciej Stachowiak .
2136
2137         Add support for the new UIDelegate webView:closeFullScreenWithListener: method:
2138         * DumpRenderTree/mac/UIDelegate.mm:
2139         (-[UIDelegate webView:closeFullScreenWithListener:]):
2140
2141         Add support for the new closeFullScreen injected bundle method:
2142         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2143         (WTR::InjectedBundlePage::InjectedBundlePage):
2144         (WTR::InjectedBundlePage::closeFullScreen):
2145         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
2146
2147 2012-05-04  Ojan Vafai  <ojan@chromium.org>
2148
2149         Fix test naming to be camel-case like the methods they are testing.
2150
2151         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2152         (test):
2153
2154 2012-05-04  Ojan Vafai  <ojan@chromium.org>
2155
2156         Show the results links in the embedded flakiness dashboard
2157         https://bugs.webkit.org/show_bug.cgi?id=85660
2158
2159         Reviewed by Adam Barth.
2160
2161         Turns out these links don't add much visual noise and it's annoying
2162         to have to go to the dashboard from garden-o-matic just to view the
2163         results for a test.
2164
2165         This is useful for cases where garden-o-matic doesn't properly show the
2166         results (e.g. missing tests). If we address all those cases, then
2167         we could easily hide the results links again if we wanted to.
2168
2169         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2170         (showUpdateInfoForTest):
2171         (htmlForIndividualTestOnAllBuildersWithResultsLinks):
2172         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2173         (test):
2174
2175 2012-05-04  Dirk Pranke  <dpranke@chromium.org>
2176
2177         Revert to using chromium --test-shell mode on SL in NRWT
2178
2179         https://bugs.webkit.org/show_bug.cgi?id=83076
2180
2181         Unreviewed, build fix.
2182
2183         We seem to be seeing a lot more flakiness on the bot since
2184         I switched NRWT to "DRT" mode. We'll try reverting back and
2185         see if this helps things.
2186
2187         * Scripts/webkitpy/layout_tests/port/chromium.py:
2188         (ChromiumDriver.__init__):
2189
2190 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
2191
2192         [EFL] Implement layoutTestController.dumpResourceResponseMIMETypes
2193         https://bugs.webkit.org/show_bug.cgi?id=84941
2194
2195         Reviewed by Martin Robinson.
2196
2197         Print MIME type detection messages in EFL's DumpRenderTree if
2198         LayoutTestController's dumpResourceResponseMIMETypes() returns true.
2199
2200         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2201         (DumpRenderTreeChrome::onResponseReceived):
2202
2203 2012-05-04  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
2204
2205         [GTK] DRT needs an implementation of LayoutTestController::setDefersLoading and ::goBack
2206         https://bugs.webkit.org/show_bug.cgi?id=85134
2207
2208         Reviewed by Martin Robinson.
2209
2210         Add missing implementation setDefersLoading and goBack to GTK's
2211         LayoutTestController so that the test case rely on these methods
2212         can be enabled.
2213
2214         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2215         (resetDefaultsToConsistentValues):
2216         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2217         (LayoutTestController::goBack):
2218         (LayoutTestController::setDefersLoading):
2219
2220 2012-05-04  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
2221
2222         [Qt][NRWT] Run each DRT in it's own xvfb
2223         https://bugs.webkit.org/show_bug.cgi?id=77335
2224
2225         Reviewed by Dirk Pranke.
2226
2227         * Scripts/webkitpy/layout_tests/port/gtk.py:
2228         (GtkPort._driver_class):
2229         * Scripts/webkitpy/layout_tests/port/qt.py:
2230         (QtPort._driver_class):
2231         * Scripts/webkitpy/layout_tests/port/xvfbdriver.py: Added.
2232         (XvfbDriver):
2233         (XvfbDriver._start):
2234         (XvfbDriver._start.x_filter):
2235         (XvfbDriver.stop):
2236
2237 2012-05-04  Zan Dobersek  <zandobersek@gmail.com>
2238
2239         Use destructuring assingment for the return values of the
2240         port.diff_image() method call.
2241
2242         Rubber-stamped by Dirk Pranke.
2243
2244         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2245         (write_test_result):
2246
2247 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2248
2249         Store fewer runs by default in the flakiness dashboard
2250         https://bugs.webkit.org/show_bug.cgi?id=85568
2251
2252         Reviewed by Dirk Pranke.
2253
2254         This will greatly shrink the size of the json files we load,
2255         resulting in a much faster initial load. People rarely care
2256         about more than 100 runs. For the cases they do, they can still click
2257         the show all runs checkbox to show 500 runs.
2258
2259         * TestResultServer/model/jsonresults.py:
2260
2261 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
2262
2263         [EFL] Frame load callbacks output is missing for redirections
2264         https://bugs.webkit.org/show_bug.cgi?id=85173
2265
2266         Reviewed by Antonio Gomes.
2267
2268         Catch the new ewk_frame signals to print the expected output in case
2269         of redirections, when LayoutTestController's dumpFrameLoadCallbacks()
2270         returns true.
2271
2272         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2273         (DumpRenderTreeChrome::createView):
2274         (DumpRenderTreeChrome::onWillSendRequest):
2275         (DumpRenderTreeChrome::onFrameCreated):
2276         (DumpRenderTreeChrome::onFrameRedirectForProvisionalLoad):
2277         (DumpRenderTreeChrome::onFrameRedirectRequested):
2278         (DumpRenderTreeChrome::onResponseReceived):
2279         * DumpRenderTree/efl/DumpRenderTreeChrome.h:
2280         (DumpRenderTreeChrome):
2281
2282 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2283
2284         [Qt] Fix up warning about missing GLib/Gio/GStreamer
2285
2286         We don't actually fall back.
2287
2288         Rubber-stamped by Csaba Osztrogonác.
2289
2290         * qmake/mkspecs/features/features.prf:
2291
2292 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2293
2294         [Qt] Clarify warning about missing GLib/Gio/GStreamer for media support
2295
2296         And only print it once.
2297
2298         Reviewed by Csaba Osztrogonác.
2299
2300         * qmake/mkspecs/features/features.prf:
2301
2302 2012-05-04  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2303
2304         [Qt] Clean up and split features.prf into a static list of defaults
2305
2306         The static list of feature defaults is used as a fallback for any
2307         feature that's not dynamically detected or overriden on the command
2308         line (though build-webkit or passing DEFINES+= to qmake).
2309
2310         The static list is complete, which allows for auto-generation based
2311         on Features.py (see bug https://bugs.webkit.org/show_bug.cgi?id=85456)
2312
2313         https://bugs.webkit.org/show_bug.cgi?id=85611
2314
2315         Reviewed by Simon Hausmann.
2316
2317         * Scripts/build-webkit:
2318         * Tools.pro:
2319         * qmake/mkspecs/features/features.prf:
2320         * qmake/mkspecs/features/features.pri: Added.
2321
2322 2012-05-04  Sheriff Bot  <webkit.review.bot@gmail.com>
2323
2324         Unreviewed, rolling out r116085, r116091, and r116095.
2325         http://trac.webkit.org/changeset/116085
2326         http://trac.webkit.org/changeset/116091
2327         http://trac.webkit.org/changeset/116095
2328         https://bugs.webkit.org/show_bug.cgi?id=85628
2329
2330         We are not ready with dependencies on all platform yet (mac) +
2331         problems with debug builds. (Requested by Zoltan on #webkit).
2332
2333         * qmake/mkspecs/features/features.prf:
2334
2335 2012-05-04  Christophe Dumez  <christophe.dumez@intel.com>
2336
2337         [EFL] LayoutTestController does not support overriding WebKitLoadSiteIconsKey
2338         https://bugs.webkit.org/show_bug.cgi?id=85171
2339
2340         Reviewed by Andreas Kling.
2341
2342         Implement support for overriding "WebKitLoadSiteIconsKey" setting in
2343         EFL's LayoutTestController.
2344
2345         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2346         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
2347         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2348         (LayoutTestController::overridePreference):
2349
2350 2012-05-03  Alexis Menard  <alexis.menard@openbossa.org>
2351
2352         [Qt] Enable fullscreen API for WebKit2.
