Make Chromium port for Android use hardware gpu path default.
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-28  Hao Zheng  <zhenghao@chromium.org>
2
3         Make Chromium port for Android use hardware gpu path default.
4         https://bugs.webkit.org/show_bug.cgi?id=82033
5
6         Reviewed by Dirk Pranke.
7
8         Skip canvas tests, which are run in virtual gpu test suites.
9
10         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
11         (ChromiumAndroidPort.__init__):
12         (ChromiumAndroidPort.skipped_tests):
13
14 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
15
16         kill-old-processes doesn't do anything on linux
17         https://bugs.webkit.org/show_bug.cgi?id=82388
18
19         Disable kill-old-processes on the Qt bots not to kill each others.
20
21         Reviewed by Ryosuke Niwa.
22
23         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
24         (Factory.__init__):
25
26 2012-03-27  Hao Zheng  <zhenghao@chromium.org>
27
28         Handle DRT crash caused by Android OOM.
29         https://bugs.webkit.org/show_bug.cgi?id=82310
30
31         Reviewed by Tony Chang.
32
33         When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
34         is stopped silently and regarded as crashed. Re-run the test for
35         such crash.
36
37         Add some debug log to NRWT.
38
39         * Scripts/webkitpy/layout_tests/port/chromium.py:
40         (ChromiumDriver.run_test):
41         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
42         (ChromiumAndroidPort.__init__):
43         (ChromiumAndroidPort.get_last_stacktrace):
44         (ChromiumAndroidDriver.__init__):
45         (ChromiumAndroidDriver._start):
46         (ChromiumAndroidDriver.run_test):
47         (ChromiumAndroidDriver._get_drt_return_value):
48
49 2012-03-27  Ojan Vafai  <ojan@chromium.org>
50
51         Fix expected results for some unittest failures.
52
53         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
54
55 2012-03-27  Adrienne Walker  <enne@google.com>
56
57         [chromium] Fix --enabled-threaded-compositing flag in DRT
58         https://bugs.webkit.org/show_bug.cgi?id=82405
59
60         Reviewed by James Robinson.
61
62         --enable-threaded-compositing wasn't working because there was no way
63         to set the value before it was used in the TestShell constructor.
64
65         To avoid adding yet another boolean parameter to TestShell's
66         constructor, add an explicit initialize function that can operate on
67         any boolean settings that are set post-constructor.
68
69         * DumpRenderTree/chromium/DumpRenderTree.cpp:
70         (main):
71         * DumpRenderTree/chromium/TestShell.cpp:
72         (TestShell::TestShell):
73         (TestShell::initialize):
74         * DumpRenderTree/chromium/TestShell.h:
75         (TestShell):
76         (TestShell::setTestShellMode):
77
78 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
79
80         new-run-webkit-tests: suppress debug output from run-safari on results file
81         https://bugs.webkit.org/show_bug.cgi?id=82400
82
83         Reviewed by Ryosuke Niwa.
84
85         Send the run-safari output to /dev/null rather than the console.
86
87         * Scripts/webkitpy/layout_tests/port/mac.py:
88         (MacPort.show_results_html_file):
89
90 2012-03-27  Simon Fraser  <simon.fraser@apple.com>
91
92         Improve error reporting in run-javascriptcore-tests
93         https://bugs.webkit.org/show_bug.cgi?id=82379
94
95         Reviewed by Jessie Berlin.
96         
97         When the script dies, have it print out the reason for dying.
98
99         * Scripts/run-javascriptcore-tests:
100
101 2012-03-27  Eric Seidel  <eric@webkit.org>
102
103         Add Dave Barton to the Contributors list so I can more easily CC him on bugs
104         https://bugs.webkit.org/show_bug.cgi?id=82373
105
106         Reviewed by Ryosuke Niwa.
107
108         Dave's committer-status in in-progress, so until then
109         list him as a Contributor so at least I can more easily CC him on bugs.
110
111         * Scripts/webkitpy/common/config/committers.py:
112
113 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
114
115         [jhbuild] Prepend jhbuildrc's directory to sys.path instead of appending.
116         https://bugs.webkit.org/show_bug.cgi?id=82252
117
118         Reviewed by Gustavo Noronha Silva.
119
120         Appending the directory containing common.py may be problematic if
121         another path in sys.path also contains either a common.py or a
122         common/ module, so prepend to select our version instead. This is
123         the case with fail2ban, which has a common/ directory which may
124         end up being in Python's default path.
125
126         * efl/jhbuildrc:
127         * gtk/jhbuildrc:
128
129 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
130
131         test-webkitpy: prepare for better test run output
132         https://bugs.webkit.org/show_bug.cgi?id=82290
133
134         Reviewed by Adam Barth.
135
136         This code basically re-implements the output of the TextTestRunner default
137         runner code from unittest, although the implementation is quite
138         different, in preparation for changing the test output to be
139         metered and possibly running in parallel.
140
141         The output is almost identical to before, except that instead of
142         logging "test_regular (webkitpy.main.RunnerTest) passed" we log
143         "webkitpy.main.RunnerTest.test_regular passed". It has always
144         annoyed me that they invert the names to be harder to read and
145         so that you can't copy & paste back to the input for
146         test-webkitpy.
147
148         This patch is provided to add a skeleton for unit tests and for
149         comparison to an upcoming patch that will actually add new
150         functionality.
151
152         * Scripts/webkitpy/test/main.py:
153         (Tester.__init__):
154         (Tester._configure_logging):
155         (Tester._run_tests):
156         * Scripts/webkitpy/test/runner.py: Added.
157         (TestRunner):
158         (TestRunner.__init__):
159         (TestRunner.test_name):
160         (TestRunner.all_test_names):
161         (TestRunner.run):
162         (TestRunner.write_result):
163         (TestRunner.write_summary):
164         * Scripts/webkitpy/test/runner_unittest.py: Added.
165         (FakeModuleSuite):
166         (FakeModuleSuite.__init__):
167         (FakeModuleSuite.__str__):
168         (FakeModuleSuite.run):
169         (FakeTopSuite):
170         (FakeTopSuite.__init__):
171         (FakeLoader):
172         (FakeLoader.__init__):
173         (FakeLoader.top_suite):
174         (FakeLoader.loadTestsFromName):
175         (RunnerTest):
176         (RunnerTest.test_regular):
177         (RunnerTest.test_verbose):
178
179 2012-03-27  Gustavo Noronha Silva  <gns@gnome.org>
180
181         [GTK] Build gnutls without p11-kit support
182         https://bugs.webkit.org/show_bug.cgi?id=82212
183
184         Reviewed by Martin Robinson.
185
186         * gtk/jhbuild.modules: pass --without-p11-kit to gnutls'
187         configure, to avoid requiring that dependency, which is not used
188         by libsoup and WebKit anyway
189
190 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
191
192         [Qt] Make debug build work on 32 bit
193         https://bugs.webkit.org/show_bug.cgi?id=82331
194
195         Use --no-keep-memory option of ld not to cache the symbol tables of input files in
196         memory to avoid memory exhaustion during the linking phase for x86-debug builds.
197
198         Reviewed by Tor Arne Vestbø.
199
200         * qmake/mkspecs/features/unix/default_post.prf:
201         * qmake/mkspecs/features/unix/default_pre.prf:
202
203 2012-03-23  Balazs Ankes  <bank@inf.u-szeged.hu>
204
205         Add png checking to check-webkit-style
206         https://bugs.webkit.org/show_bug.cgi?id=75824
207
208         Reviewed by NOBODY Tony Chang.
209
210         * Scripts/webkitpy/style/checker.py:
211         (_all_categories):
212         (FileType):
213         (CheckerDispatcher._should_skip_file_path):
214         (CheckerDispatcher._file_type):
215         (CheckerDispatcher._create_checker):
216         * Scripts/webkitpy/style/checker_unittest.py:
217         (CheckerDispatcherDispatchTest.test_none_paths):
218         * Scripts/webkitpy/style/checkers/png.py: Added.
219         (PNGChecker):
220         (PNGChecker.__init__):
221         (PNGChecker.check):
222         (PNGChecker._config_file_path):
223         * Scripts/webkitpy/style/checkers/png_unittest.py: Added.
224         (MockSCMDetector):
225         (MockSCMDetector.__init__):
226         (MockSCMDetector.display_name):
227         (MockSCMDetector.propget):
228         (PNGCheckerTest):
229         (PNGCheckerTest.test_init):
230         (PNGCheckerTest.test_init.mock_handle_style_error):
231         (PNGCheckerTest.test_check):
232         (PNGCheckerTest.test_check.mock_handle_style_error):
233         * Scripts/webkitpy/style/patchreader.py:
234         (PatchReader.check):
235
236 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
237
238         Unreviewed. Skip GTK+ unit tests that fail in the bots
239
240         * Scripts/run-gtk-tests:
241         (TestRunner): Add failing tests to the skipped list.
242
243 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
244
245         Unreviewed. Unskip all GTK+ unit test.
246
247         Currently, skipped tests contains the list of unit tests that have
248         failed in one or more bots. In this moment it's not possible to
249         skip just the test cases that fail, so the complete unit test is
250         skipped when a single test case fails. For that reason, we don't
251         know which test cases are actually failing for every skipped
252         test. We are planning to land a patch to allow skipping individual
253         test cases, but first we need to know which test cases are
254         currently failing in the bots to skip them. So, this patch simply
255         unskips all unit tests in order to watch the bots and get the list
256         of test cases failing. Failing tests will be skipped again as soon
257         as we have the new list of failing test cases.
258
259         * Scripts/run-gtk-tests:
260         (TestRunner): Reset the list of skipped tests.
261
262 2012-03-26  Scott Byer  <scottbyer@chromium.org>
263
264         Enable layout testing of the scroll animator.
265         https://bugs.webkit.org/show_bug.cgi?id=81858
266         Add a call to the InternalSettings that layout tests can use to
267         turn on scroll animation. Enable animation updates for the
268         Chromium platform DRT when scroll animation has been turned on in
269         a test. This should be a no-op for all current layout tests.
270
271         Reviewed by James Robinson.
272
273         * DumpRenderTree/chromium/WebViewHost.cpp:
274         (WebViewHost::serviceAnimation):
275         (WebViewHost::scheduleAnimation):
276         * DumpRenderTree/chromium/WebViewHost.h:
277         (WebViewHost):
278
279 2012-03-26  Dinu Jacob  <dinu.jacob@nokia.com>
280
281         [Qt][WK2] Support multi-file upload
282         https://bugs.webkit.org/show_bug.cgi?id=81589
283
284         Reviewed by Simon Hausmann.
285
286         Modified filePicker to support multi-file upload.
287
288         * MiniBrowser/qt/MiniBrowser.qrc:
289         * MiniBrowser/qt/icons/checkbox_checked.png: Added.
290         * MiniBrowser/qt/icons/checkbox_unchecked.png: Added.
291         * MiniBrowser/qt/js/MultiSelect.js: Added.
292         (values):
293         (isSelected):
294         (removeValue):
295         * MiniBrowser/qt/qml/CheckBox.qml: Added.
296         * MiniBrowser/qt/qml/FilePicker.qml:
297
298 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
299
300         Retry crashing tests serially at the end of NRWT on Apple Mac
301         https://bugs.webkit.org/show_bug.cgi?id=82233
302
303         Reviewed by Dirk Pranke.
304
305         * Scripts/webkitpy/layout_tests/controllers/manager.py:
306         (Manager.run):
307         * Scripts/webkitpy/layout_tests/port/base.py:
308         (Port.should_retry_crashes):
309         * Scripts/webkitpy/layout_tests/port/mac.py:
310         (MacPort.should_retry_crashes):
311
312 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
313
314         NRWT should retry failures serially
315         https://bugs.webkit.org/show_bug.cgi?id=82241
316
317         Reviewed by Dirk Pranke.
318
319         * Scripts/webkitpy/layout_tests/controllers/manager.py:
320         (Manager._run_tests): Add another parameter to specify the number of workers to use.
321         (Manager.run): Run the failures serially.
322
323 2012-03-26  Adrienne Walker  <enne@google.com>
324
325         Allow expectations on directories in virtual test suites
326         https://bugs.webkit.org/show_bug.cgi?id=82250
327
328         Reviewed by Dirk Pranke.
329
330         lookup_virtual_test_base was only returning valid results for files
331         and not directories. Fix by falling back to the virtual test suite
332         mapping if possible.
333
334         * Scripts/webkitpy/layout_tests/port/base.py:
335         (Port.lookup_virtual_test_base):
336         * Scripts/webkitpy/layout_tests/port/test.py:
337         (TestPort.skipped_tests):
338         (TestPort.virtual_test_suites):
339
340 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
341
342         Fix duplicated ChangeLog entry from r112171.
343
344         Unreviewed, build fix.
345
346 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
347
348         nrwt: remove --worker-model flag
349         https://bugs.webkit.org/show_bug.cgi?id=82112
350
351         Reviewed by Adam Barth.
352
353         The --worker-model=inline flag is basically redundant with the
354         --child-processes=1 flag (technically, you could one run child
355         process, but this is only useful for testing); this patch
356         removes this flag, which simplifies things and enables more
357         cleanup down the road.
358
359         * Scripts/webkitpy/layout_tests/controllers/manager.py:
360         (Manager._run_tests):
361         (Manager.print_config):
362         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
363         (get):
364         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
365         (make_broker):
366         (FunctionTests.test_get__inline):
367         (FunctionTests.test_get__processes):
368         (_TestsMixin.setUp):
369         (_TestsMixin.make_broker):
370         (InlineBrokerTests.setUp):
371         (MultiProcessBrokerTests.setUp):
372         * Scripts/webkitpy/layout_tests/port/base.py:
373         (Port.default_child_processes):
374         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
375         (ChromiumAndroidPort.check_sys_deps):
376         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
377         (ChromiumAndroidPortTest):
378         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
379         (MockDRTPortTest.make_port):
380         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
381         (PortTestCase):
382         (PortTestCase.make_port):
383         * Scripts/webkitpy/layout_tests/port/test.py:
384         (TestPort.default_child_processes):
385         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
386         (_set_up_derived_options):
387         (parse_args):
388         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
389         (parse_args):
390         (MainTest.test_batch_size):
391         (MainTest.test_child_processes_2):
392         (MainTest.test_child_processes_min):
393         (MainTest.test_exception_raised):
394         (MainTest.test_keyboard_interrupt):
395         (MainTest.test_exit_after_n_crashes):
396         (MainTest.assert_run_order):
397         (MainTest.test_run_order__inline):
398         (MainTest.test_virtual):
399
400 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
401
402         test-webkitpy: split test-finding code into a different module
403         https://bugs.webkit.org/show_bug.cgi?id=82253
404
405         Reviewed by Adam Barth.
406
407         Per suggestion from abarth, this change splits all the
408         filesystem-crawling, test-finding code in test-webkitpy out into
409         a separate module and switches to using a FileSystem object.
410         This makes things much more testable, so we also add tests :).
411
412         We also add a realpath() method to the Filesystem object, since
413         test-webkitpy needs that in order to be able to resolve symlinks
414         properly to determine whether a file is under a particular tree
415         or not.
416
417         * Scripts/webkitpy/common/system/filesystem.py:
418         (FileSystem.realpath):
419         * Scripts/webkitpy/common/system/filesystem_mock.py:
420         (MockFileSystem.realpath):
421         * Scripts/webkitpy/test/main.py:
422         (Tester.__init__):
423         (Tester.add_tree):
424         (Tester.run):
425         (Tester._run_tests):
426         (Tester._log_exception):
427         * Scripts/webkitpy/test/test_finder.py: Added.
