Use xcrun to find nm and ranlib on OS X builds
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-28  David Kilzer  <ddkilzer@apple.com>
2
3         Use xcrun to find nm and ranlib on OS X builds
4         <http://webkit.org/b/82466>
5
6         Reviewed by Dan Bernstein.
7
8         * Scripts/build-webkit: Use xcrun to find ranlib.
9         * Scripts/webkitdirs.pm:
10         (determineNmPath): Added.  Use xcrun to find nm on OS X.
11         (nmPath): Added.
12         * Scripts/webkitperl/features.pm:
13         (libraryContainsSymbol): Use nmPath().
14
15 2012-03-28  Ojan Vafai  <ojan@chromium.org>
16
17         Add history navigation to garden-o-matic
18         https://bugs.webkit.org/show_bug.cgi?id=82495
19
20         Reviewed by Dimitri Glazkov.
21
22         Also, maintain scroll offsets when returning to a tab. This makes addressing
23         expected failures considerably easier when dealing with below-the-fold tests.
24
25         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
26         -Store the scrollTop before switching to a new tab and restore the appropriate scrollTop after
27         switching.
28         -Modify window.location with the tabName in the hash. Can't use pushState because this is served
29         from a file URL and Chrome puts each different file URL in it's own origin.
30
31         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
32         Remove the href so that clicking on the accordion item does not modify window.location.hash.
33
34         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
35         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
36
37 2012-03-28  Philippe Normand  <pnormand@igalia.com>
38
39         [GTK] generate-gtk-doc doesn't cope with custom build directory
40         https://bugs.webkit.org/show_bug.cgi?id=82448
41
42         Reviewed by Martin Robinson.
43
44         Pass the product directory in generate-gtkdoc arguments from
45         build-webkit.
46
47         * Scripts/webkitdirs.pm:
48         (buildAutotoolsProject):
49         * gtk/common.py:
50         (get_build_path):
51
52 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
53
54         [GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
55         https://bugs.webkit.org/show_bug.cgi?id=82341
56
57         Reviewed by Martin Robinson.
58
59         Add --skipped=skip|ignore|only command line option.
60
61         * Scripts/run-gtk-tests:
62         (TestRunner._test_cases_to_skip): Return the list of test cases
63         to skip for the given test depending on the skipped action.
64         (TestRunner._should_run_test): Decide whether to run the test or
65         not depending on the skipped action.
66         (TestRunner.run_tests): Only show the list of skipped tests when
67         skipped action is 'skip'.
68
69 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
70
71         [GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
72         https://bugs.webkit.org/show_bug.cgi?id=82333
73
74         Reviewed by Martin Robinson.
75
76         * Scripts/run-gtk-tests:
77         (TestRunner.SkippedTest): Add SkippedTest class to store more
78         information about skipped tests.
79         (TestRunner._setup_testing_environment): Use SkippedTest to add
80         TestWebKitAccessibility to the skipped list.
81         (TestRunner._find_skipped_test): Return the SkippedTest object for
82         a given test path.
83         (TestRunner._test_cases_to_skip): Returns the list of test cases
84         that should be skipped for a given test path.
85         (TestRunner._should_run_test): Helper funtion to decide whether
86         the given test path should be run or not. Tests in the skipped
87         list that don't have a list of failing test cases are not run at
88         all. Tests in the skipped list that contain a list of test cases
89         are run with gtester -s to skip the individual test cases.
90         (TestRunner._run_test): Helper funtion to run a given test.
91         (TestRunner.run_tests): Build the list of tests to run based on
92         _should_run_test() and use helper function _run_test() to run
93         every test.
94
95 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
96
97         [Qt] Couple of tests have different results on 64 bit and/or in debug mode compared to 32 bit and/or release mode
98         https://bugs.webkit.org/show_bug.cgi?id=52810
99
100         Reviewed by Simon Hausmann.
101
102         Use SSE2 floating point math on 32 bit instead of the default
103         387 to make layout test result same on 32 and on 64 bit builds.
104
105         * qmake/mkspecs/features/unix/default_post.prf:
106
107 2012-03-28  Sergio Villar Senin  <svillar@igalia.com>
108
109         [Soup] DNS prefetching spams resolver, shoots self in the foot
110         https://bugs.webkit.org/show_bug.cgi?id=41630
111
112         Reviewed by Martin Robinson.
113
114         Bump libsoup and glib dependencies.
115
116         * efl/jhbuild.modules:
117         * gtk/jhbuild.modules:
118
119 2012-03-28  Gustavo Noronha Silva  <gns@gnome.org>
120
121         [GTK] Disable introspection for jhbuild modules
122         https://bugs.webkit.org/show_bug.cgi?id=82395
123
124         Reviewed by Philippe Normand.
125
126         * efl/jhbuild.modules: add --disable-introspection to all modules
127         that support it
128         * gtk/jhbuild.modules: ditto.
129
130 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
131
132         [Qt] Unreviewed, fix API tests after adding full platforms. (r111658)
133
134         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
135         (BuildAndTestFactory.__init__):
136
137 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
138
139         [Qt] Unreviewed typo fix after r112383.
140
141         * Scripts/run-qtwebkit-tests:
142         (Main.check_crash_occurences):
143
144 2012-03-28  Ádám Kallai  <kadam@inf.u-szeged.hu>
145
146         [Qt] run-qtwebkit-tests should report crashes
147         https://bugs.webkit.org/show_bug.cgi?id=76905
148
149         Added checking of crashing tests. Now final summary includes the total number of crashing tests too.
150         Made unit test for this modification by János Badics <jbadics@inf.u-szeged.hu>.
151
152         Reviewed by Csaba Osztrogonác.
153
154         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
155         (RunQtAPITests.commandComplete):
156         (RunQtAPITests.evaluateCommand):
157         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
158         (RunQtAPITestsTest):
159         (test_failure):
160         (test_crash):
161         * Scripts/run-qtwebkit-tests:
162         (Main.announce_results_html):
163         (Main):
164         (Main.check_crash_occurences):
165         (Main.convert_to_stdout):
166         (Main.convert_to_html):
167
168 2012-03-28  Kwonjin Jeong  <gram@company100.net>
169
170         Add support for Intel compiler on Linux QT port.
171         https://bugs.webkit.org/show_bug.cgi?id=81077
172
173         Reviewed by Tor Arne Vestbø.
174
175         * qmake/mkspecs/features/default_post.prf:
176         Don't hide symbols when building with Intel compiler.
177         The symbols are needed when building QtWebKit tests.
178
179         * qmake/mkspecs/features/unix/icc.prf:
180         Added to ignore icc-specific warnings.
181         This feature is used with command-line arguments for build-webkit script.
182
183 2012-03-28  Philippe Normand  <pnormand@igalia.com>
184
185         Unreviewed, trivial update-webkit-libs-jhbuild system call fix.
186
187         * Scripts/update-webkitgtk-libs: This script is written in
188         Perl. Not Python. So use perl vocabulary to refer to command line arguments.
189
190 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
191
192         kill-old-processes doesn't do anything on linux
193         https://bugs.webkit.org/show_bug.cgi?id=82388
194
195         Reviewed by Tony Chang.
196
197         Kill cc1plus, ld, etc... on linux. Also ps aux + grep to kill old run-webkit-tests processes.
198         This should help stabilizing Chromium linux bots.
199
200         * BuildSlaveSupport/kill-old-processes:
201
202 2012-03-28  Philippe Normand  <pnormand@igalia.com>
203
204         [GTK] jhbuild wrongly tricked into clean-up
205         https://bugs.webkit.org/show_bug.cgi?id=82216
206
207         Reviewed by Gustavo Noronha Silva.
208
209         Save jhbuild{rc,.modules}.md5sum in Dependencies instead of the
210         build tree. Also when checking if the jhbuild configuration
211         changed use absolute md5sum paths.
212
213         * Scripts/webkitdirs.pm:
214         (runAutogenForAutotoolsProjectIfNecessary):
215         (jhbuildConfigurationChanged):
216         (buildAutotoolsProject):
217
218 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
219
220         kill-old-processes doesn't do anything on linux
221         https://bugs.webkit.org/show_bug.cgi?id=82388
222
223         One more attempt to disable kill-old-processes on the Qt bots not to kill each others.
224
225         The platform in Factory comes from config.json, accordingly platform is fullplatform,
226         for example: qt-4.8, qt-5.0-wk2. We have to check only the prefix of platform.
227
228         Reviewed by Ryosuke Niwa.
229
230         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
231         (Factory.__init__):
232
233 2012-03-28  Hao Zheng  <zhenghao@chromium.org>
234
235         Make Chromium port for Android use hardware gpu path default.
236         https://bugs.webkit.org/show_bug.cgi?id=82033
237
238         Reviewed by Dirk Pranke.
239
240         Skip canvas tests, which are run in virtual gpu test suites.
241
242         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
243         (ChromiumAndroidPort.__init__):
244         (ChromiumAndroidPort.skipped_tests):
245
246 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
247
248         kill-old-processes doesn't do anything on linux
249         https://bugs.webkit.org/show_bug.cgi?id=82388
250
251         Disable kill-old-processes on the Qt bots not to kill each others.
252
253         Reviewed by Ryosuke Niwa.
254
255         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
256         (Factory.__init__):
257
258 2012-03-27  Hao Zheng  <zhenghao@chromium.org>
259
260         Handle DRT crash caused by Android OOM.
261         https://bugs.webkit.org/show_bug.cgi?id=82310
262
263         Reviewed by Tony Chang.
264
265         When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
266         is stopped silently and regarded as crashed. Re-run the test for
267         such crash.
268
269         Add some debug log to NRWT.
270
271         * Scripts/webkitpy/layout_tests/port/chromium.py:
272         (ChromiumDriver.run_test):
273         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
274         (ChromiumAndroidPort.__init__):
275         (ChromiumAndroidPort.get_last_stacktrace):
276         (ChromiumAndroidDriver.__init__):
277         (ChromiumAndroidDriver._start):
278         (ChromiumAndroidDriver.run_test):
279         (ChromiumAndroidDriver._get_drt_return_value):
280
281 2012-03-27  Ojan Vafai  <ojan@chromium.org>
282
283         Fix expected results for some unittest failures.
284
285         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
286
287 2012-03-27  Adrienne Walker  <enne@google.com>
288
289         [chromium] Fix --enabled-threaded-compositing flag in DRT
290         https://bugs.webkit.org/show_bug.cgi?id=82405
291
292         Reviewed by James Robinson.
293
294         --enable-threaded-compositing wasn't working because there was no way
295         to set the value before it was used in the TestShell constructor.
296
297         To avoid adding yet another boolean parameter to TestShell's
298         constructor, add an explicit initialize function that can operate on
299         any boolean settings that are set post-constructor.
300
301         * DumpRenderTree/chromium/DumpRenderTree.cpp:
302         (main):
303         * DumpRenderTree/chromium/TestShell.cpp:
304         (TestShell::TestShell):
305         (TestShell::initialize):
306         * DumpRenderTree/chromium/TestShell.h:
307         (TestShell):
308         (TestShell::setTestShellMode):
309
310 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
311
312         new-run-webkit-tests: suppress debug output from run-safari on results file
313         https://bugs.webkit.org/show_bug.cgi?id=82400
314
315         Reviewed by Ryosuke Niwa.
316
317         Send the run-safari output to /dev/null rather than the console.
318
319         * Scripts/webkitpy/layout_tests/port/mac.py:
320         (MacPort.show_results_html_file):
321
322 2012-03-27  Simon Fraser  <simon.fraser@apple.com>
323
324         Improve error reporting in run-javascriptcore-tests
325         https://bugs.webkit.org/show_bug.cgi?id=82379
326
327         Reviewed by Jessie Berlin.
328         
329         When the script dies, have it print out the reason for dying.
330
331         * Scripts/run-javascriptcore-tests:
332
333 2012-03-27  Eric Seidel  <eric@webkit.org>
334
335         Add Dave Barton to the Contributors list so I can more easily CC him on bugs
336         https://bugs.webkit.org/show_bug.cgi?id=82373
337
338         Reviewed by Ryosuke Niwa.
339
340         Dave's committer-status in in-progress, so until then
341         list him as a Contributor so at least I can more easily CC him on bugs.
342
343         * Scripts/webkitpy/common/config/committers.py:
344
345 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
346
347         [jhbuild] Prepend jhbuildrc's directory to sys.path instead of appending.
348         https://bugs.webkit.org/show_bug.cgi?id=82252
349
350         Reviewed by Gustavo Noronha Silva.
351
352         Appending the directory containing common.py may be problematic if
353         another path in sys.path also contains either a common.py or a
354         common/ module, so prepend to select our version instead. This is
355         the case with fail2ban, which has a common/ directory which may
356         end up being in Python's default path.
357
358         * efl/jhbuildrc:
359         * gtk/jhbuildrc:
360
361 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
362
363         test-webkitpy: prepare for better test run output
364         https://bugs.webkit.org/show_bug.cgi?id=82290
365
366         Reviewed by Adam Barth.
367
368         This code basically re-implements the output of the TextTestRunner default
369         runner code from unittest, although the implementation is quite
370         different, in preparation for changing the test output to be
371         metered and possibly running in parallel.
372
373         The output is almost identical to before, except that instead of
374         logging "test_regular (webkitpy.main.RunnerTest) passed" we log
375         "webkitpy.main.RunnerTest.test_regular passed". It has always
376         annoyed me that they invert the names to be harder to read and
377         so that you can't copy & paste back to the input for
378         test-webkitpy.
379
380         This patch is provided to add a skeleton for unit tests and for
381         comparison to an upcoming patch that will actually add new
382         functionality.
383
384         * Scripts/webkitpy/test/main.py:
385         (Tester.__init__):
386         (Tester._configure_logging):
387         (Tester._run_tests):
388         * Scripts/webkitpy/test/runner.py: Added.
389         (TestRunner):
390         (TestRunner.__init__):
391         (TestRunner.test_name):
392         (TestRunner.all_test_names):
393         (TestRunner.run):
394         (TestRunner.write_result):
395         (TestRunner.write_summary):
396         * Scripts/webkitpy/test/runner_unittest.py: Added.
397         (FakeModuleSuite):
398         (FakeModuleSuite.__init__):
399         (FakeModuleSuite.__str__):
400         (FakeModuleSuite.run):
401         (FakeTopSuite):
402         (FakeTopSuite.__init__):
403         (FakeLoader):
404         (FakeLoader.__init__):
405         (FakeLoader.top_suite):
406         (FakeLoader.loadTestsFromName):
407         (RunnerTest):
408         (RunnerTest.test_regular):
409         (RunnerTest.test_verbose):
410
411 2012-03-27  Gustavo Noronha Silva  <gns@gnome.org>
412
413         [GTK] Build gnutls without p11-kit support
414         https://bugs.webkit.org/show_bug.cgi?id=82212
415
416         Reviewed by Martin Robinson.
417
418         * gtk/jhbuild.modules: pass --without-p11-kit to gnutls'
419         configure, to avoid requiring that dependency, which is not used
420         by libsoup and WebKit anyway
421
422 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
423
424         [Qt] Make debug build work on 32 bit
425         https://bugs.webkit.org/show_bug.cgi?id=82331
426
427         Use --no-keep-memory option of ld not to cache the symbol tables of input files in
428         memory to avoid memory exhaustion during the linking phase for x86-debug builds.
429
430         Reviewed by Tor Arne Vestbø.
431
432         * qmake/mkspecs/features/unix/default_post.prf:
433         * qmake/mkspecs/features/unix/default_pre.prf:
434
435 2012-03-23  Balazs Ankes  <bank@inf.u-szeged.hu>
436
437         Add png checking to check-webkit-style
438         https://bugs.webkit.org/show_bug.cgi?id=75824
439
440         Reviewed by NOBODY Tony Chang.
441
442         * Scripts/webkitpy/style/checker.py:
443         (_all_categories):
444         (FileType):
445         (CheckerDispatcher._should_skip_file_path):
446         (CheckerDispatcher._file_type):
447         (CheckerDispatcher._create_checker):
448         * Scripts/webkitpy/style/checker_unittest.py:
449         (CheckerDispatcherDispatchTest.test_none_paths):
450         * Scripts/webkitpy/style/checkers/png.py: Added.
