[GTK][EFL] run-javascriptcore-tests should be run through jhbuild
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-29  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2
3         [GTK][EFL] run-javascriptcore-tests should be run through jhbuild
4         https://bugs.webkit.org/show_bug.cgi?id=82581
5
6         Reviewed by Martin Robinson.
7
8         Running Javascriptcore tests through jhbuild
9         for consistency with run-webkit-tests and in order to
10         avoid confusing libraries when facing regressions.
11
12         * Scripts/run-javascriptcore-tests:
13
14 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
15
16         test-webkitpy: add --timing
17         https://bugs.webkit.org/show_bug.cgi?id=82550
18
19         Reviewed by Eric Seidel.
20
21         This patch adds a --timing option that will display the time
22         each test takes. It also removes the --silent option, since
23         probably no one ever used it, and cleans up the logging
24         option parsing code to be easier to follow.
25
26         * Scripts/webkitpy/test/main.py:
27         (Tester._parse_args):
28         (Tester._configure):
29         * Scripts/webkitpy/test/runner.py:
30         (TestRunner.write_result):
31         * Scripts/webkitpy/test/runner_unittest.py:
32         (RunnerTest.test_regular):
33         (RunnerTest.test_verbose):
34         (RunnerTest):
35         (RunnerTest.test_timing):
36
37 2012-03-29  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
38
39         HashMap<>::add should return a more descriptive object
40         https://bugs.webkit.org/show_bug.cgi?id=71063
41
42         Reviewed by Ryosuke Niwa.
43
44         Update code to use AddResult instead of a pair.
45
46         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
47         (LayoutTestController::evaluateScriptInIsolatedWorld):
48         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
49         (LayoutTestController::evaluateScriptInIsolatedWorld):
50         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
51         (WTR::LayoutTestController::evaluateScriptInIsolatedWorld):
52
53 2012-03-29  Ilya Tikhonovsky  <loislo@chromium.org>
54
55         Web Inspector: chromium: DRT --no-timeout option doesn't work.
56         https://bugs.webkit.org/show_bug.cgi?id=82608
57
58         Initial value for m_timeout was initialized in constructor and was overwritten in DRT::main.
59         This was broken by r112354 and the sequence became opposite.
60
61         Reviewed by Yury Semikhatsky.
62
63         * DumpRenderTree/chromium/TestShell.cpp:
64         (TestShell::TestShell):
65         (TestShell::initialize):
66
67 2012-03-29  Vineet Chaudhary  <rgf748@motorola.com>
68
69         Remove custom bindings form Internals.idl of attribute type Array.
70         https://bugs.webkit.org/show_bug.cgi?id=82319
71
72         Reviewed by Kentaro Hara.
73
74         Remove JSInternalsCustom.cpp and V8InternalsCustom.cpp as no longer required.
75
76         * GNUmakefile.am:
77
78 2012-03-28  Mark Rowe  <mrowe@apple.com>
79
80         Fix --no-saved-state.
81
82         Ignore the flag before Lion, rather than making it specific to Lion.
83
84         Rubber-stamped by Dan Bernstein.
85
86         * Scripts/webkitdirs.pm:
87         (argumentsForRunAndDebugMacWebKitApp):
88
89 2012-03-28  Dave Tharp  <dtharp@codeaurora.org>
90
91         Add dtharp to contributors list
92         https://bugs.webkit.org/show_bug.cgi?id=82545
93
94         Reviewed by Adam Barth.
95
96         Adding dtharp as Contributor.
97
98         * Scripts/webkitpy/common/config/committers.py:
99
100 2012-03-28  Stuart Eichert  <stuart.eichert@palm.com>
101
102         build-webkit : the word "pass" is spelled incorrectly as "passs"
103         https://bugs.webkit.org/show_bug.cgi?id=82532
104
105         Reviewed by Julien Chaffraix.
106
107         Corrected spelling of the word "pass" in build-webkit.
108
109         * Scripts/build-webkit:
110
111 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
112
113         new-run-webkit-tests doesn't log correctly after retrying failures
114         https://bugs.webkit.org/show_bug.cgi?id=82541
115
116         Unreviewed, build fix.
117
118         r112189 introduced a regression where we would accidentally
119         delete the log handler after we retried any failing tests - that
120         meant that any log messages around uploading the results got
121         dropped.
122
123         * Scripts/webkitpy/layout_tests/controllers/manager.py:
124         (Manager._run_tests):
125
126 2012-03-28  Kevin Ollivier  <kevino@theolliviers.com>
127
128         [wx] Unreviewed. Build fix, move WTF back into JSCore target
129         until issues with JSCore not linking in all WTF symbols are resolved.
130
131         * waf/build/settings.py:
132
133 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
134
135         download-built-product build step should detect 404 errors
136         https://bugs.webkit.org/show_bug.cgi?id=82491
137
138         Reviewed by Ryosuke Niwa.
139         
140         Pass --fail to curl so that it reports 404 errors.
141
142         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
143         (DownloadBuiltProduct): Pass --fail to curl so that it reports errors.
144         * BuildSlaveSupport/built-product-archive:
145         (extractBuiltProduct): Remove trailing whitespace.
146
147 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
148
149         flakiness dashboard should only keep the last 500 runs
150         https://bugs.webkit.org/show_bug.cgi?id=82530
151
152         Reviewed by Ojan Vafai.
153
154         If a bot has an overly large amount of failing tests, then
155         storing 750 runs' worth takes a lot of memory and CPU time to
156         process and we end up getting killed by AppEngine; this cases
157         data corruption and confusion down the road when we have to
158         retry the upload of the data and the tests show up as multiple
159         runs.
160
161         * TestResultServer/model/jsonresults.py:
162         (JsonResults.update_file):
163
164 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
165
166         extract-built-product build step should fail if unzipping fails
167         https://bugs.webkit.org/show_bug.cgi?id=82493
168
169         Reviewed by Simon Fraser.
170
171         Forward non-zero exit code from subprocesses so that the master can detect any failures
172         and bail out early.
173
174         * BuildSlaveSupport/built-product-archive:
175         (main):
176
177 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
178
179         test-webkitpy should support files, directories, and packages as command line args
180         https://bugs.webkit.org/show_bug.cgi?id=76765
181
182         Reviewed by Adam Barth.
183
184         This patch adds support for specifying files, directories, and
185         packages to test-webkitpy along with the already existing
186         support for modules, test classes, and individual test names.
187
188         Also, fix a bug in filesystem_mock where we wouldn't normalize a
189         path containing a reference to the current directory properly,
190         for example, '/foo/./bar.py'.
191
192         * Scripts/webkitpy/common/system/filesystem_mock.py:
193         (MockFileSystem.normpath):
194         * Scripts/webkitpy/test/main.py:
195         (Tester._configure_logging):
196         (Tester._run_tests):
197         * Scripts/webkitpy/test/test_finder.py:
198         (TestDirectoryTree.find_modules):
199         (TestDirectoryTree.subpath):
200         (TestFinder.is_dotted_name):
201         (TestFinder.find_names):
202         (TestFinder):
203         (TestFinder._find_names_for_arg):
204         (TestFinder._find_in_trees):
205         (TestFinder._default_names):
206         * Scripts/webkitpy/test/test_finder_unittest.py:
207         (TestFinderTest.setUp):
208         (TestFinderTest.tearDown):
209         (TestFinderTest.check_names):
210         (TestFinderTest.test_default_names):
211         (TestFinderTest):
212         (TestFinderTest.test_paths):
213
214 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
215
216         Fix inspector tests when running in internal builds.
217
218         <rdar://problem/11050694>
219
220         Reviewed by Tim Hatcher.
221
222         * DumpRenderTree/mac/DumpRenderTree.mm:
223         (resetDefaultsToConsistentValues):
224         * WebKitTestRunner/TestController.cpp:
225         (WTR::TestController::resetStateToConsistentValues):
226
227 2012-03-28  David Kilzer  <ddkilzer@apple.com>
228
229         Use xcrun to find nm and ranlib on OS X builds
230         <http://webkit.org/b/82466>
231
232         Reviewed by Dan Bernstein.
233
234         * Scripts/build-webkit: Use xcrun to find ranlib.
235         * Scripts/webkitdirs.pm:
236         (determineNmPath): Added.  Use xcrun to find nm on OS X.
237         (nmPath): Added.
238         * Scripts/webkitperl/features.pm:
239         (libraryContainsSymbol): Use nmPath().
240
241 2012-03-28  Ojan Vafai  <ojan@chromium.org>
242
243         Add history navigation to garden-o-matic
244         https://bugs.webkit.org/show_bug.cgi?id=82495
245
246         Reviewed by Dimitri Glazkov.
247
248         Also, maintain scroll offsets when returning to a tab. This makes addressing
249         expected failures considerably easier when dealing with below-the-fold tests.