428         (TestDirectoryTree):
429         (TestDirectoryTree.__init__):
430         (TestDirectoryTree.find_modules):
431         (TestDirectoryTree.find_modules.file_filter):
432         (TestDirectoryTree.to_module):
433         (TestDirectoryTree.clean):
434         (TestFinder):
435         (TestFinder.__init__):
436         (TestFinder.add_tree):
437         (TestFinder.additional_paths):
438         (TestFinder.clean_trees):
439         (TestFinder.is_module):
440         (TestFinder.to_module):
441         (TestFinder.find_names):
442         (TestFinder._exclude):
443         * Scripts/webkitpy/test/test_finder_unittest.py: Added.
444         (TestFinderTest):
445         (TestFinderTest.setUp):
446         (TestFinderTest.tearDown):
447         (TestFinderTest.test_additional_system_paths):
448         (TestFinderTest.test_is_module):
449         (TestFinderTest.test_to_module):
450         (TestFinderTest.test_clean):
451         (TestFinderTest.test_find_names):
452
453 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
454
455         [jhbuild] Use $MAKE if it is defined to build jhbuild itself.
456         https://bugs.webkit.org/show_bug.cgi?id=82209
457
458         Reviewed by Martin Robinson.
459
460         Respecting the $MAKE environment variable when it is defined makes
461         building jhbuild more portable, as some platforms may have GNU
462         make as gmake, for example.
463
464         * jhbuild/jhbuild-wrapper:
465         (install_jhbuild):
466
467 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
468
469         add a Tree abstraction to test-webkitpy to better encapsulate things
470         https://bugs.webkit.org/show_bug.cgi?id=82158
471
472         Reviewed by Adam Barth.
473
474         There are no functional changes in this patch, but adding a
475         basic abstraction for the trees we are looking in for python
476         tests will allow me to add some features later on more easily
477         (like ignoring certain directories like webkitpy/thirdparty),
478         and it makes things slightly less hard-coded.
479
480         * Scripts/test-webkitpy:
481         * Scripts/webkitpy/test/main.py:
482         (Tester.__init__):
483         (Tester):
484         (Tester.add_tree):
485         (Tester._parse_args):
486         (Tester._configure):
487         (Tester.run):
488         (Tester._find_modules):
489         (Tester._run_tests):
490         (Tester._is_module):
491         (Tester._log_exception):
492         (TestDirectoryTree):
493         (TestDirectoryTree.__init__):
494         (TestDirectoryTree.find_modules):
495         (TestDirectoryTree.clean):
496
497 2012-03-26  Ojan Vafai  <ojan@chromium.org>
498
499         garden-o-matic should special case reftests
500         https://bugs.webkit.org/show_bug.cgi?id=81812
501
502         Reviewed by Adam Barth.
503
504         On the examine page, don't show a rebaseline button for reftests.
505         We still show it on the failure summary page because tests are grouped
506         together. Long-term, we should make sure webkit-patch rebaseline-test
507         never tries to rebaseline reftests.
508
509         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
510         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
511         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
512
513 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
514
515         clean up NRWT logging/metering, be less verbose
516         https://bugs.webkit.org/show_bug.cgi?id=73846
517
518         Reviewed by Ojan Vafai.
519
520         Re-land r112014 with a build fix for Python 2.6 compatibility.
521
522         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
523         (ManagerTest.test_http_locking):
524         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
525         (AbstractWorker.run):
526         * Scripts/webkitpy/layout_tests/controllers/worker.py:
527         (Worker.__init__):
528         (Worker._set_up_logging):
529         (Worker):
530         (Worker._set_up_host_and_port):
531         (Worker.run):
532         (Worker.cleanup):
533         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
534         (WorkerTest.test_default_platform_in_worker):
535         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
536         (run):
537         (main):
538         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
539         (MeteredStream):
540         (MeteredStream.implements):
541         (MeteredStream._erasure):
542         (MeteredStream._ensure_newline):
543         (MeteredStream.__init__):
544         (MeteredStream.__del__):
545         (MeteredStream.cleanup):
546         (MeteredStream.write_throttled_update):
547         (MeteredStream.write_update):
548         (MeteredStream.write):
549         (MeteredStream.writeln):
550         (MeteredStream._erase_last_partial_line):
551         (_LogHandler):
552         (_LogHandler.__init__):
553         (_LogHandler.emit):
554         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
555         (RegularTest):
556         (RegularTest.setUp):
557         (RegularTest.tearDown):
558         (RegularTest.time_fn):
559         (RegularTest.test_logging_not_included):
560         (RegularTest._basic):
561         (RegularTest.test_basic):
562         (RegularTest._log_after_update):
563         (RegularTest.test_log_after_update):
564         (TtyTest):
565         (TtyTest.test_basic):
566         (TtyTest.test_log_after_update):
567         (VerboseTest):
568         (VerboseTest.test_basic):
569         (VerboseTest.test_log_after_update):
570         * Scripts/webkitpy/layout_tests/views/printing.py:
571         (parse_print_options):
572         (Printer.__init__):
573         (Printer.cleanup):
574         (Printer.print_progress):
575         (Printer.print_update):
576         (Printer._write):
577         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
578         (Testprinter.get_printer):
579         (Testprinter.test_print_progress):
580         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
581         (PerfTestsRunner.__init__):
582         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
583         (test_run_test_pause_before_testing):
584
585 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
586
587         [jhbuild] Revert $MAKE environment hack introduced in r101929.
588         https://bugs.webkit.org/show_bug.cgi?id=82234
589
590         Reviewed by Martin Robinson.
591
592         The problem it was supposedly fixing was fixed in jhbuild itself
593         in git revision da8944d8ca987ca48c91b350257a530050406092, so we
594         can remove the workaround added to the script.
595
596         * jhbuild/jhbuild-wrapper:
597         (install_jhbuild):
598
599 2012-03-26  Leo Yang  <leo.yang@torchmobile.com.cn>
600
601         [BlackBerry] Skeleton code of AsyncFileSystemBlackBerry.cpp
602         https://bugs.webkit.org/show_bug.cgi?id=82157
603
604         Reviewed by Rob Buis.
605
606         * Scripts/build-webkit:
607         Make ENABLE_FILE_SYSTEM default on for blackberry.
608
609 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
610
611         [jhbuild] Explicitly disable DTrace when building glib.
612         https://bugs.webkit.org/show_bug.cgi?id=82145
613
614         Reviewed by Gustavo Noronha Silva.
615
616         DTrace support is not really needed by WebKit, so disabling it
617         helps avoid some build errors on more exotic platforms (such as
618         FreeBSD).
619
620         * efl/jhbuild.modules:
621         * gtk/jhbuild.modules:
622
623 2012-03-26  Sheriff Bot  <webkit.review.bot@gmail.com>
624
625         Unreviewed, rolling out r111993.
626         http://trac.webkit.org/changeset/111993
627         https://bugs.webkit.org/show_bug.cgi?id=82184
628
629         It broke layout and API tests and made WTR crash (Requested by
630         Ossy on #webkit).
631
632         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
633         (WTR::PlatformWebView::PlatformWebView):
634
635 2012-03-25  Kevin Ollivier  <kevino@theolliviers.com>
636
637         [wx] Unreviewed build fix. Move WTF to its own static lib build.
638
639         * waf/build/settings.py:
640         (common_configure):
641
642 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
643
644         AX: Support solution to handle invalid ax text marker
645         https://bugs.webkit.org/show_bug.cgi?id=82023
646
647         Reviewed by Oliver Hunt.
648
649         * DumpRenderTree/AccessibilityUIElement.cpp:
650         (indexForTextMarkerCallback):
651         (isTextMarkerValidCallback):
652         (textMarkerForIndexCallback):
653         (AccessibilityUIElement::indexForTextMarker):
654         (AccessibilityUIElement::isTextMarkerValid):
655         (AccessibilityUIElement::textMarkerForIndex):
656         (AccessibilityUIElement::getJSClass):
657         * DumpRenderTree/AccessibilityUIElement.h:
658         (AccessibilityUIElement):
659         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
660         (AccessibilityUIElement::indexForTextMarker):
661         (AccessibilityUIElement::textMarkerForIndex):
662         (AccessibilityUIElement::isTextMarkerValid):
663         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
664         (WTR::AccessibilityUIElement::indexForTextMarker):
665         (WTR::AccessibilityUIElement::isTextMarkerValid):
666         (WTR::AccessibilityUIElement::textMarkerForIndex):
667         (WTR):
668         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
669         (AccessibilityUIElement):
670         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
671         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
672         (WTR):
673         (WTR::AccessibilityUIElement::indexForTextMarker):
674         (WTR::AccessibilityUIElement::isTextMarkerValid):
675         (WTR::AccessibilityUIElement::textMarkerForIndex):
676
677 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
678
679         Unreviewed, rolling out r112014.
680         http://trac.webkit.org/changeset/112014
681         https://bugs.webkit.org/show_bug.cgi?id=82138
682
683         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
684         #webkit).
685
686         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
687         (ManagerTest.test_http_locking):
688         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
689         (AbstractWorker.run):
690         * Scripts/webkitpy/layout_tests/controllers/worker.py:
691         (Worker.__init__):
692         (Worker.safe_init):
693         (Worker.run):
694         (Worker.cleanup):
695         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
696         (WorkerTest.test_default_platform_in_worker):
697         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
698         (run):
699         (main):
700         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
701         (MeteredStream):
702         (MeteredStream.is):
703         (MeteredStream.__init__):
704         (MeteredStream.write):
705         (MeteredStream.update):
706         (MeteredStream._overwrite):
707         (MeteredStream._reset):
708         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
709         (TestMeteredStream):
710         (TestMeteredStream.test_regular):
711         * Scripts/webkitpy/layout_tests/views/printing.py:
712         (_configure_logging):
713         (_restore_logging):
714         (Printer.__init__):
715         (Printer.cleanup):
716         (Printer.print_progress):
717         (Printer.print_update):
718         (Printer._write):
719         (Printer):
720         (Printer._update):
721         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
722         (TestUtilityFunctions.assertEmpty):
723         (TestUtilityFunctions):
724         (TestUtilityFunctions.assertNotEmpty):
725         (TestUtilityFunctions.assertWritten):
726         (TestUtilityFunctions.test_configure_logging):
727         (Testprinter.get_printer):
728         (Testprinter.test_print_progress):
729         (Testprinter.test_print_progress.mock_time):
730         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
731         (PerfTestsRunner.__init__):
732         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
733         (test_run_test_pause_before_testing):
734
735 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
736
737         [wx] Unreviewed. Fix the build after WTF move.
738
739         * waf/build/build_utils.py:
740         (get_excludes):
741         (get_excludes_in_dirs):
742         * waf/build/settings.py:
743         (get_port_excludes):
744         (common_configure):
745
746 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
747
748         clean up NRWT logging/metering, be less verbose
749         https://bugs.webkit.org/show_bug.cgi?id=73846
750
751         Reviewed by Ojan Vafai.
752
753         This patch totally rewrites the metered_stream module and
754         changes the way it and the printing module now interact with the
755         logging subsystem.
756
757         The printing module becomes a dumb layout_tests-specific wrapper around
758         the metered stream. All of the intelligence of how and when to
759         actually write (or erase) lines and how things interact with
760         logging moves into the metered stream.
761
762         As of now, anything written to the metered stream *bypasses the
763         logging system*. If a logger is passed to the meter constructor,
764         the metered stream will also attach a log handler so that all
765         messages fed to the logger will also be written to the metered
766         stream. This detangles the logic that used to exist where
767         sometimes the meter would write to the logger and sometimes the
768         logger would write to the meter.
769
770         All of this makes the logic generic and reusable for
771         test-webkitpy (coming in another patch). It also helps clarify
772         what the mysterious "configure_logging" parameter to the printer
773         constructor used to do (that parameter itself is gone now, and
774         by default the printer and metered_stream will not ever touch
775         the logging system; this makes a bunch of tests less fragile).
776
777         All of the above should produce no user-visible changes.
778
779         However, to also fix the bug in question, we change the message
780         format in --verbose mode to %H:%M:%S.%frac %pid %message,
781         eliminating the date, file:lineno, and level parameters. I am
782         making this change at the same time because it would be
783         difficult to write things consistently between the logs and the
784         metered stream if we were still logging the file:lineno.
785
786         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
787           Remove the configure_logging parameter
788         * Scripts/webkitpy/layout_tests/controllers/worker.py:
789           Use metered stream to configure the logging (we don't need a printer)
790           and be careful about when and how we manipulate the log level
791           and handlers on the root logger.
792         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
793           Pass in the root logger so that log messages are formatted
794           properly, and set the log level outside of any routines called
795           by unit tests so as to not interfere.
796         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
797           remove a duplicate log message.
798         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
799           complete rewrite :).
800         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
801           complete rewrite :).
802         * Scripts/webkitpy/layout_tests/views/printing.py:
803           move the logging configuration into metered_stream, and also
804           move the "when should I log progress messages" into metered stream.
805         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
806           remove tests that are no longer necessary.
807         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
808           remove the configure_logging parameter.
809         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
810           update the test; since the printer no longer writes to the log,
811           we need to check what is written to the regular_output instead.
812           (test_run_test_pause_before_testing):
813
814 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
815
816         Add myself and Martin Robinson to:
817           * GStreamerGraphics
818           * GtkWebKit2PublicAPI
819           * SoupNetwork
820         and to the newly created WebKitGTKTranslations.
821
822         Rubber-stamped by Martin Robinson.
823
824         * Scripts/webkitpy/common/config/watchlist:
825
826 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
827
828         [Qt][WK2] Implement PageClient::isViewWindowActive()
829         https://bugs.webkit.org/show_bug.cgi?id=81143
830
831         Reviewed by Kenneth Rohde Christiansen.
832
833         Fix the WrapperWindow from our PlatformWebView
834         so it correctly creates the platform related
835         stuff for QWindow, which is not created unless
836         QWindow::setVisible() or QWindow::show() are called.
837
838         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
839         (WTR::PlatformWebView::PlatformWebView):
840
841 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
842
843         nrwt: don't wait for safari to exit before exiting after showing the results file
844         https://bugs.webkit.org/show_bug.cgi?id=81845
845
846         Reviewed by Ryosuke Niwa.
847
848         Prior to this change, if you ran new-run-webkit-tests and
849         displayed the HTML results file at the end, we would block
850         waiting for the user to quit the browser. There doesn't seem to
851         be a need for that, and the Chromium ports don't do this.
852
853         Also, update the mac tests to capture the output and be quiet :).
854
855         * Scripts/webkitpy/common/system/executive_mock.py:
856           Make popen() testable, implement should_log for it.
857         * Scripts/webkitpy/layout_tests/port/mac.py:
858         (MacPort.show_results_html_file):
859         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
860           Silence some of the tests via outputcapture.
861
862 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
863
864         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
865         https://bugs.webkit.org/show_bug.cgi?id=82073
866
867         Reviewed by Martin Robinson.
868
869         * Scripts/webkitdirs.pm:
870         (jhbuildConfigurationChanged): new function that isolates the
871         jhbuild configuration change checking from the autogen logic.
872         (mustReRunAutogen): code for the new function lived here.
873         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
874         itself, and cleans up the jhbuild root if configuration has
875         changed; it also forces autogen to be run in that case, to
876         make sure the new libraries are used.
877         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
878         * jhbuild/jhbuild-wrapper:
879         (ensure_jhbuild): this might be run inside a jhbuild environment,
880         in which case aclocal fails when trying to use the now deleted
881         aclocal directory in the jhbuild prefix, so work around that.
882
883 2012-03-23  Ojan Vafai  <ojan@chromium.org>
884
885         garden-o-matic scrolls to the top of the page when you click an accordion item
886         https://bugs.webkit.org/show_bug.cgi?id=82088
887
888         Reviewed by Adam Barth.
889
890         Check the scroll position and make sure that the clicked accordion item is visible.
891
892         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
893
894 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
895
896         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
897         https://bugs.webkit.org/show_bug.cgi?id=80567
898
899         Reviewed by Ryosuke Niwa.
900
901         Move clean_up_run to occur right after we finish testing and
902         before we start processing the results. Also makes set_up_run()
903         and clean_up_run() private since there's no real need for
904         callers to be aware of them.