451         (PNGChecker):
452         (PNGChecker.__init__):
453         (PNGChecker.check):
454         (PNGChecker._config_file_path):
455         * Scripts/webkitpy/style/checkers/png_unittest.py: Added.
456         (MockSCMDetector):
457         (MockSCMDetector.__init__):
458         (MockSCMDetector.display_name):
459         (MockSCMDetector.propget):
460         (PNGCheckerTest):
461         (PNGCheckerTest.test_init):
462         (PNGCheckerTest.test_init.mock_handle_style_error):
463         (PNGCheckerTest.test_check):
464         (PNGCheckerTest.test_check.mock_handle_style_error):
465         * Scripts/webkitpy/style/patchreader.py:
466         (PatchReader.check):
467
468 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
469
470         Unreviewed. Skip GTK+ unit tests that fail in the bots
471
472         * Scripts/run-gtk-tests:
473         (TestRunner): Add failing tests to the skipped list.
474
475 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
476
477         Unreviewed. Unskip all GTK+ unit test.
478
479         Currently, skipped tests contains the list of unit tests that have
480         failed in one or more bots. In this moment it's not possible to
481         skip just the test cases that fail, so the complete unit test is
482         skipped when a single test case fails. For that reason, we don't
483         know which test cases are actually failing for every skipped
484         test. We are planning to land a patch to allow skipping individual
485         test cases, but first we need to know which test cases are
486         currently failing in the bots to skip them. So, this patch simply
487         unskips all unit tests in order to watch the bots and get the list
488         of test cases failing. Failing tests will be skipped again as soon
489         as we have the new list of failing test cases.
490
491         * Scripts/run-gtk-tests:
492         (TestRunner): Reset the list of skipped tests.
493
494 2012-03-26  Scott Byer  <scottbyer@chromium.org>
495
496         Enable layout testing of the scroll animator.
497         https://bugs.webkit.org/show_bug.cgi?id=81858
498         Add a call to the InternalSettings that layout tests can use to
499         turn on scroll animation. Enable animation updates for the
500         Chromium platform DRT when scroll animation has been turned on in
501         a test. This should be a no-op for all current layout tests.
502
503         Reviewed by James Robinson.
504
505         * DumpRenderTree/chromium/WebViewHost.cpp:
506         (WebViewHost::serviceAnimation):
507         (WebViewHost::scheduleAnimation):
508         * DumpRenderTree/chromium/WebViewHost.h:
509         (WebViewHost):
510
511 2012-03-26  Dinu Jacob  <dinu.jacob@nokia.com>
512
513         [Qt][WK2] Support multi-file upload
514         https://bugs.webkit.org/show_bug.cgi?id=81589
515
516         Reviewed by Simon Hausmann.
517
518         Modified filePicker to support multi-file upload.
519
520         * MiniBrowser/qt/MiniBrowser.qrc:
521         * MiniBrowser/qt/icons/checkbox_checked.png: Added.
522         * MiniBrowser/qt/icons/checkbox_unchecked.png: Added.
523         * MiniBrowser/qt/js/MultiSelect.js: Added.
524         (values):
525         (isSelected):
526         (removeValue):
527         * MiniBrowser/qt/qml/CheckBox.qml: Added.
528         * MiniBrowser/qt/qml/FilePicker.qml:
529
530 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
531
532         Retry crashing tests serially at the end of NRWT on Apple Mac
533         https://bugs.webkit.org/show_bug.cgi?id=82233
534
535         Reviewed by Dirk Pranke.
536
537         * Scripts/webkitpy/layout_tests/controllers/manager.py:
538         (Manager.run):
539         * Scripts/webkitpy/layout_tests/port/base.py:
540         (Port.should_retry_crashes):
541         * Scripts/webkitpy/layout_tests/port/mac.py:
542         (MacPort.should_retry_crashes):
543
544 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
545
546         NRWT should retry failures serially
547         https://bugs.webkit.org/show_bug.cgi?id=82241
548
549         Reviewed by Dirk Pranke.
550
551         * Scripts/webkitpy/layout_tests/controllers/manager.py:
552         (Manager._run_tests): Add another parameter to specify the number of workers to use.
553         (Manager.run): Run the failures serially.
554
555 2012-03-26  Adrienne Walker  <enne@google.com>
556
557         Allow expectations on directories in virtual test suites
558         https://bugs.webkit.org/show_bug.cgi?id=82250
559
560         Reviewed by Dirk Pranke.
561
562         lookup_virtual_test_base was only returning valid results for files
563         and not directories. Fix by falling back to the virtual test suite
564         mapping if possible.
565
566         * Scripts/webkitpy/layout_tests/port/base.py:
567         (Port.lookup_virtual_test_base):
568         * Scripts/webkitpy/layout_tests/port/test.py:
569         (TestPort.skipped_tests):
570         (TestPort.virtual_test_suites):
571
572 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
573
574         Fix duplicated ChangeLog entry from r112171.
575
576         Unreviewed, build fix.
577
578 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
579
580         nrwt: remove --worker-model flag
581         https://bugs.webkit.org/show_bug.cgi?id=82112
582
583         Reviewed by Adam Barth.
584
585         The --worker-model=inline flag is basically redundant with the
586         --child-processes=1 flag (technically, you could one run child
587         process, but this is only useful for testing); this patch
588         removes this flag, which simplifies things and enables more
589         cleanup down the road.
590
591         * Scripts/webkitpy/layout_tests/controllers/manager.py:
592         (Manager._run_tests):
593         (Manager.print_config):
594         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
595         (get):
596         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
597         (make_broker):
598         (FunctionTests.test_get__inline):
599         (FunctionTests.test_get__processes):
600         (_TestsMixin.setUp):
601         (_TestsMixin.make_broker):
602         (InlineBrokerTests.setUp):
603         (MultiProcessBrokerTests.setUp):
604         * Scripts/webkitpy/layout_tests/port/base.py:
605         (Port.default_child_processes):
606         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
607         (ChromiumAndroidPort.check_sys_deps):
608         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
609         (ChromiumAndroidPortTest):
610         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
611         (MockDRTPortTest.make_port):
612         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
613         (PortTestCase):
614         (PortTestCase.make_port):
615         * Scripts/webkitpy/layout_tests/port/test.py:
616         (TestPort.default_child_processes):
617         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
618         (_set_up_derived_options):
619         (parse_args):
620         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
621         (parse_args):
622         (MainTest.test_batch_size):
623         (MainTest.test_child_processes_2):
624         (MainTest.test_child_processes_min):
625         (MainTest.test_exception_raised):
626         (MainTest.test_keyboard_interrupt):
627         (MainTest.test_exit_after_n_crashes):
628         (MainTest.assert_run_order):
629         (MainTest.test_run_order__inline):
630         (MainTest.test_virtual):
631
632 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
633
634         test-webkitpy: split test-finding code into a different module
635         https://bugs.webkit.org/show_bug.cgi?id=82253
636
637         Reviewed by Adam Barth.
638
639         Per suggestion from abarth, this change splits all the
640         filesystem-crawling, test-finding code in test-webkitpy out into
641         a separate module and switches to using a FileSystem object.
642         This makes things much more testable, so we also add tests :).
643
644         We also add a realpath() method to the Filesystem object, since
645         test-webkitpy needs that in order to be able to resolve symlinks
646         properly to determine whether a file is under a particular tree
647         or not.
648
649         * Scripts/webkitpy/common/system/filesystem.py:
650         (FileSystem.realpath):
651         * Scripts/webkitpy/common/system/filesystem_mock.py:
652         (MockFileSystem.realpath):
653         * Scripts/webkitpy/test/main.py:
654         (Tester.__init__):
655         (Tester.add_tree):
656         (Tester.run):
657         (Tester._run_tests):
658         (Tester._log_exception):
659         * Scripts/webkitpy/test/test_finder.py: Added.
660         (TestDirectoryTree):
661         (TestDirectoryTree.__init__):
662         (TestDirectoryTree.find_modules):
663         (TestDirectoryTree.find_modules.file_filter):
664         (TestDirectoryTree.to_module):
665         (TestDirectoryTree.clean):
666         (TestFinder):
667         (TestFinder.__init__):
668         (TestFinder.add_tree):
669         (TestFinder.additional_paths):
670         (TestFinder.clean_trees):
671         (TestFinder.is_module):
672         (TestFinder.to_module):
673         (TestFinder.find_names):
674         (TestFinder._exclude):
675         * Scripts/webkitpy/test/test_finder_unittest.py: Added.
676         (TestFinderTest):
677         (TestFinderTest.setUp):
678         (TestFinderTest.tearDown):
679         (TestFinderTest.test_additional_system_paths):
680         (TestFinderTest.test_is_module):
681         (TestFinderTest.test_to_module):
682         (TestFinderTest.test_clean):
683         (TestFinderTest.test_find_names):
684
685 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
686
687         [jhbuild] Use $MAKE if it is defined to build jhbuild itself.
688         https://bugs.webkit.org/show_bug.cgi?id=82209
689
690         Reviewed by Martin Robinson.
691
692         Respecting the $MAKE environment variable when it is defined makes
693         building jhbuild more portable, as some platforms may have GNU
694         make as gmake, for example.
695
696         * jhbuild/jhbuild-wrapper:
697         (install_jhbuild):
698
699 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
700
701         add a Tree abstraction to test-webkitpy to better encapsulate things
702         https://bugs.webkit.org/show_bug.cgi?id=82158
703
704         Reviewed by Adam Barth.
705
706         There are no functional changes in this patch, but adding a
707         basic abstraction for the trees we are looking in for python
708         tests will allow me to add some features later on more easily
709         (like ignoring certain directories like webkitpy/thirdparty),
710         and it makes things slightly less hard-coded.
711
712         * Scripts/test-webkitpy:
713         * Scripts/webkitpy/test/main.py:
714         (Tester.__init__):
715         (Tester):
716         (Tester.add_tree):
717         (Tester._parse_args):
718         (Tester._configure):
719         (Tester.run):
720         (Tester._find_modules):
721         (Tester._run_tests):
722         (Tester._is_module):
723         (Tester._log_exception):
724         (TestDirectoryTree):
725         (TestDirectoryTree.__init__):
726         (TestDirectoryTree.find_modules):
727         (TestDirectoryTree.clean):
728
729 2012-03-26  Ojan Vafai  <ojan@chromium.org>
730
731         garden-o-matic should special case reftests
732         https://bugs.webkit.org/show_bug.cgi?id=81812
733
734         Reviewed by Adam Barth.
735
736         On the examine page, don't show a rebaseline button for reftests.
737         We still show it on the failure summary page because tests are grouped
738         together. Long-term, we should make sure webkit-patch rebaseline-test
739         never tries to rebaseline reftests.
740
741         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
742         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
743         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
744
745 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
746
747         clean up NRWT logging/metering, be less verbose
748         https://bugs.webkit.org/show_bug.cgi?id=73846
749
750         Reviewed by Ojan Vafai.
751
752         Re-land r112014 with a build fix for Python 2.6 compatibility.
753
754         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
755         (ManagerTest.test_http_locking):
756         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
757         (AbstractWorker.run):
758         * Scripts/webkitpy/layout_tests/controllers/worker.py:
759         (Worker.__init__):
760         (Worker._set_up_logging):
761         (Worker):
762         (Worker._set_up_host_and_port):
763         (Worker.run):
764         (Worker.cleanup):
765         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
766         (WorkerTest.test_default_platform_in_worker):
767         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
768         (run):
769         (main):
770         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
771         (MeteredStream):
772         (MeteredStream.implements):
773         (MeteredStream._erasure):
774         (MeteredStream._ensure_newline):
775         (MeteredStream.__init__):
776         (MeteredStream.__del__):
777         (MeteredStream.cleanup):
778         (MeteredStream.write_throttled_update):
779         (MeteredStream.write_update):
780         (MeteredStream.write):
781         (MeteredStream.writeln):
782         (MeteredStream._erase_last_partial_line):
783         (_LogHandler):
784         (_LogHandler.__init__):
785         (_LogHandler.emit):
786         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
787         (RegularTest):
788         (RegularTest.setUp):
789         (RegularTest.tearDown):
790         (RegularTest.time_fn):
791         (RegularTest.test_logging_not_included):
792         (RegularTest._basic):
793         (RegularTest.test_basic):
794         (RegularTest._log_after_update):
795         (RegularTest.test_log_after_update):
796         (TtyTest):
797         (TtyTest.test_basic):
798         (TtyTest.test_log_after_update):
799         (VerboseTest):
800         (VerboseTest.test_basic):
801         (VerboseTest.test_log_after_update):
802         * Scripts/webkitpy/layout_tests/views/printing.py:
803         (parse_print_options):
804         (Printer.__init__):
805         (Printer.cleanup):
806         (Printer.print_progress):
807         (Printer.print_update):
808         (Printer._write):
809         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
810         (Testprinter.get_printer):
811         (Testprinter.test_print_progress):
812         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
813         (PerfTestsRunner.__init__):
814         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
815         (test_run_test_pause_before_testing):
816
817 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
818
819         [jhbuild] Revert $MAKE environment hack introduced in r101929.
820         https://bugs.webkit.org/show_bug.cgi?id=82234
821
822         Reviewed by Martin Robinson.
823
824         The problem it was supposedly fixing was fixed in jhbuild itself
825         in git revision da8944d8ca987ca48c91b350257a530050406092, so we
826         can remove the workaround added to the script.
827
828         * jhbuild/jhbuild-wrapper:
829         (install_jhbuild):
830
831 2012-03-26  Leo Yang  <leo.yang@torchmobile.com.cn>
832
833         [BlackBerry] Skeleton code of AsyncFileSystemBlackBerry.cpp
834         https://bugs.webkit.org/show_bug.cgi?id=82157
835
836         Reviewed by Rob Buis.
837
838         * Scripts/build-webkit:
839         Make ENABLE_FILE_SYSTEM default on for blackberry.
840
841 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
842
843         [jhbuild] Explicitly disable DTrace when building glib.
844         https://bugs.webkit.org/show_bug.cgi?id=82145
845
846         Reviewed by Gustavo Noronha Silva.
847
848         DTrace support is not really needed by WebKit, so disabling it
849         helps avoid some build errors on more exotic platforms (such as
850         FreeBSD).
851
852         * efl/jhbuild.modules:
853         * gtk/jhbuild.modules:
854
855 2012-03-26  Sheriff Bot  <webkit.review.bot@gmail.com>
856
857         Unreviewed, rolling out r111993.
858         http://trac.webkit.org/changeset/111993
859         https://bugs.webkit.org/show_bug.cgi?id=82184
860
861         It broke layout and API tests and made WTR crash (Requested by
862         Ossy on #webkit).
863
864         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
865         (WTR::PlatformWebView::PlatformWebView):
866
867 2012-03-25  Kevin Ollivier  <kevino@theolliviers.com>
868
869         [wx] Unreviewed build fix. Move WTF to its own static lib build.
870
871         * waf/build/settings.py:
872         (common_configure):
873
874 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
875
876         AX: Support solution to handle invalid ax text marker
877         https://bugs.webkit.org/show_bug.cgi?id=82023
878
879         Reviewed by Oliver Hunt.
880
881         * DumpRenderTree/AccessibilityUIElement.cpp:
882         (indexForTextMarkerCallback):
883         (isTextMarkerValidCallback):
884         (textMarkerForIndexCallback):
885         (AccessibilityUIElement::indexForTextMarker):
886         (AccessibilityUIElement::isTextMarkerValid):
887         (AccessibilityUIElement::textMarkerForIndex):
888         (AccessibilityUIElement::getJSClass):
889         * DumpRenderTree/AccessibilityUIElement.h:
890         (AccessibilityUIElement):
891         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
892         (AccessibilityUIElement::indexForTextMarker):
893         (AccessibilityUIElement::textMarkerForIndex):
894         (AccessibilityUIElement::isTextMarkerValid):
895         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
896         (WTR::AccessibilityUIElement::indexForTextMarker):
897         (WTR::AccessibilityUIElement::isTextMarkerValid):
898         (WTR::AccessibilityUIElement::textMarkerForIndex):
899         (WTR):
900         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
901         (AccessibilityUIElement):
902         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
903         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
904         (WTR):
905         (WTR::AccessibilityUIElement::indexForTextMarker):
906         (WTR::AccessibilityUIElement::isTextMarkerValid):
907         (WTR::AccessibilityUIElement::textMarkerForIndex):
908
909 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
910
911         Unreviewed, rolling out r112014.
912         http://trac.webkit.org/changeset/112014
913         https://bugs.webkit.org/show_bug.cgi?id=82138
914
915         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
916         #webkit).