250
251         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
252         -Store the scrollTop before switching to a new tab and restore the appropriate scrollTop after
253         switching.
254         -Modify window.location with the tabName in the hash. Can't use pushState because this is served
255         from a file URL and Chrome puts each different file URL in it's own origin.
256
257         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
258         Remove the href so that clicking on the accordion item does not modify window.location.hash.
259
260         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
261         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
262
263 2012-03-28  Philippe Normand  <pnormand@igalia.com>
264
265         [GTK] generate-gtk-doc doesn't cope with custom build directory
266         https://bugs.webkit.org/show_bug.cgi?id=82448
267
268         Reviewed by Martin Robinson.
269
270         Pass the product directory in generate-gtkdoc arguments from
271         build-webkit.
272
273         * Scripts/webkitdirs.pm:
274         (buildAutotoolsProject):
275         * gtk/common.py:
276         (get_build_path):
277
278 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
279
280         [GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
281         https://bugs.webkit.org/show_bug.cgi?id=82341
282
283         Reviewed by Martin Robinson.
284
285         Add --skipped=skip|ignore|only command line option.
286
287         * Scripts/run-gtk-tests:
288         (TestRunner._test_cases_to_skip): Return the list of test cases
289         to skip for the given test depending on the skipped action.
290         (TestRunner._should_run_test): Decide whether to run the test or
291         not depending on the skipped action.
292         (TestRunner.run_tests): Only show the list of skipped tests when
293         skipped action is 'skip'.
294
295 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
296
297         [GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
298         https://bugs.webkit.org/show_bug.cgi?id=82333
299
300         Reviewed by Martin Robinson.
301
302         * Scripts/run-gtk-tests:
303         (TestRunner.SkippedTest): Add SkippedTest class to store more
304         information about skipped tests.
305         (TestRunner._setup_testing_environment): Use SkippedTest to add
306         TestWebKitAccessibility to the skipped list.
307         (TestRunner._find_skipped_test): Return the SkippedTest object for
308         a given test path.
309         (TestRunner._test_cases_to_skip): Returns the list of test cases
310         that should be skipped for a given test path.
311         (TestRunner._should_run_test): Helper funtion to decide whether
312         the given test path should be run or not. Tests in the skipped
313         list that don't have a list of failing test cases are not run at
314         all. Tests in the skipped list that contain a list of test cases
315         are run with gtester -s to skip the individual test cases.
316         (TestRunner._run_test): Helper funtion to run a given test.
317         (TestRunner.run_tests): Build the list of tests to run based on
318         _should_run_test() and use helper function _run_test() to run
319         every test.
320
321 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
322
323         [Qt] Couple of tests have different results on 64 bit and/or in debug mode compared to 32 bit and/or release mode
324         https://bugs.webkit.org/show_bug.cgi?id=52810
325
326         Reviewed by Simon Hausmann.
327
328         Use SSE2 floating point math on 32 bit instead of the default
329         387 to make layout test result same on 32 and on 64 bit builds.
330
331         * qmake/mkspecs/features/unix/default_post.prf:
332
333 2012-03-28  Sergio Villar Senin  <svillar@igalia.com>
334
335         [Soup] DNS prefetching spams resolver, shoots self in the foot
336         https://bugs.webkit.org/show_bug.cgi?id=41630
337
338         Reviewed by Martin Robinson.
339
340         Bump libsoup and glib dependencies.
341
342         * efl/jhbuild.modules:
343         * gtk/jhbuild.modules:
344
345 2012-03-28  Gustavo Noronha Silva  <gns@gnome.org>
346
347         [GTK] Disable introspection for jhbuild modules
348         https://bugs.webkit.org/show_bug.cgi?id=82395
349
350         Reviewed by Philippe Normand.
351
352         * efl/jhbuild.modules: add --disable-introspection to all modules
353         that support it
354         * gtk/jhbuild.modules: ditto.
355
356 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
357
358         [Qt] Unreviewed, fix API tests after adding full platforms. (r111658)
359
360         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
361         (BuildAndTestFactory.__init__):
362
363 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
364
365         [Qt] Unreviewed typo fix after r112383.
366
367         * Scripts/run-qtwebkit-tests:
368         (Main.check_crash_occurences):
369
370 2012-03-28  Ádám Kallai  <kadam@inf.u-szeged.hu>
371
372         [Qt] run-qtwebkit-tests should report crashes
373         https://bugs.webkit.org/show_bug.cgi?id=76905
374
375         Added checking of crashing tests. Now final summary includes the total number of crashing tests too.
376         Made unit test for this modification by János Badics <jbadics@inf.u-szeged.hu>.
377
378         Reviewed by Csaba Osztrogonác.
379
380         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
381         (RunQtAPITests.commandComplete):
382         (RunQtAPITests.evaluateCommand):
383         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
384         (RunQtAPITestsTest):
385         (test_failure):
386         (test_crash):
387         * Scripts/run-qtwebkit-tests:
388         (Main.announce_results_html):
389         (Main):
390         (Main.check_crash_occurences):
391         (Main.convert_to_stdout):
392         (Main.convert_to_html):
393
394 2012-03-28  Kwonjin Jeong  <gram@company100.net>
395
396         Add support for Intel compiler on Linux QT port.
397         https://bugs.webkit.org/show_bug.cgi?id=81077
398
399         Reviewed by Tor Arne Vestbø.
400
401         * qmake/mkspecs/features/default_post.prf:
402         Don't hide symbols when building with Intel compiler.
403         The symbols are needed when building QtWebKit tests.
404
405         * qmake/mkspecs/features/unix/icc.prf:
406         Added to ignore icc-specific warnings.
407         This feature is used with command-line arguments for build-webkit script.
408
409 2012-03-28  Philippe Normand  <pnormand@igalia.com>
410
411         Unreviewed, trivial update-webkit-libs-jhbuild system call fix.
412
413         * Scripts/update-webkitgtk-libs: This script is written in
414         Perl. Not Python. So use perl vocabulary to refer to command line arguments.
415
416 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
417
418         kill-old-processes doesn't do anything on linux
419         https://bugs.webkit.org/show_bug.cgi?id=82388
420
421         Reviewed by Tony Chang.
422
423         Kill cc1plus, ld, etc... on linux. Also ps aux + grep to kill old run-webkit-tests processes.
424         This should help stabilizing Chromium linux bots.
425
426         * BuildSlaveSupport/kill-old-processes:
427
428 2012-03-28  Philippe Normand  <pnormand@igalia.com>
429
430         [GTK] jhbuild wrongly tricked into clean-up
431         https://bugs.webkit.org/show_bug.cgi?id=82216
432
433         Reviewed by Gustavo Noronha Silva.
434
435         Save jhbuild{rc,.modules}.md5sum in Dependencies instead of the
436         build tree. Also when checking if the jhbuild configuration
437         changed use absolute md5sum paths.
438
439         * Scripts/webkitdirs.pm:
440         (runAutogenForAutotoolsProjectIfNecessary):
441         (jhbuildConfigurationChanged):
442         (buildAutotoolsProject):
443
444 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
445
446         kill-old-processes doesn't do anything on linux
447         https://bugs.webkit.org/show_bug.cgi?id=82388
448
449         One more attempt to disable kill-old-processes on the Qt bots not to kill each others.
450
451         The platform in Factory comes from config.json, accordingly platform is fullplatform,
452         for example: qt-4.8, qt-5.0-wk2. We have to check only the prefix of platform.
453
454         Reviewed by Ryosuke Niwa.
455
456         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
457         (Factory.__init__):
458
459 2012-03-28  Hao Zheng  <zhenghao@chromium.org>
460
461         Make Chromium port for Android use hardware gpu path default.
462         https://bugs.webkit.org/show_bug.cgi?id=82033
463
464         Reviewed by Dirk Pranke.
465
466         Skip canvas tests, which are run in virtual gpu test suites.
467
468         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
469         (ChromiumAndroidPort.__init__):
470         (ChromiumAndroidPort.skipped_tests):
471
472 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
473
474         kill-old-processes doesn't do anything on linux
475         https://bugs.webkit.org/show_bug.cgi?id=82388
476
477         Disable kill-old-processes on the Qt bots not to kill each others.
478
479         Reviewed by Ryosuke Niwa.
480
481         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
482         (Factory.__init__):
483
484 2012-03-27  Hao Zheng  <zhenghao@chromium.org>
485
486         Handle DRT crash caused by Android OOM.
487         https://bugs.webkit.org/show_bug.cgi?id=82310
488
489         Reviewed by Tony Chang.
490
491         When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
492         is stopped silently and regarded as crashed. Re-run the test for
493         such crash.