905
906         * Scripts/webkitpy/layout_tests/controllers/manager.py:
907         (Manager._set_up_run):
908         (Manager.run):
909         (Manager._clean_up_run):
910         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
911         (ManagerTest.test_http_locking):
912         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
913         (run):
914
915 2012-03-23  Dean Jackson  <dino@apple.com>
916
917         Disable CSS_SHADERS in Apple builds
918         https://bugs.webkit.org/show_bug.cgi?id=81996
919
920         Reviewed by Simon Fraser.
921
922         Default enableCSSShaders to be 0 on all builds that use
923         this script. Use the flag to turn it back on.
924
925         * Scripts/build-webkit:
926
927 2012-03-23  Tony Chang  <tony@chromium.org>
928
929         [chromium] rename newwtf target back to wtf
930         https://bugs.webkit.org/show_bug.cgi?id=82064
931
932         Reviewed by Adam Barth.
933
934         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
935
936 2012-03-23  Jessie Berlin  <jberlin@apple.com>
937
938         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
939         the Lion Intel WebKit2 Debug bots.
940         https://bugs.webkit.org/show_bug.cgi?id=81012.
941
942         Rubber-stamped by Dan Bernstein.
943
944         Add a step to kill the pasteboard server while we investigate what is causing these
945         assertion failures (to get the bots green).
946
947         * BuildSlaveSupport/kill-old-processes:
948
949 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
950
951         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
952         https://bugs.webkit.org/show_bug.cgi?id=81986
953
954         Reviewed by David Levin.
955
956         Track whether or not errors are filtered in error handlers. If an
957         alphabetical sorting error is reported for a header and filtered, try
958         reporting it for the other potential source of the error. This ensures
959         that the style bot can still find new sorting errors in both potential
960         situations:
961
962             #include <foo.h> <-- 1. edited line
963             #include <baz.h> <-- 2. edited line
964
965         * Scripts/webkitpy/style/checkers/cpp.py:
966         (check_include_line): Potentially report the error for both lines,
967         if the first error is filtered.
968         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
969         test that allows only reporting errors for certain lines. This verifies that
970         when one line of an ordering error is filtered, the error is reported on the
971         other line.
972         (ErrorCollector.__init__):
973         (ErrorCollector.__call__):
974         (CppStyleTestBase.perform_lint):
975         (CppStyleTestBase.perform_language_rules_check):
976         (CppStyleTestBase.assert_language_rules_check):
977         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
978         Added a new test that verifies that when one line of the two lines of an ordering
979         error occur, the error is reported on the other line.
980         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
981         (MockErrorHandler.__call__): Report True because the error is handled.
982         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
983         (ErrorCollector.__call__): Ditto.
984         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
985         (MockErrorHandler.__call__): Ditto.
986         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
987         (TestErrorHandler.__call__): Ditto.
988         * Scripts/webkitpy/style/checkers/xml_unittest.py:
989         (MockErrorHandler.__call__): Ditto.
990         * Scripts/webkitpy/style/error_handlers.py:
991         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
992         and False if it is filtered.
993
994 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
995
996         Build fix for WinCE after r111778.
997
998         * WinCELauncher/CMakeLists.txt:
999
1000 2012-03-23  Peter Beverloo  <peter@chromium.org>
1001
1002         Unreviewed buildfix for the Chromium Android builder.
1003
1004         Disable the ImageDiff target for now. Android needs to build this for
1005         host rather than for target (as other configurations do), but since the
1006         WTF move the dependency chain got significantly more complicated.
1007
1008         I'll work out a proper fix in the following bug, but since several
1009         things are broken due to this, this buildfix should be landed first.
1010         https://bugs.webkit.org/show_bug.cgi?id=82039
1011
1012         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1013
1014 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
1015
1016         [EFL][DRT] Implement LayoutTestController's methods related with editing.
1017         https://bugs.webkit.org/show_bug.cgi?id=80136
1018
1019         Reviewed by Ryosuke Niwa.
1020
1021         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
1022         setSelectTrailingWhitespaceEnabled.
1023
1024         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1025         (LayoutTestController::setSmartInsertDeleteEnabled):
1026         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
1027
1028 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1029
1030         Convert hasGrammarMarker to use Internals interface
1031         https://bugs.webkit.org/show_bug.cgi?id=82004
1032
1033         Reviewed by Ryosuke Niwa.
1034
1035         Remove hasGrammarMarker functions, because it is able to work in the
1036         cross-port way through the Internals interface.
1037
1038         * DumpRenderTree/LayoutTestController.cpp:
1039         (LayoutTestController::staticFunctions):
1040         * DumpRenderTree/LayoutTestController.h:
1041         (LayoutTestController):
1042         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1043         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1044         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1045         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1046         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1047         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1048
1049 2012-03-22  Adam Barth  <abarth@webkit.org>
1050
1051         sync-master-with-upstream creates non-linear git history
1052         https://bugs.webkit.org/show_bug.cgi?id=82006
1053
1054         Reviewed by Julien Chaffraix.
1055
1056         Previously this script would create a merge commit at the HEAD of
1057         origin/master, causing your fork of WebKit to diverge from upstream.
1058         The new version of this script pushes directly from upstream to origin
1059         (without round-tripping through any local branches), avoiding this
1060         problem.
1061
1062         * Scripts/sync-master-with-upstream:
1063
1064 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1065
1066         Unreviewed build fix. Add knowledge of the WTF project to
1067         the autotools build system scripts so that build-jsc works
1068         correctly for GTK+.
1069
1070         * Scripts/webkitdirs.pm:
1071         (buildAutotoolsProject): handle the WTF project, build the
1072         libWTF.la target.
1073         (buildGtkProject): add WTF to the list of supported projects.
1074
1075 2012-03-22  Eric Seidel  <eric@webkit.org>
1076
1077         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
1078         https://bugs.webkit.org/show_bug.cgi?id=81984
1079
1080         Reviewed by Adam Barth.
1081
1082         * Scripts/VCSUtils.pm:
1083         (adjustPathForRecentRenamings):
1084
1085 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1086
1087         [CMake] Fix build-jsc after r111778 and r111797.
1088         https://bugs.webkit.org/show_bug.cgi?id=82001
1089
1090         Reviewed by Filip Pizlo.
1091
1092         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
1093         does return if no error occurred, and build-jsc failed with
1094         "building not defined for this platform".
1095
1096         * Scripts/build-jsc: Always exit after caling
1097         buildCMakeProjectOrExit().
1098         * Scripts/webkitdirs.pm:
1099         (buildCMakeProjectOrExit): Return 0 if no error occurred.
1100
1101 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1102
1103         [CMake] Unreviewed build fix after r111778.
1104
1105         * DumpRenderTree/efl/CMakeLists.txt: Replace
1106         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
1107         paths.
1108         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
1109         wtf include.
1110         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
1111         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
1112         with ${WTF_DIR}/wtf in the include paths.
1113
1114 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
1115
1116         Tools/Scripts/build-jsc doesn't work after WTF move
1117         https://bugs.webkit.org/show_bug.cgi?id=81999
1118
1119         Reviewed by Eric Seidel.
1120         
1121         For ports that require WTF and JSC to be built as two projects and where
1122         the machinery to build both does not already exist, teach the tool to do
1123         so for us.
1124
1125         * Scripts/build-jsc:
1126         (buildMyProject):
1127
1128 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1129
1130         Fix accidental missing line in r111793.
1131
1132         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1133         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
1134         Remove base.values. It was unused and duplicates the builtin Object.values.
1135         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1136         Accidentally deleted this line before committing r111793.
1137
1138 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1139
1140         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
1141         https://bugs.webkit.org/show_bug.cgi?id=81983
1142
1143         Reviewed by Adam Barth.
1144
1145         The dialogs get in the way if you are doing multiple rebaselines. Also,
1146         a number of people didn't realize that you could do multiple rebaselines
1147         in parallel. A non-modal dialog is just confusing.
1148
1149         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1150         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1151         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
1152         show the number of tests being rebaselined.
1153
1154         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1155         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
1156         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
1157         into a single area.
1158
1159         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
1160         also serves to clear all teh content in the StatusArea. 
1161
1162         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1163         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1164
1165 2012-03-22  Tony Chang  <tony@chromium.org>
1166
1167         Unreviewed, fix chromium build after wtf move.
1168
1169         Only use newwtf, remove references to wtf.
1170
1171         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1172
1173 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1174
1175         Actually move WTF files to their new home
1176         https://bugs.webkit.org/show_bug.cgi?id=81844
1177
1178         [Qt] Unreviewed buildfix after r111778.
1179
1180         * DumpRenderTree/qt/DumpRenderTree.pro:
1181
1182 2012-03-22  Dan Bernstein  <mitz@apple.com>
1183
1184         Font fallback in WebKitTestRunner depends on user settings
1185         https://bugs.webkit.org/show_bug.cgi?id=81978
1186
1187         Reviewed by Jessie Berlin.
1188
1189         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
1190         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
1191         with a value consisting of the array ("en") to the override defaults
1192         dictionary. This is the same value used in DumpRenderTree, and it
1193         ensures that font fallback follows a consistent path that is not
1194         affected by the language preferences of the user running the test.
1195
1196 2012-03-22  Andy Estes  <aestes@apple.com>
1197
1198         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
1199         https://bugs.webkit.org/show_bug.cgi?id=81971
1200
1201         Reviewed by Dan Bernstein.
1202
1203         Xcode will automatically link ObjC binaries against libarclite in some
1204         cases, which defines a class called __ARCLite__.
1205         check-for-inappropriate-objc-class-names shouldn't complain about this
1206         class name.
1207
1208         * Scripts/check-for-inappropriate-objc-class-names:
1209
1210 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1211
1212         new-run-webkit-tests blames the wrong test for crashing
1213         https://bugs.webkit.org/show_bug.cgi?id=81951
1214
1215         Reviewed by Adam Barth.
1216
1217         NRWT wasn't extracting the WebProcess pid correct; should've
1218         used re.search instead of re.match :(. Add more tests!
1219
1220         * Scripts/webkitpy/layout_tests/port/mac.py:
1221         (MacPort._get_crash_log):
1222         * Scripts/webkitpy/layout_tests/port/webkit.py:
1223         (WebKitDriver.has_crashed):
1224         (WebKitDriver._check_for_driver_crash):
1225         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1226         (WebKitDriverTest.test_no_timeout):
1227         (WebKitDriverTest):
1228         (WebKitDriverTest.test_check_for_driver_crash):
1229         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
1230         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
1231         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
1232         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
1233         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
1234         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
1235
1236 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1237
1238         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
1239         https://bugs.webkit.org/show_bug.cgi?id=81512
1240
1241         Adding missing implementation markerTextForListItem to EFL's
1242         LayoutTestController so that we can unskip related tests from
1243         the skip list.
1244
1245         Reviewed by Antonio Gomes.
1246
1247         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1248         (LayoutTestController::markerTextForListItem): Implemented.
1249
1250 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1251
1252         [EFL] Enable view mode media feature layout tests
1253         https://bugs.webkit.org/show_bug.cgi?id=81205
1254
1255         Adding missing implementation setViewModeMediaFeature to EFL's 
1256         LayoutTestController so that we can unskip view mode 
1257         media feature layout tests.
1258
1259         Reviewed by Antonio Gomes.
1260
1261         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1262         (LayoutTestController::setViewModeMediaFeature): Implemented.
1263
1264 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1265
1266         nrwt: crash while stopping layout test helper on apple mac lion
1267         https://bugs.webkit.org/show_bug.cgi?id=81949
1268
1269         Reviewed by Eric Seidel.
1270
1271         stop_helper() didn't work right if the helper process wasn't
1272         still running or didn't respond properly to being shut down. Fix
1273         this and add some tests. Also add code to MockExecutive() to be
1274         able to mock executive.popen() properly.
1275
1276         * Scripts/webkitpy/common/system/executive_mock.py:
1277         (MockProcess.__init__):
1278         (MockProcess.wait):
1279         (MockExecutive.__init__):
1280         (MockExecutive.popen):
1281         * Scripts/webkitpy/common/system/systemhost_mock.py:
1282         (MockSystemHost.__init__):
1283         * Scripts/webkitpy/layout_tests/port/mac.py:
1284         (MacPort.start_helper):
1285         (MacPort.stop_helper):
1286         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1287         (test_get_crash_log):
1288         (test_helper_starts):
1289         (test_helper_fails_to_start):
1290         (test_helper_fails_to_stop):
1291         (test_helper_fails_to_stop.bad_waiter):
1292
1293 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1294
1295         [GTK] Add make to the jhbuild moduleset
1296         https://bugs.webkit.org/show_bug.cgi?id=81769
1297
1298         Reviewed by Martin Robinson.
1299
1300         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
1301         Add GNU make with custom patches to fix several issues:
1302
1303         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
1304           fix the argument list limit of GNU make when running make
1305           distcheck.
1306         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
1307           and Jeremy Devenport to fix parallel build issues present in
1308           GNU make 3.82
1309
1310         * gtk/jhbuild.modules: Added make with patches.
1311         * gtk/patches/make-3.82-arg-list-length.patch: Added.
1312         * gtk/patches/make-3.82-parallel-build.patch: Added.
1313
1314 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
1315
1316         [wx] Unreviewed. Adding Source/WTF to the build and updating
1317         waf port list.
1318
1319         * DumpRenderTree/wscript:
1320         * waf/build/settings.py:
1321
1322 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1323
1324         Re-land fix for r81603 yet again (!) with latest build fix.
1325         https://bugs.webkit.org/show_bug.cgi?id=81603
1326
1327         Unreviewed, build fix.
1328
1329         The latest problem was a bad sprintf() in
1330         BasePort._get_crash_log(). I've reworked that routine and added
1331         unit tests for coverage.
1332
1333         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1334         (SingleTestRunner._handle_error):
1335         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1336         (write_test_result):
1337         (TestResultWriter.write_crash_log):
1338         * Scripts/webkitpy/layout_tests/port/base.py:
1339         (Port.diff_text):
1340         (Port._get_crash_log):
1341         * Scripts/webkitpy/layout_tests/port/chromium.py:
1342         (ChromiumDriver.run_test):
1343         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1344         (ChromiumDriverTest.test_crash_log):
1345         * Scripts/webkitpy/layout_tests/port/driver.py:
1346         (DriverOutput.__init__):
1347         * Scripts/webkitpy/layout_tests/port/mac.py:
1348         (MacPort.is_lion):
1349         (MacPort._get_crash_log):
1350         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1351         (test_default_child_processes):
1352         (test_get_crash_log):
1353         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1354         (MockDRTPortTest.test_get_crash_log):
1355         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1356         (PortTestCase.test_get_crash_log):
1357         * Scripts/webkitpy/layout_tests/port/test.py:
1358         (TestDriver.run_test):
1359         * Scripts/webkitpy/layout_tests/port/webkit.py:
1360         (WebKitDriver.__init__):
1361         (WebKitDriver._start):
1362         (WebKitDriver.has_crashed):
1363         (WebKitDriver._check_for_driver_crash):
1364         (WebKitDriver.run_test):
1365
1366 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1367
1368         [Qt] Enable support for handling database quota's in the MiniBrowser
1369         https://bugs.webkit.org/show_bug.cgi?id=81928
1370
1371         Reviewed by Simon Hausmann.
1372
1373         * MiniBrowser/qt/qml/BrowserWindow.qml:
1374
1375 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1376
1377         [EFL] Implement LayoutTestController::setEditingBehavior
1378         https://bugs.webkit.org/show_bug.cgi?id=81124
1379
1380         Adding missing implementation setEditingBehavior to EFL's
1381         LayoutTestController. This implementation allows us to
1382         unskip some tests from the skip list.
1383
1384         Reviewed by Martin Robinson.
1385
1386         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1387         (LayoutTestController::setEditingBehavior): Implemented.