917
918         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
919         (ManagerTest.test_http_locking):
920         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
921         (AbstractWorker.run):
922         * Scripts/webkitpy/layout_tests/controllers/worker.py:
923         (Worker.__init__):
924         (Worker.safe_init):
925         (Worker.run):
926         (Worker.cleanup):
927         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
928         (WorkerTest.test_default_platform_in_worker):
929         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
930         (run):
931         (main):
932         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
933         (MeteredStream):
934         (MeteredStream.is):
935         (MeteredStream.__init__):
936         (MeteredStream.write):
937         (MeteredStream.update):
938         (MeteredStream._overwrite):
939         (MeteredStream._reset):
940         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
941         (TestMeteredStream):
942         (TestMeteredStream.test_regular):
943         * Scripts/webkitpy/layout_tests/views/printing.py:
944         (_configure_logging):
945         (_restore_logging):
946         (Printer.__init__):
947         (Printer.cleanup):
948         (Printer.print_progress):
949         (Printer.print_update):
950         (Printer._write):
951         (Printer):
952         (Printer._update):
953         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
954         (TestUtilityFunctions.assertEmpty):
955         (TestUtilityFunctions):
956         (TestUtilityFunctions.assertNotEmpty):
957         (TestUtilityFunctions.assertWritten):
958         (TestUtilityFunctions.test_configure_logging):
959         (Testprinter.get_printer):
960         (Testprinter.test_print_progress):
961         (Testprinter.test_print_progress.mock_time):
962         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
963         (PerfTestsRunner.__init__):
964         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
965         (test_run_test_pause_before_testing):
966
967 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
968
969         [wx] Unreviewed. Fix the build after WTF move.
970
971         * waf/build/build_utils.py:
972         (get_excludes):
973         (get_excludes_in_dirs):
974         * waf/build/settings.py:
975         (get_port_excludes):
976         (common_configure):
977
978 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
979
980         clean up NRWT logging/metering, be less verbose
981         https://bugs.webkit.org/show_bug.cgi?id=73846
982
983         Reviewed by Ojan Vafai.
984
985         This patch totally rewrites the metered_stream module and
986         changes the way it and the printing module now interact with the
987         logging subsystem.
988
989         The printing module becomes a dumb layout_tests-specific wrapper around
990         the metered stream. All of the intelligence of how and when to
991         actually write (or erase) lines and how things interact with
992         logging moves into the metered stream.
993
994         As of now, anything written to the metered stream *bypasses the
995         logging system*. If a logger is passed to the meter constructor,
996         the metered stream will also attach a log handler so that all
997         messages fed to the logger will also be written to the metered
998         stream. This detangles the logic that used to exist where
999         sometimes the meter would write to the logger and sometimes the
1000         logger would write to the meter.
1001
1002         All of this makes the logic generic and reusable for
1003         test-webkitpy (coming in another patch). It also helps clarify
1004         what the mysterious "configure_logging" parameter to the printer
1005         constructor used to do (that parameter itself is gone now, and
1006         by default the printer and metered_stream will not ever touch
1007         the logging system; this makes a bunch of tests less fragile).
1008
1009         All of the above should produce no user-visible changes.
1010
1011         However, to also fix the bug in question, we change the message
1012         format in --verbose mode to %H:%M:%S.%frac %pid %message,
1013         eliminating the date, file:lineno, and level parameters. I am
1014         making this change at the same time because it would be
1015         difficult to write things consistently between the logs and the
1016         metered stream if we were still logging the file:lineno.
1017
1018         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1019           Remove the configure_logging parameter
1020         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1021           Use metered stream to configure the logging (we don't need a printer)
1022           and be careful about when and how we manipulate the log level
1023           and handlers on the root logger.
1024         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1025           Pass in the root logger so that log messages are formatted
1026           properly, and set the log level outside of any routines called
1027           by unit tests so as to not interfere.
1028         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
1029           remove a duplicate log message.
1030         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1031           complete rewrite :).
1032         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1033           complete rewrite :).
1034         * Scripts/webkitpy/layout_tests/views/printing.py:
1035           move the logging configuration into metered_stream, and also
1036           move the "when should I log progress messages" into metered stream.
1037         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1038           remove tests that are no longer necessary.
1039         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1040           remove the configure_logging parameter.
1041         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1042           update the test; since the printer no longer writes to the log,
1043           we need to check what is written to the regular_output instead.
1044           (test_run_test_pause_before_testing):
1045
1046 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
1047
1048         Add myself and Martin Robinson to:
1049           * GStreamerGraphics
1050           * GtkWebKit2PublicAPI
1051           * SoupNetwork
1052         and to the newly created WebKitGTKTranslations.
1053
1054         Rubber-stamped by Martin Robinson.
1055
1056         * Scripts/webkitpy/common/config/watchlist:
1057
1058 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
1059
1060         [Qt][WK2] Implement PageClient::isViewWindowActive()
1061         https://bugs.webkit.org/show_bug.cgi?id=81143
1062
1063         Reviewed by Kenneth Rohde Christiansen.
1064
1065         Fix the WrapperWindow from our PlatformWebView
1066         so it correctly creates the platform related
1067         stuff for QWindow, which is not created unless
1068         QWindow::setVisible() or QWindow::show() are called.
1069
1070         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1071         (WTR::PlatformWebView::PlatformWebView):
1072
1073 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1074
1075         nrwt: don't wait for safari to exit before exiting after showing the results file
1076         https://bugs.webkit.org/show_bug.cgi?id=81845
1077
1078         Reviewed by Ryosuke Niwa.
1079
1080         Prior to this change, if you ran new-run-webkit-tests and
1081         displayed the HTML results file at the end, we would block
1082         waiting for the user to quit the browser. There doesn't seem to
1083         be a need for that, and the Chromium ports don't do this.
1084
1085         Also, update the mac tests to capture the output and be quiet :).
1086
1087         * Scripts/webkitpy/common/system/executive_mock.py:
1088           Make popen() testable, implement should_log for it.
1089         * Scripts/webkitpy/layout_tests/port/mac.py:
1090         (MacPort.show_results_html_file):
1091         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1092           Silence some of the tests via outputcapture.
1093
1094 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
1095
1096         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
1097         https://bugs.webkit.org/show_bug.cgi?id=82073
1098
1099         Reviewed by Martin Robinson.
1100
1101         * Scripts/webkitdirs.pm:
1102         (jhbuildConfigurationChanged): new function that isolates the
1103         jhbuild configuration change checking from the autogen logic.
1104         (mustReRunAutogen): code for the new function lived here.
1105         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
1106         itself, and cleans up the jhbuild root if configuration has
1107         changed; it also forces autogen to be run in that case, to
1108         make sure the new libraries are used.
1109         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
1110         * jhbuild/jhbuild-wrapper:
1111         (ensure_jhbuild): this might be run inside a jhbuild environment,
1112         in which case aclocal fails when trying to use the now deleted
1113         aclocal directory in the jhbuild prefix, so work around that.
1114
1115 2012-03-23  Ojan Vafai  <ojan@chromium.org>
1116
1117         garden-o-matic scrolls to the top of the page when you click an accordion item
1118         https://bugs.webkit.org/show_bug.cgi?id=82088
1119
1120         Reviewed by Adam Barth.
1121
1122         Check the scroll position and make sure that the clicked accordion item is visible.
1123
1124         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1125
1126 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1127
1128         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
1129         https://bugs.webkit.org/show_bug.cgi?id=80567
1130
1131         Reviewed by Ryosuke Niwa.
1132
1133         Move clean_up_run to occur right after we finish testing and
1134         before we start processing the results. Also makes set_up_run()
1135         and clean_up_run() private since there's no real need for
1136         callers to be aware of them.
1137
1138         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1139         (Manager._set_up_run):
1140         (Manager.run):
1141         (Manager._clean_up_run):
1142         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1143         (ManagerTest.test_http_locking):
1144         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1145         (run):
1146
1147 2012-03-23  Dean Jackson  <dino@apple.com>
1148
1149         Disable CSS_SHADERS in Apple builds
1150         https://bugs.webkit.org/show_bug.cgi?id=81996
1151
1152         Reviewed by Simon Fraser.
1153
1154         Default enableCSSShaders to be 0 on all builds that use
1155         this script. Use the flag to turn it back on.
1156
1157         * Scripts/build-webkit:
1158
1159 2012-03-23  Tony Chang  <tony@chromium.org>
1160
1161         [chromium] rename newwtf target back to wtf
1162         https://bugs.webkit.org/show_bug.cgi?id=82064
1163
1164         Reviewed by Adam Barth.
1165
1166         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1167
1168 2012-03-23  Jessie Berlin  <jberlin@apple.com>
1169
1170         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
1171         the Lion Intel WebKit2 Debug bots.
1172         https://bugs.webkit.org/show_bug.cgi?id=81012.
1173
1174         Rubber-stamped by Dan Bernstein.
1175
1176         Add a step to kill the pasteboard server while we investigate what is causing these
1177         assertion failures (to get the bots green).
1178
1179         * BuildSlaveSupport/kill-old-processes:
1180
1181 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
1182
1183         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
1184         https://bugs.webkit.org/show_bug.cgi?id=81986
1185
1186         Reviewed by David Levin.
1187
1188         Track whether or not errors are filtered in error handlers. If an
1189         alphabetical sorting error is reported for a header and filtered, try
1190         reporting it for the other potential source of the error. This ensures
1191         that the style bot can still find new sorting errors in both potential
1192         situations:
1193
1194             #include <foo.h> <-- 1. edited line
1195             #include <baz.h> <-- 2. edited line
1196
1197         * Scripts/webkitpy/style/checkers/cpp.py:
1198         (check_include_line): Potentially report the error for both lines,
1199         if the first error is filtered.
1200         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
1201         test that allows only reporting errors for certain lines. This verifies that
1202         when one line of an ordering error is filtered, the error is reported on the
1203         other line.
1204         (ErrorCollector.__init__):
1205         (ErrorCollector.__call__):
1206         (CppStyleTestBase.perform_lint):
1207         (CppStyleTestBase.perform_language_rules_check):
1208         (CppStyleTestBase.assert_language_rules_check):
1209         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
1210         Added a new test that verifies that when one line of the two lines of an ordering
1211         error occur, the error is reported on the other line.
1212         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
1213         (MockErrorHandler.__call__): Report True because the error is handled.
1214         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1215         (ErrorCollector.__call__): Ditto.
1216         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
1217         (MockErrorHandler.__call__): Ditto.
1218         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
1219         (TestErrorHandler.__call__): Ditto.
1220         * Scripts/webkitpy/style/checkers/xml_unittest.py:
1221         (MockErrorHandler.__call__): Ditto.
1222         * Scripts/webkitpy/style/error_handlers.py:
1223         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
1224         and False if it is filtered.
1225
1226 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
1227
1228         Build fix for WinCE after r111778.
1229
1230         * WinCELauncher/CMakeLists.txt:
1231
1232 2012-03-23  Peter Beverloo  <peter@chromium.org>
1233
1234         Unreviewed buildfix for the Chromium Android builder.
1235
1236         Disable the ImageDiff target for now. Android needs to build this for
1237         host rather than for target (as other configurations do), but since the
1238         WTF move the dependency chain got significantly more complicated.
1239
1240         I'll work out a proper fix in the following bug, but since several
1241         things are broken due to this, this buildfix should be landed first.
1242         https://bugs.webkit.org/show_bug.cgi?id=82039
1243
1244         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1245
1246 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
1247
1248         [EFL][DRT] Implement LayoutTestController's methods related with editing.
1249         https://bugs.webkit.org/show_bug.cgi?id=80136
1250
1251         Reviewed by Ryosuke Niwa.
1252
1253         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
1254         setSelectTrailingWhitespaceEnabled.
1255
1256         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1257         (LayoutTestController::setSmartInsertDeleteEnabled):
1258         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
1259
1260 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1261
1262         Convert hasGrammarMarker to use Internals interface
1263         https://bugs.webkit.org/show_bug.cgi?id=82004
1264
1265         Reviewed by Ryosuke Niwa.
1266
1267         Remove hasGrammarMarker functions, because it is able to work in the
1268         cross-port way through the Internals interface.
1269
1270         * DumpRenderTree/LayoutTestController.cpp:
1271         (LayoutTestController::staticFunctions):
1272         * DumpRenderTree/LayoutTestController.h:
1273         (LayoutTestController):
1274         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1275         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1276         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1277         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1278         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1279         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1280
1281 2012-03-22  Adam Barth  <abarth@webkit.org>
1282
1283         sync-master-with-upstream creates non-linear git history
1284         https://bugs.webkit.org/show_bug.cgi?id=82006
1285
1286         Reviewed by Julien Chaffraix.
1287
1288         Previously this script would create a merge commit at the HEAD of
1289         origin/master, causing your fork of WebKit to diverge from upstream.
1290         The new version of this script pushes directly from upstream to origin
1291         (without round-tripping through any local branches), avoiding this
1292         problem.
1293
1294         * Scripts/sync-master-with-upstream:
1295
1296 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1297
1298         Unreviewed build fix. Add knowledge of the WTF project to
1299         the autotools build system scripts so that build-jsc works
1300         correctly for GTK+.
1301
1302         * Scripts/webkitdirs.pm:
1303         (buildAutotoolsProject): handle the WTF project, build the
1304         libWTF.la target.
1305         (buildGtkProject): add WTF to the list of supported projects.
1306
1307 2012-03-22  Eric Seidel  <eric@webkit.org>
1308
1309         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
1310         https://bugs.webkit.org/show_bug.cgi?id=81984
1311
1312         Reviewed by Adam Barth.
1313
1314         * Scripts/VCSUtils.pm:
1315         (adjustPathForRecentRenamings):
1316
1317 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1318
1319         [CMake] Fix build-jsc after r111778 and r111797.
1320         https://bugs.webkit.org/show_bug.cgi?id=82001
1321
1322         Reviewed by Filip Pizlo.
1323
1324         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
1325         does return if no error occurred, and build-jsc failed with
1326         "building not defined for this platform".
1327
1328         * Scripts/build-jsc: Always exit after caling
1329         buildCMakeProjectOrExit().
1330         * Scripts/webkitdirs.pm:
1331         (buildCMakeProjectOrExit): Return 0 if no error occurred.
1332
1333 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1334
1335         [CMake] Unreviewed build fix after r111778.
1336
1337         * DumpRenderTree/efl/CMakeLists.txt: Replace
1338         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
1339         paths.
1340         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
1341         wtf include.
1342         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
1343         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
1344         with ${WTF_DIR}/wtf in the include paths.
1345
1346 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
1347
1348         Tools/Scripts/build-jsc doesn't work after WTF move
1349         https://bugs.webkit.org/show_bug.cgi?id=81999
1350
1351         Reviewed by Eric Seidel.
1352         
1353         For ports that require WTF and JSC to be built as two projects and where
1354         the machinery to build both does not already exist, teach the tool to do
1355         so for us.
1356
1357         * Scripts/build-jsc:
1358         (buildMyProject):
1359
1360 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1361
1362         Fix accidental missing line in r111793.
1363
1364         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1365         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
1366         Remove base.values. It was unused and duplicates the builtin Object.values.
1367         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1368         Accidentally deleted this line before committing r111793.