494
495         Add some debug log to NRWT.
496
497         * Scripts/webkitpy/layout_tests/port/chromium.py:
498         (ChromiumDriver.run_test):
499         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
500         (ChromiumAndroidPort.__init__):
501         (ChromiumAndroidPort.get_last_stacktrace):
502         (ChromiumAndroidDriver.__init__):
503         (ChromiumAndroidDriver._start):
504         (ChromiumAndroidDriver.run_test):
505         (ChromiumAndroidDriver._get_drt_return_value):
506
507 2012-03-27  Ojan Vafai  <ojan@chromium.org>
508
509         Fix expected results for some unittest failures.
510
511         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
512
513 2012-03-27  Adrienne Walker  <enne@google.com>
514
515         [chromium] Fix --enabled-threaded-compositing flag in DRT
516         https://bugs.webkit.org/show_bug.cgi?id=82405
517
518         Reviewed by James Robinson.
519
520         --enable-threaded-compositing wasn't working because there was no way
521         to set the value before it was used in the TestShell constructor.
522
523         To avoid adding yet another boolean parameter to TestShell's
524         constructor, add an explicit initialize function that can operate on
525         any boolean settings that are set post-constructor.
526
527         * DumpRenderTree/chromium/DumpRenderTree.cpp:
528         (main):
529         * DumpRenderTree/chromium/TestShell.cpp:
530         (TestShell::TestShell):
531         (TestShell::initialize):
532         * DumpRenderTree/chromium/TestShell.h:
533         (TestShell):
534         (TestShell::setTestShellMode):
535
536 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
537
538         new-run-webkit-tests: suppress debug output from run-safari on results file
539         https://bugs.webkit.org/show_bug.cgi?id=82400
540
541         Reviewed by Ryosuke Niwa.
542
543         Send the run-safari output to /dev/null rather than the console.
544
545         * Scripts/webkitpy/layout_tests/port/mac.py:
546         (MacPort.show_results_html_file):
547
548 2012-03-27  Simon Fraser  <simon.fraser@apple.com>
549
550         Improve error reporting in run-javascriptcore-tests
551         https://bugs.webkit.org/show_bug.cgi?id=82379
552
553         Reviewed by Jessie Berlin.
554         
555         When the script dies, have it print out the reason for dying.
556
557         * Scripts/run-javascriptcore-tests:
558
559 2012-03-27  Eric Seidel  <eric@webkit.org>
560
561         Add Dave Barton to the Contributors list so I can more easily CC him on bugs
562         https://bugs.webkit.org/show_bug.cgi?id=82373
563
564         Reviewed by Ryosuke Niwa.
565
566         Dave's committer-status in in-progress, so until then
567         list him as a Contributor so at least I can more easily CC him on bugs.
568
569         * Scripts/webkitpy/common/config/committers.py:
570
571 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
572
573         [jhbuild] Prepend jhbuildrc's directory to sys.path instead of appending.
574         https://bugs.webkit.org/show_bug.cgi?id=82252
575
576         Reviewed by Gustavo Noronha Silva.
577
578         Appending the directory containing common.py may be problematic if
579         another path in sys.path also contains either a common.py or a
580         common/ module, so prepend to select our version instead. This is
581         the case with fail2ban, which has a common/ directory which may
582         end up being in Python's default path.
583
584         * efl/jhbuildrc:
585         * gtk/jhbuildrc:
586
587 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
588
589         test-webkitpy: prepare for better test run output
590         https://bugs.webkit.org/show_bug.cgi?id=82290
591
592         Reviewed by Adam Barth.
593
594         This code basically re-implements the output of the TextTestRunner default
595         runner code from unittest, although the implementation is quite
596         different, in preparation for changing the test output to be
597         metered and possibly running in parallel.
598
599         The output is almost identical to before, except that instead of
600         logging "test_regular (webkitpy.main.RunnerTest) passed" we log
601         "webkitpy.main.RunnerTest.test_regular passed". It has always
602         annoyed me that they invert the names to be harder to read and
603         so that you can't copy & paste back to the input for
604         test-webkitpy.
605
606         This patch is provided to add a skeleton for unit tests and for
607         comparison to an upcoming patch that will actually add new
608         functionality.
609
610         * Scripts/webkitpy/test/main.py:
611         (Tester.__init__):
612         (Tester._configure_logging):
613         (Tester._run_tests):
614         * Scripts/webkitpy/test/runner.py: Added.
615         (TestRunner):
616         (TestRunner.__init__):
617         (TestRunner.test_name):
618         (TestRunner.all_test_names):
619         (TestRunner.run):
620         (TestRunner.write_result):
621         (TestRunner.write_summary):
622         * Scripts/webkitpy/test/runner_unittest.py: Added.
623         (FakeModuleSuite):
624         (FakeModuleSuite.__init__):
625         (FakeModuleSuite.__str__):
626         (FakeModuleSuite.run):
627         (FakeTopSuite):
628         (FakeTopSuite.__init__):
629         (FakeLoader):
630         (FakeLoader.__init__):
631         (FakeLoader.top_suite):
632         (FakeLoader.loadTestsFromName):
633         (RunnerTest):
634         (RunnerTest.test_regular):
635         (RunnerTest.test_verbose):
636
637 2012-03-27  Gustavo Noronha Silva  <gns@gnome.org>
638
639         [GTK] Build gnutls without p11-kit support
640         https://bugs.webkit.org/show_bug.cgi?id=82212
641
642         Reviewed by Martin Robinson.
643
644         * gtk/jhbuild.modules: pass --without-p11-kit to gnutls'
645         configure, to avoid requiring that dependency, which is not used
646         by libsoup and WebKit anyway
647
648 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
649
650         [Qt] Make debug build work on 32 bit
651         https://bugs.webkit.org/show_bug.cgi?id=82331
652
653         Use --no-keep-memory option of ld not to cache the symbol tables of input files in
654         memory to avoid memory exhaustion during the linking phase for x86-debug builds.
655
656         Reviewed by Tor Arne Vestbø.
657
658         * qmake/mkspecs/features/unix/default_post.prf:
659         * qmake/mkspecs/features/unix/default_pre.prf:
660
661 2012-03-23  Balazs Ankes  <bank@inf.u-szeged.hu>
662
663         Add png checking to check-webkit-style
664         https://bugs.webkit.org/show_bug.cgi?id=75824
665
666         Reviewed by NOBODY Tony Chang.
667
668         * Scripts/webkitpy/style/checker.py:
669         (_all_categories):
670         (FileType):
671         (CheckerDispatcher._should_skip_file_path):
672         (CheckerDispatcher._file_type):
673         (CheckerDispatcher._create_checker):
674         * Scripts/webkitpy/style/checker_unittest.py:
675         (CheckerDispatcherDispatchTest.test_none_paths):
676         * Scripts/webkitpy/style/checkers/png.py: Added.
677         (PNGChecker):
678         (PNGChecker.__init__):
679         (PNGChecker.check):
680         (PNGChecker._config_file_path):
681         * Scripts/webkitpy/style/checkers/png_unittest.py: Added.
682         (MockSCMDetector):
683         (MockSCMDetector.__init__):
684         (MockSCMDetector.display_name):
685         (MockSCMDetector.propget):
686         (PNGCheckerTest):
687         (PNGCheckerTest.test_init):
688         (PNGCheckerTest.test_init.mock_handle_style_error):
689         (PNGCheckerTest.test_check):
690         (PNGCheckerTest.test_check.mock_handle_style_error):
691         * Scripts/webkitpy/style/patchreader.py:
692         (PatchReader.check):
693
694 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
695
696         Unreviewed. Skip GTK+ unit tests that fail in the bots
697
698         * Scripts/run-gtk-tests:
699         (TestRunner): Add failing tests to the skipped list.
700
701 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
702
703         Unreviewed. Unskip all GTK+ unit test.
704
705         Currently, skipped tests contains the list of unit tests that have
706         failed in one or more bots. In this moment it's not possible to
707         skip just the test cases that fail, so the complete unit test is
708         skipped when a single test case fails. For that reason, we don't
709         know which test cases are actually failing for every skipped
710         test. We are planning to land a patch to allow skipping individual
711         test cases, but first we need to know which test cases are
712         currently failing in the bots to skip them. So, this patch simply
713         unskips all unit tests in order to watch the bots and get the list
714         of test cases failing. Failing tests will be skipped again as soon
715         as we have the new list of failing test cases.
716
717         * Scripts/run-gtk-tests:
718         (TestRunner): Reset the list of skipped tests.