1388
1389 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
1390
1391         [GTK] Use the angle-bracket form to include wtf headers
1392         https://bugs.webkit.org/show_bug.cgi?id=81884
1393
1394         Reviewed by Eric Seidel.
1395
1396         Use #include <wtf/foo> instead of #include "foo".
1397
1398         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1399         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1400         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1401         * DumpRenderTree/gtk/EditingCallbacks.cpp:
1402         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
1403         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
1404
1405 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1406
1407         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
1408
1409         * Scripts/webkitpy/common/config/committers.py:
1410
1411 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1412
1413         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
1414         https://bugs.webkit.org/show_bug.cgi?id=81889
1415
1416         Reviewed by Kenneth Rohde Christiansen.
1417
1418         Enable FAST_MOBILE_SCROLLING feature.
1419
1420         * qmake/mkspecs/features/features.prf:
1421
1422 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1423
1424         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
1425         https://bugs.webkit.org/show_bug.cgi?id=81900
1426
1427         Reviewed by Gustavo Noronha Silva.
1428
1429         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
1430         path which assumed a certain directory structure for the build
1431         directory which may noy be true.
1432
1433         Instead, hardcode the path to the downloaded fonts starting from the
1434         top-level source directory. While this solution is also far from
1435         perfect, it makes it more consistent and reliable, and easier to
1436         tackle at once with the rest of bug 81475.
1437
1438         * DumpRenderTree/efl/CMakeLists.txt:
1439
1440 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
1441
1442         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
1443         https://bugs.webkit.org/show_bug.cgi?id=81888
1444
1445         Reviewed by Philippe Normand.
1446
1447         * Scripts/update-webkit-libs-jhbuild:
1448         (runJhbuild): Helper function to run jhbuild with a given command.
1449         Instead of running jhbuild directly, use the jhbuild-wrapper, so
1450         that we don't need to run it first to check whether jhbuild is
1451         installed or not.
1452         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
1453         * gtk/run-with-jhbuild: Ditto.
1454         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
1455         not only run.
1456
1457 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1458
1459         Unreviewed, test-webkitpy fix after r111661.
1460
1461         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1462         (MainTest.test_crash_log):
1463         (MainTest.test_web_process_crash_log):
1464
1465 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1466
1467         Unreviewed, test-webkitpy fix after r111642.
1468
1469         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1470         (ChromiumDriverTest.setUp):
1471         (ChromiumDriverTest.test_two_drivers):
1472
1473 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1474
1475         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1476         https://bugs.webkit.org/show_bug.cgi?id=81603
1477
1478         Unreviewed rolling out r111609 and part of r111615,
1479         because it broke NRWT on Qt-WK2 platform.
1480
1481         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1482         (SingleTestRunner._handle_error):
1483         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1484         (write_test_result):
1485         (TestResultWriter.write_crash_report):
1486         * Scripts/webkitpy/layout_tests/port/base.py:
1487         (Port.is_crash_reporter):
1488         (Port._driver_class):
1489         * Scripts/webkitpy/layout_tests/port/chromium.py:
1490         (ChromiumDriver.run_test):
1491         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1492         (ChromiumDriverTest.test_crashed_process_name):
1493         * Scripts/webkitpy/layout_tests/port/driver.py:
1494         (DriverOutput.__init__):
1495         * Scripts/webkitpy/layout_tests/port/mac.py:
1496         (MacPort):
1497         (MacPort.is_crash_reporter):
1498         (MacPort.release_http_lock):
1499         * Scripts/webkitpy/layout_tests/port/test.py:
1500         (TestDriver.run_test):
1501         * Scripts/webkitpy/layout_tests/port/webkit.py:
1502         (WebKitDriver.__init__):
1503         (WebKitDriver._start):
1504         (WebKitDriver.has_crashed):
1505         (WebKitDriver._check_for_driver_crash):
1506         (WebKitDriver.run_test):
1507
1508 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1509
1510         [Qt] Add full platforms to Qt buildslaves
1511         https://bugs.webkit.org/show_bug.cgi?id=81877
1512
1513         Reviewed by Ryosuke Niwa.
1514
1515         * BuildSlaveSupport/build.webkit.org-config/config.json:
1516
1517 2012-03-21  Johnny Ding  <jnd@chromium.org>
1518
1519         Customize layout test timeout value for different ports.
1520         https://bugs.webkit.org/show_bug.cgi?id=79859
1521
1522         Reviewed by Ojan Vafai.
1523
1524         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1525         (Manager):
1526         * Scripts/webkitpy/layout_tests/port/base.py:
1527         (Port.default_test_timeout_ms):
1528         * Scripts/webkitpy/layout_tests/port/chromium.py:
1529         (ChromiumDriver.stop):
1530         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1531         (ChromiumAndroidPort.default_test_timeout_ms):
1532         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1533         (_set_up_derived_options):
1534
1535 2012-03-21  Mark Rowe  <mrowe@apple.com>
1536
1537         Fix the build.
1538
1539         Ensure that our test tools don't try and use RTTI unnecessarily.
1540
1541         * DumpRenderTree/mac/Configurations/Base.xcconfig:
1542         * TestWebKitAPI/Configurations/Base.xcconfig:
1543         * WebKitTestRunner/Configurations/Base.xcconfig:
1544
1545 2012-03-21  Adam Barth  <abarth@webkit.org>
1546
1547         WebKit should have some scripts to smooth interactions with GitHub
1548         https://bugs.webkit.org/show_bug.cgi?id=81842
1549
1550         Reviewed by Eric Seidel.
1551
1552         These scripts support the workflow in
1553         https://trac.webkit.org/wiki/UsingGitHub by automating the
1554         configuration and syncing operations needed for a GitHub fork of WebKit
1555         to track the main WebKit repository.
1556
1557         * Scripts/configure-github-as-upstream: Added.
1558         * Scripts/sync-master-with-upstream: Added.
1559
1560 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1561
1562         WTR - log the pid of a crashing WebProcess
1563         https://bugs.webkit.org/show_bug.cgi?id=81575
1564
1565         Reviewed by Alexey Proskuryakov.
1566
1567         This allows run-webkit-tests to find the right crash log for a
1568         given crashed WebProcess in WebKit2.
1569
1570         * WebKitTestRunner/TestController.cpp:
1571         (WTR::exceededDatabaseQuota):
1572         (WTR::TestController::createOtherPage):
1573         (WTR::TestController::initialize):
1574         (WTR::TestController::resetStateToConsistentValues):
1575         (WTR::TestController::runTest):
1576         (WTR::TestController::processDidCrash):
1577
1578 2012-03-21  Enrica Casucci  <enrica@apple.com>
1579
1580         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
1581         https://bugs.webkit.org/show_bug.cgi?id=81835
1582         <rdar://problem/11082749>
1583
1584         Reviewed by Brady Eidson.
1585
1586         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
1587         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
1588         (TestWebKitAPI::contextMenuCopyLink):
1589
1590 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
1591
1592         [CSS Shaders] Make CSS Shaders compile on Chromium
1593         https://bugs.webkit.org/show_bug.cgi?id=81435
1594
1595         Reviewed by Stephen White.
1596         
1597         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
1598
1599         * DumpRenderTree/chromium/LayoutTestController.cpp:
1600         (LayoutTestController::overridePreference):
1601         * DumpRenderTree/chromium/WebPreferences.cpp:
1602         (WebPreferences::reset):
1603         (WebPreferences::applyTo):
1604         * DumpRenderTree/chromium/WebPreferences.h:
1605         (WebPreferences):
1606
1607 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1608
1609         fix regression on gtk port introduced by r111609
1610         https://bugs.webkit.org/show_bug.cgi?id=81839
1611
1612         Reviewed by Philippe Normand.
1613
1614         r111609 should've initialized a couple of missing fields in
1615         WebKitDriver, and also should've checked for method overrides
1616         in the other ports :(.
1617
1618         * Scripts/webkitpy/layout_tests/port/gtk.py:
1619         (GtkDriver._start):
1620         * Scripts/webkitpy/layout_tests/port/webkit.py:
1621         (WebKitDriver.__init__):
1622         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1623         (MainTest.test_crash_log):
1624         (MainTest.test_web_process_crash_log):
1625
1626 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1627
1628         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1629         https://bugs.webkit.org/show_bug.cgi?id=81603
1630
1631         Reviewed by Adam Barth.
1632
1633         Add a new _get_crash_log() overridable method on the Port
1634         object to customize how we fetch crash logs per port.
1635         Mac will now slow-spin with a timeout waiting for the crash log, rather
1636         than waiting for the ReportCrash process to exit (it appears
1637         that the mac will manage multiple crashes with a single
1638         ReportCrash process, the process waits around longer than
1639         necessary, presumably to avoid thrashing if processes are
1640         repeatedly crashing).
1641
1642         Also, add the DriverOutput should contain the crash log and other info,
1643         which is created in a port-specific manner but can then be
1644         treated generically. Previously single_test_runner would get
1645         told that something crashed and attempt to do something to get
1646         the crash log, but it didn't have the information it needed to
1647         od the right thing; better to make the driver hand back the
1648         right info.
1649
1650         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1651         (SingleTestRunner._handle_error):
1652         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1653         (write_test_result):
1654         (TestResultWriter.write_crash_log):
1655         * Scripts/webkitpy/layout_tests/port/base.py:
1656         (Port.diff_text):
1657         (Port._get_crash_log):
1658         * Scripts/webkitpy/layout_tests/port/chromium.py:
1659         (ChromiumDriver.run_test):
1660         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1661         (ChromiumDriverTest.test_crash_log):
1662         * Scripts/webkitpy/layout_tests/port/driver.py:
1663         (DriverOutput.__init__):
1664         * Scripts/webkitpy/layout_tests/port/mac.py:
1665         (MacPort.is_lion):
1666         (MacPort._get_crash_log):
1667         * Scripts/webkitpy/layout_tests/port/test.py:
1668         (TestDriver.run_test):
1669         * Scripts/webkitpy/layout_tests/port/webkit.py:
1670         (WebKitDriver.__init__):
1671         (WebKitDriver._start):
1672         (WebKitDriver.has_crashed):
1673         (WebKitDriver._check_for_driver_crash):
1674         (WebKitDriver.run_test):
1675
1676 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1677
1678         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
1679         https://bugs.webkit.org/show_bug.cgi?id=81507
1680
1681         Now that we're only loading from one directory, no need to
1682         iterate over a list of directories anymore.
1683
1684         Reviewed by Martin Robinson.
1685
1686         * DumpRenderTree/efl/CMakeLists.txt:
1687         * DumpRenderTree/efl/FontManagement.cpp:
1688         (addFontDirectory):
1689         (addFontsToEnvironment):
1690
1691 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1692
1693         webkitpy: get ServerProcess out of the reportcrash business
1694         https://bugs.webkit.org/show_bug.cgi?id=81600
1695
1696         Unreviewed, build fix.
1697
1698         Re-land r111307 and r111293 with another fix for a crash in NRWT;
1699         we need to check if the driver has crashed before attempting to
1700         read from it.
1701
1702
1703         * Scripts/webkitpy/layout_tests/port/server_process.py:
1704         (ServerProcess._reset):
1705         (ServerProcess._handle_possible_interrupt):
1706         (ServerProcess.write):
1707         (ServerProcess.read_stdout):
1708         (ServerProcess.has_crashed):
1709         (ServerProcess._read):
1710         (ServerProcess.stop):
1711         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1712         (TrivialMockPort.check_for_leaks):
1713         (TestServerProcess.test_broken_pipe):
1714         * Scripts/webkitpy/layout_tests/port/webkit.py:
1715         (WebKitPort._read_image_diff):
1716         (WebKitDriver.has_crashed):
1717         (WebKitDriver._check_for_driver_crash):
1718         (WebKitDriver.run_test):
1719         (WebKitDriver._read_block):
1720         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1721         (MockServerProcess.__init__):
1722         (MockServerProcess):
1723         (MockServerProcess.has_crashed):
1724
1725 2012-03-21  Zeno Albisser  <zeno@webkit.org>
1726
1727         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
1728         https://bugs.webkit.org/show_bug.cgi?id=81750
1729
1730         Building WebCore on mac currently creates an archive that is bigger than
1731         4GB. But ranlib fails to create a symbol table for such a big archive,
1732         even on 64bit machines.
1733         Therefore we start using SVGAllInOne.cpp when building debug
1734         on mac. This reduces the size of the archive by about 300MB.
1735
1736         Reviewed by Tor Arne Vestbø.
1737
1738         * qmake/mkspecs/features/mac/default_pre.prf:
1739
1740 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
1741
1742         Updated the address list of committers.py.
1743
1744         Reviewed by Csaba Osztrogonác.
1745
1746         * Scripts/webkitpy/common/config/committers.py:
1747
1748 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
1749
1750         Updating my nick in committers file.
1751
1752         Unreviewed.
1753
1754         * Scripts/webkitpy/common/config/committers.py:
1755
1756 2012-03-21  Peter Beverloo  <peter@chromium.org>
1757
1758         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
1759         https://bugs.webkit.org/show_bug.cgi?id=80861
1760
1761         Reviewed by Tony Chang.
1762
1763         Remove most Android-specific logic from webkitdirs.pm in favor of
1764         handling this during project generation time.
1765
1766         * Scripts/webkitdirs.pm:
1767         (buildChromiumMakefile):
1768
1769 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1770
1771         [GTK] Allow running javascript from location bar in MiniBrowser
1772         https://bugs.webkit.org/show_bug.cgi?id=81331
1773
1774         Reviewed by Martin Robinson.
1775
1776         * MiniBrowser/gtk/BrowserWindow.c:
1777         (activateUriEntryCallback): Use browser_window_load_uri().
1778         (browser_window_load_uri): Check whether uri starts with
1779         javascript: and run the given script, or just call
1780         webkit_web_view_load_uri() to load the given uri.
1781         * MiniBrowser/gtk/BrowserWindow.h:
1782         * MiniBrowser/gtk/main.c:
1783         (createBrowserWindow): Use browser_window_load_uri().
1784
1785 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
1786
1787         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
1788         https://bugs.webkit.org/show_bug.cgi?id=75543
1789
1790         Reviewed by Martin Robinson.
1791
1792         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
1793         MiniBrowser CPP flags.
1794
1795 2012-03-20  Eric Seidel  <eric@webkit.org>
1796
1797         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
1798         https://bugs.webkit.org/show_bug.cgi?id=80911
1799
1800         Reviewed by Adam Barth.
1801
1802         Update include paths for Source/WTF.
1803
1804         * DumpRenderTree/efl/CMakeLists.txt:
1805         * DumpRenderTree/qt/DumpRenderTree.pro:
1806         * QtTestBrowser/QtTestBrowser.pro:
1807         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
1808
1809 2012-03-20  Adam Barth  <abarth@webkit.org>
1810
1811         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
1812         https://bugs.webkit.org/show_bug.cgi?id=81567
1813
1814         Reviewed by Eric Seidel.
1815
1816         This temporary measure should help us avoid disrupting the folks
1817         working on WebAudio when we move the webaudio directory to Modules.
1818
1819         * Scripts/VCSUtils.pm:
1820         (adjustPathForRecentRenamings):
1821         (parseGitDiffHeader):
1822         (parseSvnDiffHeader):
1823
1824 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1825
1826         Unreviewed, rolling out r111453.
1827         http://trac.webkit.org/changeset/111453
1828         https://bugs.webkit.org/show_bug.cgi?id=81600
1829
1830         still broken
1831
1832         * Scripts/webkitpy/layout_tests/port/server_process.py:
1833         (ServerProcess._reset):
1834         (ServerProcess.handle_interrupt):
1835         (ServerProcess.write):
1836         (ServerProcess._check_for_crash):
1837         (ServerProcess._handle_timeout):
1838         (ServerProcess._check_for_abort):
1839         (ServerProcess._read):
1840         (ServerProcess.stop):
1841         (ServerProcess):
1842         (ServerProcess.set_crashed):
1843         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1844         (TrivialMockPort.is_crash_reporter):
1845         (TestServerProcess.test_broken_pipe):
1846         * Scripts/webkitpy/layout_tests/port/webkit.py:
1847         (WebKitPort._read_image_diff):
1848         (WebKitDriver.has_crashed):
1849         (WebKitDriver._check_for_driver_crash):
1850         (WebKitDriver):
1851         (WebKitDriver._detected_crash):
1852         (WebKitDriver._subprocess_crashed):
1853         (WebKitDriver._crashed_process_name):
1854         (WebKitDriver.run_test):
1855         (WebKitDriver._read_block):
1856         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1857         (MockServerProcess.__init__):
1858
1859 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
1860
1861         webkitpy: get ServerProcess out of the reportcrash business
1862         https://bugs.webkit.org/show_bug.cgi?id=81600
1863
1864         Unreviewed, build fix.
1865
1866         Re-land r111307 and r111293 with fix for crash in NRWT; I
1867         was failing to set the crashing process name.