1369
1370 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1371
1372         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
1373         https://bugs.webkit.org/show_bug.cgi?id=81983
1374
1375         Reviewed by Adam Barth.
1376
1377         The dialogs get in the way if you are doing multiple rebaselines. Also,
1378         a number of people didn't realize that you could do multiple rebaselines
1379         in parallel. A non-modal dialog is just confusing.
1380
1381         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1382         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1383         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
1384         show the number of tests being rebaselined.
1385
1386         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1387         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
1388         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
1389         into a single area.
1390
1391         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
1392         also serves to clear all teh content in the StatusArea. 
1393
1394         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1395         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1396
1397 2012-03-22  Tony Chang  <tony@chromium.org>
1398
1399         Unreviewed, fix chromium build after wtf move.
1400
1401         Only use newwtf, remove references to wtf.
1402
1403         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1404
1405 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1406
1407         Actually move WTF files to their new home
1408         https://bugs.webkit.org/show_bug.cgi?id=81844
1409
1410         [Qt] Unreviewed buildfix after r111778.
1411
1412         * DumpRenderTree/qt/DumpRenderTree.pro:
1413
1414 2012-03-22  Dan Bernstein  <mitz@apple.com>
1415
1416         Font fallback in WebKitTestRunner depends on user settings
1417         https://bugs.webkit.org/show_bug.cgi?id=81978
1418
1419         Reviewed by Jessie Berlin.
1420
1421         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
1422         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
1423         with a value consisting of the array ("en") to the override defaults
1424         dictionary. This is the same value used in DumpRenderTree, and it
1425         ensures that font fallback follows a consistent path that is not
1426         affected by the language preferences of the user running the test.
1427
1428 2012-03-22  Andy Estes  <aestes@apple.com>
1429
1430         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
1431         https://bugs.webkit.org/show_bug.cgi?id=81971
1432
1433         Reviewed by Dan Bernstein.
1434
1435         Xcode will automatically link ObjC binaries against libarclite in some
1436         cases, which defines a class called __ARCLite__.
1437         check-for-inappropriate-objc-class-names shouldn't complain about this
1438         class name.
1439
1440         * Scripts/check-for-inappropriate-objc-class-names:
1441
1442 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1443
1444         new-run-webkit-tests blames the wrong test for crashing
1445         https://bugs.webkit.org/show_bug.cgi?id=81951
1446
1447         Reviewed by Adam Barth.
1448
1449         NRWT wasn't extracting the WebProcess pid correct; should've
1450         used re.search instead of re.match :(. Add more tests!
1451
1452         * Scripts/webkitpy/layout_tests/port/mac.py:
1453         (MacPort._get_crash_log):
1454         * Scripts/webkitpy/layout_tests/port/webkit.py:
1455         (WebKitDriver.has_crashed):
1456         (WebKitDriver._check_for_driver_crash):
1457         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1458         (WebKitDriverTest.test_no_timeout):
1459         (WebKitDriverTest):
1460         (WebKitDriverTest.test_check_for_driver_crash):
1461         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
1462         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
1463         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
1464         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
1465         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
1466         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
1467
1468 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1469
1470         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
1471         https://bugs.webkit.org/show_bug.cgi?id=81512
1472
1473         Adding missing implementation markerTextForListItem to EFL's
1474         LayoutTestController so that we can unskip related tests from
1475         the skip list.
1476
1477         Reviewed by Antonio Gomes.
1478
1479         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1480         (LayoutTestController::markerTextForListItem): Implemented.
1481
1482 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1483
1484         [EFL] Enable view mode media feature layout tests
1485         https://bugs.webkit.org/show_bug.cgi?id=81205
1486
1487         Adding missing implementation setViewModeMediaFeature to EFL's 
1488         LayoutTestController so that we can unskip view mode 
1489         media feature layout tests.
1490
1491         Reviewed by Antonio Gomes.
1492
1493         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1494         (LayoutTestController::setViewModeMediaFeature): Implemented.
1495
1496 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1497
1498         nrwt: crash while stopping layout test helper on apple mac lion
1499         https://bugs.webkit.org/show_bug.cgi?id=81949
1500
1501         Reviewed by Eric Seidel.
1502
1503         stop_helper() didn't work right if the helper process wasn't
1504         still running or didn't respond properly to being shut down. Fix
1505         this and add some tests. Also add code to MockExecutive() to be
1506         able to mock executive.popen() properly.
1507
1508         * Scripts/webkitpy/common/system/executive_mock.py:
1509         (MockProcess.__init__):
1510         (MockProcess.wait):
1511         (MockExecutive.__init__):
1512         (MockExecutive.popen):
1513         * Scripts/webkitpy/common/system/systemhost_mock.py:
1514         (MockSystemHost.__init__):
1515         * Scripts/webkitpy/layout_tests/port/mac.py:
1516         (MacPort.start_helper):
1517         (MacPort.stop_helper):
1518         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1519         (test_get_crash_log):
1520         (test_helper_starts):
1521         (test_helper_fails_to_start):
1522         (test_helper_fails_to_stop):
1523         (test_helper_fails_to_stop.bad_waiter):
1524
1525 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1526
1527         [GTK] Add make to the jhbuild moduleset
1528         https://bugs.webkit.org/show_bug.cgi?id=81769
1529
1530         Reviewed by Martin Robinson.
1531
1532         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
1533         Add GNU make with custom patches to fix several issues:
1534
1535         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
1536           fix the argument list limit of GNU make when running make
1537           distcheck.
1538         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
1539           and Jeremy Devenport to fix parallel build issues present in
1540           GNU make 3.82
1541
1542         * gtk/jhbuild.modules: Added make with patches.
1543         * gtk/patches/make-3.82-arg-list-length.patch: Added.
1544         * gtk/patches/make-3.82-parallel-build.patch: Added.
1545
1546 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
1547
1548         [wx] Unreviewed. Adding Source/WTF to the build and updating
1549         waf port list.
1550
1551         * DumpRenderTree/wscript:
1552         * waf/build/settings.py:
1553
1554 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1555
1556         Re-land fix for r81603 yet again (!) with latest build fix.
1557         https://bugs.webkit.org/show_bug.cgi?id=81603
1558
1559         Unreviewed, build fix.
1560
1561         The latest problem was a bad sprintf() in
1562         BasePort._get_crash_log(). I've reworked that routine and added
1563         unit tests for coverage.
1564
1565         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1566         (SingleTestRunner._handle_error):
1567         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1568         (write_test_result):
1569         (TestResultWriter.write_crash_log):
1570         * Scripts/webkitpy/layout_tests/port/base.py:
1571         (Port.diff_text):
1572         (Port._get_crash_log):
1573         * Scripts/webkitpy/layout_tests/port/chromium.py:
1574         (ChromiumDriver.run_test):
1575         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1576         (ChromiumDriverTest.test_crash_log):
1577         * Scripts/webkitpy/layout_tests/port/driver.py:
1578         (DriverOutput.__init__):
1579         * Scripts/webkitpy/layout_tests/port/mac.py:
1580         (MacPort.is_lion):
1581         (MacPort._get_crash_log):
1582         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1583         (test_default_child_processes):
1584         (test_get_crash_log):
1585         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1586         (MockDRTPortTest.test_get_crash_log):
1587         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1588         (PortTestCase.test_get_crash_log):
1589         * Scripts/webkitpy/layout_tests/port/test.py:
1590         (TestDriver.run_test):
1591         * Scripts/webkitpy/layout_tests/port/webkit.py:
1592         (WebKitDriver.__init__):
1593         (WebKitDriver._start):
1594         (WebKitDriver.has_crashed):
1595         (WebKitDriver._check_for_driver_crash):
1596         (WebKitDriver.run_test):
1597
1598 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1599
1600         [Qt] Enable support for handling database quota's in the MiniBrowser
1601         https://bugs.webkit.org/show_bug.cgi?id=81928
1602
1603         Reviewed by Simon Hausmann.
1604
1605         * MiniBrowser/qt/qml/BrowserWindow.qml:
1606
1607 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1608
1609         [EFL] Implement LayoutTestController::setEditingBehavior
1610         https://bugs.webkit.org/show_bug.cgi?id=81124
1611
1612         Adding missing implementation setEditingBehavior to EFL's
1613         LayoutTestController. This implementation allows us to
1614         unskip some tests from the skip list.
1615
1616         Reviewed by Martin Robinson.
1617
1618         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1619         (LayoutTestController::setEditingBehavior): Implemented.
1620
1621 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
1622
1623         [GTK] Use the angle-bracket form to include wtf headers
1624         https://bugs.webkit.org/show_bug.cgi?id=81884
1625
1626         Reviewed by Eric Seidel.
1627
1628         Use #include <wtf/foo> instead of #include "foo".
1629
1630         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1631         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1632         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1633         * DumpRenderTree/gtk/EditingCallbacks.cpp:
1634         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
1635         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
1636
1637 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1638
1639         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
1640
1641         * Scripts/webkitpy/common/config/committers.py:
1642
1643 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1644
1645         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
1646         https://bugs.webkit.org/show_bug.cgi?id=81889
1647
1648         Reviewed by Kenneth Rohde Christiansen.
1649
1650         Enable FAST_MOBILE_SCROLLING feature.
1651
1652         * qmake/mkspecs/features/features.prf:
1653
1654 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1655
1656         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
1657         https://bugs.webkit.org/show_bug.cgi?id=81900
1658
1659         Reviewed by Gustavo Noronha Silva.
1660
1661         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
1662         path which assumed a certain directory structure for the build
1663         directory which may noy be true.
1664
1665         Instead, hardcode the path to the downloaded fonts starting from the
1666         top-level source directory. While this solution is also far from
1667         perfect, it makes it more consistent and reliable, and easier to
1668         tackle at once with the rest of bug 81475.
1669
1670         * DumpRenderTree/efl/CMakeLists.txt:
1671
1672 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
1673
1674         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
1675         https://bugs.webkit.org/show_bug.cgi?id=81888
1676
1677         Reviewed by Philippe Normand.
1678
1679         * Scripts/update-webkit-libs-jhbuild:
1680         (runJhbuild): Helper function to run jhbuild with a given command.
1681         Instead of running jhbuild directly, use the jhbuild-wrapper, so
1682         that we don't need to run it first to check whether jhbuild is
1683         installed or not.
1684         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
1685         * gtk/run-with-jhbuild: Ditto.
1686         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
1687         not only run.
1688
1689 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1690
1691         Unreviewed, test-webkitpy fix after r111661.
1692
1693         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1694         (MainTest.test_crash_log):
1695         (MainTest.test_web_process_crash_log):
1696
1697 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1698
1699         Unreviewed, test-webkitpy fix after r111642.
1700
1701         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1702         (ChromiumDriverTest.setUp):
1703         (ChromiumDriverTest.test_two_drivers):
1704
1705 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1706
1707         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1708         https://bugs.webkit.org/show_bug.cgi?id=81603
1709
1710         Unreviewed rolling out r111609 and part of r111615,
1711         because it broke NRWT on Qt-WK2 platform.
1712
1713         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1714         (SingleTestRunner._handle_error):
1715         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1716         (write_test_result):
1717         (TestResultWriter.write_crash_report):
1718         * Scripts/webkitpy/layout_tests/port/base.py:
1719         (Port.is_crash_reporter):
1720         (Port._driver_class):
1721         * Scripts/webkitpy/layout_tests/port/chromium.py:
1722         (ChromiumDriver.run_test):
1723         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1724         (ChromiumDriverTest.test_crashed_process_name):
1725         * Scripts/webkitpy/layout_tests/port/driver.py:
1726         (DriverOutput.__init__):
1727         * Scripts/webkitpy/layout_tests/port/mac.py:
1728         (MacPort):
1729         (MacPort.is_crash_reporter):
1730         (MacPort.release_http_lock):
1731         * Scripts/webkitpy/layout_tests/port/test.py:
1732         (TestDriver.run_test):
1733         * Scripts/webkitpy/layout_tests/port/webkit.py:
1734         (WebKitDriver.__init__):
1735         (WebKitDriver._start):
1736         (WebKitDriver.has_crashed):
1737         (WebKitDriver._check_for_driver_crash):
1738         (WebKitDriver.run_test):
1739
1740 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1741
1742         [Qt] Add full platforms to Qt buildslaves
1743         https://bugs.webkit.org/show_bug.cgi?id=81877
1744
1745         Reviewed by Ryosuke Niwa.
1746
1747         * BuildSlaveSupport/build.webkit.org-config/config.json:
1748
1749 2012-03-21  Johnny Ding  <jnd@chromium.org>
1750
1751         Customize layout test timeout value for different ports.
1752         https://bugs.webkit.org/show_bug.cgi?id=79859
1753
1754         Reviewed by Ojan Vafai.
1755
1756         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1757         (Manager):
1758         * Scripts/webkitpy/layout_tests/port/base.py:
1759         (Port.default_test_timeout_ms):
1760         * Scripts/webkitpy/layout_tests/port/chromium.py:
1761         (ChromiumDriver.stop):
1762         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1763         (ChromiumAndroidPort.default_test_timeout_ms):
1764         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1765         (_set_up_derived_options):
1766
1767 2012-03-21  Mark Rowe  <mrowe@apple.com>
1768
1769         Fix the build.
1770
1771         Ensure that our test tools don't try and use RTTI unnecessarily.
1772
1773         * DumpRenderTree/mac/Configurations/Base.xcconfig:
1774         * TestWebKitAPI/Configurations/Base.xcconfig:
1775         * WebKitTestRunner/Configurations/Base.xcconfig:
1776
1777 2012-03-21  Adam Barth  <abarth@webkit.org>
1778
1779         WebKit should have some scripts to smooth interactions with GitHub
1780         https://bugs.webkit.org/show_bug.cgi?id=81842
1781
1782         Reviewed by Eric Seidel.
1783
1784         These scripts support the workflow in
1785         https://trac.webkit.org/wiki/UsingGitHub by automating the
1786         configuration and syncing operations needed for a GitHub fork of WebKit
1787         to track the main WebKit repository.
1788
1789         * Scripts/configure-github-as-upstream: Added.
1790         * Scripts/sync-master-with-upstream: Added.
1791
1792 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1793
1794         WTR - log the pid of a crashing WebProcess
1795         https://bugs.webkit.org/show_bug.cgi?id=81575
1796
1797         Reviewed by Alexey Proskuryakov.
1798
1799         This allows run-webkit-tests to find the right crash log for a
1800         given crashed WebProcess in WebKit2.
1801
1802         * WebKitTestRunner/TestController.cpp:
1803         (WTR::exceededDatabaseQuota):
1804         (WTR::TestController::createOtherPage):
1805         (WTR::TestController::initialize):
1806         (WTR::TestController::resetStateToConsistentValues):
1807         (WTR::TestController::runTest):
1808         (WTR::TestController::processDidCrash):
1809
1810 2012-03-21  Enrica Casucci  <enrica@apple.com>
1811
1812         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
1813         https://bugs.webkit.org/show_bug.cgi?id=81835
1814         <rdar://problem/11082749>
1815
1816         Reviewed by Brady Eidson.
1817
1818         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
1819         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
1820         (TestWebKitAPI::contextMenuCopyLink):
1821
1822 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
1823
1824         [CSS Shaders] Make CSS Shaders compile on Chromium
1825         https://bugs.webkit.org/show_bug.cgi?id=81435
1826
1827         Reviewed by Stephen White.
1828         
1829         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
1830
1831         * DumpRenderTree/chromium/LayoutTestController.cpp:
1832         (LayoutTestController::overridePreference):
1833         * DumpRenderTree/chromium/WebPreferences.cpp:
1834         (WebPreferences::reset):
1835         (WebPreferences::applyTo):
1836         * DumpRenderTree/chromium/WebPreferences.h:
1837         (WebPreferences):
1838
1839 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1840
1841         fix regression on gtk port introduced by r111609
1842         https://bugs.webkit.org/show_bug.cgi?id=81839
1843
1844         Reviewed by Philippe Normand.
1845
1846         r111609 should've initialized a couple of missing fields in
1847         WebKitDriver, and also should've checked for method overrides
1848         in the other ports :(.