719
720 2012-03-26  Scott Byer  <scottbyer@chromium.org>
721
722         Enable layout testing of the scroll animator.
723         https://bugs.webkit.org/show_bug.cgi?id=81858
724         Add a call to the InternalSettings that layout tests can use to
725         turn on scroll animation. Enable animation updates for the
726         Chromium platform DRT when scroll animation has been turned on in
727         a test. This should be a no-op for all current layout tests.
728
729         Reviewed by James Robinson.
730
731         * DumpRenderTree/chromium/WebViewHost.cpp:
732         (WebViewHost::serviceAnimation):
733         (WebViewHost::scheduleAnimation):
734         * DumpRenderTree/chromium/WebViewHost.h:
735         (WebViewHost):
736
737 2012-03-26  Dinu Jacob  <dinu.jacob@nokia.com>
738
739         [Qt][WK2] Support multi-file upload
740         https://bugs.webkit.org/show_bug.cgi?id=81589
741
742         Reviewed by Simon Hausmann.
743
744         Modified filePicker to support multi-file upload.
745
746         * MiniBrowser/qt/MiniBrowser.qrc:
747         * MiniBrowser/qt/icons/checkbox_checked.png: Added.
748         * MiniBrowser/qt/icons/checkbox_unchecked.png: Added.
749         * MiniBrowser/qt/js/MultiSelect.js: Added.
750         (values):
751         (isSelected):
752         (removeValue):
753         * MiniBrowser/qt/qml/CheckBox.qml: Added.
754         * MiniBrowser/qt/qml/FilePicker.qml:
755
756 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
757
758         Retry crashing tests serially at the end of NRWT on Apple Mac
759         https://bugs.webkit.org/show_bug.cgi?id=82233
760
761         Reviewed by Dirk Pranke.
762
763         * Scripts/webkitpy/layout_tests/controllers/manager.py:
764         (Manager.run):
765         * Scripts/webkitpy/layout_tests/port/base.py:
766         (Port.should_retry_crashes):
767         * Scripts/webkitpy/layout_tests/port/mac.py:
768         (MacPort.should_retry_crashes):
769
770 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
771
772         NRWT should retry failures serially
773         https://bugs.webkit.org/show_bug.cgi?id=82241
774
775         Reviewed by Dirk Pranke.
776
777         * Scripts/webkitpy/layout_tests/controllers/manager.py:
778         (Manager._run_tests): Add another parameter to specify the number of workers to use.
779         (Manager.run): Run the failures serially.
780
781 2012-03-26  Adrienne Walker  <enne@google.com>
782
783         Allow expectations on directories in virtual test suites
784         https://bugs.webkit.org/show_bug.cgi?id=82250
785
786         Reviewed by Dirk Pranke.
787
788         lookup_virtual_test_base was only returning valid results for files
789         and not directories. Fix by falling back to the virtual test suite
790         mapping if possible.
791
792         * Scripts/webkitpy/layout_tests/port/base.py:
793         (Port.lookup_virtual_test_base):
794         * Scripts/webkitpy/layout_tests/port/test.py:
795         (TestPort.skipped_tests):
796         (TestPort.virtual_test_suites):
797
798 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
799
800         Fix duplicated ChangeLog entry from r112171.
801
802         Unreviewed, build fix.
803
804 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
805
806         nrwt: remove --worker-model flag
807         https://bugs.webkit.org/show_bug.cgi?id=82112
808
809         Reviewed by Adam Barth.
810
811         The --worker-model=inline flag is basically redundant with the
812         --child-processes=1 flag (technically, you could one run child
813         process, but this is only useful for testing); this patch
814         removes this flag, which simplifies things and enables more
815         cleanup down the road.
816
817         * Scripts/webkitpy/layout_tests/controllers/manager.py:
818         (Manager._run_tests):
819         (Manager.print_config):
820         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
821         (get):
822         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
823         (make_broker):
824         (FunctionTests.test_get__inline):
825         (FunctionTests.test_get__processes):
826         (_TestsMixin.setUp):
827         (_TestsMixin.make_broker):
828         (InlineBrokerTests.setUp):
829         (MultiProcessBrokerTests.setUp):
830         * Scripts/webkitpy/layout_tests/port/base.py:
831         (Port.default_child_processes):
832         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
833         (ChromiumAndroidPort.check_sys_deps):
834         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
835         (ChromiumAndroidPortTest):
836         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
837         (MockDRTPortTest.make_port):
838         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
839         (PortTestCase):
840         (PortTestCase.make_port):
841         * Scripts/webkitpy/layout_tests/port/test.py:
842         (TestPort.default_child_processes):
843         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
844         (_set_up_derived_options):
845         (parse_args):
846         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
847         (parse_args):
848         (MainTest.test_batch_size):
849         (MainTest.test_child_processes_2):
850         (MainTest.test_child_processes_min):
851         (MainTest.test_exception_raised):
852         (MainTest.test_keyboard_interrupt):
853         (MainTest.test_exit_after_n_crashes):
854         (MainTest.assert_run_order):
855         (MainTest.test_run_order__inline):
856         (MainTest.test_virtual):
857
858 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
859
860         test-webkitpy: split test-finding code into a different module
861         https://bugs.webkit.org/show_bug.cgi?id=82253
862
863         Reviewed by Adam Barth.
864
865         Per suggestion from abarth, this change splits all the
866         filesystem-crawling, test-finding code in test-webkitpy out into
867         a separate module and switches to using a FileSystem object.
868         This makes things much more testable, so we also add tests :).
869
870         We also add a realpath() method to the Filesystem object, since
871         test-webkitpy needs that in order to be able to resolve symlinks
872         properly to determine whether a file is under a particular tree
873         or not.
874
875         * Scripts/webkitpy/common/system/filesystem.py:
876         (FileSystem.realpath):
877         * Scripts/webkitpy/common/system/filesystem_mock.py:
878         (MockFileSystem.realpath):
879         * Scripts/webkitpy/test/main.py:
880         (Tester.__init__):
881         (Tester.add_tree):
882         (Tester.run):
883         (Tester._run_tests):
884         (Tester._log_exception):
885         * Scripts/webkitpy/test/test_finder.py: Added.
886         (TestDirectoryTree):
887         (TestDirectoryTree.__init__):
888         (TestDirectoryTree.find_modules):
889         (TestDirectoryTree.find_modules.file_filter):
890         (TestDirectoryTree.to_module):
891         (TestDirectoryTree.clean):
892         (TestFinder):
893         (TestFinder.__init__):
894         (TestFinder.add_tree):
895         (TestFinder.additional_paths):
896         (TestFinder.clean_trees):
897         (TestFinder.is_module):
898         (TestFinder.to_module):
899         (TestFinder.find_names):
900         (TestFinder._exclude):
901         * Scripts/webkitpy/test/test_finder_unittest.py: Added.
902         (TestFinderTest):
903         (TestFinderTest.setUp):
904         (TestFinderTest.tearDown):
905         (TestFinderTest.test_additional_system_paths):
906         (TestFinderTest.test_is_module):
907         (TestFinderTest.test_to_module):
908         (TestFinderTest.test_clean):
909         (TestFinderTest.test_find_names):
910
911 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
912
913         [jhbuild] Use $MAKE if it is defined to build jhbuild itself.
914         https://bugs.webkit.org/show_bug.cgi?id=82209
915
916         Reviewed by Martin Robinson.
917
918         Respecting the $MAKE environment variable when it is defined makes
919         building jhbuild more portable, as some platforms may have GNU
920         make as gmake, for example.
921
922         * jhbuild/jhbuild-wrapper:
923         (install_jhbuild):
924
925 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
926
927         add a Tree abstraction to test-webkitpy to better encapsulate things
928         https://bugs.webkit.org/show_bug.cgi?id=82158
929
930         Reviewed by Adam Barth.
931
932         There are no functional changes in this patch, but adding a
933         basic abstraction for the trees we are looking in for python
934         tests will allow me to add some features later on more easily
935         (like ignoring certain directories like webkitpy/thirdparty),
936         and it makes things slightly less hard-coded.
937
938         * Scripts/test-webkitpy:
939         * Scripts/webkitpy/test/main.py:
940         (Tester.__init__):
941         (Tester):
942         (Tester.add_tree):
943         (Tester._parse_args):
944         (Tester._configure):
945         (Tester.run):
946         (Tester._find_modules):
947         (Tester._run_tests):
948         (Tester._is_module):
949         (Tester._log_exception):
950         (TestDirectoryTree):
951         (TestDirectoryTree.__init__):
952         (TestDirectoryTree.find_modules):
953         (TestDirectoryTree.clean):
954
955 2012-03-26  Ojan Vafai  <ojan@chromium.org>
956
957         garden-o-matic should special case reftests
958         https://bugs.webkit.org/show_bug.cgi?id=81812
959
960         Reviewed by Adam Barth.
961
962         On the examine page, don't show a rebaseline button for reftests.
963         We still show it on the failure summary page because tests are grouped
964         together. Long-term, we should make sure webkit-patch rebaseline-test
965         never tries to rebaseline reftests.