2353         https://bugs.webkit.org/show_bug.cgi?id=85498
2354
2355         Reviewed by Simon Hausmann.
2356
2357         Enable the fullscreen API on the MiniBrowser and turn
2358         it default on WebKit2, disable it on WK1.
2359
2360         * MiniBrowser/qt/BrowserWindow.cpp:
2361         (BrowserWindow::BrowserWindow):
2362         * MiniBrowser/qt/qml/BrowserWindow.qml:
2363         * Scripts/webkitperl/FeatureList.pm:
2364         * qmake/mkspecs/features/features.prf:
2365
2366 2012-05-04  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2367
2368         REGRESSION(r116008) old-run-webkit-tests broken
2369         https://bugs.webkit.org/show_bug.cgi?id=85597
2370
2371         Reviewed by Simon Hausmann.
2372
2373         Reinstate libraryContainsSymbol.
2374
2375         * Scripts/webkitperl/features.pm:
2376         (libraryContainsSymbol):
2377
2378 2012-05-04  Zoltan Horvath  <zoltan@webkit.org>
2379
2380         [Qt] Set WebCore imagedecoders as default and add fallback to QImageDecoder
2381         https://bugs.webkit.org/show_bug.cgi?id=80400
2382
2383         Remove WTF_USE_QT_IMAGE_DECODER macro.
2384
2385         Reviewed by Simon Hausmann.
2386
2387         * qmake/mkspecs/features/features.prf:
2388
2389 2012-05-03  Christophe Dumez  <christophe.dumez@intel.com>
2390
2391         [EFL] DRT should obey layoutTestController's addURLToRedirect()
2392         https://bugs.webkit.org/show_bug.cgi?id=82722
2393
2394         Reviewed by Maciej Stachowiak.
2395
2396         Implement support for layoutTestController's addURLToRedirect()
2397         in EFL's DRT so that it redirects properly when indicated.
2398
2399         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2400         (DumpRenderTreeChrome::onWillSendRequest):
2401
2402 2012-05-03  Christophe Dumez  <christophe.dumez@intel.com>
2403
2404         [EFL] DRT should support LayoutTestController's willSendRequestReturnsNullOnRedirect()
2405         https://bugs.webkit.org/show_bug.cgi?id=82704
2406
2407         Reviewed by Antonio Gomes.
2408
2409         EFL's DRT needs to obey LayoutTestController's
2410         willSendRequestReturnsNullOnRedirect().
2411
2412         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
2413         (DumpRenderTreeChrome::onWillSendRequest):
2414
2415 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2416
2417         git-add-reviewer: Do not put the bug title and its URL in the same line in the commit message.
2418         https://bugs.webkit.org/show_bug.cgi?id=85570
2419
2420         Reviewed by Adam Roben.
2421
2422         The standard format of commit messages and ChangeLog entries has a
2423         short summary in the first line, followed by the bug URL in
2424         Bugzilla. Passing "%s" in the pretty format line to git-rev-list
2425         merges these two lines into a single one separated by a space
2426         character, which is not what we want.
2427
2428         Since "%s%n%n%b" is in practice "%B" (raw message), use it
2429         instead, as the first two lines do not end up being merged
2430         together.
2431
2432         * Scripts/git-add-reviewer:
2433         (writeCommitMessageToFile):
2434
2435 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2436
2437         Improve UI for garden-o-matic examine view when there are fewer than 5 tests.
2438         https://bugs.webkit.org/show_bug.cgi?id=85566
2439
2440         Reviewed by Adam Barth.
2441
2442         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2443         (.):
2444         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
2445         (.):
2446         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2447
2448 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2449
2450         webkitpy: Recognize FreeBSD as a valid platform.
2451         https://bugs.webkit.org/show_bug.cgi?id=81467
2452
2453         Reviewed by Adam Barth.
2454
2455         Add PlatformInfo.is_freebsd() so that trying to use webkit-patch
2456         on FreeBSD does not raise an "unrecognized platform" exception.
2457
2458         Most of this patch contains some unit tests for the mentioned
2459         method, as currently platform-specific code is needed only on
2460         Windows.
2461
2462         * Scripts/webkitpy/common/system/platforminfo.py:
2463         (PlatformInfo.__init__):
2464         (PlatformInfo.is_freebsd):
2465         (PlatformInfo._determine_os_name):
2466         * Scripts/webkitpy/common/system/platforminfo_mock.py:
2467         (MockPlatformInfo.is_freebsd):
2468         * Scripts/webkitpy/common/system/platforminfo_unittest.py:
2469         (fake_platform):
2470         (fake_platform.FakePlatformModule.release):
2471         (TestPlatformInfo.test_real_code):
2472         (TestPlatformInfo.test_os_name_and_wrappers):
2473         (TestPlatformInfo.test_os_version):
2474         (TestPlatformInfo.test_display_name):
2475         (TestPlatformInfo.test_total_bytes_memory):
2476         (TestPlatformInfo.test_free_bytes_memory):
2477         * Scripts/webkitpy/layout_tests/port/factory.py:
2478         (PortFactory._default_port):
2479
2480 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2481
2482         Make the table headers white-space:nowrap. This decreases the vertical size,
2483         which makes for more minimal embedding in garden-o-matic.
2484
2485         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
2486         (.table-header-content *):
2487
2488 2012-05-03  Eric Seidel  <eric@webkit.org>
2489
2490         Split build-webkit's feature option list off into its own module in preparation for autogeneration
2491         https://bugs.webkit.org/show_bug.cgi?id=85548
2492
2493         Reviewed by Daniel Bates.
2494
2495         My Perl-fu is very weak.  This moves the feature option list off into a separate module
2496         file so that I can easily autogenerate just that file with the fancy new generate-feature-files command.
2497         It's a bit odd to have an array which we return out to build-webkit to modify, but it turns out
2498         not to matter.  Also, build-webkit doesn't actually use any of the variables we're using to
2499         store the results from the option parse, so those end up inaccessible from build-webkit (except by reference)
2500         which is actually cleaner.  Callers can't assume that they would get a new array every time, but someone
2501         with more Perl-fu than I could create something fancier than this if we need in the future.
2502
2503         * Scripts/build-webkit:
2504         * Scripts/webkitperl/FeatureList.pm: Added.
2505         (getFeatureOptionList):
2506
2507 2012-05-03  Dirk Pranke  <dpranke@chromium.org>
2508
2509         REGRESSION: run-webkit-tests failing on Chromium Mac
2510         https://bugs.webkit.org/show_bug.cgi?id=85459
2511
2512         Unreviewed, build fix.
2513
2514         Turns out select doesn't throw IOErrors, it has it's own kind of
2515         error :(. Trap that instead.
2516
2517         * Scripts/webkitpy/layout_tests/port/server_process.py:
2518         (ServerProcess._wait_for_data_and_update_buffers_using_select):
2519
2520 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2521
2522         [CMake] Rewrite FindCairo.cmake.
2523         https://bugs.webkit.org/show_bug.cgi?id=84895
2524
2525         Reviewed by Daniel Bates.
2526
2527         The old approach relied on pkg-config for finding Cairo (which
2528         introduced a dependency on pkg-config that could be avoided), used
2529         the LibFindMacros code that we should probably remove in the
2530         future and did not use the FindPackageHandleStandardArguments
2531         module.
2532
2533         Change all that by rewriting the module.
2534         - Use the pkg-config output optionally instead of requiring it
2535         like LibFindMacros did.