1868
1869         * Scripts/webkitpy/layout_tests/port/server_process.py:
1870         (ServerProcess._reset):
1871         (ServerProcess._handle_possible_interrupt):
1872         (ServerProcess.write):
1873         (ServerProcess.read_stdout):
1874         (ServerProcess.has_crashed):
1875         (ServerProcess._read):
1876         (ServerProcess.stop):
1877         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1878         (TrivialMockPort.check_for_leaks):
1879         (TestServerProcess.test_broken_pipe):
1880         * Scripts/webkitpy/layout_tests/port/webkit.py:
1881         (WebKitPort._read_image_diff):
1882         (WebKitDriver.has_crashed):
1883         (WebKitDriver._check_for_driver_crash):
1884         (WebKitDriver.run_test):
1885         (WebKitDriver._read_block):
1886         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1887         (MockServerProcess.__init__):
1888         (MockServerProcess):
1889         (MockServerProcess.has_crashed):
1890
1891 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1892
1893         [mac] Restore color space switching code to run-webkit-tests
1894         https://bugs.webkit.org/show_bug.cgi?id=80571
1895         <rdar://problem/11008529>
1896
1897         Reviewed by Simon Fraser.
1898
1899         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
1900         
1901         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
1902         didn't exist until Lion.
1903
1904         * DumpRenderTree/mac/LayoutTestHelper.m:
1905         (installLayoutTestColorProfile):
1906         (restoreUserColorProfile):
1907
1908 2012-03-20  Adele Peterson  <adele@apple.com>
1909
1910         Update the last test to use EXPECT_WK_STREQ.
1911
1912         Reviewed by Dan Bernstein.
1913
1914         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
1915
1916 2012-03-20  Adele Peterson  <adele@apple.com>
1917
1918         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
1919         https://bugs.webkit.org/show_bug.cgi?id=81630
1920         <rdar://problem/10650660>
1921
1922         Reviewed by Alexey Proskuryakov.
1923
1924         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1925         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
1926         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
1927         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
1928         (TestWebKitAPI):
1929         (TestWebKitAPI::didFinishLoadForFrame):
1930         (TestWebKitAPI::TEST):
1931         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
1932
1933 2012-03-20  Eric Seidel  <eric@webkit.org>
1934
1935         Add link on queues.webkit.org to kov's fancy queue-depth graph
1936         https://bugs.webkit.org/show_bug.cgi?id=81690
1937
1938         Reviewed by Adam Barth.
1939
1940         * QueueStatusServer/templates/recentstatus.html:
1941
1942 2012-03-20  Tim Horton  <timothy_horton@apple.com>
1943
1944         [mac] Restore color space switching code to run-webkit-tests
1945         https://bugs.webkit.org/show_bug.cgi?id=80571
1946         <rdar://problem/11008529>
1947
1948         Reviewed by Simon Fraser.
1949
1950         Make use of the layout test helper hooks to restore the Mac system-wide
1951         color profile switching code. This makes WkTR and DRT results match, and
1952         causes the main display's color profile to not affect results on Lion.
1953
1954         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1955         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
1956         (installLayoutTestColorProfile):
1957         (restoreUserColorProfile):
1958         (simpleSignalHandler):
1959         (main):
1960         * Scripts/webkitpy/layout_tests/port/mac.py:
1961         (MacPort.release_http_lock):
1962         (MacPort):
1963         (MacPort._path_to_helper):
1964         (MacPort.start_helper):
1965         (MacPort.stop_helper):
1966         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
1967         (WTR::PlatformWebView::PlatformWebView):
1968         (WTR::PlatformWebView::windowSnapshotImage):
1969
1970 2012-02-24  Darin Fisher  <darin@chromium.org>
1971
1972         Add more Chromium WebKit API reviewers
1973         https://bugs.webkit.org/show_bug.cgi?id=81576
1974
1975         Reviewed by Adam Barth and David Levin.
1976
1977         Simplify watchlist to lump all WebKit API into one.
1978
1979         * Scripts/webkitpy/common/config/watchlist:
1980
1981 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1982
1983         Convert hasSpellingMarker to use Internals interface.
1984         https://bugs.webkit.org/show_bug.cgi?id=81300
1985
1986         Reviewed by Ryosuke Niwa.
1987
1988         Remove hasSpellingMarker functions, because it is able to work in the
1989         cross-port way through the Internals interface.
1990
1991         * DumpRenderTree/LayoutTestController.cpp:
1992         (LayoutTestController::staticFunctions):
1993         * DumpRenderTree/LayoutTestController.h:
1994         (LayoutTestController):
1995         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1996         * DumpRenderTree/chromium/LayoutTestController.cpp:
1997         (LayoutTestController::LayoutTestController):
1998         * DumpRenderTree/chromium/LayoutTestController.h:
1999         (LayoutTestController):
2000         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2001         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2002         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2003         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2004         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2005         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2006         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2007
2008 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2009
2010         Add new Qt-WK2 performance tester bot to build.webkit.org
2011         https://bugs.webkit.org/show_bug.cgi?id=80842
2012
2013         Reviewed by Ryosuke Niwa.
2014
2015         * BuildSlaveSupport/build.webkit.org-config/config.json:
2016         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2017         (RunAndUploadPerfTestsWebKit2):
2018         (RunAndUploadPerfTestsWebKit2.start):
2019         (BuildAndPerfTestWebKit2Factory):
2020         (BuildAndPerfTestWebKit2Factory.__init__):
2021         (DownloadAndPerfTestWebKit2Factory):
2022         (DownloadAndPerfTestWebKit2Factory.__init__):
2023
2024 2012-03-20  Peter Beverloo  <peter@chromium.org>
2025
2026         [Chromium] Update to Android NDK r7b
2027         https://bugs.webkit.org/show_bug.cgi?id=81005
2028
2029         Reviewed by Adam Barth.
2030
2031         Update to version r7b of the Android NDK, release notes of which are
2032         available on the following page:
2033         http://developer.android.com/sdk/ndk/index.html
2034
2035         * Scripts/update-webkit-chromium:
2036
2037 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2038
2039         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
2040         https://bugs.webkit.org/show_bug.cgi?id=81502
2041
2042         Unreviewed.
2043
2044         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
2045
2046 2012-03-19  Jessie Berlin  <jberlin@apple.com>
2047
2048         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
2049         because they broke running the layout tests on mac.
2050
2051         * Scripts/webkitpy/layout_tests/port/webkit.py:
2052         (WebKitDriver._check_for_driver_crash):
2053
2054 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
2055
2056         [GTK] Allow running run-gtk-tests during 'make distcheck'
2057         https://bugs.webkit.org/show_bug.cgi?id=81415
2058
2059         Reviewed by Philippe Normand.
2060
2061         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
2062
2063         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
2064         main command run during 'make check.'
2065         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
2066         a requirement to run it. If jhbuild isn't around it will simply not start the
2067         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
2068         main loop and simply use the main loop in a synchronous method to start SPI
2069         daemons. Did a bunch of general cleanup around these tasks.
2070
2071 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2072
2073         jhbuild: Use the multiprocessing module to get the number of CPUs.
2074         https://bugs.webkit.org/show_bug.cgi?id=81585
2075
2076         Reviewed by Martin Robinson.
2077
2078         Now that support for Python 2.5 has been dropped, we can safely
2079         use multiprocessing.cpu_count() to obtain the current number of
2080         CPUs in the system instead of running a webkitperl script for
2081         that.
2082
2083         * efl/common.py:
2084         (top_level_path):
2085         * efl/jhbuildrc:
2086         * gtk/common.py:
2087         (build_path):
2088         * gtk/jhbuildrc:
2089
2090 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2091
2092         We should only add a live event listener once, not once per TestSelector.
2093         Adding one per TestSelector would cause accordion('option', 'active') to
2094         incorrectly be "false" if you examine multiple sets of tests.
2095
2096         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2097
2098 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2099
2100         Fix typo introduced in r111293.
2101
2102         Unreviewed, build fix.
2103
2104         * Scripts/webkitpy/layout_tests/port/webkit.py:
2105         (WebKitDriver._check_for_driver_crash):
2106
2107 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2108
2109         webkitpy: get ServerProcess out of the reportcrash business
2110         https://bugs.webkit.org/show_bug.cgi?id=81600
2111
2112         Reviewed by Adam Barth.
2113
2114         ServerProcess was half-aware that ReportCrash might run
2115         sometimes, and that the process ServerProcess was talking to
2116         might have its own crashing subprocesses; neither of these
2117         things really worked right and it made the logic convoluted, so
2118         this change makes handling crashes completely separate from the
2119         server_process code, so that it can focus on just I/O to the
2120         subprocess.
2121
2122         There should be no functional changes resulting from this patch.
2123
2124         * Scripts/webkitpy/layout_tests/port/server_process.py:
2125         (ServerProcess._reset):
2126         (ServerProcess._handle_possible_interrupt):
2127         (ServerProcess.write):
2128         (ServerProcess.read_stdout):
2129         (ServerProcess.has_crashed):
2130         (ServerProcess._read):
2131         (ServerProcess.stop):
2132         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2133         (TrivialMockPort.check_for_leaks):
2134         (TestServerProcess.test_broken_pipe):
2135         * Scripts/webkitpy/layout_tests/port/webkit.py:
2136         (WebKitPort._read_image_diff):
2137         (WebKitDriver.has_crashed):
2138         (WebKitDriver._check_for_driver_crash):
2139         (WebKitDriver.run_test):
2140         (WebKitDriver._read_block):
2141         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2142         (MockServerProcess.__init__):
2143         (MockServerProcess):
2144         (MockServerProcess.has_crashed):
2145
2146 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2147
2148         webkitpy: crashlog parsing is broken
2149         https://bugs.webkit.org/show_bug.cgi?id=81586
2150
2151         Reviewed by Adam Barth.
2152
2153         Seeking on a file opened through codecs() appears to not work
2154         correctly (at least on SL).
2155         
2156         The code was more complicated than it needed to be, so I have
2157         revamped it to just read a full crashlog at once and then look
2158         for matches. 
2159         
2160         I have also added the ability to optionally propagate
2161         errors back (which can be helpful to debug races when ReportCrash is
2162         still running and you need to see that you couldn't open some
2163         files, rather than just ignoring them).
2164
2165         * Scripts/webkitpy/common/system/crashlogs.py:
2166         (CrashLogs.find_newest_log):
2167         (CrashLogs._find_newest_log_darwin):
2168         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2169         (CrashLogsTest.test_find_log_darwin):
2170
2171 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2172
2173         webkitpy: clean up some port._filesystem references to not reference private members
2174         https://bugs.webkit.org/show_bug.cgi?id=81595
2175
2176         Reviewed by Eric Seidel.
2177
2178         There should be no functional changes in this patch, just some
2179         refactoring. Also, fixes a minor nit in write_image_diff_files(()
2180
2181         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2182         (SingleTestRunner.__init__):
2183         (SingleTestRunner._run_compare_test):
2184         (SingleTestRunner._run_rebaseline):
2185         (SingleTestRunner._save_baseline_data):
2186         (SingleTestRunner._handle_error):
2187         (SingleTestRunner._run_reftest):
2188         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2189         (write_test_result):
2190         (TestResultWriter.__init__):
2191         (TestResultWriter._make_output_directory):
2192         (TestResultWriter.output_filename):
2193         (TestResultWriter._output_testname):
2194         (TestResultWriter.write_output_files):
2195         (TestResultWriter.write_stderr):
2196         (TestResultWriter.write_crash_report):
2197         (TestResultWriter.create_text_diff_and_write_result):
2198         (TestResultWriter.write_image_diff_files):
2199         (copy_file):
2200         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
2201         (TestResultWriterTest.test_reftest_diff_image):
2202
2203 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2204
2205         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
2206         https://bugs.webkit.org/show_bug.cgi?id=81598
2207
2208         Reviewed by Eric Seidel.
2209
2210         "DumpRenderTree" shouldn't be hard-coded here, since some other
2211         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
2212         
2213         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2214         (TestExpectations):
2215
2216 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2217
2218         webkitpy: allow instance data in TestFailure objects
2219         https://bugs.webkit.org/show_bug.cgi?id=81580
2220
2221         Reviewed by Tony Chang.
2222
2223         Current TestFailures are basically enums, but it would be nice,
2224         especially for crashes, if they contained the name and pid of
2225         the process that crashed.
2226
2227         * Scripts/webkitpy/layout_tests/models/test_failures.py:
2228         (TestFailure.message):
2229         (FailureTimeout.message):
2230         (FailureCrash):
2231         (FailureCrash.__init__):
2232         (FailureCrash.message):
2233         (FailureMissingResult.message):
2234         (FailureTextMismatch.message):
2235         (FailureMissingImageHash.message):
2236         (FailureMissingImage.message):
2237         (FailureImageHashMismatch.message):
2238         (FailureImageHashIncorrect.message):
2239         (FailureReftestMismatch.message):
2240         (FailureReftestMismatchDidNotOccur.message):
2241         (FailureReftestNoImagesGenerated.message):
2242         (FailureMissingAudio.message):
2243         (FailureAudioMismatch.message):
2244         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
2245         (TestFailuresTest.test_equals):
2246         (TestFailuresTest):
2247         (TestFailuresTest.test_crashes):
2248
2249 2012-03-19  Brady Eidson  <beidson@apple.com>
2250
2251         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
2252         Crash in 3rd party WebKit apps under XHR/Cache code
2253
2254         Reviewed by Antti Koivisto.
2255
2256         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2257         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
2258         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
2259         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
2260         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
2261         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
2262         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
2263         (TestWebKitAPI):
2264         (TestWebKitAPI::TEST):
2265
2266 2012-03-19  Brady Eidson  <beidson@apple.com>
2267
2268         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
2269         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
2270
2271         Reviewed by Enrica Casucci.
2272
2273         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
2274         are in the correct format.
2275
2276         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2277         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
2278         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
2279         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
2280         (TestWebKitAPI::contextMenuCopyLink):
2281         (TestWebKitAPI::TEST):
2282
2283 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2284
2285         Make the selected item in the accordion a link to the flakiness dashboard
2286         https://bugs.webkit.org/show_bug.cgi?id=81571
2287
2288         Reviewed by Dimitri Glazkov.
2289
2290         Also, make it so that you can select the text of the link, instead of the current behavior
2291         of acting as if you had clicked on the accordion header.
2292
2293         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2294         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2295
2296 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2297
2298         NRWT runs some tests that are skipped with -i command line option
2299         https://bugs.webkit.org/show_bug.cgi?id=81535
2300
2301         Reviewed by Ojan Vafai.
2302
2303         This change modifies the interaction of Skipped files and
2304         test_expectations files so that entries in Skipped files (and
2305         the -i command line) override *everything* in the expectations
2306         file.
2307
2308         Specifically, a directory in a Skipped file will cause all of
2309         the tests in the dir to be skipped even if individual tests in
2310         the dir are listed in the test_expectations.txt.
2311
2312         Skipped files also override anything in an overrides files.
2313
2314         This seems to make more intuitive sense, since if you list
2315         something in the Skips file (and even more specify it on the
2316         command line) you probably want it to be universally applied.