966
967         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
968         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
969         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
970
971 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
972
973         clean up NRWT logging/metering, be less verbose
974         https://bugs.webkit.org/show_bug.cgi?id=73846
975
976         Reviewed by Ojan Vafai.
977
978         Re-land r112014 with a build fix for Python 2.6 compatibility.
979
980         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
981         (ManagerTest.test_http_locking):
982         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
983         (AbstractWorker.run):
984         * Scripts/webkitpy/layout_tests/controllers/worker.py:
985         (Worker.__init__):
986         (Worker._set_up_logging):
987         (Worker):
988         (Worker._set_up_host_and_port):
989         (Worker.run):
990         (Worker.cleanup):
991         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
992         (WorkerTest.test_default_platform_in_worker):
993         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
994         (run):
995         (main):
996         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
997         (MeteredStream):
998         (MeteredStream.implements):
999         (MeteredStream._erasure):
1000         (MeteredStream._ensure_newline):
1001         (MeteredStream.__init__):
1002         (MeteredStream.__del__):
1003         (MeteredStream.cleanup):
1004         (MeteredStream.write_throttled_update):
1005         (MeteredStream.write_update):
1006         (MeteredStream.write):
1007         (MeteredStream.writeln):
1008         (MeteredStream._erase_last_partial_line):
1009         (_LogHandler):
1010         (_LogHandler.__init__):
1011         (_LogHandler.emit):
1012         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1013         (RegularTest):
1014         (RegularTest.setUp):
1015         (RegularTest.tearDown):
1016         (RegularTest.time_fn):
1017         (RegularTest.test_logging_not_included):
1018         (RegularTest._basic):
1019         (RegularTest.test_basic):
1020         (RegularTest._log_after_update):
1021         (RegularTest.test_log_after_update):
1022         (TtyTest):
1023         (TtyTest.test_basic):
1024         (TtyTest.test_log_after_update):
1025         (VerboseTest):
1026         (VerboseTest.test_basic):
1027         (VerboseTest.test_log_after_update):
1028         * Scripts/webkitpy/layout_tests/views/printing.py:
1029         (parse_print_options):
1030         (Printer.__init__):
1031         (Printer.cleanup):
1032         (Printer.print_progress):
1033         (Printer.print_update):
1034         (Printer._write):
1035         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1036         (Testprinter.get_printer):
1037         (Testprinter.test_print_progress):
1038         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1039         (PerfTestsRunner.__init__):
1040         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1041         (test_run_test_pause_before_testing):
1042
1043 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1044
1045         [jhbuild] Revert $MAKE environment hack introduced in r101929.
1046         https://bugs.webkit.org/show_bug.cgi?id=82234
1047
1048         Reviewed by Martin Robinson.
1049
1050         The problem it was supposedly fixing was fixed in jhbuild itself
1051         in git revision da8944d8ca987ca48c91b350257a530050406092, so we
1052         can remove the workaround added to the script.
1053
1054         * jhbuild/jhbuild-wrapper:
1055         (install_jhbuild):
1056
1057 2012-03-26  Leo Yang  <leo.yang@torchmobile.com.cn>
1058
1059         [BlackBerry] Skeleton code of AsyncFileSystemBlackBerry.cpp
1060         https://bugs.webkit.org/show_bug.cgi?id=82157
1061
1062         Reviewed by Rob Buis.
1063
1064         * Scripts/build-webkit:
1065         Make ENABLE_FILE_SYSTEM default on for blackberry.
1066
1067 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1068
1069         [jhbuild] Explicitly disable DTrace when building glib.
1070         https://bugs.webkit.org/show_bug.cgi?id=82145
1071
1072         Reviewed by Gustavo Noronha Silva.
1073
1074         DTrace support is not really needed by WebKit, so disabling it
1075         helps avoid some build errors on more exotic platforms (such as
1076         FreeBSD).
1077
1078         * efl/jhbuild.modules:
1079         * gtk/jhbuild.modules:
1080
1081 2012-03-26  Sheriff Bot  <webkit.review.bot@gmail.com>
1082
1083         Unreviewed, rolling out r111993.
1084         http://trac.webkit.org/changeset/111993
1085         https://bugs.webkit.org/show_bug.cgi?id=82184
1086
1087         It broke layout and API tests and made WTR crash (Requested by
1088         Ossy on #webkit).
1089
1090         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1091         (WTR::PlatformWebView::PlatformWebView):
1092
1093 2012-03-25  Kevin Ollivier  <kevino@theolliviers.com>
1094
1095         [wx] Unreviewed build fix. Move WTF to its own static lib build.
1096
1097         * waf/build/settings.py:
1098         (common_configure):
1099
1100 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
1101
1102         AX: Support solution to handle invalid ax text marker
1103         https://bugs.webkit.org/show_bug.cgi?id=82023
1104
1105         Reviewed by Oliver Hunt.
1106
1107         * DumpRenderTree/AccessibilityUIElement.cpp:
1108         (indexForTextMarkerCallback):
1109         (isTextMarkerValidCallback):
1110         (textMarkerForIndexCallback):
1111         (AccessibilityUIElement::indexForTextMarker):
1112         (AccessibilityUIElement::isTextMarkerValid):
1113         (AccessibilityUIElement::textMarkerForIndex):
1114         (AccessibilityUIElement::getJSClass):
1115         * DumpRenderTree/AccessibilityUIElement.h:
1116         (AccessibilityUIElement):
1117         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1118         (AccessibilityUIElement::indexForTextMarker):
1119         (AccessibilityUIElement::textMarkerForIndex):
1120         (AccessibilityUIElement::isTextMarkerValid):
1121         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1122         (WTR::AccessibilityUIElement::indexForTextMarker):
1123         (WTR::AccessibilityUIElement::isTextMarkerValid):
1124         (WTR::AccessibilityUIElement::textMarkerForIndex):
1125         (WTR):
1126         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1127         (AccessibilityUIElement):
1128         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1129         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1130         (WTR):
1131         (WTR::AccessibilityUIElement::indexForTextMarker):
1132         (WTR::AccessibilityUIElement::isTextMarkerValid):
1133         (WTR::AccessibilityUIElement::textMarkerForIndex):
1134
1135 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
1136
1137         Unreviewed, rolling out r112014.
1138         http://trac.webkit.org/changeset/112014
1139         https://bugs.webkit.org/show_bug.cgi?id=82138
1140
1141         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
1142         #webkit).
1143
1144         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1145         (ManagerTest.test_http_locking):
1146         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1147         (AbstractWorker.run):
1148         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1149         (Worker.__init__):
1150         (Worker.safe_init):
1151         (Worker.run):
1152         (Worker.cleanup):
1153         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
1154         (WorkerTest.test_default_platform_in_worker):
1155         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1156         (run):
1157         (main):
1158         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1159         (MeteredStream):
1160         (MeteredStream.is):
1161         (MeteredStream.__init__):
1162         (MeteredStream.write):
1163         (MeteredStream.update):
1164         (MeteredStream._overwrite):
1165         (MeteredStream._reset):
1166         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1167         (TestMeteredStream):
1168         (TestMeteredStream.test_regular):
1169         * Scripts/webkitpy/layout_tests/views/printing.py:
1170         (_configure_logging):
1171         (_restore_logging):
1172         (Printer.__init__):
1173         (Printer.cleanup):
1174         (Printer.print_progress):
1175         (Printer.print_update):
1176         (Printer._write):
1177         (Printer):
1178         (Printer._update):
1179         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1180         (TestUtilityFunctions.assertEmpty):
1181         (TestUtilityFunctions):
1182         (TestUtilityFunctions.assertNotEmpty):
1183         (TestUtilityFunctions.assertWritten):
1184         (TestUtilityFunctions.test_configure_logging):
1185         (Testprinter.get_printer):
1186         (Testprinter.test_print_progress):
1187         (Testprinter.test_print_progress.mock_time):
1188         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1189         (PerfTestsRunner.__init__):
1190         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1191         (test_run_test_pause_before_testing):
1192
1193 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
1194
1195         [wx] Unreviewed. Fix the build after WTF move.