2536         - Remove the implicit dependency on FreeType which often found it
2537         the wrong way via pkg-config and without considering
2538         CMAKE_PREFIX_PATH.
2539         - Retrieve the Cairo version by looking at cairo-version.h instead
2540         of relying on pkg-config. It requires some additional code for
2541         checking if the desired version has been found, but that will not
2542         be needed once we start depending on CMake 2.8.3 or later.
2543
2544         The only downside is that FPHSA sets <UPPERCASED_NAME>_FOUND
2545         instead of <Name>_FOUND, and to keep things consistent
2546         Cairo_LIBRARIES and Cairo_INCLUDE_DIRS have become CAIRO_LIBRARIES
2547         and CAIRO_INCLUDE_DIRS.
2548
2549         * DumpRenderTree/efl/CMakeLists.txt: Use CAIRO_FOO instead of
2550         Cairo_FOO.
2551         * EWebLauncher/CMakeLists.txt: Ditto.
2552
2553 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2554
2555         Fix modifier and bug sorting on the flakiness dashboard
2556         https://bugs.webkit.org/show_bug.cgi?id=85554
2557
2558         Reviewed by Dirk Pranke.
2559
2560         The sorting assumes that the column header name matches a field
2561         in the testResults object and uses that field for sorting.
2562         Filter the bugs out of the modifiers list and put them in a bugs field
2563         and only put the non-bug modifiers in the modifiers list.
2564
2565         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2566         (populateExpectationsData):
2567         (htmlForSingleTestRow):
2568         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2569
2570 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2571
2572         Fix some state resetting when running unittests. HtmlForSingleTestRow
2573         was flaky.
2574
2575         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2576         (htmlForSingleTestRow):
2577         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2578         (resetGlobals):
2579
2580 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2581
2582         webkitpy: Use PlatformInfo wherever possible in the User class.
2583         https://bugs.webkit.org/show_bug.cgi?id=85549
2584
2585         Reviewed by Dirk Pranke.
2586
2587         Instead of relying on sys.platform to detect whether the current
2588         platform is a Mac, use the PlatformInfo class instead.
2589
2590         * Scripts/webkitpy/common/system/user.py:
2591         (User.__init__):
2592         (User.edit_changelog):
2593
2594 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2595
2596         Remove the percent failed column from the flakiness dashboard
2597         https://bugs.webkit.org/show_bug.cgi?id=85552
2598
2599         Reviewed by Dirk Pranke.
2600
2601         This column isn't useful enough to justify the visual noise.
2602
2603         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2604         (processMissingAndExtraExpectations):
2605         (tableHeaders):
2606         (htmlForSingleTestRow):
2607         (sortTests):
2608         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2609         (test):
2610
2611 2012-05-03  Eric Seidel  <eric@webkit.org>
2612
2613         --coverage should not not use the "feature" options system in build-webkit, similarly remove special-casing for WTF_URL
2614         https://bugs.webkit.org/show_bug.cgi?id=85542
2615
2616         Reviewed by Daniel Bates.
2617
2618         Remove these last two odd-men-out, so that we can generate the default feature options
2619         for build-webkit using the new generate-feature-files command.
2620         The Mac build system has some special casing for ENABLE_ defines which gets them
2621         defined as -D on the build command, even though initially they start out as
2622         environment variables ENABLE_FOO = ENABLE_FOO.  The Feature options in build-webkit
2623         depend on this support, but it doesn't work for WTF_USE_ defines like WTF_USE_WTF_URL.
2624         I'm told that WTF_URL is not actively being worked on, so just removing it for now.
2625         It should probably come back to life as an ENABLE_, or build-webkit or the Xcode projects
2626         should be made generically smarter about these USE_WTF defines.
2627         (There is still WTF_USE_TILED_BACKING_STORE in the feature list, but it only seems
2628         used by Qt.  If someone tried build-webkit --tiled-backing-store on Mac today it
2629         would not work!)
2630
2631         * Scripts/build-webkit:
2632
2633 2012-05-03  Eric Seidel  <eric@webkit.org>
2634
2635         Remove the last bits of Leopard support from our perl scripts
2636         https://bugs.webkit.org/show_bug.cgi?id=85545
2637
2638         Reviewed by Daniel Bates.
2639
2640         Chromium still supports Leopard (for a little while yet), but AppleMacWebKit dropped
2641         Leopard support several months ago.  Remove the last bits of isLeopard() branching
2642         in our Perl scripts, since no one is using this (Chromium has never used this Perl code).
2643
2644         * Scripts/build-webkit:
2645         * Scripts/old-run-webkit-tests:
2646         (countAndPrintLeaks):
2647         (captureSavedCrashLog):
2648         * Scripts/webkitdirs.pm:
2649         (determineArchitecture):
2650         (argumentsForRunAndDebugMacWebKitApp):
2651
2652 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2653
2654         [Chromium] Don't show the WebKit Linux ASAN bot on the flakiness dashboard.
2655         https://bugs.webkit.org/show_bug.cgi?id=85551
2656
2657         Reviewed by Dirk Pranke.
2658
2659         It's not uploading results and in the short term we don't plan to have it start.
2660
2661         * TestResultServer/static-dashboards/builders.js:
2662         (isChromiumWebkitTipOfTreeTestRunner):
2663         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2664
2665 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2666
2667         Show the test name in the dashboard if showChrome==false and we're showing multiple tests
2668         https://bugs.webkit.org/show_bug.cgi?id=85243
2669
2670         Reviewed by Dirk Pranke.
2671
2672         This happens for cases where a test name matches multiple tests.
2673         In practice, this only happens for virtual test suites (e.g. the canvas tests).
2674
2675         * TestResultServer/static-dashboards/flakiness_dashboard.js:
2676         (htmlForIndividualTestOnAllBuildersWithChrome):
2677         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2678         (testHtmlForIndividualTestOnAllBuildersWithChromeNonexistant):
2679         (testHtmlForIndividualTestOnAllBuildersWithChrome):
2680         (testHtmlForIndividualTestOnAllBuildersWithChromeWebkitMaster):
2681         (testHtmlForIndividualTests):
2682
2683 2012-05-03  Ojan Vafai  <ojan@chromium.org>
2684
2685         Port the flakiness dashboard unittests to qunit
2686         https://bugs.webkit.org/show_bug.cgi?id=85544
2687
2688         Reviewed by Adam Barth.
2689
2690         * TestResultServer/static-dashboards/dashboard_base.js:
2691         (appendJSONScriptElements):
2692         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
2693         (resetGlobals):
2694         (runExpectationsTest):
2695         (test):
2696         * TestResultServer/static-dashboards/run-unittests.html: Added.
2697
2698 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2699
2700         webkitpy: Remove check for readline on Mac.
2701         https://bugs.webkit.org/show_bug.cgi?id=85547
2702
2703         Reviewed by Eric Seidel.
2704
2705         Python 2.5 is not supported by the WebKit project anymore, and
2706         Python's readline documentation does not mention the module not
2707         being available in the supported versions.
2708
2709         * Scripts/webkitpy/common/system/user.py:
2710
2711 2012-05-03  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2712
2713         [NRWT] Run performance tests with lock
2714         https://bugs.webkit.org/show_bug.cgi?id=78628
2715
2716         Reviewed by Tony Chang.
2717
2718         Locking performance tests (like we do for http tests) will force them
2719         to run in serial. This reduces the load of the machine when running perf
2720         tests and minimizes the chances of the tests to fail, specially
2721         timeouts.
2722
2723         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2724         (Manager.__init__):
2725         (Manager._is_perf_test):
2726         (Manager._test_requires_lock):
2727         * Scripts/webkitpy/layout_tests/port/http_lock.py:
2728
2729 2012-05-03  Dirk Pranke  <dpranke@chromium.org>
2730
2731         Attempt to fix a crash on SL ...
2732
2733         Unreviewed, build fix.
2734
2735         It looks like we may be getting an IOError raised during
2736         select() on chromium mac, and that error isn't being caught.