1849
1850         * Scripts/webkitpy/layout_tests/port/gtk.py:
1851         (GtkDriver._start):
1852         * Scripts/webkitpy/layout_tests/port/webkit.py:
1853         (WebKitDriver.__init__):
1854         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1855         (MainTest.test_crash_log):
1856         (MainTest.test_web_process_crash_log):
1857
1858 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1859
1860         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1861         https://bugs.webkit.org/show_bug.cgi?id=81603
1862
1863         Reviewed by Adam Barth.
1864
1865         Add a new _get_crash_log() overridable method on the Port
1866         object to customize how we fetch crash logs per port.
1867         Mac will now slow-spin with a timeout waiting for the crash log, rather
1868         than waiting for the ReportCrash process to exit (it appears
1869         that the mac will manage multiple crashes with a single
1870         ReportCrash process, the process waits around longer than
1871         necessary, presumably to avoid thrashing if processes are
1872         repeatedly crashing).
1873
1874         Also, add the DriverOutput should contain the crash log and other info,
1875         which is created in a port-specific manner but can then be
1876         treated generically. Previously single_test_runner would get
1877         told that something crashed and attempt to do something to get
1878         the crash log, but it didn't have the information it needed to
1879         od the right thing; better to make the driver hand back the
1880         right info.
1881
1882         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1883         (SingleTestRunner._handle_error):
1884         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1885         (write_test_result):
1886         (TestResultWriter.write_crash_log):
1887         * Scripts/webkitpy/layout_tests/port/base.py:
1888         (Port.diff_text):
1889         (Port._get_crash_log):
1890         * Scripts/webkitpy/layout_tests/port/chromium.py:
1891         (ChromiumDriver.run_test):
1892         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1893         (ChromiumDriverTest.test_crash_log):
1894         * Scripts/webkitpy/layout_tests/port/driver.py:
1895         (DriverOutput.__init__):
1896         * Scripts/webkitpy/layout_tests/port/mac.py:
1897         (MacPort.is_lion):
1898         (MacPort._get_crash_log):
1899         * Scripts/webkitpy/layout_tests/port/test.py:
1900         (TestDriver.run_test):
1901         * Scripts/webkitpy/layout_tests/port/webkit.py:
1902         (WebKitDriver.__init__):
1903         (WebKitDriver._start):
1904         (WebKitDriver.has_crashed):
1905         (WebKitDriver._check_for_driver_crash):
1906         (WebKitDriver.run_test):
1907
1908 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
1909
1910         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
1911         https://bugs.webkit.org/show_bug.cgi?id=81507
1912
1913         Now that we're only loading from one directory, no need to
1914         iterate over a list of directories anymore.
1915
1916         Reviewed by Martin Robinson.
1917
1918         * DumpRenderTree/efl/CMakeLists.txt:
1919         * DumpRenderTree/efl/FontManagement.cpp:
1920         (addFontDirectory):
1921         (addFontsToEnvironment):
1922
1923 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
1924
1925         webkitpy: get ServerProcess out of the reportcrash business
1926         https://bugs.webkit.org/show_bug.cgi?id=81600
1927
1928         Unreviewed, build fix.
1929
1930         Re-land r111307 and r111293 with another fix for a crash in NRWT;
1931         we need to check if the driver has crashed before attempting to
1932         read from it.
1933
1934
1935         * Scripts/webkitpy/layout_tests/port/server_process.py:
1936         (ServerProcess._reset):
1937         (ServerProcess._handle_possible_interrupt):
1938         (ServerProcess.write):
1939         (ServerProcess.read_stdout):
1940         (ServerProcess.has_crashed):
1941         (ServerProcess._read):
1942         (ServerProcess.stop):
1943         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
1944         (TrivialMockPort.check_for_leaks):
1945         (TestServerProcess.test_broken_pipe):
1946         * Scripts/webkitpy/layout_tests/port/webkit.py:
1947         (WebKitPort._read_image_diff):
1948         (WebKitDriver.has_crashed):
1949         (WebKitDriver._check_for_driver_crash):
1950         (WebKitDriver.run_test):
1951         (WebKitDriver._read_block):
1952         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1953         (MockServerProcess.__init__):
1954         (MockServerProcess):
1955         (MockServerProcess.has_crashed):
1956
1957 2012-03-21  Zeno Albisser  <zeno@webkit.org>
1958
1959         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
1960         https://bugs.webkit.org/show_bug.cgi?id=81750
1961
1962         Building WebCore on mac currently creates an archive that is bigger than
1963         4GB. But ranlib fails to create a symbol table for such a big archive,
1964         even on 64bit machines.
1965         Therefore we start using SVGAllInOne.cpp when building debug
1966         on mac. This reduces the size of the archive by about 300MB.
1967
1968         Reviewed by Tor Arne Vestbø.
1969
1970         * qmake/mkspecs/features/mac/default_pre.prf:
1971
1972 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
1973
1974         Updated the address list of committers.py.
1975
1976         Reviewed by Csaba Osztrogonác.
1977
1978         * Scripts/webkitpy/common/config/committers.py:
1979
1980 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
1981
1982         Updating my nick in committers file.
1983
1984         Unreviewed.
1985
1986         * Scripts/webkitpy/common/config/committers.py:
1987
1988 2012-03-21  Peter Beverloo  <peter@chromium.org>
1989
1990         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
1991         https://bugs.webkit.org/show_bug.cgi?id=80861
1992
1993         Reviewed by Tony Chang.
1994
1995         Remove most Android-specific logic from webkitdirs.pm in favor of
1996         handling this during project generation time.
1997
1998         * Scripts/webkitdirs.pm:
1999         (buildChromiumMakefile):
2000
2001 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2002
2003         [GTK] Allow running javascript from location bar in MiniBrowser
2004         https://bugs.webkit.org/show_bug.cgi?id=81331
2005
2006         Reviewed by Martin Robinson.
2007
2008         * MiniBrowser/gtk/BrowserWindow.c:
2009         (activateUriEntryCallback): Use browser_window_load_uri().
2010         (browser_window_load_uri): Check whether uri starts with
2011         javascript: and run the given script, or just call
2012         webkit_web_view_load_uri() to load the given uri.
2013         * MiniBrowser/gtk/BrowserWindow.h:
2014         * MiniBrowser/gtk/main.c:
2015         (createBrowserWindow): Use browser_window_load_uri().
2016
2017 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2018
2019         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
2020         https://bugs.webkit.org/show_bug.cgi?id=75543
2021
2022         Reviewed by Martin Robinson.
2023
2024         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
2025         MiniBrowser CPP flags.
2026
2027 2012-03-20  Eric Seidel  <eric@webkit.org>
2028
2029         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
2030         https://bugs.webkit.org/show_bug.cgi?id=80911
2031
2032         Reviewed by Adam Barth.
2033
2034         Update include paths for Source/WTF.
2035
2036         * DumpRenderTree/efl/CMakeLists.txt:
2037         * DumpRenderTree/qt/DumpRenderTree.pro:
2038         * QtTestBrowser/QtTestBrowser.pro:
2039         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
2040
2041 2012-03-20  Adam Barth  <abarth@webkit.org>
2042
2043         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
2044         https://bugs.webkit.org/show_bug.cgi?id=81567
2045
2046         Reviewed by Eric Seidel.
2047
2048         This temporary measure should help us avoid disrupting the folks
2049         working on WebAudio when we move the webaudio directory to Modules.
2050
2051         * Scripts/VCSUtils.pm:
2052         (adjustPathForRecentRenamings):
2053         (parseGitDiffHeader):
2054         (parseSvnDiffHeader):
2055
2056 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2057
2058         Unreviewed, rolling out r111453.
2059         http://trac.webkit.org/changeset/111453
2060         https://bugs.webkit.org/show_bug.cgi?id=81600
2061
2062         still broken
2063
2064         * Scripts/webkitpy/layout_tests/port/server_process.py:
2065         (ServerProcess._reset):
2066         (ServerProcess.handle_interrupt):
2067         (ServerProcess.write):
2068         (ServerProcess._check_for_crash):
2069         (ServerProcess._handle_timeout):
2070         (ServerProcess._check_for_abort):
2071         (ServerProcess._read):
2072         (ServerProcess.stop):
2073         (ServerProcess):
2074         (ServerProcess.set_crashed):
2075         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2076         (TrivialMockPort.is_crash_reporter):
2077         (TestServerProcess.test_broken_pipe):
2078         * Scripts/webkitpy/layout_tests/port/webkit.py:
2079         (WebKitPort._read_image_diff):
2080         (WebKitDriver.has_crashed):
2081         (WebKitDriver._check_for_driver_crash):
2082         (WebKitDriver):
2083         (WebKitDriver._detected_crash):
2084         (WebKitDriver._subprocess_crashed):
2085         (WebKitDriver._crashed_process_name):
2086         (WebKitDriver.run_test):
2087         (WebKitDriver._read_block):
2088         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2089         (MockServerProcess.__init__):
2090
2091 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2092
2093         webkitpy: get ServerProcess out of the reportcrash business
2094         https://bugs.webkit.org/show_bug.cgi?id=81600
2095
2096         Unreviewed, build fix.
2097
2098         Re-land r111307 and r111293 with fix for crash in NRWT; I
2099         was failing to set the crashing process name.
2100
2101         * Scripts/webkitpy/layout_tests/port/server_process.py:
2102         (ServerProcess._reset):
2103         (ServerProcess._handle_possible_interrupt):
2104         (ServerProcess.write):
2105         (ServerProcess.read_stdout):
2106         (ServerProcess.has_crashed):
2107         (ServerProcess._read):
2108         (ServerProcess.stop):
2109         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2110         (TrivialMockPort.check_for_leaks):
2111         (TestServerProcess.test_broken_pipe):
2112         * Scripts/webkitpy/layout_tests/port/webkit.py:
2113         (WebKitPort._read_image_diff):
2114         (WebKitDriver.has_crashed):
2115         (WebKitDriver._check_for_driver_crash):
2116         (WebKitDriver.run_test):
2117         (WebKitDriver._read_block):
2118         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2119         (MockServerProcess.__init__):
2120         (MockServerProcess):
2121         (MockServerProcess.has_crashed):
2122
2123 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2124
2125         [mac] Restore color space switching code to run-webkit-tests
2126         https://bugs.webkit.org/show_bug.cgi?id=80571
2127         <rdar://problem/11008529>
2128
2129         Reviewed by Simon Fraser.
2130
2131         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
2132         
2133         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
2134         didn't exist until Lion.
2135
2136         * DumpRenderTree/mac/LayoutTestHelper.m:
2137         (installLayoutTestColorProfile):
2138         (restoreUserColorProfile):
2139
2140 2012-03-20  Adele Peterson  <adele@apple.com>
2141
2142         Update the last test to use EXPECT_WK_STREQ.
2143
2144         Reviewed by Dan Bernstein.
2145
2146         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
2147
2148 2012-03-20  Adele Peterson  <adele@apple.com>
2149
2150         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
2151         https://bugs.webkit.org/show_bug.cgi?id=81630
2152         <rdar://problem/10650660>
2153
2154         Reviewed by Alexey Proskuryakov.
2155
2156         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2157         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
2158         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
2159         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
2160         (TestWebKitAPI):
2161         (TestWebKitAPI::didFinishLoadForFrame):
2162         (TestWebKitAPI::TEST):
2163         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
2164
2165 2012-03-20  Eric Seidel  <eric@webkit.org>
2166
2167         Add link on queues.webkit.org to kov's fancy queue-depth graph
2168         https://bugs.webkit.org/show_bug.cgi?id=81690
2169
2170         Reviewed by Adam Barth.
2171
2172         * QueueStatusServer/templates/recentstatus.html:
2173
2174 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2175
2176         [mac] Restore color space switching code to run-webkit-tests
2177         https://bugs.webkit.org/show_bug.cgi?id=80571
2178         <rdar://problem/11008529>
2179
2180         Reviewed by Simon Fraser.
2181
2182         Make use of the layout test helper hooks to restore the Mac system-wide
2183         color profile switching code. This makes WkTR and DRT results match, and
2184         causes the main display's color profile to not affect results on Lion.
2185
2186         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2187         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
2188         (installLayoutTestColorProfile):
2189         (restoreUserColorProfile):
2190         (simpleSignalHandler):
2191         (main):
2192         * Scripts/webkitpy/layout_tests/port/mac.py:
2193         (MacPort.release_http_lock):
2194         (MacPort):
2195         (MacPort._path_to_helper):
2196         (MacPort.start_helper):
2197         (MacPort.stop_helper):
2198         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
2199         (WTR::PlatformWebView::PlatformWebView):
2200         (WTR::PlatformWebView::windowSnapshotImage):
2201
2202 2012-02-24  Darin Fisher  <darin@chromium.org>
2203
2204         Add more Chromium WebKit API reviewers
2205         https://bugs.webkit.org/show_bug.cgi?id=81576
2206
2207         Reviewed by Adam Barth and David Levin.
2208
2209         Simplify watchlist to lump all WebKit API into one.
2210
2211         * Scripts/webkitpy/common/config/watchlist:
2212
2213 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2214
2215         Convert hasSpellingMarker to use Internals interface.
2216         https://bugs.webkit.org/show_bug.cgi?id=81300
2217
2218         Reviewed by Ryosuke Niwa.
2219
2220         Remove hasSpellingMarker functions, because it is able to work in the
2221         cross-port way through the Internals interface.
2222
2223         * DumpRenderTree/LayoutTestController.cpp:
2224         (LayoutTestController::staticFunctions):
2225         * DumpRenderTree/LayoutTestController.h:
2226         (LayoutTestController):
2227         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2228         * DumpRenderTree/chromium/LayoutTestController.cpp:
2229         (LayoutTestController::LayoutTestController):
2230         * DumpRenderTree/chromium/LayoutTestController.h:
2231         (LayoutTestController):
2232         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2233         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2234         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2235         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2236         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2237         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2238         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2239
2240 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2241
2242         Add new Qt-WK2 performance tester bot to build.webkit.org
2243         https://bugs.webkit.org/show_bug.cgi?id=80842
2244
2245         Reviewed by Ryosuke Niwa.
2246
2247         * BuildSlaveSupport/build.webkit.org-config/config.json:
2248         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2249         (RunAndUploadPerfTestsWebKit2):
2250         (RunAndUploadPerfTestsWebKit2.start):
2251         (BuildAndPerfTestWebKit2Factory):
2252         (BuildAndPerfTestWebKit2Factory.__init__):
2253         (DownloadAndPerfTestWebKit2Factory):
2254         (DownloadAndPerfTestWebKit2Factory.__init__):
2255
2256 2012-03-20  Peter Beverloo  <peter@chromium.org>
2257
2258         [Chromium] Update to Android NDK r7b
2259         https://bugs.webkit.org/show_bug.cgi?id=81005
2260
2261         Reviewed by Adam Barth.
2262
2263         Update to version r7b of the Android NDK, release notes of which are
2264         available on the following page:
2265         http://developer.android.com/sdk/ndk/index.html
2266
2267         * Scripts/update-webkit-chromium:
2268
2269 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2270
2271         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
2272         https://bugs.webkit.org/show_bug.cgi?id=81502
2273
2274         Unreviewed.
2275
2276         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
2277
2278 2012-03-19  Jessie Berlin  <jberlin@apple.com>
2279
2280         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
2281         because they broke running the layout tests on mac.
2282
2283         * Scripts/webkitpy/layout_tests/port/webkit.py:
2284         (WebKitDriver._check_for_driver_crash):
2285
2286 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
2287
2288         [GTK] Allow running run-gtk-tests during 'make distcheck'
2289         https://bugs.webkit.org/show_bug.cgi?id=81415
2290
2291         Reviewed by Philippe Normand.
2292
2293         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
2294
2295         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
2296         main command run during 'make check.'
2297         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
2298         a requirement to run it. If jhbuild isn't around it will simply not start the
2299         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
2300         main loop and simply use the main loop in a synchronous method to start SPI
2301         daemons. Did a bunch of general cleanup around these tasks.
2302
2303 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2304
2305         jhbuild: Use the multiprocessing module to get the number of CPUs.