2317
2318         Theoretically we could add more precedence levels and have full
2319         paths in an expectations file override dirs in a Skipped file
2320         (but not the command line), but I don't know yet that that level
2321         of complexity is justified.
2322         
2323         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2324         (TestExpectations.__init__):
2325         (TestExpectations._add_skipped_tests):
2326         (TestExpectations._add_test):
2327         (TestExpectations._add_expectations):
2328         (TestExpectations._add_expectation_line):
2329         (TestExpectations._already_seen_better_match):
2330         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2331         (test_more_specific_override_resets_skip):
2332         (SkippedTests):
2333         (SkippedTests.get_exp):
2334         (SkippedTests.assert_exp):
2335         (SkippedTests.test_skipped_tests_work):
2336         (SkippedTests.test_duplicate_skipped_test_fails_lint):
2337         (SkippedTests.test_skipped_file_overrides_expectations):
2338         (SkippedTests.test_skipped_dir_overrides_expectations):
2339         (SkippedTests.test_skipped_file_overrides_overrides):
2340         (SkippedTests.test_skipped_dir_overrides_overrides):
2341
2342 2012-03-19  Eric Seidel  <eric@webkit.org>
2343
2344         Fix WTF header include discipline in Chromium WebKit
2345         https://bugs.webkit.org/show_bug.cgi?id=81281
2346
2347         Reviewed by James Robinson.
2348
2349         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2350         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
2351
2352 2012-03-19  Andy Chen  <andchen@rim.com>
2353
2354         [BlackBerry] Add new layout test for text match marker
2355         https://bugs.webkit.org/show_bug.cgi?id=81531
2356
2357         Make the LayoutTestControllerBlackBerry::findString to test the
2358         BlackBerry port implementation.
2359
2360         Reviewed by Rob Buis.
2361
2362         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2363         (LayoutTestController::findString):
2364
2365 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2366
2367         Don't show WONTFIX tests in garden-o-matic
2368         https://bugs.webkit.org/show_bug.cgi?id=81554
2369
2370         Reviewed by Dimitri Glazkov.
2371
2372         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2373         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2374
2375 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
2376
2377         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
2378         https://bugs.webkit.org/show_bug.cgi?id=81530
2379
2380         Reviewed by Jessie Berlin.
2381
2382         Add rowAtIndex to WebKitTestRunner.
2383
2384         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
2385         (WTR::AccessibilityUIElement::rowAtIndex):
2386         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
2387         (AccessibilityUIElement):
2388         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
2389         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2390         (WTR):
2391         (WTR::AccessibilityUIElement::rowAtIndex):
2392
2393 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2394
2395         Change garden-o-matic All Failures tab to Expected Failures
2396         https://bugs.webkit.org/show_bug.cgi?id=81540
2397
2398         Reviewed by Dimitri Glazkov.
2399
2400         Also, fix some broken unittests that had just gotten stale.
2401
2402         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
2403         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
2404         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
2405         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2406         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2407         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2408         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2409         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
2410         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
2411         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2412
2413 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
2414
2415         [chromium] DRT crashes at shutdown.
2416         https://bugs.webkit.org/show_bug.cgi?id=81504
2417
2418         Reviewed by James Robinson.
2419
2420         WebCompositor must be destroyed after WebView is properly destroyed,
2421         or some compositor objects like CCLayerTreeHost may not be destroyed.
2422         However, it is really hard to make the destruction order correct
2423         explicitly in ~TestShell. So move it out to DumpRenderTree.
2424
2425         * DumpRenderTree/chromium/DumpRenderTree.cpp:
2426         (main):
2427         * DumpRenderTree/chromium/TestShell.cpp:
2428         (TestShell::~TestShell):
2429
2430 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2431
2432         add dpranke to webkitpy watchlist
2433         https://bugs.webkit.org/show_bug.cgi?id=81433
2434
2435         Reviewed by Benjamin Poulain.
2436
2437         * Scripts/webkitpy/common/config/watchlist:
2438
2439 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2440
2441         test-webkitpy is failing on SL/Python 2.6.1
2442         https://bugs.webkit.org/show_bug.cgi?id=81533
2443
2444         Reviewed by Ojan Vafai.
2445
2446         It appears that itertools.combinations() is buggy in 2.6.1; this
2447         backports the implementation from 2.7.
2448
2449         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2450         (TestConfigurationConverter.collapse_macros):
2451         (TestConfigurationConverter):
2452         (TestConfigurationConverter.combinations):
2453
2454 2012-03-19  David Michael Barr  <davidbarr@google.com>
2455
2456         Optimize the cleanup sequence in the start-queue.sh script
2457         https://github.com/abarth/webkit/pull/1
2458
2459         Reviewed by Adam Barth.
2460
2461         Updating the working copy is an expensive operation
2462         for a repository the size of WebKit.
2463         Attempt to reset only to the target revision.
2464
2465         * EWSTools/start-queue.sh:
2466
2467 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
2468
2469         Adjust architecture type of the EFL build bot.
2470         https://bugs.webkit.org/show_bug.cgi?id=81452
2471
2472         Reviewed by Tony Chang.
2473
2474         * BuildSlaveSupport/build.webkit.org-config/config.json: The
2475         buildslave is a 64-bit machine, therefore describe it as being
2476         x86_64, not i386.
2477
2478 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2479
2480         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
2481         https://bugs.webkit.org/show_bug.cgi?id=81514
2482
2483         Reviewed by Simon Hausmann.
2484
2485         * MiniBrowser/qt/qml/BrowserWindow.qml:
2486
2487 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2488
2489         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
2490         https://bugs.webkit.org/show_bug.cgi?id=81491
2491
2492         Reviewed by Philippe Normand.
2493
2494         For the find_path(), find_library() etc CMake calls to find the
2495         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
2496         set, otherwise CMake cannot know the dependencies exist and will only
2497         look in the standard directories.
2498
2499         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
2500         the jhbuild Root directory.
2501
2502 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2503
2504         [EFL] Add XDG env vars to jhbuildrc
2505         https://bugs.webkit.org/show_bug.cgi?id=81510
2506
2507         Needed to ensure correct glib behaviour in jhbuild environment.
2508
2509         Reviewed by Philippe Normand.
2510
2511         * efl/jhbuildrc:
2512
2513 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2514
2515         Select best target for tap gesture.
2516         https://bugs.webkit.org/show_bug.cgi?id=78801
2517
2518         Reviewed by Kenneth Rohde Christiansen.
2519
2520         Add TOUCH_ADJUSTMENT to enabled features.
2521
2522         * qmake/mkspecs/features/features.prf:
2523
2524 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
2525
2526         [chromium] synthesize wheel events for fling on main thread
2527         https://bugs.webkit.org/show_bug.cgi?id=81458
2528
2529         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
2530
2531         Reviewed by James Robinson.
2532
2533         * DumpRenderTree/chromium/EventSender.cpp:
2534         (EventSender::EventSender):
2535         (EventSender::gestureFlingCancel):
2536         (EventSender::gestureFlingStart):
2537         * DumpRenderTree/chromium/EventSender.h:
2538         (EventSender):
2539
2540 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2541
2542         [Qt] Add command line option to set the user agent in the MiniBrowser
2543         https://bugs.webkit.org/show_bug.cgi?id=81508
2544
2545         Reviewed by Simon Hausmann.
2546
2547         * MiniBrowser/qt/BrowserWindow.cpp:
2548         (BrowserWindow::BrowserWindow):
2549         (BrowserWindow::webViewExperimental):
2550         * MiniBrowser/qt/BrowserWindow.h:
2551         (BrowserWindow):
2552         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2553         (printHelp):
2554         (MiniBrowserApplication::handleUserOptions):
2555         * MiniBrowser/qt/MiniBrowserApplication.h:
2556         (WindowOptions::WindowOptions):
2557         (WindowOptions::setUserAgent):
2558         (WindowOptions::userAgent):
2559         (WindowOptions):
2560
2561 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2562
2563         [Qt] Remove dead user agent code from the MiniBrowser.
2564
2565         Reviewed by Simon Hausmann.
2566
2567         * MiniBrowser/qt/BrowserWindow.cpp:
2568         * MiniBrowser/qt/BrowserWindow.h:
2569         (BrowserWindow):
2570         * MiniBrowser/qt/MiniBrowser.qrc:
2571         * MiniBrowser/qt/useragentlist.txt: Removed.
2572
2573 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2574
2575         jhbuild: Call jhbuild's autogen.sh via bash.
2576         https://bugs.webkit.org/show_bug.cgi?id=81485
2577
2578         Reviewed by Philippe Normand.
2579
2580         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
2581         this is usually true only on Linux systems, and even then bash for
2582         some reason might not be available in that location. It is safer to
2583         call bash directly and pass autogen.sh to it instead.
2584
2585         * jhbuild/jhbuild-wrapper:
2586         (install_jhbuild):
2587
2588 2012-03-19  Philippe Normand  <pnormand@igalia.com>
2589
2590         [GTK] Prefix run-launcher with run-with-jhbuild
2591         https://bugs.webkit.org/show_bug.cgi?id=81503
2592
2593         Reviewed by Hajime Morita.
2594
2595         * Scripts/run-launcher:
2596
2597 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2598
2599         [EFL] Prefix run-launcher with run-with-jhbuild
2600         https://bugs.webkit.org/show_bug.cgi?id=80026
2601
2602         Reviewed by Hajime Morita.
2603
2604         * Scripts/run-launcher:
2605
2606 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
2607
2608         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
2609         https://bugs.webkit.org/show_bug.cgi?id=81498
2610
2611         * qmake/mkspecs/features/unix/default_post.prf:
2612
2613 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2614
2615         Unreviewed; switch to using my FreeBSD.org email address, the
2616         other one is going to disappear soon.
2617
2618         * Scripts/webkitpy/common/config/committers.py:
2619         * Scripts/webkitpy/common/config/watchlist:
2620
2621 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
2622
2623         Unreviewed; fix webkitpy unit test failures after r111116.
2624
2625         It turns out that cmd_line in the patch in bug 80025 had the right
2626         signature when it was submitted, but r109242 added a few more
2627         required parameters, and since the method (and EflDriver) did not
2628         exist at the time we ended up committing a broken method.
2629
2630         * Scripts/webkitpy/layout_tests/port/efl.py:
2631         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
2632         arguments.
2633
2634 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
2635
2636         [BlackBerry] Enable Web Timing for performance profiling and improvement.
2637         https://bugs.webkit.org/show_bug.cgi?id=81085
2638
2639         Reviewed by Rob Buis.
2640
2641         * Scripts/build-webkit:
2642
2643 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2644
2645         Add a dependencies installation step for EFL to buildbot master.cfg
2646         https://bugs.webkit.org/show_bug.cgi?id=81337
2647
2648         Reviewed by Gustavo Noronha Silva.
2649
2650         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2651         (InstallEflDependencies):
2652         (Factory.__init__):
2653
2654 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2655
2656         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
2657         https://bugs.webkit.org/show_bug.cgi?id=80025
2658
2659         Reviewed by Gustavo Noronha Silva.
2660
2661         * Scripts/webkitpy/layout_tests/port/efl.py:
2662         (EflDriver):
2663         (EflDriver.cmd_line):
2664         (EflPort._driver_class):
2665
2666 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2667
2668         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
2669         https://bugs.webkit.org/show_bug.cgi?id=79904
2670
2671         build-webkit --efl will use jhbuild for fetching and building dependencies.
2672
2673         Reviewed by Gustavo Noronha Silva.
2674
2675         * Scripts/update-webkitefl-libs: Added.
2676         * Scripts/webkitdirs.pm:
2677         (jhbuildWrapperPrefixIfNeeded):
2678         (generateBuildSystemFromCMakeProject):
2679         (buildCMakeGeneratedProject):
2680         * efl/common.py: Added.
2681         (script_path):
2682         (top_level_path):
2683         (number_of_cpus):
2684         * efl/jhbuild.modules: Added.
2685         * efl/jhbuildrc: Added.
2686         * efl/run-with-jhbuild: Added.
2687
2688 2012-03-16  Stephanie Lewis  <slewis@apple.com>
2689
2690         https://bugs.webkit.org/show_bug.cgi?id=81065
2691         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
2692         Fix crash preventing WebKitTestRunner from working in Release.  
2693         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
2694         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
2695         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
2696
2697         I removed ostringstream in favor of StringBuilder.  
2698
2699         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
2700
2701         Reviewed by Geoff Garen.
2702
2703         Add DEBUG_DEFINES to set NDEBUG on release builds.
2704         * WebKitTestRunner/Configurations/Base.xcconfig:
2705         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2706
2707         Remove ostringstream member in favor of StringBuilder.
2708         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2709         (WTR::InjectedBundle::initialize):
2710         (WTR::InjectedBundle::beginTesting):
2711         (WTR::InjectedBundle::done):
2712         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2713         (WTF):
2714         (WTR::InjectedBundle::stringBuilder):
2715         (InjectedBundle):
2716
2717         Append output to StringBuilder.
2718         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2719         (WTR::hasPrefix):
2720         (WTR::dumpPath):
2721         (WTR::rangeToStr):
2722         (WTR::styleDecToStr):
2723         (WTR::frameToStr):
2724         (WTR::dumpFrameScrollPosition):
2725         (WTR::dumpFrameText):
2726         (WTR::dumpDescendantFramesText):
2727         (WTR::InjectedBundlePage::dump):
2728         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
2729         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
2730         (WTR::InjectedBundlePage::willSendRequestForFrame):
2731         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
2732         (WTR::lastFileURLPathComponent):
2733         (WTR::InjectedBundlePage::willAddMessageToConsole):
2734         (WTR::InjectedBundlePage::willSetStatusbarText):
2735         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
2736         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
2737         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
2738         (WTR::InjectedBundlePage::shouldBeginEditing):
2739         (WTR::InjectedBundlePage::shouldEndEditing):
2740         (WTR::InjectedBundlePage::shouldInsertNode):
2741         (WTR::InjectedBundlePage::shouldInsertText):
2742         (WTR::InjectedBundlePage::shouldDeleteRange):
2743         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
2744         (WTR::InjectedBundlePage::shouldApplyStyle):
2745         (WTR::InjectedBundlePage::didBeginEditing):
2746         (WTR::InjectedBundlePage::didEndEditing):
2747         (WTR::InjectedBundlePage::didChange):
2748         (WTR::InjectedBundlePage::didChangeSelection):
2749         (WTR::InjectedBundlePage::supportsFullScreen):
2750         (WTR::InjectedBundlePage::enterFullScreenForElement):
2751         (WTR::InjectedBundlePage::exitFullScreenForElement):
2752         (WTR::InjectedBundlePage::beganEnterFullScreen):
2753         (WTR::InjectedBundlePage::beganExitFullScreen):
2754         (WTR::dumpBackForwardListItem):
2755         (WTR::InjectedBundlePage::dumpBackForwardList):
2756         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2757         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
2758         (WTR::LayoutTestController::dumpConfigurationForViewport):
2759         
2760         Convert between WTF::String and WKStringRef.
2761         * WebKitTestRunner/StringFunctions.h:
2762         (WTR::toWTFString):
2763         (WTR):
2764         * WebKitTestRunner/TestInvocation.cpp:
2765         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2766
2767 2012-03-16  Eric Seidel  <eric@webkit.org>
2768
2769         Add a land-from-url command
2770         https://bugs.webkit.org/show_bug.cgi?id=81411
2771
2772         Reviewed by Adam Barth.
2773
2774         Very basic so far.  The goal is to make it landing smarter
2775         so that users don't have to bother to parse out the ids out of the urls.