2737         Pushing it into the try/catch block will hopefully catch this
2738         and allow us to see why the subprocess is crashing.
2739
2740         * Scripts/webkitpy/layout_tests/port/server_process.py:
2741         (ServerProcess._wait_for_data_and_update_buffers_using_select):
2742
2743 2012-05-02  Jon Honeycutt  <jhoneycutt@apple.com>
2744
2745         WebFrameLoaderClient::dispatchWillSendSubmitEvent() needs to be
2746         implemented for WebKit2
2747         https://bugs.webkit.org/show_bug.cgi?id=84304
2748
2749         Reviewed by Jessie Berlin.
2750
2751         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2752         Added new files to project.
2753
2754         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent.cpp: Added.
2755         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
2756         Look for the "DidReceiveWillSendSubmitEvent" message. The message body
2757         will be the dictionary of form control names and values passed to the
2758         injected bundle's willSendSubmitEvent handler; look for the pairs
2759         "textField" -> "text field" and "passwordField" -> "password field".
2760         Check that "hiddenField" is not included.
2761         (TestWebKitAPI::setInjectedBundleClient):
2762         (TestWebKitAPI::TEST):
2763
2764         * TestWebKitAPI/Tests/WebKit2/WillSendSubmitEvent_Bundle.cpp: Added.
2765         (TestWebKitAPI::willSendSubmitEvent):
2766         Post a message to the TestWebKitAPI process with the dictionary of form
2767         control names and values that we received.
2768         (TestWebKitAPI::WillSendSubmitEventTest::WillSendSubmitEventTest):
2769         (TestWebKitAPI::WillSendSubmitEventTest::didCreatePage):
2770         Set up the page's form client.
2771
2772         * TestWebKitAPI/Tests/WebKit2/auto-submitting-form.html: Added.
2773
2774 2012-05-03  Eric Seidel  <eric@webkit.org>
2775
2776         Remove dead code from build-webkit for working around old XCode dependency bug
2777         https://bugs.webkit.org/show_bug.cgi?id=85535
2778
2779         Reviewed by Adam Barth.
2780
2781         Xcode used to have a bug whereby it didn't notice if you changed the #defines
2782         passed on the command line to xcodebuild.  I wrote code to work around this
2783         while bringing up SVG support 5-6 years ago.  I think we can finally remove this
2784         code from build-webkit, since SVG is the only feature flag to ever use it
2785         (modern in-progress features like MathML don't seem to be using the support)
2786         an I suspect the XCode bug has long been fixed.
2787
2788         * Scripts/build-webkit:
2789         * Scripts/webkitperl/features.pm:
2790
2791 2012-05-03  Andy Estes  <aestes@apple.com>
2792
2793         InjectedBundleControllerMac.mm needs to include config.h
2794         https://bugs.webkit.org/show_bug.cgi?id=85531
2795
2796         Reviewed by Andreas Kling.
2797
2798         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm:
2799
2800 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2801
2802         webkitpy: Use PlatformInfo whenever possible in server_process.py.
2803         https://bugs.webkit.org/show_bug.cgi?id=85516
2804
2805         Reviewed by Eric Seidel.
2806
2807         Eliminate direct access to sys.platform wherever possible by using
2808         PlatformInfo.is_win() instead.
2809
2810         * Scripts/webkitpy/layout_tests/port/server_process.py:
2811         (ServerProcess.__init__): Remove the executive parameter as it was
2812         not passed by any caller, obtain a SystemHost from port_obj
2813         instead.
2814         (ServerProcess._log): Style, remove extra empty line.
2815         (ServerProcess._start):
2816         (ServerProcess.stop):
2817         (ServerProcess.kill):
2818         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2819         (TrivialMockPort.__init__): Create a MockSystemHost used by
2820         ServerProcess.
2821         (TestServerProcess.test_broken_pipe): Test different platforms.
2822
2823 2012-05-03  Thiago Marcos P. Santos  <thiago.santos@intel.com>
2824
2825         webkitpy: Style checker broken when committing a PNG
2826         https://bugs.webkit.org/show_bug.cgi?id=85504
2827
2828         Reviewed by Dirk Pranke.
2829
2830         * Scripts/webkitpy/style/checkers/png.py:
2831         (PNGChecker.__init__):
2832
2833 2012-05-03  Philippe Normand  <pnormand@igalia.com>
2834
2835         Unreviewed, fix another call to join(), similar to the ones landed
2836         in r115975.
2837
2838         * Scripts/webkitdirs.pm:
2839         (jhbuildConfigurationChanged):
2840
2841 2012-05-03  Raphael Kubo da Costa  <rakuco@webkit.org>
2842
2843         webkitpy: Remove sys.platform accesses in CrashLogs.
2844         https://bugs.webkit.org/show_bug.cgi?id=85436
2845
2846         Reviewed by Dirk Pranke.
2847
2848         Instead of directly checking for sys.platform == 'darwin', use a
2849         SystemHost and call PlatformInfo.is_mac().
2850
2851         * Scripts/webkitpy/common/system/crashlogs.py:
2852         (CrashLogs.__init__):
2853         (CrashLogs.find_newest_log):
2854         (CrashLogs._log_directory_darwin):
2855         (CrashLogs._find_newest_log_darwin):
2856         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2857         (CrashLogsTest.test_find_log_darwin):
2858         * Scripts/webkitpy/layout_tests/port/mac.py:
2859         (MacPort._get_crash_log):
2860         * Scripts/webkitpy/layout_tests/port/test.py:
2861         (TestDriver.run_test):
2862         * Scripts/webkitpy/tool/commands/queries.py:
2863         (execute):
2864
2865 2012-05-03  Philippe Normand  <pnormand@igalia.com>
2866
2867         [GTK] join() is wrongly used in webkitdirs.pm since r115532
2868         https://bugs.webkit.org/show_bug.cgi?id=85501
2869
2870         Reviewed by Martin Robinson.
2871
2872         * Scripts/webkitdirs.pm:
2873         (getJhbuildPath):
2874         (buildAutotoolsProject):
2875
2876 2012-05-03  Zan Dobersek  <zandobersek@gmail.com>
2877
2878         [Gtk] ImageDiff crashes when it can't open a display
2879         https://bugs.webkit.org/show_bug.cgi?id=85476
2880
2881         Reviewed by Martin Robinson.
2882
2883         Replace gdk_init() initialization call with the g_type_init()
2884         call. GdkPixbufLoader doesn't actually need the current call to be
2885         made and this change will also remove crashes when ImageDiff
2886         is started in an environment without a DISPLAY env.
2887
2888         * DumpRenderTree/gtk/ImageDiff.cpp:
2889         (main):
2890
2891 2012-05-03  Zan Dobersek  <zandobersek@gmail.com>
2892
2893         ImageDiff should be run inside a properly established environment
2894         https://bugs.webkit.org/show_bug.cgi?id=85292
2895
2896         Reviewed by Martin Robinson.
2897
2898         When creating the ImageDiff server process, pass along a freshly-set-up environment
2899         in which the process should be executed in.
2900
2901         * Scripts/webkitpy/layout_tests/port/webkit.py:
2902         (WebKitPort._start_image_diff_process):
2903
2904 2012-05-03  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2905
2906         [Qt] Allow the web process and WTR to be paused on startup
2907
2908         Makes it easier to debug the web process or run-webkit-tests -2, as you
2909         have ample time to attach gdb to the process.