2306         https://bugs.webkit.org/show_bug.cgi?id=81585
2307
2308         Reviewed by Martin Robinson.
2309
2310         Now that support for Python 2.5 has been dropped, we can safely
2311         use multiprocessing.cpu_count() to obtain the current number of
2312         CPUs in the system instead of running a webkitperl script for
2313         that.
2314
2315         * efl/common.py:
2316         (top_level_path):
2317         * efl/jhbuildrc:
2318         * gtk/common.py:
2319         (build_path):
2320         * gtk/jhbuildrc:
2321
2322 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2323
2324         We should only add a live event listener once, not once per TestSelector.
2325         Adding one per TestSelector would cause accordion('option', 'active') to
2326         incorrectly be "false" if you examine multiple sets of tests.
2327
2328         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2329
2330 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2331
2332         Fix typo introduced in r111293.
2333
2334         Unreviewed, build fix.
2335
2336         * Scripts/webkitpy/layout_tests/port/webkit.py:
2337         (WebKitDriver._check_for_driver_crash):
2338
2339 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2340
2341         webkitpy: get ServerProcess out of the reportcrash business
2342         https://bugs.webkit.org/show_bug.cgi?id=81600
2343
2344         Reviewed by Adam Barth.
2345
2346         ServerProcess was half-aware that ReportCrash might run
2347         sometimes, and that the process ServerProcess was talking to
2348         might have its own crashing subprocesses; neither of these
2349         things really worked right and it made the logic convoluted, so
2350         this change makes handling crashes completely separate from the
2351         server_process code, so that it can focus on just I/O to the
2352         subprocess.
2353
2354         There should be no functional changes resulting from this patch.
2355
2356         * Scripts/webkitpy/layout_tests/port/server_process.py:
2357         (ServerProcess._reset):
2358         (ServerProcess._handle_possible_interrupt):
2359         (ServerProcess.write):
2360         (ServerProcess.read_stdout):
2361         (ServerProcess.has_crashed):
2362         (ServerProcess._read):
2363         (ServerProcess.stop):
2364         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2365         (TrivialMockPort.check_for_leaks):
2366         (TestServerProcess.test_broken_pipe):
2367         * Scripts/webkitpy/layout_tests/port/webkit.py:
2368         (WebKitPort._read_image_diff):
2369         (WebKitDriver.has_crashed):
2370         (WebKitDriver._check_for_driver_crash):
2371         (WebKitDriver.run_test):
2372         (WebKitDriver._read_block):
2373         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2374         (MockServerProcess.__init__):
2375         (MockServerProcess):
2376         (MockServerProcess.has_crashed):
2377
2378 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2379
2380         webkitpy: crashlog parsing is broken
2381         https://bugs.webkit.org/show_bug.cgi?id=81586
2382
2383         Reviewed by Adam Barth.
2384
2385         Seeking on a file opened through codecs() appears to not work
2386         correctly (at least on SL).
2387         
2388         The code was more complicated than it needed to be, so I have
2389         revamped it to just read a full crashlog at once and then look
2390         for matches. 
2391         
2392         I have also added the ability to optionally propagate
2393         errors back (which can be helpful to debug races when ReportCrash is
2394         still running and you need to see that you couldn't open some
2395         files, rather than just ignoring them).
2396
2397         * Scripts/webkitpy/common/system/crashlogs.py:
2398         (CrashLogs.find_newest_log):
2399         (CrashLogs._find_newest_log_darwin):
2400         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2401         (CrashLogsTest.test_find_log_darwin):
2402
2403 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2404
2405         webkitpy: clean up some port._filesystem references to not reference private members
2406         https://bugs.webkit.org/show_bug.cgi?id=81595
2407
2408         Reviewed by Eric Seidel.
2409
2410         There should be no functional changes in this patch, just some
2411         refactoring. Also, fixes a minor nit in write_image_diff_files(()
2412
2413         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2414         (SingleTestRunner.__init__):
2415         (SingleTestRunner._run_compare_test):
2416         (SingleTestRunner._run_rebaseline):
2417         (SingleTestRunner._save_baseline_data):
2418         (SingleTestRunner._handle_error):
2419         (SingleTestRunner._run_reftest):
2420         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2421         (write_test_result):
2422         (TestResultWriter.__init__):
2423         (TestResultWriter._make_output_directory):
2424         (TestResultWriter.output_filename):
2425         (TestResultWriter._output_testname):
2426         (TestResultWriter.write_output_files):
2427         (TestResultWriter.write_stderr):
2428         (TestResultWriter.write_crash_report):
2429         (TestResultWriter.create_text_diff_and_write_result):
2430         (TestResultWriter.write_image_diff_files):
2431         (copy_file):
2432         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
2433         (TestResultWriterTest.test_reftest_diff_image):
2434
2435 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2436
2437         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
2438         https://bugs.webkit.org/show_bug.cgi?id=81598
2439
2440         Reviewed by Eric Seidel.
2441
2442         "DumpRenderTree" shouldn't be hard-coded here, since some other
2443         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
2444         
2445         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2446         (TestExpectations):
2447
2448 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2449
2450         webkitpy: allow instance data in TestFailure objects
2451         https://bugs.webkit.org/show_bug.cgi?id=81580
2452
2453         Reviewed by Tony Chang.
2454
2455         Current TestFailures are basically enums, but it would be nice,
2456         especially for crashes, if they contained the name and pid of
2457         the process that crashed.
2458
2459         * Scripts/webkitpy/layout_tests/models/test_failures.py:
2460         (TestFailure.message):
2461         (FailureTimeout.message):
2462         (FailureCrash):
2463         (FailureCrash.__init__):
2464         (FailureCrash.message):
2465         (FailureMissingResult.message):
2466         (FailureTextMismatch.message):
2467         (FailureMissingImageHash.message):
2468         (FailureMissingImage.message):
2469         (FailureImageHashMismatch.message):
2470         (FailureImageHashIncorrect.message):
2471         (FailureReftestMismatch.message):
2472         (FailureReftestMismatchDidNotOccur.message):
2473         (FailureReftestNoImagesGenerated.message):
2474         (FailureMissingAudio.message):
2475         (FailureAudioMismatch.message):
2476         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
2477         (TestFailuresTest.test_equals):
2478         (TestFailuresTest):
2479         (TestFailuresTest.test_crashes):
2480
2481 2012-03-19  Brady Eidson  <beidson@apple.com>
2482
2483         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
2484         Crash in 3rd party WebKit apps under XHR/Cache code
2485
2486         Reviewed by Antti Koivisto.
2487
2488         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2489         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
2490         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
2491         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
2492         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
2493         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
2494         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
2495         (TestWebKitAPI):
2496         (TestWebKitAPI::TEST):
2497
2498 2012-03-19  Brady Eidson  <beidson@apple.com>
2499
2500         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
2501         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
2502
2503         Reviewed by Enrica Casucci.
2504
2505         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
2506         are in the correct format.
2507
2508         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2509         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
2510         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
2511         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
2512         (TestWebKitAPI::contextMenuCopyLink):
2513         (TestWebKitAPI::TEST):
2514
2515 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2516
2517         Make the selected item in the accordion a link to the flakiness dashboard
2518         https://bugs.webkit.org/show_bug.cgi?id=81571
2519
2520         Reviewed by Dimitri Glazkov.
2521
2522         Also, make it so that you can select the text of the link, instead of the current behavior
2523         of acting as if you had clicked on the accordion header.
2524
2525         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2526         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2527
2528 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2529
2530         NRWT runs some tests that are skipped with -i command line option
2531         https://bugs.webkit.org/show_bug.cgi?id=81535
2532
2533         Reviewed by Ojan Vafai.
2534
2535         This change modifies the interaction of Skipped files and
2536         test_expectations files so that entries in Skipped files (and
2537         the -i command line) override *everything* in the expectations
2538         file.
2539
2540         Specifically, a directory in a Skipped file will cause all of
2541         the tests in the dir to be skipped even if individual tests in
2542         the dir are listed in the test_expectations.txt.
2543
2544         Skipped files also override anything in an overrides files.
2545
2546         This seems to make more intuitive sense, since if you list
2547         something in the Skips file (and even more specify it on the
2548         command line) you probably want it to be universally applied.
2549
2550         Theoretically we could add more precedence levels and have full
2551         paths in an expectations file override dirs in a Skipped file
2552         (but not the command line), but I don't know yet that that level
2553         of complexity is justified.
2554         
2555         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2556         (TestExpectations.__init__):
2557         (TestExpectations._add_skipped_tests):
2558         (TestExpectations._add_test):
2559         (TestExpectations._add_expectations):
2560         (TestExpectations._add_expectation_line):
2561         (TestExpectations._already_seen_better_match):
2562         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2563         (test_more_specific_override_resets_skip):
2564         (SkippedTests):
2565         (SkippedTests.get_exp):
2566         (SkippedTests.assert_exp):
2567         (SkippedTests.test_skipped_tests_work):
2568         (SkippedTests.test_duplicate_skipped_test_fails_lint):
2569         (SkippedTests.test_skipped_file_overrides_expectations):
2570         (SkippedTests.test_skipped_dir_overrides_expectations):
2571         (SkippedTests.test_skipped_file_overrides_overrides):
2572         (SkippedTests.test_skipped_dir_overrides_overrides):
2573
2574 2012-03-19  Eric Seidel  <eric@webkit.org>
2575
2576         Fix WTF header include discipline in Chromium WebKit
2577         https://bugs.webkit.org/show_bug.cgi?id=81281
2578
2579         Reviewed by James Robinson.
2580
2581         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2582         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
2583
2584 2012-03-19  Andy Chen  <andchen@rim.com>
2585
2586         [BlackBerry] Add new layout test for text match marker
2587         https://bugs.webkit.org/show_bug.cgi?id=81531
2588
2589         Make the LayoutTestControllerBlackBerry::findString to test the
2590         BlackBerry port implementation.
2591
2592         Reviewed by Rob Buis.
2593
2594         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2595         (LayoutTestController::findString):
2596
2597 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2598
2599         Don't show WONTFIX tests in garden-o-matic
2600         https://bugs.webkit.org/show_bug.cgi?id=81554
2601
2602         Reviewed by Dimitri Glazkov.
2603
2604         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2605         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2606
2607 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
2608
2609         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
2610         https://bugs.webkit.org/show_bug.cgi?id=81530
2611
2612         Reviewed by Jessie Berlin.
2613
2614         Add rowAtIndex to WebKitTestRunner.
2615
2616         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
2617         (WTR::AccessibilityUIElement::rowAtIndex):
2618         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
2619         (AccessibilityUIElement):
2620         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
2621         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2622         (WTR):
2623         (WTR::AccessibilityUIElement::rowAtIndex):
2624
2625 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2626
2627         Change garden-o-matic All Failures tab to Expected Failures
2628         https://bugs.webkit.org/show_bug.cgi?id=81540
2629
2630         Reviewed by Dimitri Glazkov.
2631
2632         Also, fix some broken unittests that had just gotten stale.
2633
2634         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
2635         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
2636         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
2637         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2638         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2639         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2640         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2641         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
2642         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
2643         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2644
2645 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
2646
2647         [chromium] DRT crashes at shutdown.
2648         https://bugs.webkit.org/show_bug.cgi?id=81504
2649
2650         Reviewed by James Robinson.
2651
2652         WebCompositor must be destroyed after WebView is properly destroyed,
2653         or some compositor objects like CCLayerTreeHost may not be destroyed.
2654         However, it is really hard to make the destruction order correct
2655         explicitly in ~TestShell. So move it out to DumpRenderTree.
2656
2657         * DumpRenderTree/chromium/DumpRenderTree.cpp:
2658         (main):
2659         * DumpRenderTree/chromium/TestShell.cpp:
2660         (TestShell::~TestShell):
2661
2662 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2663
2664         add dpranke to webkitpy watchlist
2665         https://bugs.webkit.org/show_bug.cgi?id=81433
2666
2667         Reviewed by Benjamin Poulain.
2668
2669         * Scripts/webkitpy/common/config/watchlist:
2670
2671 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2672
2673         test-webkitpy is failing on SL/Python 2.6.1
2674         https://bugs.webkit.org/show_bug.cgi?id=81533
2675
2676         Reviewed by Ojan Vafai.
2677
2678         It appears that itertools.combinations() is buggy in 2.6.1; this
2679         backports the implementation from 2.7.
2680
2681         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2682         (TestConfigurationConverter.collapse_macros):
2683         (TestConfigurationConverter):
2684         (TestConfigurationConverter.combinations):
2685
2686 2012-03-19  David Michael Barr  <davidbarr@google.com>
2687
2688         Optimize the cleanup sequence in the start-queue.sh script
2689         https://github.com/abarth/webkit/pull/1
2690
2691         Reviewed by Adam Barth.
2692
2693         Updating the working copy is an expensive operation
2694         for a repository the size of WebKit.
2695         Attempt to reset only to the target revision.
2696
2697         * EWSTools/start-queue.sh:
2698
2699 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
2700
2701         Adjust architecture type of the EFL build bot.
2702         https://bugs.webkit.org/show_bug.cgi?id=81452
2703
2704         Reviewed by Tony Chang.
2705
2706         * BuildSlaveSupport/build.webkit.org-config/config.json: The
2707         buildslave is a 64-bit machine, therefore describe it as being
2708         x86_64, not i386.
2709
2710 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2711
2712         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
2713         https://bugs.webkit.org/show_bug.cgi?id=81514
2714
2715         Reviewed by Simon Hausmann.
2716
2717         * MiniBrowser/qt/qml/BrowserWindow.qml:
2718
2719 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2720
2721         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
2722         https://bugs.webkit.org/show_bug.cgi?id=81491
2723
2724         Reviewed by Philippe Normand.
2725
2726         For the find_path(), find_library() etc CMake calls to find the
2727         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
2728         set, otherwise CMake cannot know the dependencies exist and will only
2729         look in the standard directories.
2730
2731         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
2732         the jhbuild Root directory.
2733
2734 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2735
2736         [EFL] Add XDG env vars to jhbuildrc
2737         https://bugs.webkit.org/show_bug.cgi?id=81510
2738
2739         Needed to ensure correct glib behaviour in jhbuild environment.
2740
2741         Reviewed by Philippe Normand.
2742
2743         * efl/jhbuildrc:
2744
2745 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2746
2747         Select best target for tap gesture.
2748         https://bugs.webkit.org/show_bug.cgi?id=78801
2749
2750         Reviewed by Kenneth Rohde Christiansen.
2751
2752         Add TOUCH_ADJUSTMENT to enabled features.
2753
2754         * qmake/mkspecs/features/features.prf:
2755
2756 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
2757
2758         [chromium] synthesize wheel events for fling on main thread
2759         https://bugs.webkit.org/show_bug.cgi?id=81458
2760
2761         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
2762
2763         Reviewed by James Robinson.
2764
2765         * DumpRenderTree/chromium/EventSender.cpp:
2766         (EventSender::EventSender):
2767         (EventSender::gestureFlingCancel):
2768         (EventSender::gestureFlingStart):
2769         * DumpRenderTree/chromium/EventSender.h:
2770         (EventSender):
2771
2772 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2773
2774         [Qt] Add command line option to set the user agent in the MiniBrowser
2775         https://bugs.webkit.org/show_bug.cgi?id=81508
2776
2777         Reviewed by Simon Hausmann.
2778
2779         * MiniBrowser/qt/BrowserWindow.cpp:
2780         (BrowserWindow::BrowserWindow):
2781         (BrowserWindow::webViewExperimental):
2782         * MiniBrowser/qt/BrowserWindow.h:
2783         (BrowserWindow):
2784         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2785         (printHelp):
2786         (MiniBrowserApplication::handleUserOptions):
2787         * MiniBrowser/qt/MiniBrowserApplication.h:
2788         (WindowOptions::WindowOptions):
2789         (WindowOptions::setUserAgent):
2790         (WindowOptions::userAgent):
2791         (WindowOptions):
2792
2793 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2794
2795         [Qt] Remove dead user agent code from the MiniBrowser.