1196
1197         * waf/build/build_utils.py:
1198         (get_excludes):
1199         (get_excludes_in_dirs):
1200         * waf/build/settings.py:
1201         (get_port_excludes):
1202         (common_configure):
1203
1204 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
1205
1206         clean up NRWT logging/metering, be less verbose
1207         https://bugs.webkit.org/show_bug.cgi?id=73846
1208
1209         Reviewed by Ojan Vafai.
1210
1211         This patch totally rewrites the metered_stream module and
1212         changes the way it and the printing module now interact with the
1213         logging subsystem.
1214
1215         The printing module becomes a dumb layout_tests-specific wrapper around
1216         the metered stream. All of the intelligence of how and when to
1217         actually write (or erase) lines and how things interact with
1218         logging moves into the metered stream.
1219
1220         As of now, anything written to the metered stream *bypasses the
1221         logging system*. If a logger is passed to the meter constructor,
1222         the metered stream will also attach a log handler so that all
1223         messages fed to the logger will also be written to the metered
1224         stream. This detangles the logic that used to exist where
1225         sometimes the meter would write to the logger and sometimes the
1226         logger would write to the meter.
1227
1228         All of this makes the logic generic and reusable for
1229         test-webkitpy (coming in another patch). It also helps clarify
1230         what the mysterious "configure_logging" parameter to the printer
1231         constructor used to do (that parameter itself is gone now, and
1232         by default the printer and metered_stream will not ever touch
1233         the logging system; this makes a bunch of tests less fragile).
1234
1235         All of the above should produce no user-visible changes.
1236
1237         However, to also fix the bug in question, we change the message
1238         format in --verbose mode to %H:%M:%S.%frac %pid %message,
1239         eliminating the date, file:lineno, and level parameters. I am
1240         making this change at the same time because it would be
1241         difficult to write things consistently between the logs and the
1242         metered stream if we were still logging the file:lineno.
1243
1244         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1245           Remove the configure_logging parameter
1246         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1247           Use metered stream to configure the logging (we don't need a printer)
1248           and be careful about when and how we manipulate the log level
1249           and handlers on the root logger.
1250         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1251           Pass in the root logger so that log messages are formatted
1252           properly, and set the log level outside of any routines called
1253           by unit tests so as to not interfere.
1254         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
1255           remove a duplicate log message.
1256         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1257           complete rewrite :).
1258         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1259           complete rewrite :).
1260         * Scripts/webkitpy/layout_tests/views/printing.py:
1261           move the logging configuration into metered_stream, and also
1262           move the "when should I log progress messages" into metered stream.
1263         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1264           remove tests that are no longer necessary.
1265         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1266           remove the configure_logging parameter.
1267         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1268           update the test; since the printer no longer writes to the log,
1269           we need to check what is written to the regular_output instead.
1270           (test_run_test_pause_before_testing):
1271
1272 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
1273
1274         Add myself and Martin Robinson to:
1275           * GStreamerGraphics
1276           * GtkWebKit2PublicAPI
1277           * SoupNetwork
1278         and to the newly created WebKitGTKTranslations.
1279
1280         Rubber-stamped by Martin Robinson.
1281
1282         * Scripts/webkitpy/common/config/watchlist:
1283
1284 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
1285
1286         [Qt][WK2] Implement PageClient::isViewWindowActive()
1287         https://bugs.webkit.org/show_bug.cgi?id=81143
1288
1289         Reviewed by Kenneth Rohde Christiansen.
1290
1291         Fix the WrapperWindow from our PlatformWebView
1292         so it correctly creates the platform related
1293         stuff for QWindow, which is not created unless
1294         QWindow::setVisible() or QWindow::show() are called.
1295
1296         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1297         (WTR::PlatformWebView::PlatformWebView):
1298
1299 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1300
1301         nrwt: don't wait for safari to exit before exiting after showing the results file
1302         https://bugs.webkit.org/show_bug.cgi?id=81845
1303
1304         Reviewed by Ryosuke Niwa.
1305
1306         Prior to this change, if you ran new-run-webkit-tests and
1307         displayed the HTML results file at the end, we would block
1308         waiting for the user to quit the browser. There doesn't seem to
1309         be a need for that, and the Chromium ports don't do this.
1310
1311         Also, update the mac tests to capture the output and be quiet :).
1312
1313         * Scripts/webkitpy/common/system/executive_mock.py:
1314           Make popen() testable, implement should_log for it.
1315         * Scripts/webkitpy/layout_tests/port/mac.py:
1316         (MacPort.show_results_html_file):
1317         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1318           Silence some of the tests via outputcapture.
1319
1320 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
1321
1322         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
1323         https://bugs.webkit.org/show_bug.cgi?id=82073
1324
1325         Reviewed by Martin Robinson.
1326
1327         * Scripts/webkitdirs.pm:
1328         (jhbuildConfigurationChanged): new function that isolates the
1329         jhbuild configuration change checking from the autogen logic.
1330         (mustReRunAutogen): code for the new function lived here.
1331         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
1332         itself, and cleans up the jhbuild root if configuration has
1333         changed; it also forces autogen to be run in that case, to
1334         make sure the new libraries are used.
1335         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
1336         * jhbuild/jhbuild-wrapper:
1337         (ensure_jhbuild): this might be run inside a jhbuild environment,
1338         in which case aclocal fails when trying to use the now deleted
1339         aclocal directory in the jhbuild prefix, so work around that.
1340
1341 2012-03-23  Ojan Vafai  <ojan@chromium.org>
1342
1343         garden-o-matic scrolls to the top of the page when you click an accordion item
1344         https://bugs.webkit.org/show_bug.cgi?id=82088
1345
1346         Reviewed by Adam Barth.
1347
1348         Check the scroll position and make sure that the clicked accordion item is visible.
1349
1350         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1351
1352 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1353
1354         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
1355         https://bugs.webkit.org/show_bug.cgi?id=80567
1356
1357         Reviewed by Ryosuke Niwa.
1358
1359         Move clean_up_run to occur right after we finish testing and
1360         before we start processing the results. Also makes set_up_run()
1361         and clean_up_run() private since there's no real need for
1362         callers to be aware of them.
1363
1364         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1365         (Manager._set_up_run):
1366         (Manager.run):
1367         (Manager._clean_up_run):
1368         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1369         (ManagerTest.test_http_locking):
1370         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1371         (run):
1372
1373 2012-03-23  Dean Jackson  <dino@apple.com>
1374
1375         Disable CSS_SHADERS in Apple builds
1376         https://bugs.webkit.org/show_bug.cgi?id=81996
1377
1378         Reviewed by Simon Fraser.
1379
1380         Default enableCSSShaders to be 0 on all builds that use
1381         this script. Use the flag to turn it back on.
1382
1383         * Scripts/build-webkit:
1384
1385 2012-03-23  Tony Chang  <tony@chromium.org>
1386
1387         [chromium] rename newwtf target back to wtf
1388         https://bugs.webkit.org/show_bug.cgi?id=82064
1389
1390         Reviewed by Adam Barth.
1391
1392         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1393
1394 2012-03-23  Jessie Berlin  <jberlin@apple.com>
1395
1396         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
1397         the Lion Intel WebKit2 Debug bots.
1398         https://bugs.webkit.org/show_bug.cgi?id=81012.
1399
1400         Rubber-stamped by Dan Bernstein.
1401
1402         Add a step to kill the pasteboard server while we investigate what is causing these
1403         assertion failures (to get the bots green).
1404
1405         * BuildSlaveSupport/kill-old-processes:
1406
1407 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
1408
1409         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
1410         https://bugs.webkit.org/show_bug.cgi?id=81986
1411
1412         Reviewed by David Levin.
1413
1414         Track whether or not errors are filtered in error handlers. If an
1415         alphabetical sorting error is reported for a header and filtered, try
1416         reporting it for the other potential source of the error. This ensures
1417         that the style bot can still find new sorting errors in both potential
1418         situations:
1419
1420             #include <foo.h> <-- 1. edited line
1421             #include <baz.h> <-- 2. edited line
1422
1423         * Scripts/webkitpy/style/checkers/cpp.py:
1424         (check_include_line): Potentially report the error for both lines,
1425         if the first error is filtered.
1426         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
1427         test that allows only reporting errors for certain lines. This verifies that
1428         when one line of an ordering error is filtered, the error is reported on the
1429         other line.
1430         (ErrorCollector.__init__):
1431         (ErrorCollector.__call__):
1432         (CppStyleTestBase.perform_lint):
1433         (CppStyleTestBase.perform_language_rules_check):
1434         (CppStyleTestBase.assert_language_rules_check):
1435         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
1436         Added a new test that verifies that when one line of the two lines of an ordering
1437         error occur, the error is reported on the other line.
1438         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
1439         (MockErrorHandler.__call__): Report True because the error is handled.
1440         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1441         (ErrorCollector.__call__): Ditto.
1442         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
1443         (MockErrorHandler.__call__): Ditto.