2910
2911         Reviewed by Simon Hausmann.
2912
2913         * Scripts/webkitpy/layout_tests/port/qt.py:
2914         (QtPort.setup_environ_for_server):
2915         * WebKitTestRunner/qt/main.cpp:
2916         (main):
2917
2918 2012-05-03  Csaba Osztrogonác  <ossy@webkit.org>
2919
2920         Password protect "Stop build" button
2921         https://bugs.webkit.org/show_bug.cgi?id=81982
2922
2923         Reviewed by Ryosuke Niwa.
2924
2925         Only authenticated users should be able to use stopBuild, stopAllBuilds, cancelPendingBuild
2926         features to prevent killing builds on build.webkit.org by SPAM bots.
2927
2928         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2929
2930 2012-05-03  Dongwoo Im  <dw.im@samsung.com>
2931
2932         [DRT] Remove the name of parameter, visibility, from the setPageVisibility stub implementations.
2933         https://bugs.webkit.org/show_bug.cgi?id=85468
2934
2935         Reviewed by Nikolas Zimmermann.
2936
2937         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2938         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2939         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2940         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2941         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2942         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2943         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2944         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2945         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2946         (LayoutTestController::setPageVisibility): Remove the name of parameter, visibility.
2947
2948 2012-05-03  Stephanie Lewis  <slewis@apple.com>
2949
2950         https://bugs.webkit.org/show_bug.cgi?id=85450 unbounded growth of JSDOMWindowShells loading pages in the same window
2951         <rdar://problem/11320059> REGRESSION (r115083): PLT3 shows linear memory growth and gets slower with each run
2952
2953         Reviewed by Brady Eidson.
2954
2955         The API added for DOMWindowExtension, didCreateGlobalObjectForFrame, would create a global object
2956         for every world, even those that did not need the callback.  This had the side effect of creating a
2957         JSDOMWindowShell that the associated world didn't necessarily know to clean up.  Instead of creating
2958         unnecessary objects change the API to globalObjectIsAvailableForFrame and do not pass the global object
2959         in the API.  The object can be accessed later by those worlds which require it.
2960
2961         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic.cpp:
2962         (TestWebKitAPI):
2963         (TestWebKitAPI::didReceiveMessageFromInjectedBundle):
2964         * TestWebKitAPI/Tests/WebKit2/DOMWindowExtensionBasic_Bundle.cpp:
2965         (TestWebKitAPI):
2966         (DOMWindowExtensionBasic):
2967         (TestWebKitAPI::DOMWindowExtensionBasic::didCreatePage):
2968         (TestWebKitAPI::DOMWindowExtensionBasic::globalObjectIsAvailableForFrame):
2969         (TestWebKitAPI::globalObjectIsAvailableForFrameCallback):
2970
2971 2012-05-03  Nikolas Zimmermann  <nzimmermann@rim.com>
2972
2973         Not reviewed. Fix mac build, it was missing setPageVisibility/resetPageVisibility stub implementations.
2974
2975         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2976         (LayoutTestController::resetPageVisibility):
2977         (LayoutTestController::setPageVisibility):
2978
2979 2012-05-02  Alexander Færøy  <ahf@0x90.dk>
2980
2981         Rename deviceDPI to devicePixelRatio
2982         https://bugs.webkit.org/show_bug.cgi?id=85049
2983
2984         Reviewed by Kenneth Rohde Christiansen.
2985
2986         Use 1.5 as device pixel ratio for the Qt MiniBrowser.
2987
2988         * MiniBrowser/qt/qml/BrowserWindow.qml:
2989
2990 2012-05-03  Dongwoo Im  <dw.im@samsung.com>
2991
2992         [EFL][DRT] Implement the LayoutTestController's methods related to the Page Visibility API.
2993         https://bugs.webkit.org/show_bug.cgi?id=85347
2994
2995         Reviewed by Nikolas Zimmermann.
2996
2997         * DumpRenderTree/LayoutTestController.h:
2998         (LayoutTestController):
2999         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
3000         (LayoutTestController::resetPageVisibility): Not implemented.
3001         (LayoutTestController::setPageVisibility): Not implemented.
3002         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3003         (LayoutTestController::resetPageVisibility): Reset the page visibility.
3004         (LayoutTestController::setPageVisibility): Set the page visibility.
3005         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3006         (LayoutTestController::resetPageVisibility): Not implemented.
3007         (LayoutTestController::setPageVisibility): Not implemented.
3008         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3009         (LayoutTestController::resetPageVisibility): Not implemented.
3010         (LayoutTestController::setPageVisibility): Not implemented.
3011         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3012         (LayoutTestController::resetPageVisibility): Not implemented.
3013         (LayoutTestController::setPageVisibility): Not implemented.
3014         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
3015         (LayoutTestController::resetPageVisibility): Not implemented.
3016         (LayoutTestController::setPageVisibility): Not implemented.
3017
3018 2012-05-02  Sheriff Bot  <webkit.review.bot@gmail.com>
3019
3020         Unreviewed, rolling out r115918.
3021         http://trac.webkit.org/changeset/115918
3022         https://bugs.webkit.org/show_bug.cgi?id=85452
3023
3024         "Broke test-webkitpy on Apple Lion bots" (Requested by estes
3025         on #webkit).
3026
3027         * Scripts/webkitpy/common/system/crashlogs.py:
3028         (CrashLogs.__init__):
3029         (CrashLogs.find_newest_log):
3030         (CrashLogs._log_directory_darwin):
3031         (CrashLogs._find_newest_log_darwin):
3032         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
3033         (CrashLogsTest.test_find_log_darwin):
3034         * Scripts/webkitpy/layout_tests/port/mac.py:
3035         (MacPort._get_crash_log):
3036         * Scripts/webkitpy/layout_tests/port/test.py:
3037         (TestDriver.run_test):
3038         * Scripts/webkitpy/tool/commands/queries.py:
3039         (execute):
3040
3041 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3042
3043         webkitpy: Remove sys.platform accesses in CrashLogs.
3044         https://bugs.webkit.org/show_bug.cgi?id=85436
3045
3046         Reviewed by Dirk Pranke.
3047
3048         Instead of directly checking for sys.platform == 'darwin', use a
3049         SystemHost and call PlatformInfo.is_mac().
3050
3051         * Scripts/webkitpy/common/system/crashlogs.py:
3052         (CrashLogs.__init__):
3053         (CrashLogs.find_newest_log):
3054         (CrashLogs._log_directory_darwin):
3055         (CrashLogs._find_newest_log_darwin):
3056         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
3057         (CrashLogsTest.test_find_log_darwin):
3058         * Scripts/webkitpy/layout_tests/port/mac.py:
3059         (MacPort._get_crash_log):
3060         * Scripts/webkitpy/layout_tests/port/test.py:
3061         (TestDriver.run_test):
3062         * Scripts/webkitpy/tool/commands/queries.py:
3063         (execute):
3064
3065 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3066
3067         webkitpy: Special-case win32 calls in file_lock and executive.
3068         https://bugs.webkit.org/show_bug.cgi?id=85431
3069
3070         Reviewed by Dirk Pranke.
3071
3072         Simplify accesses to sys.platform by only checking if sys.platform
3073         is "win32" and removing explicit checks for the else case.
3074
3075         * Scripts/webkitpy/common/system/executive.py:
3076         (Executive.check_running_pid):
3077         * Scripts/webkitpy/common/system/file_lock.py:
3078         (FileLock._create_lock):
3079         (FileLock._remove_lock):
3080
3081 2012-05-02  Ojan Vafai  <ojan@chromium.org>
3082
3083         Remove pink from flakiness dashboard
3084         https://bugs.webkit.org/show_bug.cgi?id=85422
3085
3086         Reviewed by Dirk Pranke.
3087
3088         The pink indicates that the test doesn't match it's listing in test_expectations.txt.
3089         In practice, this just added noise and confused people.
3090
3091         * TestResultServer/static-dashboards/flakiness_dashboard.js:
3092         (htmlForSingleTestRow):
3093         (showLegend):
3094         * TestResultServer/static-dashboards/flakiness_dashboard_tests.css:
3095         (#legend-contents *):
3096         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
3097         (testHtmlForSingleTestRow):
3098
3099 2012-05-02  Gustavo Noronha Silva  <gns@gnome.org>
3100
3101         Unreviewed. Fix mistake I committed when applying review comments.
3102
3103         * jhbuild/jhbuild-wrapper: dependencies_path was getting 'Root'
3104         appended to its path, so jhbuild itself was being cloned and
3105         installed in the wrong place when using WEBKITOUTPUTDIR.