2776
2777         * Scripts/webkitpy/common/config/urls.py:
2778         (parse_bug_id):
2779         (parse_attachment_id):
2780         * Scripts/webkitpy/common/config/urls_unittest.py:
2781         (URLsTest.test_parse_bug_id):
2782         (URLsTest):
2783         (URLsTest.test_parse_attachment_id):
2784         * Scripts/webkitpy/tool/commands/download.py:
2785         (ProcessURLsMixin):
2786         (ProcessURLsMixin._fetch_list_of_patches_to_process):
2787         (LandFromURL):
2788         * Scripts/webkitpy/tool/commands/download_unittest.py:
2789         (test_land_from_bug):
2790         (test_land_from_url):
2791
2792 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
2793
2794         build-webkit launcher instructions use wrong relative path for Qt
2795         https://bugs.webkit.org/show_bug.cgi?id=81357
2796
2797         Reviewed by Eric Seidel.
2798
2799         Ensure we are at the proper directory before generating run-launcher
2800         instructions.
2801
2802         * Scripts/build-webkit:
2803
2804 2012-03-16  Brady Eidson  <beidson@apple.com>
2805
2806         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
2807         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
2808
2809         Reviewed by Geoff Garen.
2810
2811         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2812         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
2813
2814 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2815
2816         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
2817         https://bugs.webkit.org/show_bug.cgi?id=81407
2818
2819         Reviewed by Adam Barth.
2820
2821         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2822         (summarize_results):
2823         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2824         (ResultSummaryTest.summarized_results):
2825         (ResultSummaryTest.test_no_svn_revision):
2826         (ResultSummaryTest.test_svn_revision):
2827         (ResultSummaryTest):
2828         (ResultSummaryTest.test_summarized_results_wontfix):
2829
2830 2012-03-16  Ojan Vafai  <ojan@chromium.org>
2831
2832         Have webkit-patch rebaseline-test update test_expectations.txt
2833         https://bugs.webkit.org/show_bug.cgi?id=81054
2834
2835         Recommit http://trac.webkit.org/changeset/110783 now that
2836         http://trac.webkit.org/changeset/111034 has landed.
2837
2838         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2839         (TestExpectations.remove_configuration_from_test):
2840         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2841         (RemoveConfigurationsTest):
2842         (RemoveConfigurationsTest.test_remove):
2843         (test_remove_line):
2844         * Scripts/webkitpy/tool/commands/rebaseline.py:
2845         (RebaselineTest._update_expectations_file):
2846         (RebaselineTest._rebaseline_test_and_update_expectations):
2847         (RebaselineTest):
2848         (RebaselineTest.execute):
2849         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2850         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2851         (test_rebaseline_updates_expectations_file):
2852
2853 2012-03-15  Ojan Vafai  <ojan@chromium.org>
2854
2855         Specifier collapsing when writing test expectations lines gets a number of cases wrong
2856         https://bugs.webkit.org/show_bug.cgi?id=81309
2857
2858         Reviewed by Dimitri Glazkov.
2859
2860         I've run this over all the lines in the current Chromium test_expectations.txt file,
2861         so I'm relatively confident we now cover all the cases.
2862
2863         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2864         (TestConfigurationConverter.__init__):
2865         (TestConfigurationConverter.collapse_macros):
2866         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
2867         (TestConfigurationConverter):
2868         (TestConfigurationConverter.intersect_combination):
2869         (TestConfigurationConverter.symmetric_difference):
2870         (TestConfigurationConverter.to_specifiers_list):
2871         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
2872         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
2873         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2874         (make_mock_all_test_configurations_set):
2875         (TestConfigurationConverterTest.test_symmetric_difference):
2876         (TestConfigurationConverterTest.test_to_config_set):
2877         (TestConfigurationConverterTest.test_macro_expansion):
2878         (TestConfigurationConverterTest.test_to_specifier_lists):
2879         (TestConfigurationConverterTest.test_converter_macro_collapsing):
2880         * Scripts/webkitpy/layout_tests/port/chromium.py:
2881         (ChromiumPort):
2882         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2883         (ChromiumAndroidPort.__init__):
2884         The android port uses "arm" as it's architecture, which is technically correct,
2885         but considerably complicates making collapsing work. We probably should kill
2886         the concept of architecture entirely. The benefits are not worth the code
2887         complexity.
2888
2889         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2890         (ChromiumPortTest.test_all_test_configurations):
2891         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
2892         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
2893         (test_rebaseline_updates_expectations_file):
2894
2895 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
2896
2897         [Qt][Wk2] Assertion Failure and crash on file upload
2898         https://bugs.webkit.org/show_bug.cgi?id=80854
2899
2900         Reviewed by Simon Hausmann.
2901
2902         Added filePicker to WebView using experimental API.
2903
2904         * MiniBrowser/qt/MiniBrowser.qrc:
2905         * MiniBrowser/qt/icons/folder.png: Added.
2906         * MiniBrowser/qt/icons/titlebar.png: Added.
2907         * MiniBrowser/qt/icons/up.png: Added.
2908         * MiniBrowser/qt/qml/BrowserWindow.qml:
2909         * MiniBrowser/qt/qml/FilePicker.qml: Added.
2910
2911 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
2912
2913         Updating email for committer.py script.
2914
2915         * Scripts/webkitpy/common/config/committers.py:
2916
2917 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
2918
2919         [chromium] DRT: Add support for sending gesture scrolls without a begin
2920         https://bugs.webkit.org/show_bug.cgi?id=81292
2921
2922         Reviewed by Kent Tamura.
2923
2924         * DumpRenderTree/chromium/EventSender.cpp:
2925         (EventSender::EventSender):
2926         (EventSender::gestureScrollFirstPoint):
2927         * DumpRenderTree/chromium/EventSender.h:
2928         (EventSender):
2929
2930 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
2931
2932         [GTK] WebKitIconDatabase doesn't keep icons cached
2933         https://bugs.webkit.org/show_bug.cgi?id=56200
2934
2935         Reviewed by Martin Robinson.
2936
2937         Add support for favicons in GtkLauncher.
2938
2939         * GtkLauncher/main.c:
2940         (activateUriEntryCb):
2941         (iconLoadedCb):
2942         (createBrowser):
2943
2944 2012-03-16  Adam Barth  <abarth@webkit.org>
2945
2946         Remove sheriff-bot's last-green-revision command
2947         https://bugs.webkit.org/show_bug.cgi?id=81314
2948
2949         Reviewed by Eric Seidel.
2950
2951         This command is a cruel joke.  WebKit doesn't really ever have a
2952         last-green-revision.
2953
2954         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2955         (BuildBot._find_green_revision):
2956         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2957         (MockBuildBot.builder_statuses):
2958         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2959         (test_find_green_revision):
2960         * Scripts/webkitpy/tool/bot/irc_command.py:
2961         (IRCCommand.execute):
2962         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2963         (SheriffIRCBotTest.test_help):
2964         * Scripts/webkitpy/tool/commands/queries.py:
2965         (PatchesToReview.execute):
2966
2967 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2968
2969         [Qt] Make gccdepends test aware of broken icecream deps generation
2970
2971         Icecream seems to build the files without passing -o to gcc with
2972         the original destination path, which means the target name in the
2973         dependency file is wrong.
2974
2975         Also, echo something to the final file instead of just touching it,
2976         so it's not deleted on every build.
2977
2978         Reviewed by Simon Hausmann.
2979
2980         * qmake/config.tests/gccdepends/gccdepends.pro:
2981
2982 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
2983
2984         Support for Battery Status API
2985         https://bugs.webkit.org/show_bug.cgi?id=62698
2986
2987         Add include directory and file.
2988         http://www.w3.org/TR/battery-status/
2989
2990         Reviewed by Adam Barth.
2991
2992         * Scripts/build-webkit:
2993
2994 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
2995
2996         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
2997         https://bugs.webkit.org/show_bug.cgi?id=81007
2998
2999         Reviewed by Adam Barth.
3000
3001         A dedidcated build slave is available for efl on build.webkit.org. 
3002         This change add's an entry for the elf port in list of available port builders.
3003         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
3004
3005         * Scripts/webkitpy/layout_tests/port/builders.py:
3006
3007 2012-03-16  Adam Barth  <abarth@webkit.org>
3008
3009         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
3010         https://bugs.webkit.org/show_bug.cgi?id=81302
3011
3012         Reviewed by Eric Seidel.
3013
3014         This is a trivial patch, but the variable name turned up in a grep of
3015         webkitpy and looks like copy pasta.
3016
3017         * Scripts/webkitpy/common/net/statusserver.py:
3018         (StatusServer.next_work_item):
3019
3020 2012-03-16  Andrew Lo  <anlo@rim.com>
3021
3022         [BlackBerry] Enable PAGE_VISIBILITY_API
3023         https://bugs.webkit.org/show_bug.cgi?id=81257
3024
3025         Enable Page Visibility API for BlackBerry.
3026
3027         Reviewed by Rob Buis.
3028
3029         * Scripts/build-webkit:
3030
3031 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
3032
3033         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
3034         https://bugs.webkit.org/show_bug.cgi?id=81232
3035
3036         Reviewed by Philippe Normand.
3037
3038         Use the newly-added helper method in DumpRenderTreeSupportGtk.
3039
3040         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3041         (LayoutTestController::computedStyleIncludingVisitedInfo):
3042
3043 2012-03-15  Eric Seidel  <eric@webkit.org>
3044
3045         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
3046         https://bugs.webkit.org/show_bug.cgi?id=81313
3047
3048         Reviewed by Adam Barth.
3049
3050         * Scripts/webkitpy/common/checkout/changelog.py:
3051         (parse_bug_id_from_changelog):
3052         * Scripts/webkitpy/common/config/urls.py:
3053         (parse_bug_id):
3054         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
3055         (URLsTest):
3056         (URLsTest.test_parse_bug_id):
3057         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
3058         (test_parse_bug_id):
3059         * Scripts/webkitpy/tool/bot/irc_command.py:
3060         (_post_error_and_check_for_bug_url):
3061         * Scripts/webkitpy/tool/bot/sheriff.py:
3062         (Sheriff.post_rollout_patch):
3063         (Sheriff.post_chromium_deps_roll):
3064
3065 2012-03-15  Adam Barth  <abarth@webkit.org>
3066
3067         The commit-queue should fast-track patches that have already passed the testing EWS bots
3068         https://bugs.webkit.org/show_bug.cgi?id=81305
3069
3070         Reviewed by Eric Seidel.
3071
3072         If a patch has already passed the testing EWS bots (currently only the
3073         chromium-ews bot), we should be able to land the patch without running
3074         the tests again.  This patch is somewhat risky in that the tests might
3075         have been broken between when the EWS bots ran and when the
3076         commit-queue is processing the patch.  If that turns out to be a
3077         problem in practice, we'll tighten up this fast-track in some way,
3078         likely by making sure that the pass status is recent.
3079
3080         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
3081         (CommitQueueTaskDelegate.did_pass_testing_ews):
3082         (CommitQueueTask._did_pass_tests_recently):
3083         (CommitQueueTask.run):
3084         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3085         (MockCommitQueue.did_pass_testing_ews):
3086         (test_fast_success_case):
3087         * Scripts/webkitpy/tool/commands/queues.py:
3088         (CommitQueue.did_pass_testing_ews):
3089         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3090         (test_did_pass_testing_ews):
3091
3092 2012-03-15  Kentaro Hara  <haraken@chromium.org>
3093
3094         [PerformanceTests] run-perf-tests should output correct units
3095         https://bugs.webkit.org/show_bug.cgi?id=78303
3096
3097         Reviewed by Ryosuke Niwa.
3098
3099         Previously run-perf-tests just supported "ms" for units.
3100         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
3101         This patch fixes run-perf-tests so that they can support custom units.
3102         The default unit is "ms".
3103
3104         The test result looks like this:
3105
3106         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
3107         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
3108         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
3109
3110         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3111         (PerfTestsRunner._process_parser_test_result):
3112
3113 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
3114
3115         [WinCairo] Unreviewed build change after wtf path changes.
3116
3117         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
3118         directories to look at the private header directory to find 'wtf'.
3119
3120 2012-03-15  Kent Tamura  <tkent@chromium.org>
3121
3122         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
3123         https://bugs.webkit.org/show_bug.cgi?id=81293
3124
3125         Reviewed by Adam Barth.
3126
3127         r110595 introduced new format of the author field of ChangeLog entry:
3128         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
3129         Updte chanelog.py to support it.
3130
3131         * Scripts/webkitpy/common/checkout/changelog.py:
3132         (ChangeLogEntry._split_contributor_names):
3133         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
3134         (test_parse_authors):
3135
3136 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
3137
3138         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
3139
3140         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
3141
3142         Reviewed by Jessie Berlin.
3143
3144         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3145         (LayoutTestController::setValueForUser):
3146         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3147         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3148         (WTR::LayoutTestController::setValueForUser):
3149         (WTR):
3150         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3151         (LayoutTestController):
3152
3153 2012-03-15  Adam Barth  <abarth@webkit.org>
3154
3155         Update the start-queue.sh script to match what we're using on the bots
3156         https://bugs.webkit.org/show_bug.cgi?id=81277
3157
3158         Reviewed by Eric Seidel.
3159
3160         Over time we've added more magic git commands to clean up the local repo after
3161         it has been borked in various ways. The idea behind this sequence is to clear
3162         out any changes in the working copy and then to rebuild the master branch from
3163         origin/master.
3164
3165         This patch also removes the watchdog timer that we've stopped using on the
3166         bots.
3167
3168         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
3169         Note: Eric and I are experimenting with GitHub to see if it might be
3170         useful for WebKit.  If you have thoughts on this topic, please feel
3171         free to grab Eric or me on #webkit to chat.  I assume that he and I are
3172         the only ones who care about this file so using this alternative
3173         process for this patch shouldn't cause any trouble.
3174
3175         * EWSTools/start-queue.sh:
3176
3177 2012-03-15  Adam Barth  <abarth@webkit.org>
3178
3179         The commit-queue can hang when the test_expectations.txt style check fails during commit
3180         https://bugs.webkit.org/show_bug.cgi?id=81251
3181
3182         Reviewed by Eric Seidel.
3183
3184         We forgot to check the --non-interactive flag, so we end up prompting
3185         the "user" who doesn't exist on the bot.  We might consider teaching
3186         the User class about non-interactive if we have this sort of bug again.
3187
3188         * Scripts/webkitpy/tool/steps/commit.py:
3189         (Commit.options):
3190         (Commit._check_test_expectations):
3191         (Commit.run):
3192         * Scripts/webkitpy/tool/steps/commit_unittest.py:
3193         (CommitTest.test_check_test_expectations):
3194
3195 2012-03-15  Jessie Berlin  <jberlin@apple.com>
3196
3197         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
3198         Mac WK2 because WebAudio is not enabled
3199         https://bugs.webkit.org/show_bug.cgi?id=81259
3200
3201         Reviewed by Dan Bernstein.
3202
3203         Enable WebAudio so that it appears in the window object in the tests.
3204         The webaudio tests are all still skipped in the Mac Skipped list.
3205
3206         * WebKitTestRunner/TestController.cpp:
3207         (WTR::TestController::resetStateToConsistentValues):
3208
3209 2012-03-15  David Levin  <levin@chromium.org>
3210
3211         Various autoresize fixes.
3212         https://bugs.webkit.org/show_bug.cgi?id=81199
3213
3214         Reviewed by Darin Fisher.
3215
3216         * DumpRenderTree/chromium/LayoutTestController.cpp:
3217         (LayoutTestController::reset): Stop using the deprecated api that was removed.
3218
3219 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3220
3221         [GTK] Actually use --release and --debug command line options in run-gtk-test
3222         https://bugs.webkit.org/show_bug.cgi?id=81234
3223
3224         Reviewed by Philippe Normand.
3225
3226         Actually handle --release and --debug command line options to
3227         create the build directory, falling back to current hack to detect
3228         whether it's a Release or Debug build when both options are missed.
3229
3230         * Scripts/run-gtk-tests:
3231         (TestRunner.__init__): Receive options too.
3232         (TestRunner._get_top_level_directory): Helper function to get the
3233         top level directory.