1444         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
1445         (TestErrorHandler.__call__): Ditto.
1446         * Scripts/webkitpy/style/checkers/xml_unittest.py:
1447         (MockErrorHandler.__call__): Ditto.
1448         * Scripts/webkitpy/style/error_handlers.py:
1449         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
1450         and False if it is filtered.
1451
1452 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
1453
1454         Build fix for WinCE after r111778.
1455
1456         * WinCELauncher/CMakeLists.txt:
1457
1458 2012-03-23  Peter Beverloo  <peter@chromium.org>
1459
1460         Unreviewed buildfix for the Chromium Android builder.
1461
1462         Disable the ImageDiff target for now. Android needs to build this for
1463         host rather than for target (as other configurations do), but since the
1464         WTF move the dependency chain got significantly more complicated.
1465
1466         I'll work out a proper fix in the following bug, but since several
1467         things are broken due to this, this buildfix should be landed first.
1468         https://bugs.webkit.org/show_bug.cgi?id=82039
1469
1470         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1471
1472 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
1473
1474         [EFL][DRT] Implement LayoutTestController's methods related with editing.
1475         https://bugs.webkit.org/show_bug.cgi?id=80136
1476
1477         Reviewed by Ryosuke Niwa.
1478
1479         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
1480         setSelectTrailingWhitespaceEnabled.
1481
1482         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1483         (LayoutTestController::setSmartInsertDeleteEnabled):
1484         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
1485
1486 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1487
1488         Convert hasGrammarMarker to use Internals interface
1489         https://bugs.webkit.org/show_bug.cgi?id=82004
1490
1491         Reviewed by Ryosuke Niwa.
1492
1493         Remove hasGrammarMarker functions, because it is able to work in the
1494         cross-port way through the Internals interface.
1495
1496         * DumpRenderTree/LayoutTestController.cpp:
1497         (LayoutTestController::staticFunctions):
1498         * DumpRenderTree/LayoutTestController.h:
1499         (LayoutTestController):
1500         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1501         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1502         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1503         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1504         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1505         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1506
1507 2012-03-22  Adam Barth  <abarth@webkit.org>
1508
1509         sync-master-with-upstream creates non-linear git history
1510         https://bugs.webkit.org/show_bug.cgi?id=82006
1511
1512         Reviewed by Julien Chaffraix.
1513
1514         Previously this script would create a merge commit at the HEAD of
1515         origin/master, causing your fork of WebKit to diverge from upstream.
1516         The new version of this script pushes directly from upstream to origin
1517         (without round-tripping through any local branches), avoiding this
1518         problem.
1519
1520         * Scripts/sync-master-with-upstream:
1521
1522 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1523
1524         Unreviewed build fix. Add knowledge of the WTF project to
1525         the autotools build system scripts so that build-jsc works
1526         correctly for GTK+.
1527
1528         * Scripts/webkitdirs.pm:
1529         (buildAutotoolsProject): handle the WTF project, build the
1530         libWTF.la target.
1531         (buildGtkProject): add WTF to the list of supported projects.
1532
1533 2012-03-22  Eric Seidel  <eric@webkit.org>
1534
1535         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
1536         https://bugs.webkit.org/show_bug.cgi?id=81984
1537
1538         Reviewed by Adam Barth.
1539
1540         * Scripts/VCSUtils.pm:
1541         (adjustPathForRecentRenamings):
1542
1543 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1544
1545         [CMake] Fix build-jsc after r111778 and r111797.
1546         https://bugs.webkit.org/show_bug.cgi?id=82001
1547
1548         Reviewed by Filip Pizlo.
1549
1550         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
1551         does return if no error occurred, and build-jsc failed with
1552         "building not defined for this platform".
1553
1554         * Scripts/build-jsc: Always exit after caling
1555         buildCMakeProjectOrExit().
1556         * Scripts/webkitdirs.pm:
1557         (buildCMakeProjectOrExit): Return 0 if no error occurred.
1558
1559 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1560
1561         [CMake] Unreviewed build fix after r111778.
1562
1563         * DumpRenderTree/efl/CMakeLists.txt: Replace
1564         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
1565         paths.
1566         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
1567         wtf include.
1568         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
1569         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
1570         with ${WTF_DIR}/wtf in the include paths.
1571
1572 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
1573
1574         Tools/Scripts/build-jsc doesn't work after WTF move
1575         https://bugs.webkit.org/show_bug.cgi?id=81999
1576
1577         Reviewed by Eric Seidel.
1578         
1579         For ports that require WTF and JSC to be built as two projects and where
1580         the machinery to build both does not already exist, teach the tool to do
1581         so for us.
1582
1583         * Scripts/build-jsc:
1584         (buildMyProject):
1585
1586 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1587
1588         Fix accidental missing line in r111793.
1589
1590         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1591         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
1592         Remove base.values. It was unused and duplicates the builtin Object.values.
1593         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1594         Accidentally deleted this line before committing r111793.
1595
1596 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1597
1598         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
1599         https://bugs.webkit.org/show_bug.cgi?id=81983
1600
1601         Reviewed by Adam Barth.
1602
1603         The dialogs get in the way if you are doing multiple rebaselines. Also,
1604         a number of people didn't realize that you could do multiple rebaselines
1605         in parallel. A non-modal dialog is just confusing.
1606
1607         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1608         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1609         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
1610         show the number of tests being rebaselined.
1611
1612         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1613         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
1614         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
1615         into a single area.
1616
1617         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
1618         also serves to clear all teh content in the StatusArea. 
1619
1620         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1621         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1622
1623 2012-03-22  Tony Chang  <tony@chromium.org>
1624
1625         Unreviewed, fix chromium build after wtf move.
1626
1627         Only use newwtf, remove references to wtf.
1628
1629         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1630
1631 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1632
1633         Actually move WTF files to their new home
1634         https://bugs.webkit.org/show_bug.cgi?id=81844
1635
1636         [Qt] Unreviewed buildfix after r111778.
1637
1638         * DumpRenderTree/qt/DumpRenderTree.pro:
1639
1640 2012-03-22  Dan Bernstein  <mitz@apple.com>
1641
1642         Font fallback in WebKitTestRunner depends on user settings
1643         https://bugs.webkit.org/show_bug.cgi?id=81978
1644
1645         Reviewed by Jessie Berlin.
1646
1647         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
1648         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
1649         with a value consisting of the array ("en") to the override defaults
1650         dictionary. This is the same value used in DumpRenderTree, and it
1651         ensures that font fallback follows a consistent path that is not
1652         affected by the language preferences of the user running the test.
1653
1654 2012-03-22  Andy Estes  <aestes@apple.com>
1655
1656         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
1657         https://bugs.webkit.org/show_bug.cgi?id=81971
1658
1659         Reviewed by Dan Bernstein.
1660
1661         Xcode will automatically link ObjC binaries against libarclite in some
1662         cases, which defines a class called __ARCLite__.
1663         check-for-inappropriate-objc-class-names shouldn't complain about this
1664         class name.
1665
1666         * Scripts/check-for-inappropriate-objc-class-names:
1667
1668 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1669
1670         new-run-webkit-tests blames the wrong test for crashing
1671         https://bugs.webkit.org/show_bug.cgi?id=81951
1672
1673         Reviewed by Adam Barth.
1674
1675         NRWT wasn't extracting the WebProcess pid correct; should've
1676         used re.search instead of re.match :(. Add more tests!
1677
1678         * Scripts/webkitpy/layout_tests/port/mac.py:
1679         (MacPort._get_crash_log):
1680         * Scripts/webkitpy/layout_tests/port/webkit.py:
1681         (WebKitDriver.has_crashed):
1682         (WebKitDriver._check_for_driver_crash):
1683         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1684         (WebKitDriverTest.test_no_timeout):
1685         (WebKitDriverTest):
1686         (WebKitDriverTest.test_check_for_driver_crash):
1687         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
1688         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
1689         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
1690         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
1691         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
1692         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
1693
1694 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1695
1696         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
1697         https://bugs.webkit.org/show_bug.cgi?id=81512
1698
1699         Adding missing implementation markerTextForListItem to EFL's
1700         LayoutTestController so that we can unskip related tests from
1701         the skip list.
1702
1703         Reviewed by Antonio Gomes.
1704
1705         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1706         (LayoutTestController::markerTextForListItem): Implemented.
1707
1708 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1709
1710         [EFL] Enable view mode media feature layout tests
1711         https://bugs.webkit.org/show_bug.cgi?id=81205
1712
1713         Adding missing implementation setViewModeMediaFeature to EFL's 
1714         LayoutTestController so that we can unskip view mode 
1715         media feature layout tests.