3106
3107 2012-05-02  Dirk Pranke  <dpranke@chromium.org>
3108
3109         nrwt: make ServerProcess work on chromium win
3110         https://bugs.webkit.org/show_bug.cgi?id=85333
3111
3112         Reviewed by Ojan Vafai.
3113
3114         This change implements a poor man's select() that will
3115         slow-spin doing non-blocking reads on the stdout and stderr
3116         named pipes connecting the worker to the driver. Seems to work
3117         and I have yet to see much overhead or ill effects but it
3118         probably needs more testing.
3119
3120         * Scripts/webkitpy/layout_tests/port/chromium.py:
3121         (ChromiumDriver.__init__):
3122         * Scripts/webkitpy/layout_tests/port/server_process.py:
3123         (ServerProcess._start):
3124         (ServerProcess._wait_for_data_and_update_buffers_using_select):
3125         (ServerProcess._wait_for_data_and_update_buffers_using_win32_apis):
3126         (ServerProcess):
3127         (ServerProcess._non_blocking_read_win32):
3128         (ServerProcess._read):
3129         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
3130         (TestServerProcess.test_basic): Added.
3131
3132 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3133
3134         webkitpy: Use a SystemHost instead of PlatformInfo in the PNG checker.
3135         https://bugs.webkit.org/show_bug.cgi?id=85426
3136
3137         Reviewed by Dirk Pranke.
3138
3139         r115880 replaced the direct use of sys.platform with a
3140         PlatformInfo object. Dirk Pranke later told me SystemHost objects
3141         should be preferred, as nothing out of webkitpy.common.system
3142         should create its own PlatformInfo instances.
3143
3144         * Scripts/webkitpy/common/system/systemhost_mock.py:
3145         (MockSystemHost.__init__): Make it possible to specify a custom
3146         filesystem object.
3147         * Scripts/webkitpy/style/checkers/png.py:
3148         (PNGChecker.__init__): Accept a host object instead of a
3149         filesystem one, which is now obtained from host.
3150         (PNGChecker._config_file_path): Retrieve PlatformInfo from
3151         SystemHost.
3152         * Scripts/webkitpy/style/checkers/png_unittest.py:
3153         (PNGCheckerTest.test_init): Adjust calls to PNGChecker to its new
3154         constructor signature.
3155         (PNGCheckerTest.test_check): Ditto.
3156
3157 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3158
3159         webkitpy: Remove direct checks for sys.platform == 'darwin' in the layout test classes.
3160         https://bugs.webkit.org/show_bug.cgi?id=85424
3161
3162         Reviewed by Tony Chang.
3163
3164         We should reduce direct usage of sys.platform in webkitpy and use
3165         PlatformInfo instead.
3166
3167         * Scripts/webkitpy/layout_tests/servers/http_server.py:
3168         (Lighttpd._prepare_config): Call PlatformInfo.is_mac() instead of
3169         checking for 'darwin' directly.
3170         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
3171         (HttpServerBase.__init__): Ditto.
3172
3173 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3174
3175         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
3176         https://bugs.webkit.org/show_bug.cgi?id=85419
3177
3178         Reviewed by Dirk Pranke.
3179
3180         As pointed out by Dirk Pranke, a PlatformInfo is not supposed to
3181         be created directly, we should should a SystemHost instead.
3182
3183         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3184         (MainTest.setUp): Create a SystemHost and obtain the PlatformInfo
3185         object from it instead of creating PlatformInfo directly.
3186
3187 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3188
3189         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
3190         https://bugs.webkit.org/show_bug.cgi?id=85419
3191
3192         Rubber-stamped by Dirk Pranke.
3193
3194         Follow-up patch to r115887 which fixes some capitalization problems
3195         pointed out in comments 2 and 4.
3196
3197         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3198         (MainTest.setUp):
3199         (MainTest.test_child_processes_2):
3200         (MainTest.test_child_processes_min):
3201         (MainTest.test_exception_raised):
3202         (MainTest.test_keyboard_interrupt):
3203
3204 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3205
3206         webkitpy: Remove direct sys.platform usage in run_webkit_tests_integrationtest.
3207         https://bugs.webkit.org/show_bug.cgi?id=85419
3208
3209         Reviewed by Dirk Pranke.
3210
3211         Another part of the work towards replacing direct sys.platform
3212         usage with PlatformInfo.
3213
3214         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3215         (MainTest.setUp): Move the setting of SHOULD_TEST_PROCESSES
3216         here. Even though it is costlier, unittest.setUpClass was
3217         introduced in Python 2.7.
3218         (MainTest.test_child_processes_2): Use self.SHOULD_TEST_PROCESSES.
3219         (MainTest.test_child_processes_min): Ditto.
3220         (MainTest.test_exception_raised): Ditto.
3221         (MainTest.test_keyboard_interrupt): Ditto.
3222         (MainTest.test_crash_log): Check for PlatformInfo.is_mac().
3223         (MainTest.test_web_process_crash_log): Ditto, and remove duplicate
3224         check at the end.
3225         (MainTest.test_no_http_tests): Remove check for (cygwin && Python
3226         2.5.2), as we only support Python 2.6+.
3227
3228 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3229
3230         webkitpy: Remove direct sys.platform usage in the PNG checker.
3231         https://bugs.webkit.org/show_bug.cgi?id=85407
3232
3233         Reviewed by Tony Chang.
3234
3235         Work towards replacing direct sys.platform usage with
3236         PlatformInfo.
3237
3238         * Scripts/webkitpy/common/system/platforminfo_mock.py:
3239         (MockPlatformInfo.__init__): Accept os_name and os_version as
3240         arguments.
3241         * Scripts/webkitpy/style/checkers/png.py:
3242         (PNGChecker.__init__): Create a PlatformInfo object instead of
3243         using sys.platform when needed.
3244         (PNGChecker._config_file_path): Check for the current platform via
3245         PlatformInfo.
3246         * Scripts/webkitpy/style/checkers/png_unittest.py:
3247         (PNGCheckerTest.test_check): Update tests after the changes above.
3248
3249 2012-05-02  Raphael Kubo da Costa  <rakuco@webkit.org>
3250
3251         [watchlist] Update watchlist rules for EFL-related patches.
3252         http://bugs.webkit.org/show_bug.cgi?id=85160
3253
3254         Reviewed by Eric Seidel.
3255
3256         * Scripts/webkitpy/common/config/watchlist: Monitor
3257         Source/WTF/wtf/efl and LayoutTests/platform/efl as well.
3258
3259 2012-05-02  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3260
3261         Revert r115191 - "Make the web view's url property follow the active url"
3262
3263         It uncovered/caused issues in the icon-implementation that can't be
3264         easily fixed, so rolling out instead.
3265
3266         https://bugs.webkit.org/show_bug.cgi?id=77554
3267
3268         Rubber-stamped by Simon Hausmann.
3269
3270         * MiniBrowser/qt/qml/BrowserWindow.qml:
3271         * TestWebKitAPI/Tests/WebKit2/PageLoadBasic.cpp:
3272         (TestWebKitAPI::TEST):
3273
3274 2012-05-02  Zan Dobersek  <zandobersek@gmail.com>
3275
3276         Reftest mismatches are providing undefined diff percent in results.html
3277         https://bugs.webkit.org/show_bug.cgi?id=85283
3278
3279         Reviewed by Ojan Vafai.
3280
3281         When writing test results for a reftest mismatch, also take into account the
3282         difference percent that is produced when generating image diff. This value is
3283         then used when creating test results in the form of dictionaries, stored under
3284         the 'image_diff_percent' key. It ultimately ends up in JSON results data,
3285         providing additional information about the reftest difference.