3234         (TestRunner._get_build_directory): Helper function to get the
3235         build directory depending on --release/--debug command line
3236         options or gessing it if both options are missing.
3237         (TestRunner._ensure_accessibility_daemon_is_running): Use
3238         self-options now.
3239         (TestRunner.run): Ditto.
3240         (TestRunner.run.run_tests): Pass options to the constructor
3241         instead of run().
3242
3243 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3244
3245         [GTK] Run unit tests with --verbose in the bots
3246         https://bugs.webkit.org/show_bug.cgi?id=81230
3247
3248         Reviewed by Martin Robinson.
3249
3250         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3251         (RunGtkAPITests): Add --verbose to run-gtk-test command.
3252
3253 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
3254
3255         Move wx port to using export macros
3256         https://bugs.webkit.org/show_bug.cgi?id=77279
3257
3258         Reviewed by Hajime Morita.
3259
3260         * waf/build/settings.py:
3261         (common_configure):
3262
3263 2012-03-15  Philippe Normand  <pnormand@igalia.com>
3264
3265         Unreviewed, GTK build fix attempt after r110845.
3266
3267         * Scripts/run-gtk-tests: Added --release and --debug options.
3268
3269 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3270
3271         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
3272         https://bugs.webkit.org/show_bug.cgi?id=79673
3273
3274         Preparatory work for bringing jhbuild enabled build to EFL port.
3275
3276         Reviewed by Philippe Normand.
3277
3278         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
3279         * Scripts/update-webkitgtk-libs:
3280         * gtk/run-with-jhbuild:
3281         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
3282         (top_level_path):
3283         (jhbuild_installed):
3284         (jhbuild_cloned):
3285         (jhbuild_at_expected_revision):
3286         (update_jhbuild):
3287         (clone_jhbuild):
3288         (install_jhbuild):
3289         (update_webkit_libs_jhbuild):
3290         (determine_platform):
3291         (ensure_jhbuild):
3292
3293 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3294
3295         [GTK] Add command line options to run-gtk-tests script
3296         https://bugs.webkit.org/show_bug.cgi?id=81218
3297
3298         Reviewed by Philippe Normand.
3299
3300         Following command line options have been added:
3301
3302         -v, --verbose: to run gtester with --verbose. That would be used
3303         by the bots to make it easier to know which test case it failing.
3304
3305         --display: to use a different display for Xvfb
3306
3307         It's also possible to pass one or more tests as arguments.
3308
3309         * Scripts/run-gtk-tests:
3310         (TestRunner.__init__): Receive command line arguments.
3311         (TestRunner._get_tests): Get all the tests or the ones passed by
3312         command line arguments.
3313         (TestRunner._ensure_accessibility_daemon_is_running): Use display
3314         command line argument.
3315         (TestRunner.run): Pas command line options.
3316         (TestRunner.run.run_tests): Pass command line arguments.
3317
3318 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3319
3320         [GTK] WebKit2APITests/TestPrinting timeout in the bots
3321         https://bugs.webkit.org/show_bug.cgi?id=81211
3322
3323         Reviewed by Philippe Normand.
3324
3325         Add gtk+ patch to fix timeout in the bots. A similar patch has
3326         already landed in updatream gtk+, see
3327         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
3328
3329         * gtk/jhbuild.modules:
3330         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
3331
3332 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
3333
3334         sheriffbot should also be addressable with a comma in addition to colon
3335         https://bugs.webkit.org/show_bug.cgi?id=81101
3336
3337         Reviewed by Adam Barth.
3338
3339         * Scripts/webkitpy/common/net/irc/ircbot.py:
3340         (IRCBot.on_pubmsg):
3341
3342 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
3343
3344         [Qt] MiniBrowser scroll indicators should not scroll out of view
3345         https://bugs.webkit.org/show_bug.cgi?id=81131
3346
3347         Reviewed by Kenneth Rohde Christiansen.
3348
3349         * MiniBrowser/qt/qml/ScrollIndicator.qml:
3350
3351 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
3352
3353         AX: ARIA tables need to support selected rows attribute/select rows changed notification
3354         https://bugs.webkit.org/show_bug.cgi?id=79821
3355
3356         Reviewed by Beth Dakin.
3357
3358         * DumpRenderTree/AccessibilityUIElement.cpp:
3359         (rowAtIndexCallback):
3360         (AccessibilityUIElement::getJSClass):
3361         * DumpRenderTree/AccessibilityUIElement.h:
3362         (AccessibilityUIElement):
3363         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
3364         (AccessibilityUIElement::rowAtIndex):
3365         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3366         (AccessibilityUIElement::rowAtIndex):
3367         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
3368         (AccessibilityUIElement::rowAtIndex):
3369         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
3370         (AccessibilityUIElement::rowAtIndex):
3371
3372 2012-03-14  Tim Horton  <timothy_horton@apple.com>
3373
3374         NRWT should respect --additional-platform-directory's Skipped list
3375         https://bugs.webkit.org/show_bug.cgi?id=81172
3376
3377         Reviewed by Timothy Hatcher.
3378
3379         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
3380         from said path are already used, but the Skipped file is ignored.
3381
3382         * Scripts/webkitpy/layout_tests/port/webkit.py:
3383         (WebKitPort._skipped_file_search_paths):
3384         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3385         (test_skipped_file_search_paths):
3386
3387 2012-03-14  Ojan Vafai  <ojan@chromium.org>
3388
3389         Unreviewed, rolling out r110783.
3390         http://trac.webkit.org/changeset/110783
3391         https://bugs.webkit.org/show_bug.cgi?id=81054
3392
3393         Caused existing test expectation lines to expand to one for
3394         each test configuration.
3395
3396         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3397         (TestExpectations.has_warnings):
3398         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3399         (test_macro_overrides):
3400         * Scripts/webkitpy/tool/commands/rebaseline.py:
3401         (RebaselineTest._save_baseline):
3402         (RebaselineTest._rebaseline_test):
3403         (RebaselineTest.execute):
3404         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3405         (TestRebaseline.test_tests_to_update):
3406
3407 2012-03-13  Jon Lee  <jonlee@apple.com>
3408
3409         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
3410         https://bugs.webkit.org/show_bug.cgi?id=80922
3411         <rdar://problem/11035082>
3412
3413         Reviewed by Jian Li.
3414
3415         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
3416         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
3417         new API. Therefore, APIs that are common between the two will have:
3418         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
3419
3420         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
3421         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
3422
3423         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
3424         * DumpRenderTree/chromium/LayoutTestController.cpp:
3425         (LayoutTestController::grantDesktopNotificationPermission):
3426         (LayoutTestController::simulateDesktopNotificationClick):
3427         * DumpRenderTree/chromium/NotificationPresenter.cpp:
3428         * DumpRenderTree/chromium/TestShell.cpp:
3429         (TestShell::TestShell):
3430         (TestShell::resetTestController):
3431         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
3432         (DRTDesktopNotificationPresenter::checkNotificationPermission):
3433
3434 2012-03-13  Ojan Vafai  <ojan@chromium.org>
3435
3436         Have webkit-patch rebaseline-test update test_expectations.txt
3437         https://bugs.webkit.org/show_bug.cgi?id=81054
3438
3439         Reviewed by Dirk Pranke.
3440
3441         This will allow us to use the All Failures tab in garden-o-matic
3442         to rebaseline tests without making test_expectations.txt stale.
3443
3444         It's conservative and only removes lines if the test is
3445         not flaky and fails in a way the rebaseline would fix (e.g.
3446         CRASH/TIMEOUT will not be touched).
3447
3448         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3449         (TestExpectations.remove_configuration_from_test):
3450         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3451         (RemoveConfigurationsTest):
3452         (RemoveConfigurationsTest.test_remove):
3453         (test_remove_line):
3454         * Scripts/webkitpy/tool/commands/rebaseline.py:
3455         (RebaselineTest._is_supported_port):
3456         (RebaselineTest):
3457         (RebaselineTest._update_expectations_file):
3458         (RebaselineTest._rebaseline_test_and_update_expectations):
3459         (RebaselineTest.execute):
3460         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3461         (TestRebaseline.test_rebaseline_updates_expectations_file):
3462
3463 2012-03-14  Jessie Berlin  <jberlin@apple.com>
3464
3465         Crash in WKTR under addChromeInputField due to using 0 as a key in a HashMap
3466         https://bugs.webkit.org/show_bug.cgi?id=81167
3467
3468         Reviewed by Sam Weinig.
3469
3470         Do not use 0 as a key in a HashMap.
3471
3472         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3473         cacheLayoutTestControllerCallback uses the enum as keys for the callback map.
3474
3475 2012-03-14  Ojan Vafai  <ojan@chromium.org>
3476
3477         Add the chromium lion bot to garden-o-matic and remove
3478         the platform_move_to in builders.py since the Chromium Lion
3479         port is now a normal port.
3480
3481         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
3482         * Scripts/webkitpy/layout_tests/port/builders.py:
3483
3484 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
3485
3486         Fix regression/crash introduced in r110724 - bad type passed from skipped_tests() by default.
3487
3488         Unreviewed, build fix.
3489
3490         * Scripts/webkitpy/layout_tests/port/base.py:
3491         (Port.skipped_tests):
3492
3493 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
3494
3495         Please add a way to manually skip some tests in NRWT
3496         https://bugs.webkit.org/show_bug.cgi?id=81019
3497
3498         Reviewed by Ojan Vafai.
3499
3500         This patch implements the -i / --ignore-tests flag from ORWT
3501         and refactors the test_expectations.py code slightly to handle
3502         it.
3503
3504         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3505         (Manager.parse_expectations):
3506         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3507         (TestExpectations.__init__):
3508         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3509         (test_add_skipped_tests):
3510         (test_add_skipped_tests_duplicate):
3511         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
3512         (parse_args):
3513         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3514         (MainTest.test_ignore_tests):
3515         (MainTest.test_ignore_tests.assert_ignored):
3516
3517 2012-03-14  Dirk Pranke  <dpranke@chromium.org>
3518
3519         nrwt: run with no args on lion, is putting baselines in platform/mac-lion, not platform/mac
3520         https://bugs.webkit.org/show_bug.cgi?id=81028
3521
3522         Reviewed by Ojan Vafai.
3523
3524         The mac port had incorrect lookups for 'mac-lion' and
3525         'mac-future' in some places. This patch fixes that and adds a
3526         bunch of tests.
3527
3528         * Scripts/webkitpy/layout_tests/port/mac.py:
3529         (MacPort.baseline_path):
3530         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3531         (MacTest.assert_skipped_file_search_paths):
3532         (MacTest.test_skipped_file_search_paths):
3533         (_assert_search_path):
3534         (test_baseline_search_path):
3535         * Scripts/webkitpy/layout_tests/port/webkit.py:
3536         (WebKitPort._skipped_file_search_paths):
3537
3538 2012-03-13  Jer Noble  <jer.noble@apple.com>
3539
3540         Lion Intel Debug WebKit2 Tests crashing under [WKFullScreenWindowController _startEnterFullScreenAnimationWithDuration:]
3541         https://bugs.webkit.org/show_bug.cgi?id=81056
3542
3543         Reviewed by Jessie Berlin.
3544
3545         Intercept beganEnterFullScreen and beganExitFullScreen and turn them into no-ops (plus logging).
3546
3547         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
3548         (WTR::InjectedBundlePage::InjectedBundlePage):
3549         (WTR::InjectedBundlePage::beganEnterFullScreen):
3550         (WTR::InjectedBundlePage::beganExitFullScreen):
3551
3552 2012-03-14  Alexander Færøy  <alexander.faeroy@nokia.com>
3553
3554         The width and height of the scroll indicators in the MiniBrowser does
3555         not consider the offset margins.
3556         https://bugs.webkit.org/show_bug.cgi?id=81112
3557
3558         Reviewed by Kenneth Rohde Christiansen.
3559
3560         This patch fixes the issue by withdrawing two times the anchor
3561         margins.
3562
3563         * MiniBrowser/qt/qml/ScrollIndicator.qml:
3564
3565 2012-03-14  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
3566
3567         [EFL] Implement LayoutTestController::clearBackForwardList
3568         https://bugs.webkit.org/show_bug.cgi?id=80142
3569
3570         Adding missing implementation clearBackForwardList to EFL's
3571         LayoutTestController. This implementation allows us to unskip 
3572         tests connected with history.
3573
3574         Reviewed by Antonio Gomes.
3575
3576         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3577         (LayoutTestController::clearBackForwardList): Implemented.
3578
3579 2012-03-13  Thiago Marcos P. Santos  <thiago.santos@intel.com>
3580
3581         [EFL][DRT] Expose window.internals object
3582         https://bugs.webkit.org/show_bug.cgi?id=79900
3583
3584         Exposes window.internals for EFL's DRT. Makes possible to
3585         remove some tests from the skip list.
3586
3587         Reviewed by Antonio Gomes.
3588
3589         * DumpRenderTree/efl/CMakeLists.txt:
3590         * DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
3591         (DumpRenderTreeChrome::onWindowObjectCleared):
3592
3593 2012-03-13  Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>
3594
3595         Always enable ENABLE(CLIENT_BASED_GEOLOCATION)
3596         https://bugs.webkit.org/show_bug.cgi?id=78853
3597
3598         Reviewed by Adam Barth.
3599
3600         * Scripts/build-webkit:
3601         * qmake/mkspecs/features/features.prf:
3602
3603 2012-03-12  Antonio Gomes  <agomes@rim.com>
3604
3605         Convert nodesFromRect tests to use Internals interface
3606         https://bugs.webkit.org/show_bug.cgi?id=80886
3607
3608         Reviewed by Ryosuke Niwa.
3609
3610         This clean up allows to remove port specific bits of each
3611         DRT implementation that supports testing Document::nodesFromRect,
3612         in favor of making use of the 'Internals' interface to test it.
3613
3614         This makes its code conveniently more expansible and cleaner.
3615
3616         Also removed stubs from Win and EFL DRTs.
3617
3618         * DumpRenderTree/LayoutTestController.cpp:
3619         (LayoutTestController::staticFunctions):
3620         * DumpRenderTree/LayoutTestController.h:
3621         (LayoutTestController):
3622         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
3623         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3624         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
3625         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
3626         * DumpRenderTree/qt/LayoutTestControllerQt.h:
3627         (LayoutTestController):
3628         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3629
3630 2012-03-13  Ojan Vafai  <ojan@chromium.org>
3631
3632         Rebaselining for a new port doesn't work right with multiple fallback ports
3633         https://bugs.webkit.org/show_bug.cgi?id=80932
3634
3635         Reviewed by Adam Barth.
3636
3637         Since chromium-leopard also falls back to mac-leopard, we need to
3638         copy the existing result for both chromium-leopard and chromium-snowleopard
3639         before doing lion rebaselines.
3640
3641         * Scripts/webkitpy/layout_tests/port/builders.py:
3642         (fallback_port_names_for_new_port):
3643         * Scripts/webkitpy/tool/commands/rebaseline.py:
3644         (RebaselineTest):
3645         (RebaselineTest._copy_existing_baseline):
3646         (RebaselineTest._rebaseline_test):
3647         (RebaselineTest.execute):
3648         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3649         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3650         (GardeningHTTPRequestHandler.rebaseline):
3651         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3652         (GardeningServerTest.test_rebaseline_new_port):
3653
3654 2012-03-13  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3655
3656         [Qt] Tweak minibrowser UI
3657
3658         Add some margins to the scroll indicators, and set a dark background
3659         color so you see the page dimensions.
3660
3661         Reviewed by Antonio Gomes.
3662
3663         * MiniBrowser/qt/qml/BrowserWindow.qml:
3664         * MiniBrowser/qt/qml/ScrollIndicator.qml:
3665
3666 2012-03-13  ChangSeok Oh  <shivamidow@gmail.com>
3667
3668         [EFL] [DRT] Implement scheduleAsynchronousKeyDown.
3669         https://bugs.webkit.org/show_bug.cgi?id=79601
3670