2796
2797         Reviewed by Simon Hausmann.
2798
2799         * MiniBrowser/qt/BrowserWindow.cpp:
2800         * MiniBrowser/qt/BrowserWindow.h:
2801         (BrowserWindow):
2802         * MiniBrowser/qt/MiniBrowser.qrc:
2803         * MiniBrowser/qt/useragentlist.txt: Removed.
2804
2805 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2806
2807         jhbuild: Call jhbuild's autogen.sh via bash.
2808         https://bugs.webkit.org/show_bug.cgi?id=81485
2809
2810         Reviewed by Philippe Normand.
2811
2812         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
2813         this is usually true only on Linux systems, and even then bash for
2814         some reason might not be available in that location. It is safer to
2815         call bash directly and pass autogen.sh to it instead.
2816
2817         * jhbuild/jhbuild-wrapper:
2818         (install_jhbuild):
2819
2820 2012-03-19  Philippe Normand  <pnormand@igalia.com>
2821
2822         [GTK] Prefix run-launcher with run-with-jhbuild
2823         https://bugs.webkit.org/show_bug.cgi?id=81503
2824
2825         Reviewed by Hajime Morita.
2826
2827         * Scripts/run-launcher:
2828
2829 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2830
2831         [EFL] Prefix run-launcher with run-with-jhbuild
2832         https://bugs.webkit.org/show_bug.cgi?id=80026
2833
2834         Reviewed by Hajime Morita.
2835
2836         * Scripts/run-launcher:
2837
2838 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
2839
2840         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
2841         https://bugs.webkit.org/show_bug.cgi?id=81498
2842
2843         * qmake/mkspecs/features/unix/default_post.prf:
2844
2845 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2846
2847         Unreviewed; switch to using my FreeBSD.org email address, the
2848         other one is going to disappear soon.
2849
2850         * Scripts/webkitpy/common/config/committers.py:
2851         * Scripts/webkitpy/common/config/watchlist:
2852
2853 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
2854
2855         Unreviewed; fix webkitpy unit test failures after r111116.
2856
2857         It turns out that cmd_line in the patch in bug 80025 had the right
2858         signature when it was submitted, but r109242 added a few more
2859         required parameters, and since the method (and EflDriver) did not
2860         exist at the time we ended up committing a broken method.
2861
2862         * Scripts/webkitpy/layout_tests/port/efl.py:
2863         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
2864         arguments.
2865
2866 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
2867
2868         [BlackBerry] Enable Web Timing for performance profiling and improvement.
2869         https://bugs.webkit.org/show_bug.cgi?id=81085
2870
2871         Reviewed by Rob Buis.
2872
2873         * Scripts/build-webkit:
2874
2875 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2876
2877         Add a dependencies installation step for EFL to buildbot master.cfg
2878         https://bugs.webkit.org/show_bug.cgi?id=81337
2879
2880         Reviewed by Gustavo Noronha Silva.
2881
2882         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2883         (InstallEflDependencies):
2884         (Factory.__init__):
2885
2886 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2887
2888         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
2889         https://bugs.webkit.org/show_bug.cgi?id=80025
2890
2891         Reviewed by Gustavo Noronha Silva.
2892
2893         * Scripts/webkitpy/layout_tests/port/efl.py:
2894         (EflDriver):
2895         (EflDriver.cmd_line):
2896         (EflPort._driver_class):
2897
2898 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2899
2900         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
2901         https://bugs.webkit.org/show_bug.cgi?id=79904
2902
2903         build-webkit --efl will use jhbuild for fetching and building dependencies.
2904
2905         Reviewed by Gustavo Noronha Silva.
2906
2907         * Scripts/update-webkitefl-libs: Added.
2908         * Scripts/webkitdirs.pm:
2909         (jhbuildWrapperPrefixIfNeeded):
2910         (generateBuildSystemFromCMakeProject):
2911         (buildCMakeGeneratedProject):
2912         * efl/common.py: Added.
2913         (script_path):
2914         (top_level_path):
2915         (number_of_cpus):
2916         * efl/jhbuild.modules: Added.
2917         * efl/jhbuildrc: Added.
2918         * efl/run-with-jhbuild: Added.
2919
2920 2012-03-16  Stephanie Lewis  <slewis@apple.com>
2921
2922         https://bugs.webkit.org/show_bug.cgi?id=81065
2923         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
2924         Fix crash preventing WebKitTestRunner from working in Release.  
2925         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
2926         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
2927         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
2928
2929         I removed ostringstream in favor of StringBuilder.  
2930
2931         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
2932
2933         Reviewed by Geoff Garen.
2934
2935         Add DEBUG_DEFINES to set NDEBUG on release builds.
2936         * WebKitTestRunner/Configurations/Base.xcconfig:
2937         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
2938
2939         Remove ostringstream member in favor of StringBuilder.
2940         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
2941         (WTR::InjectedBundle::initialize):
2942         (WTR::InjectedBundle::beginTesting):
2943         (WTR::InjectedBundle::done):
2944         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
2945         (WTF):
2946         (WTR::InjectedBundle::stringBuilder):
2947         (InjectedBundle):
2948
2949         Append output to StringBuilder.
2950         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2951         (WTR::hasPrefix):
2952         (WTR::dumpPath):
2953         (WTR::rangeToStr):
2954         (WTR::styleDecToStr):
2955         (WTR::frameToStr):
2956         (WTR::dumpFrameScrollPosition):
2957         (WTR::dumpFrameText):
2958         (WTR::dumpDescendantFramesText):
2959         (WTR::InjectedBundlePage::dump):
2960         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
2961         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
2962         (WTR::InjectedBundlePage::willSendRequestForFrame):
2963         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
2964         (WTR::lastFileURLPathComponent):
2965         (WTR::InjectedBundlePage::willAddMessageToConsole):
2966         (WTR::InjectedBundlePage::willSetStatusbarText):
2967         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
2968         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
2969         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
2970         (WTR::InjectedBundlePage::shouldBeginEditing):
2971         (WTR::InjectedBundlePage::shouldEndEditing):
2972         (WTR::InjectedBundlePage::shouldInsertNode):
2973         (WTR::InjectedBundlePage::shouldInsertText):
2974         (WTR::InjectedBundlePage::shouldDeleteRange):
2975         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
2976         (WTR::InjectedBundlePage::shouldApplyStyle):
2977         (WTR::InjectedBundlePage::didBeginEditing):
2978         (WTR::InjectedBundlePage::didEndEditing):
2979         (WTR::InjectedBundlePage::didChange):
2980         (WTR::InjectedBundlePage::didChangeSelection):
2981         (WTR::InjectedBundlePage::supportsFullScreen):
2982         (WTR::InjectedBundlePage::enterFullScreenForElement):
2983         (WTR::InjectedBundlePage::exitFullScreenForElement):
2984         (WTR::InjectedBundlePage::beganEnterFullScreen):
2985         (WTR::InjectedBundlePage::beganExitFullScreen):
2986         (WTR::dumpBackForwardListItem):
2987         (WTR::InjectedBundlePage::dumpBackForwardList):
2988         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
2989         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
2990         (WTR::LayoutTestController::dumpConfigurationForViewport):
2991         
2992         Convert between WTF::String and WKStringRef.
2993         * WebKitTestRunner/StringFunctions.h:
2994         (WTR::toWTFString):
2995         (WTR):
2996         * WebKitTestRunner/TestInvocation.cpp:
2997         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
2998
2999 2012-03-16  Eric Seidel  <eric@webkit.org>
3000
3001         Add a land-from-url command
3002         https://bugs.webkit.org/show_bug.cgi?id=81411
3003
3004         Reviewed by Adam Barth.
3005
3006         Very basic so far.  The goal is to make it landing smarter
3007         so that users don't have to bother to parse out the ids out of the urls.
3008
3009         * Scripts/webkitpy/common/config/urls.py:
3010         (parse_bug_id):
3011         (parse_attachment_id):
3012         * Scripts/webkitpy/common/config/urls_unittest.py:
3013         (URLsTest.test_parse_bug_id):
3014         (URLsTest):
3015         (URLsTest.test_parse_attachment_id):
3016         * Scripts/webkitpy/tool/commands/download.py:
3017         (ProcessURLsMixin):
3018         (ProcessURLsMixin._fetch_list_of_patches_to_process):
3019         (LandFromURL):
3020         * Scripts/webkitpy/tool/commands/download_unittest.py:
3021         (test_land_from_bug):
3022         (test_land_from_url):
3023
3024 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
3025
3026         build-webkit launcher instructions use wrong relative path for Qt
3027         https://bugs.webkit.org/show_bug.cgi?id=81357
3028
3029         Reviewed by Eric Seidel.
3030
3031         Ensure we are at the proper directory before generating run-launcher
3032         instructions.
3033
3034         * Scripts/build-webkit:
3035
3036 2012-03-16  Brady Eidson  <beidson@apple.com>
3037
3038         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
3039         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
3040
3041         Reviewed by Geoff Garen.
3042
3043         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3044         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
3045
3046 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3047
3048         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
3049         https://bugs.webkit.org/show_bug.cgi?id=81407
3050
3051         Reviewed by Adam Barth.
3052
3053         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3054         (summarize_results):
3055         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
3056         (ResultSummaryTest.summarized_results):
3057         (ResultSummaryTest.test_no_svn_revision):
3058         (ResultSummaryTest.test_svn_revision):
3059         (ResultSummaryTest):
3060         (ResultSummaryTest.test_summarized_results_wontfix):
3061
3062 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3063
3064         Have webkit-patch rebaseline-test update test_expectations.txt
3065         https://bugs.webkit.org/show_bug.cgi?id=81054
3066
3067         Recommit http://trac.webkit.org/changeset/110783 now that
3068         http://trac.webkit.org/changeset/111034 has landed.
3069
3070         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3071         (TestExpectations.remove_configuration_from_test):
3072         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3073         (RemoveConfigurationsTest):
3074         (RemoveConfigurationsTest.test_remove):
3075         (test_remove_line):
3076         * Scripts/webkitpy/tool/commands/rebaseline.py:
3077         (RebaselineTest._update_expectations_file):
3078         (RebaselineTest._rebaseline_test_and_update_expectations):
3079         (RebaselineTest):
3080         (RebaselineTest.execute):
3081         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3082         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3083         (test_rebaseline_updates_expectations_file):
3084
3085 2012-03-15  Ojan Vafai  <ojan@chromium.org>
3086
3087         Specifier collapsing when writing test expectations lines gets a number of cases wrong
3088         https://bugs.webkit.org/show_bug.cgi?id=81309
3089
3090         Reviewed by Dimitri Glazkov.
3091
3092         I've run this over all the lines in the current Chromium test_expectations.txt file,
3093         so I'm relatively confident we now cover all the cases.
3094
3095         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3096         (TestConfigurationConverter.__init__):
3097         (TestConfigurationConverter.collapse_macros):
3098         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
3099         (TestConfigurationConverter):
3100         (TestConfigurationConverter.intersect_combination):
3101         (TestConfigurationConverter.symmetric_difference):
3102         (TestConfigurationConverter.to_specifiers_list):
3103         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
3104         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
3105         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3106         (make_mock_all_test_configurations_set):
3107         (TestConfigurationConverterTest.test_symmetric_difference):
3108         (TestConfigurationConverterTest.test_to_config_set):
3109         (TestConfigurationConverterTest.test_macro_expansion):
3110         (TestConfigurationConverterTest.test_to_specifier_lists):
3111         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3112         * Scripts/webkitpy/layout_tests/port/chromium.py:
3113         (ChromiumPort):
3114         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3115         (ChromiumAndroidPort.__init__):
3116         The android port uses "arm" as it's architecture, which is technically correct,
3117         but considerably complicates making collapsing work. We probably should kill
3118         the concept of architecture entirely. The benefits are not worth the code
3119         complexity.
3120
3121         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3122         (ChromiumPortTest.test_all_test_configurations):
3123         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3124         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3125         (test_rebaseline_updates_expectations_file):
3126
3127 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
3128
3129         [Qt][Wk2] Assertion Failure and crash on file upload
3130         https://bugs.webkit.org/show_bug.cgi?id=80854
3131
3132         Reviewed by Simon Hausmann.
3133
3134         Added filePicker to WebView using experimental API.
3135
3136         * MiniBrowser/qt/MiniBrowser.qrc:
3137         * MiniBrowser/qt/icons/folder.png: Added.
3138         * MiniBrowser/qt/icons/titlebar.png: Added.
3139         * MiniBrowser/qt/icons/up.png: Added.
3140         * MiniBrowser/qt/qml/BrowserWindow.qml:
3141         * MiniBrowser/qt/qml/FilePicker.qml: Added.
3142
3143 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
3144
3145         Updating email for committer.py script.
3146
3147         * Scripts/webkitpy/common/config/committers.py:
3148
3149 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
3150
3151         [chromium] DRT: Add support for sending gesture scrolls without a begin
3152         https://bugs.webkit.org/show_bug.cgi?id=81292
3153
3154         Reviewed by Kent Tamura.
3155
3156         * DumpRenderTree/chromium/EventSender.cpp:
3157         (EventSender::EventSender):
3158         (EventSender::gestureScrollFirstPoint):
3159         * DumpRenderTree/chromium/EventSender.h:
3160         (EventSender):
3161
3162 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
3163
3164         [GTK] WebKitIconDatabase doesn't keep icons cached
3165         https://bugs.webkit.org/show_bug.cgi?id=56200
3166
3167         Reviewed by Martin Robinson.
3168
3169         Add support for favicons in GtkLauncher.
3170
3171         * GtkLauncher/main.c:
3172         (activateUriEntryCb):
3173         (iconLoadedCb):
3174         (createBrowser):
3175
3176 2012-03-16  Adam Barth  <abarth@webkit.org>
3177
3178         Remove sheriff-bot's last-green-revision command
3179         https://bugs.webkit.org/show_bug.cgi?id=81314
3180
3181         Reviewed by Eric Seidel.
3182
3183         This command is a cruel joke.  WebKit doesn't really ever have a
3184         last-green-revision.
3185
3186         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3187         (BuildBot._find_green_revision):
3188         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3189         (MockBuildBot.builder_statuses):
3190         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3191         (test_find_green_revision):
3192         * Scripts/webkitpy/tool/bot/irc_command.py:
3193         (IRCCommand.execute):
3194         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
3195         (SheriffIRCBotTest.test_help):
3196         * Scripts/webkitpy/tool/commands/queries.py:
3197         (PatchesToReview.execute):
3198
3199 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3200
3201         [Qt] Make gccdepends test aware of broken icecream deps generation
3202
3203         Icecream seems to build the files without passing -o to gcc with
3204         the original destination path, which means the target name in the
3205         dependency file is wrong.
3206
3207         Also, echo something to the final file instead of just touching it,
3208         so it's not deleted on every build.
3209
3210         Reviewed by Simon Hausmann.
3211
3212         * qmake/config.tests/gccdepends/gccdepends.pro:
3213
3214 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
3215
3216         Support for Battery Status API
3217         https://bugs.webkit.org/show_bug.cgi?id=62698
3218
3219         Add include directory and file.
3220         http://www.w3.org/TR/battery-status/
3221
3222         Reviewed by Adam Barth.
3223
3224         * Scripts/build-webkit:
3225
3226 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
3227
3228         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
3229         https://bugs.webkit.org/show_bug.cgi?id=81007
3230
3231         Reviewed by Adam Barth.
3232
3233         A dedidcated build slave is available for efl on build.webkit.org. 
3234         This change add's an entry for the elf port in list of available port builders.
3235         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
3236
3237         * Scripts/webkitpy/layout_tests/port/builders.py:
3238
3239 2012-03-16  Adam Barth  <abarth@webkit.org>
3240
3241         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
3242         https://bugs.webkit.org/show_bug.cgi?id=81302
3243
3244         Reviewed by Eric Seidel.
3245
3246         This is a trivial patch, but the variable name turned up in a grep of
3247         webkitpy and looks like copy pasta.