3286
3287         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3288         (interpret_test_failures): Store the diff percent for reftest failure under the
3289         'image_diff_percent' key.
3290         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
3291         (write_test_result): Also set the new diff_percent attribute of the reftest
3292         mismatch failure to the value of the difference percent ImageDiff outputs.
3293         * Scripts/webkitpy/layout_tests/models/test_failures.py:
3294         (FailureReftestMismatch.__init__): Add the new diff_percent attribute with the
3295         default value of None.
3296         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3297         (EndToEndTest.test_reftest_with_two_notrefs): Update the test case expected results.
3298
3299 2012-05-02  Terry Anderson  <tdanderson@chromium.org>
3300
3301         Addition of a contributor to committers.py
3302         https://bugs.webkit.org/show_bug.cgi?id=85324
3303
3304         Reviewed by Unreviewed.
3305
3306         * Scripts/webkitpy/common/config/committers.py:
3307
3308 2012-05-02  Mikhail Pozdnyakov  <mikhail.pozdnyakov@intel.com>
3309
3310         [EFL] EFL's LayoutTestController overridePreference implementation
3311         https://bugs.webkit.org/show_bug.cgi?id=83007
3312
3313         Reviewed by Antonio Gomes.
3314
3315         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3316         (DumpRenderTreeChrome::resetDefaultsToConsistentValues):
3317         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3318         (toBool): aux function converting JSStringRef to bool
3319         (toInt): aux function converting JSStringRef to int
3320         (LayoutTestController::overridePreference): added implementation
3321
3322 2012-05-01  Raphael Kubo da Costa  <rakuco@webkit.org>
3323
3324         webkitpy: Fix sys.platform check in find_files_unittest.
3325         https://bugs.webkit.org/show_bug.cgi?id=85339
3326
3327         Reviewed by Ryosuke Niwa.
3328
3329         Follow-up to r101618 -- 'win' is not a valid sys.platform value,
3330         'win32' should be used instead. The check always returned false
3331         due to this, so assert_normalize() was never called.
3332
3333         * Scripts/webkitpy/common/find_files_unittest.py:
3334         (TestWinNormalize.test_win): Add the missing import for the
3335         FileSystem class and fix the sys.platform check.
3336
3337 2012-05-02  Zan Dobersek  <zandobersek@gmail.com>
3338
3339         [GTK] drag&drop related failures
3340         https://bugs.webkit.org/show_bug.cgi?id=84675
3341
3342         Reviewed by Philippe Normand.
3343
3344         Rather than checking options to see whether we're running pixel tests
3345         the pixel_tests argument passed to GtkDriver._start should be used as
3346         it also covers cases when pixel tests should be run because of testing
3347         reftests. This again creates different Xvfb processes with proper
3348         display IDs for a worker that's performing both ordinary and pixel
3349         tests, as was the case before changes made in r114788.
3350
3351         * Scripts/webkitpy/layout_tests/port/gtk.py:
3352         (GtkDriver._start):
3353
3354 2012-05-02  Simon Hausmann  <simon.hausmann@nokia.com>
3355
3356         [Qt] Enable device orientation/motion with Qt 5
3357
3358         Reviewed by Tor Arne Vestbø.
3359
3360         Enable device orientation/motion features with Qt 5 if Qt Sensors is available.
3361
3362         * qmake/mkspecs/features/features.prf:
3363
3364 2012-05-01  Raphael Kubo da Costa  <rakuco@webkit.org>
3365
3366         webkitpy: Remove unused platform imports.
3367         https://bugs.webkit.org/show_bug.cgi?id=85341
3368
3369         Reviewed by Eric Seidel.
3370
3371         Remove a few uses of `import platform' which were either rendered
3372         unneeded after r105931 or were never needed at all.
3373
3374         * Scripts/webkitpy/common/system/executive.py:
3375         * Scripts/webkitpy/tool/bot/flakytestreporter.py:
3376
3377 2012-05-01  Terry Anderson  <tdanderson@chromium.org>
3378
3379         [chromium] Accept four parameters when dispatching a WebInputEvent::GestureTap in chromium DRT eventSender
3380         https://bugs.webkit.org/show_bug.cgi?id=85289
3381
3382         Reviewed by Eric Seidel.
3383
3384         * DumpRenderTree/chromium/EventSender.cpp:
3385         (EventSender::gestureEvent):
3386             This change will allow for testing of WK85101. Currently
3387             no tests use the extra parameters deltaX and deltaY. Added
3388             a check for the size of |arguments|.
3389
3390 2012-05-01  Eric Seidel  <eric@webkit.org>
3391
3392         Add myself to the watchlist to watch all rendering changes, as well as sort the watchers in the list
3393         https://bugs.webkit.org/show_bug.cgi?id=85305
3394
3395         Reviewed by Adam Barth.
3396
3397         * Scripts/webkitpy/common/config/watchlist:
3398
3399 2012-05-01  Sheriff Bot  <webkit.review.bot@gmail.com>
3400
3401         Unreviewed, rolling out r115735.
3402         http://trac.webkit.org/changeset/115735
3403         https://bugs.webkit.org/show_bug.cgi?id=85314
3404
3405         cause two tests to crash due to an ASSERTION failure
3406         (Requested by zhenyao on #webkit).
3407
3408         * DumpRenderTree/chromium/EventSender.cpp:
3409         (EventSender::gestureEvent):
3410
3411 2012-05-01  Terry Anderson  <tdanderson@chromium.org>
3412
3413         [chromium] Accept four parameters when dispatching a WebInputEvent::GestureTap in chromium DRT eventSender
3414         https://bugs.webkit.org/show_bug.cgi?id=85289
3415
3416         Reviewed by Eric Seidel.
3417
3418         * DumpRenderTree/chromium/EventSender.cpp:
3419         (EventSender::gestureEvent):
3420             This change will allow for testing of WK85101. Currently no tests use the extra
3421             parameters deltaX and deltaY.
3422
3423 2012-05-01  Zan Dobersek  <zandobersek@gmail.com>
3424
3425         Print out stderr output of ImageDiff if it is present
3426         https://bugs.webkit.org/show_bug.cgi?id=85285
3427
3428         Reviewed by Philippe Normand.
3429
3430         Log the stderr output of ImageDiff as a warning if present. This is
3431         especially helpful when ImageDiff crashes and stderr data might contain
3432         relevant information about what caused the crash.
3433
3434         * Scripts/webkitpy/layout_tests/port/webkit.py:
3435         (WebKitPort._read_image_diff):
3436
3437 2012-05-01  Kenneth Rohde Christiansen  <kenneth@webkit.org>
3438
3439         [Qt] Add an experimental extension to set the min. contents width
3440         https://bugs.webkit.org/show_bug.cgi?id=85281
3441
3442         Reviewed by Antonio Gomes.
3443
3444         * MiniBrowser/qt/qml/BrowserWindow.qml:
3445
3446         Set the value of preferredMinimumContentsWidth to 980.
3447
3448 2012-04-30  Ojan Vafai  <ojan@chromium.org>
3449
3450         Fix tests to not depend on the actual list of builders at build.chromium.org.
3451
3452         * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
3453         (testHtmlForIndividualTestOnAllBuildersWithChrome):
3454         (testHtmlForIndividualTestOnAllBuildersWithChromeWebkitMaster):
3455
3456 2012-04-29  Sam Weinig  <sam@webkit.org>
3457
3458         Remove BlobBuilder
3459         https://bugs.webkit.org/show_bug.cgi?id=84036
3460
3461         Reviewed by Anders Carlsson.
3462
3463         * Scripts/build-webkit:
3464         * qmake/mkspecs/features/features.prf:
3465         Make exposing the WebKitBlobBuilder JS object conditional on a new
3466         ENABLE_LEGACY_WEBKIT_BLOB_BUILDER flag. Don't enable this for the Mac,
3467         but do for everyone else.
3468
3469 2012-04-30  Amruth Raj Padmanabhuni  <amruthraj@motorola.com>
3470
3471         [GTK] Add glib as a dependency to build glib-networking
3472         https://bugs.webkit.org/show_bug.cgi?id=85181
3473
3474         Reviewed by Martin Robinson.
3475
3476         * gtk/jhbuild.modules:
3477
3478 2012-04-30  Alexis Menard  <alexis.menard@openbossa.org>
3479
3480         Move myself to reviewers.
3481
3482         Reviewed by Andreas Kling.
3483
3484         * Scripts/webkitpy/common/config/committers.py:
3485
3486 2012-04-30  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>
3487
3488         Unreviewed, move myself to reviewers
3489
3490         * Scripts/webkitpy/common/config/committers.py:
3491
3492 2012-04-30  Carlos Garcia Campos  <cgarcia@igalia.com>
3493
3494         Unreviewed. Fix make distcheck.
3495
3496         * TestWebKitAPI/GNUmakefile.am: Add missing headers to compilation
3497         and resources to EXTRA_DIST.
3498
3499 2012-04-30  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3500
3501         [GTK] DRT support for layoutTestController.removeOriginAccessWhitelistEntry
3502         https://bugs.webkit.org/show_bug.cgi?id=85156
3503
3504         Reviewed by Martin Robinson.
3505
3506         Add missing implementation removeOriginAccessWhitelistEntry to
3507         GTK's LayoutTestController.
3508
3509         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3510         (LayoutTestController::removeOriginAccessWhitelistEntry):
3511
3512 2012-04-29  Maciej Stachowiak  <mjs@apple.com>
3513
3514         REGRESSION: On Lion, run-webkit-tests changes the display color profile even when not running pixel tests
3515         https://bugs.webkit.org/show_bug.cgi?id=81729
3516
3517         Reviewed by Dan Bernstein.
3518
3519         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3520         (Manager._set_up_run): Only launch LayoutTestHelper when running pixel tests.
3521
3522 2012-04-29  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3523
3524         [GTK] DRT needs an implementation of layoutTestController.setDomainRelaxationForbiddenForURLScheme
3525         https://bugs.webkit.org/show_bug.cgi?id=85131
3526
3527         Reviewed by Martin Robinson.
3528
3529         Add missing implementation setDomainRelaxationForbiddenForURLScheme to
3530         GTK's LayoutTestController.
3531
3532         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3533         (LayoutTestController::setDomainRelaxationForbiddenForURLScheme):
3534
3535 2012-04-28  Sam Weinig  <sam@webkit.org>
3536
3537         Smooth scrolling needs a new key
3538         <rdar://problem/11331632>
3539
3540         Reviewed by Dan Bernstein.
3541
3542         * DumpRenderTree/mac/DumpRenderTree.mm:
3543         (resetDefaultsToConsistentValues):
3544         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm:
3545         (TestWebKitAPI::InjectedBundleController::platformInitialize):
3546         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
3547         (WTR::InjectedBundle::platformInitialize):
3548         Update for new key.
3549
3550 2012-04-28  Zan Dobersek  <zandobersek@gmail.com>
3551
3552         [WK2] http/tests/navigation/anchor-frames-gbk.html fails
3553         https://bugs.webkit.org/show_bug.cgi?id=76896
3554
3555         Reviewed by Darin Adler.
3556
3557         Only dump as text if currently dumping render tree. This ensures that
3558         calling dumpAsText in a test with dumpChildFramesAsText already being
3559         called doesn't override the first decision, possibly making tests fail.
3560
3561         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3562         (WTR::LayoutTestController::dumpAsText):
3563
3564 2012-04-28  Dan Winship  <danw@gnome.org>
3565
3566         Unreviewed; Add me to the SoupNetwork watchlist
3567
3568         * Scripts/webkitpy/common/config/watchlist:
3569
3570 2012-04-27  Gustavo Noronha Silva  <gns@gnome.org>
3571
3572         [GTK] jhbuild cloning is not following WEBKITOUTPUTDIR.
3573         https://bugs.webkit.org/show_bug.cgi?id=76161
3574
3575         Reviewed by Martin Robinson.
3576
3577         * Scripts/webkitdirs.pm:
3578         (getJhbuildPath): New method to obtain the jhbuild base directory,
3579         using the product base directory
3580         (jhbuildConfigurationChanged): Use the new method
3581         (buildAutotoolsProject): Ditto.
3582         * efl/jhbuildrc: Use WEBKITOUTPUTDIR when calculating the path.
3583         * gtk/jhbuildrc: Ditto.
3584         * jhbuild/jhbuild-wrapper: Ditto.
3585
3586 2012-04-27  Scott Graham  <scottmg@chromium.org>
3587
3588         [chromium] make case of npTestNetscapePlugIn.dll match case in copy rule
3589         https://bugs.webkit.org/show_bug.cgi?id=85083
3590
3591         Reviewed by Dirk Pranke.
3592
3593         Make output product_name for npTestNetscapePlugIn.dll match the case
3594         of the copy_TestNetscapePlugIn rule. This is required for ninja, which
3595         is more particular about case matching in rules than the VS build.
3596
3597         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
3598
3599 2012-04-27  Dirk Pranke  <dpranke@chromium.org>
3600
3601         [chromium] use "drt-style" output, not "test-shell-style" output, on mac and linux DRT
3602         https://bugs.webkit.org/show_bug.cgi?id=84917
3603
3604         Unreviewed, build fix.
3605
3606         Reland the change in r115453 with a fix for chromium win.
3607         The logic in ChromiumDriver was busted on windows, causing DRT
3608         to get launched without the --test-shell flag.
3609
3610         * Scripts/webkitpy/layout_tests/port/chromium.py:
3611         (ChromiumDriver):
3612         (ChromiumDriver.__init__):
3613         (ChromiumDriver._wrapper_options):
3614         (ChromiumDriver.cmd_line):
3615         (ChromiumDriver._start):
3616         (ChromiumDriver.has_crashed):
3617         (ChromiumDriver.run_test):
3618         (ChromiumDriver.stop):
3619         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3620         (ChromiumDriverTest.setUp):
3621         (ChromiumDriverTest.test_stop):
3622         (ChromiumDriverTest.test_two_drivers.MockDriver.__init__):
3623         (ChromiumDriverTest.test_two_drivers):
3624
3625 2012-04-27  Peter Beverloo  <peter@chromium.org>
3626
3627         Add John Grabowski as a non-committer to committers.py
3628         https://bugs.webkit.org/show_bug.cgi?id=85080
3629
3630         Reviewed by Dirk Pranke.
3631
3632         He's being CC'ed often enough because of his involvement in the Chromium
3633         side of Chrome for Android upstreaming.
3634
3635         * Scripts/webkitpy/common/config/committers.py:
3636           add jrg@chromium.org
3637
3638 2012-04-27  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
3639
3640         [Qt] Fix minimal build.
3641         https://bugs.webkit.org/show_bug.cgi?id=85045
3642
3643         Reviewed by Tor Arne Vestbø.
3644
3645         Do not set conflicting values of ENABLE_XSLT.
3646
3647         * qmake/mkspecs/features/features.prf:
3648
3649 2012-04-27  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3650
3651         [EFL][DRT] EFL should sanitize pulseaudio module configuration too
3652         https://bugs.webkit.org/show_bug.cgi?id=85047
3653
3654         Reviewed by Dirk Pranke.
3655
3656         PulseA