3248
3249         * Scripts/webkitpy/common/net/statusserver.py:
3250         (StatusServer.next_work_item):
3251
3252 2012-03-16  Andrew Lo  <anlo@rim.com>
3253
3254         [BlackBerry] Enable PAGE_VISIBILITY_API
3255         https://bugs.webkit.org/show_bug.cgi?id=81257
3256
3257         Enable Page Visibility API for BlackBerry.
3258
3259         Reviewed by Rob Buis.
3260
3261         * Scripts/build-webkit:
3262
3263 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
3264
3265         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
3266         https://bugs.webkit.org/show_bug.cgi?id=81232
3267
3268         Reviewed by Philippe Normand.
3269
3270         Use the newly-added helper method in DumpRenderTreeSupportGtk.
3271
3272         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3273         (LayoutTestController::computedStyleIncludingVisitedInfo):
3274
3275 2012-03-15  Eric Seidel  <eric@webkit.org>
3276
3277         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
3278         https://bugs.webkit.org/show_bug.cgi?id=81313
3279
3280         Reviewed by Adam Barth.
3281
3282         * Scripts/webkitpy/common/checkout/changelog.py:
3283         (parse_bug_id_from_changelog):
3284         * Scripts/webkitpy/common/config/urls.py:
3285         (parse_bug_id):
3286         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
3287         (URLsTest):
3288         (URLsTest.test_parse_bug_id):
3289         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
3290         (test_parse_bug_id):
3291         * Scripts/webkitpy/tool/bot/irc_command.py:
3292         (_post_error_and_check_for_bug_url):
3293         * Scripts/webkitpy/tool/bot/sheriff.py:
3294         (Sheriff.post_rollout_patch):
3295         (Sheriff.post_chromium_deps_roll):
3296
3297 2012-03-15  Adam Barth  <abarth@webkit.org>
3298
3299         The commit-queue should fast-track patches that have already passed the testing EWS bots
3300         https://bugs.webkit.org/show_bug.cgi?id=81305
3301
3302         Reviewed by Eric Seidel.
3303
3304         If a patch has already passed the testing EWS bots (currently only the
3305         chromium-ews bot), we should be able to land the patch without running
3306         the tests again.  This patch is somewhat risky in that the tests might
3307         have been broken between when the EWS bots ran and when the
3308         commit-queue is processing the patch.  If that turns out to be a
3309         problem in practice, we'll tighten up this fast-track in some way,
3310         likely by making sure that the pass status is recent.
3311
3312         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
3313         (CommitQueueTaskDelegate.did_pass_testing_ews):
3314         (CommitQueueTask._did_pass_tests_recently):
3315         (CommitQueueTask.run):
3316         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3317         (MockCommitQueue.did_pass_testing_ews):
3318         (test_fast_success_case):
3319         * Scripts/webkitpy/tool/commands/queues.py:
3320         (CommitQueue.did_pass_testing_ews):
3321         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3322         (test_did_pass_testing_ews):
3323
3324 2012-03-15  Kentaro Hara  <haraken@chromium.org>
3325
3326         [PerformanceTests] run-perf-tests should output correct units
3327         https://bugs.webkit.org/show_bug.cgi?id=78303
3328
3329         Reviewed by Ryosuke Niwa.
3330
3331         Previously run-perf-tests just supported "ms" for units.
3332         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
3333         This patch fixes run-perf-tests so that they can support custom units.
3334         The default unit is "ms".
3335
3336         The test result looks like this:
3337
3338         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
3339         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
3340         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
3341
3342         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3343         (PerfTestsRunner._process_parser_test_result):
3344
3345 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
3346
3347         [WinCairo] Unreviewed build change after wtf path changes.
3348
3349         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
3350         directories to look at the private header directory to find 'wtf'.
3351
3352 2012-03-15  Kent Tamura  <tkent@chromium.org>
3353
3354         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
3355         https://bugs.webkit.org/show_bug.cgi?id=81293
3356
3357         Reviewed by Adam Barth.
3358
3359         r110595 introduced new format of the author field of ChangeLog entry:
3360         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
3361         Updte chanelog.py to support it.
3362
3363         * Scripts/webkitpy/common/checkout/changelog.py:
3364         (ChangeLogEntry._split_contributor_names):
3365         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
3366         (test_parse_authors):
3367
3368 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
3369
3370         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
3371
3372         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
3373
3374         Reviewed by Jessie Berlin.
3375
3376         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3377         (LayoutTestController::setValueForUser):
3378         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3379         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3380         (WTR::LayoutTestController::setValueForUser):
3381         (WTR):
3382         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3383         (LayoutTestController):
3384
3385 2012-03-15  Adam Barth  <abarth@webkit.org>
3386
3387         Update the start-queue.sh script to match what we're using on the bots
3388         https://bugs.webkit.org/show_bug.cgi?id=81277
3389
3390         Reviewed by Eric Seidel.
3391
3392         Over time we've added more magic git commands to clean up the local repo after
3393         it has been borked in various ways. The idea behind this sequence is to clear
3394         out any changes in the working copy and then to rebuild the master branch from
3395         origin/master.
3396
3397         This patch also removes the watchdog timer that we've stopped using on the
3398         bots.
3399
3400         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
3401         Note: Eric and I are experimenting with GitHub to see if it might be
3402         useful for WebKit.  If you have thoughts on this topic, please feel
3403         free to grab Eric or me on #webkit to chat.  I assume that he and I are
3404         the only ones who care about this file so using this alternative
3405         process for this patch shouldn't cause any trouble.
3406
3407         * EWSTools/start-queue.sh:
3408
3409 2012-03-15  Adam Barth  <abarth@webkit.org>
3410
3411         The commit-queue can hang when the test_expectations.txt style check fails during commit
3412         https://bugs.webkit.org/show_bug.cgi?id=81251
3413
3414         Reviewed by Eric Seidel.
3415
3416         We forgot to check the --non-interactive flag, so we end up prompting
3417         the "user" who doesn't exist on the bot.  We might consider teaching
3418         the User class about non-interactive if we have this sort of bug again.
3419
3420         * Scripts/webkitpy/tool/steps/commit.py:
3421         (Commit.options):
3422         (Commit._check_test_expectations):
3423         (Commit.run):
3424         * Scripts/webkitpy/tool/steps/commit_unittest.py:
3425         (CommitTest.test_check_test_expectations):
3426
3427 2012-03-15  Jessie Berlin  <jberlin@apple.com>
3428
3429         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
3430         Mac WK2 because WebAudio is not enabled
3431         https://bugs.webkit.org/show_bug.cgi?id=81259
3432
3433         Reviewed by Dan Bernstein.
3434
3435         Enable WebAudio so that it appears in the window object in the tests.
3436         The webaudio tests are all still skipped in the Mac Skipped list.
3437
3438         * WebKitTestRunner/TestController.cpp:
3439         (WTR::TestController::resetStateToConsistentValues):
3440
3441 2012-03-15  David Levin  <levin@chromium.org>
3442
3443         Various autoresize fixes.
3444         https://bugs.webkit.org/show_bug.cgi?id=81199
3445
3446         Reviewed by Darin Fisher.
3447
3448         * DumpRenderTree/chromium/LayoutTestController.cpp:
3449         (LayoutTestController::reset): Stop using the deprecated api that was removed.
3450
3451 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3452
3453         [GTK] Actually use --release and --debug command line options in run-gtk-test
3454         https://bugs.webkit.org/show_bug.cgi?id=81234
3455
3456         Reviewed by Philippe Normand.
3457
3458         Actually handle --release and --debug command line options to
3459         create the build directory, falling back to current hack to detect
3460         whether it's a Release or Debug build when both options are missed.
3461
3462         * Scripts/run-gtk-tests:
3463         (TestRunner.__init__): Receive options too.
3464         (TestRunner._get_top_level_directory): Helper function to get the
3465         top level directory.
3466         (TestRunner._get_build_directory): Helper function to get the
3467         build directory depending on --release/--debug command line
3468         options or gessing it if both options are missing.
3469         (TestRunner._ensure_accessibility_daemon_is_running): Use
3470         self-options now.
3471         (TestRunner.run): Ditto.
3472         (TestRunner.run.run_tests): Pass options to the constructor
3473         instead of run().
3474
3475 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3476
3477         [GTK] Run unit tests with --verbose in the bots
3478         https://bugs.webkit.org/show_bug.cgi?id=81230
3479
3480         Reviewed by Martin Robinson.
3481
3482         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3483         (RunGtkAPITests): Add --verbose to run-gtk-test command.
3484
3485 2012-03-15  Kevin Ollivier  <kevino@theolliviers.com>
3486
3487         Move wx port to using export macros
3488         https://bugs.webkit.org/show_bug.cgi?id=77279
3489
3490         Reviewed by Hajime Morita.
3491
3492         * waf/build/settings.py:
3493         (common_configure):
3494
3495 2012-03-15  Philippe Normand  <pnormand@igalia.com>
3496
3497         Unreviewed, GTK build fix attempt after r110845.
3498
3499         * Scripts/run-gtk-tests: Added --release and --debug options.
3500
3501 2012-03-15  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3502
3503         Parametrize run-with-jhbuild and update-webkitgtk-libs with platform --gtk/--efl
3504         https://bugs.webkit.org/show_bug.cgi?id=79673
3505
3506         Preparatory work for bringing jhbuild enabled build to EFL port.
3507
3508         Reviewed by Philippe Normand.
3509
3510         * Scripts/update-webkit-libs-jhbuild: Copied from Tools/Scripts/update-webkitgtk-libs.
3511         * Scripts/update-webkitgtk-libs:
3512         * gtk/run-with-jhbuild:
3513         * jhbuild/jhbuild-wrapper: Copied from Tools/gtk/run-with-jhbuild.
3514         (top_level_path):
3515         (jhbuild_installed):
3516         (jhbuild_cloned):
3517         (jhbuild_at_expected_revision):
3518         (update_jhbuild):
3519         (clone_jhbuild):
3520         (install_jhbuild):
3521         (update_webkit_libs_jhbuild):
3522         (determine_platform):
3523         (ensure_jhbuild):
3524
3525 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3526
3527         [GTK] Add command line options to run-gtk-tests script
3528         https://bugs.webkit.org/show_bug.cgi?id=81218
3529
3530         Reviewed by Philippe Normand.
3531
3532         Following command line options have been added:
3533
3534         -v, --verbose: to run gtester with --verbose. That would be used
3535         by the bots to make it easier to know which test case it failing.
3536
3537         --display: to use a different display for Xvfb
3538
3539         It's also possible to pass one or more tests as arguments.
3540
3541         * Scripts/run-gtk-tests:
3542         (TestRunner.__init__): Receive command line arguments.
3543         (TestRunner._get_tests): Get all the tests or the ones passed by
3544         command line arguments.
3545         (TestRunner._ensure_accessibility_daemon_is_running): Use display
3546         command line argument.
3547         (TestRunner.run): Pas command line options.
3548         (TestRunner.run.run_tests): Pass command line arguments.
3549
3550 2012-03-15  Carlos Garcia Campos  <cgarcia@igalia.com>
3551
3552         [GTK] WebKit2APITests/TestPrinting timeout in the bots
3553         https://bugs.webkit.org/show_bug.cgi?id=81211
3554
3555         Reviewed by Philippe Normand.
3556
3557         Add gtk+ patch to fix timeout in the bots. A similar patch has
3558         already landed in updatream gtk+, see
3559         http://git.gnome.org/browse/gtk+/commit/?id=5ebfad5d091f4c9d9c6e60364fad4c60842f027f
3560
3561         * gtk/jhbuild.modules:
3562         * gtk/patches/gtk+-3.0.12-gtkprinter.patch: Added.
3563
3564 2012-03-14  Gustavo Noronha Silva  <gns@gnome.org>
3565
3566         sheriffbot should also be addressable with a comma in addition to colon
3567         https://bugs.webkit.org/show_bug.cgi?id=81101
3568
3569         Reviewed by Adam Barth.
3570
3571         * Scripts/webkitpy/common/net/irc/ircbot.py:
3572         (IRCBot.on_pubmsg):
3573
3574 2012-03-15  Alexander Færøy  <alexander.faeroy@nokia.com>
3575
3576         [Qt] MiniBrowser scroll indicators should not scroll out of view
3577         https://bugs.webkit.org/show_bug.cgi?id=81131
3578
3579         Reviewed by Kenneth Rohde Christiansen.
3580
3581         * MiniBrowser/qt/qml/ScrollIndicator.qml:
3582
3583 2012-03-14  Chris Fleizach  <cfleizach@apple.com>
3584
3585         AX: ARIA tables need to support selected rows attribute/select rows changed notification
3586         https://bugs.webkit.org/show_bug.cgi?id=79821
3587
3588         Reviewed by Beth Dakin.
3589
3590         * DumpRenderTree/AccessibilityUIElement.cpp:
3591         (rowAtIndexCallback):
3592         (AccessibilityUIElement::getJSClass):
3593         * DumpRenderTree/AccessibilityUIElement.h:
3594         (AccessibilityUIElement):
3595         * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
3596         (AccessibilityUIElement::rowAtIndex):
3597         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3598         (AccessibilityUIElement::rowAtIndex):
3599         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
3600         (AccessibilityUIElement::rowAtIndex):
3601         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
3602         (AccessibilityUIElement::rowAtIndex):
3603
3604 2012-03-14  Tim Horton  <timothy_horton@apple.com>
3605
3606         NRWT should respect --additional-platform-directory's Skipped list
3607         https://bugs.webkit.org/show_bug.cgi?id=81172
3608
3609         Reviewed by Timothy Hatcher.
3610
3611         Respect the Skipped file inside all paths included via the --additional-platform-directory option. Test results
3612         from said path are already used, but the Skipped file is ignored.
3613
3614         * Scripts/webkitpy/layout_tests/port/webkit.py:
3615         (WebKitPort._skipped_file_search_paths):
3616         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3617         (test_skipped_file_search_paths):
3618
3619 2012-03-14  Ojan Vafai  <ojan@chromium.org>
3620
3621         Unreviewed, rolling out r110783.
3622         http://trac.webkit.org/changeset/110783
3623         https://bugs.webkit.org/show_bug.cgi?id=81054
3624
3625         Caused existing test expectation lines to expand to one for
3626         each test configuration.
3627
3628         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3629         (TestExpectations.has_warnings):
3630         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3631         (test_macro_overrides):
3632         * Scripts/webkitpy/tool/commands/rebaseline.py:
3633         (RebaselineTest._save_baseline):
3634         (RebaselineTest._rebaseline_test):
3635         (RebaselineTest.execute):
3636         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3637         (TestRebaseline.test_tests_to_update):
3638
3639 2012-03-13  Jon Lee  <jonlee@apple.com>
3640
3641         Separate NOTIFICATIONS and LEGACY_NOTIFICATIONS
3642         https://bugs.webkit.org/show_bug.cgi?id=80922
3643         <rdar://problem/11035082>
3644
3645         Reviewed by Jian Li.
3646
3647         You can include either NOTIFICATIONS or LEGACY_NOTIFICATIONS and have a complete API.
3648         LEGACY_NOTIFICATIONS should cover all of the previous functionality, and NOTIFICATIONS will cover the
3649         new API. Therefore, APIs that are common between the two will have:
3650         #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
3651
3652         This patch initially sets both to be exactly the same. As other bugs with patches begin to migrate to
3653         the new API, the defines will begin to split. This allows ports to decide which set of APIs to include.
3654
3655         Update everything to be #if ENABLE(NOTIFICATIONS) || ENABLE(LEGACY_NOTIFICATIONS)
3656         * DumpRenderTree/chromium/LayoutTestController.cpp:
3657         (LayoutTestController::grantDesktopNotificationPermission):
3658         (LayoutTestController::simulateDesktopNotificationClick):
3659         * DumpRenderTree/chromium/NotificationPresenter.cpp:
3660         * DumpRenderTree/chromium/TestShell.cpp:
3661         (TestShell::TestShell):
3662         (TestShell::resetTestController):
3663         * DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
3664         (DRTDesktopNotificationPresenter::checkNotificationPermission):
3665
3666 2012-03-13  Ojan Vafai  <ojan@chromium.org>
3667