1716
1717         Reviewed by Antonio Gomes.
1718
1719         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1720         (LayoutTestController::setViewModeMediaFeature): Implemented.
1721
1722 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1723
1724         nrwt: crash while stopping layout test helper on apple mac lion
1725         https://bugs.webkit.org/show_bug.cgi?id=81949
1726
1727         Reviewed by Eric Seidel.
1728
1729         stop_helper() didn't work right if the helper process wasn't
1730         still running or didn't respond properly to being shut down. Fix
1731         this and add some tests. Also add code to MockExecutive() to be
1732         able to mock executive.popen() properly.
1733
1734         * Scripts/webkitpy/common/system/executive_mock.py:
1735         (MockProcess.__init__):
1736         (MockProcess.wait):
1737         (MockExecutive.__init__):
1738         (MockExecutive.popen):
1739         * Scripts/webkitpy/common/system/systemhost_mock.py:
1740         (MockSystemHost.__init__):
1741         * Scripts/webkitpy/layout_tests/port/mac.py:
1742         (MacPort.start_helper):
1743         (MacPort.stop_helper):
1744         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1745         (test_get_crash_log):
1746         (test_helper_starts):
1747         (test_helper_fails_to_start):
1748         (test_helper_fails_to_stop):
1749         (test_helper_fails_to_stop.bad_waiter):
1750
1751 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1752
1753         [GTK] Add make to the jhbuild moduleset
1754         https://bugs.webkit.org/show_bug.cgi?id=81769
1755
1756         Reviewed by Martin Robinson.
1757
1758         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
1759         Add GNU make with custom patches to fix several issues:
1760
1761         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
1762           fix the argument list limit of GNU make when running make
1763           distcheck.
1764         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
1765           and Jeremy Devenport to fix parallel build issues present in
1766           GNU make 3.82
1767
1768         * gtk/jhbuild.modules: Added make with patches.
1769         * gtk/patches/make-3.82-arg-list-length.patch: Added.
1770         * gtk/patches/make-3.82-parallel-build.patch: Added.
1771
1772 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
1773
1774         [wx] Unreviewed. Adding Source/WTF to the build and updating
1775         waf port list.
1776
1777         * DumpRenderTree/wscript:
1778         * waf/build/settings.py:
1779
1780 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1781
1782         Re-land fix for r81603 yet again (!) with latest build fix.
1783         https://bugs.webkit.org/show_bug.cgi?id=81603
1784
1785         Unreviewed, build fix.
1786
1787         The latest problem was a bad sprintf() in
1788         BasePort._get_crash_log(). I've reworked that routine and added
1789         unit tests for coverage.
1790
1791         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1792         (SingleTestRunner._handle_error):
1793         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1794         (write_test_result):
1795         (TestResultWriter.write_crash_log):
1796         * Scripts/webkitpy/layout_tests/port/base.py:
1797         (Port.diff_text):
1798         (Port._get_crash_log):
1799         * Scripts/webkitpy/layout_tests/port/chromium.py:
1800         (ChromiumDriver.run_test):
1801         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1802         (ChromiumDriverTest.test_crash_log):
1803         * Scripts/webkitpy/layout_tests/port/driver.py:
1804         (DriverOutput.__init__):
1805         * Scripts/webkitpy/layout_tests/port/mac.py:
1806         (MacPort.is_lion):
1807         (MacPort._get_crash_log):
1808         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1809         (test_default_child_processes):
1810         (test_get_crash_log):
1811         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1812         (MockDRTPortTest.test_get_crash_log):
1813         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1814         (PortTestCase.test_get_crash_log):
1815         * Scripts/webkitpy/layout_tests/port/test.py:
1816         (TestDriver.run_test):
1817         * Scripts/webkitpy/layout_tests/port/webkit.py:
1818         (WebKitDriver.__init__):
1819         (WebKitDriver._start):
1820         (WebKitDriver.has_crashed):
1821         (WebKitDriver._check_for_driver_crash):
1822         (WebKitDriver.run_test):
1823
1824 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1825
1826         [Qt] Enable support for handling database quota's in the MiniBrowser
1827         https://bugs.webkit.org/show_bug.cgi?id=81928
1828
1829         Reviewed by Simon Hausmann.
1830
1831         * MiniBrowser/qt/qml/BrowserWindow.qml:
1832
1833 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1834
1835         [EFL] Implement LayoutTestController::setEditingBehavior
1836         https://bugs.webkit.org/show_bug.cgi?id=81124
1837
1838         Adding missing implementation setEditingBehavior to EFL's
1839         LayoutTestController. This implementation allows us to
1840         unskip some tests from the skip list.
1841
1842         Reviewed by Martin Robinson.
1843
1844         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1845         (LayoutTestController::setEditingBehavior): Implemented.
1846
1847 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
1848
1849         [GTK] Use the angle-bracket form to include wtf headers
1850         https://bugs.webkit.org/show_bug.cgi?id=81884
1851
1852         Reviewed by Eric Seidel.
1853
1854         Use #include <wtf/foo> instead of #include "foo".
1855
1856         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1857         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1858         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1859         * DumpRenderTree/gtk/EditingCallbacks.cpp:
1860         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
1861         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
1862
1863 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1864
1865         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
1866
1867         * Scripts/webkitpy/common/config/committers.py:
1868
1869 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1870
1871         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
1872         https://bugs.webkit.org/show_bug.cgi?id=81889
1873
1874         Reviewed by Kenneth Rohde Christiansen.
1875
1876         Enable FAST_MOBILE_SCROLLING feature.
1877
1878         * qmake/mkspecs/features/features.prf:
1879
1880 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1881
1882         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
1883         https://bugs.webkit.org/show_bug.cgi?id=81900
1884
1885         Reviewed by Gustavo Noronha Silva.
1886
1887         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
1888         path which assumed a certain directory structure for the build
1889         directory which may noy be true.
1890
1891         Instead, hardcode the path to the downloaded fonts starting from the
1892         top-level source directory. While this solution is also far from
1893         perfect, it makes it more consistent and reliable, and easier to
1894         tackle at once with the rest of bug 81475.
1895
1896         * DumpRenderTree/efl/CMakeLists.txt:
1897
1898 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
1899
1900         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
1901         https://bugs.webkit.org/show_bug.cgi?id=81888
1902
1903         Reviewed by Philippe Normand.
1904
1905         * Scripts/update-webkit-libs-jhbuild:
1906         (runJhbuild): Helper function to run jhbuild with a given command.
1907         Instead of running jhbuild directly, use the jhbuild-wrapper, so
1908         that we don't need to run it first to check whether jhbuild is
1909         installed or not.
1910         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
1911         * gtk/run-with-jhbuild: Ditto.
1912         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
1913         not only run.
1914
1915 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1916
1917         Unreviewed, test-webkitpy fix after r111661.
1918
1919         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1920         (MainTest.test_crash_log):
1921         (MainTest.test_web_process_crash_log):
1922
1923 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1924
1925         Unreviewed, test-webkitpy fix after r111642.
1926
1927         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1928         (ChromiumDriverTest.setUp):
1929         (ChromiumDriverTest.test_two_drivers):
1930
1931 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1932
1933         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1934         https://bugs.webkit.org/show_bug.cgi?id=81603
1935
1936         Unreviewed rolling out r111609 and part of r111615,
1937         because it broke NRWT on Qt-WK2 platform.
1938
1939         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1940         (SingleTestRunner._handle_error):
1941         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1942         (write_test_result):
1943         (TestResultWriter.write_crash_report):
1944         * Scripts/webkitpy/layout_tests/port/base.py:
1945         (Port.is_crash_reporter):
1946         (Port._driver_class):
1947         * Scripts/webkitpy/layout_tests/port/chromium.py:
1948         (ChromiumDriver.run_test):
1949         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1950         (ChromiumDriverTest.test_crashed_process_name):
1951         * Scripts/webkitpy/layout_tests/port/driver.py:
1952         (DriverOutput.__init__):
1953         * Scripts/webkitpy/layout_tests/port/mac.py:
1954         (MacPort):
1955         (MacPort.is_crash_reporter):
1956         (MacPort.release_http_lock):
1957         * Scripts/webkitpy/layout_tests/port/test.py:
1958         (TestDriver.run_test):
1959         * Scripts/webkitpy/layout_tests/port/webkit.py:
1960         (WebKitDriver.__init__):
1961         (WebKitDriver._start):
1962         (WebKitDriver.has_crashed):
1963         (WebKitDriver._check_for_driver_crash):
1964         (WebKitDriver.run_test):
1965
1966 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1967
1968         [Qt] Add full platforms to Qt buildslaves
1969         https://bugs.webkit.org/show_bug.cgi?id=81877
1970
1971         Reviewed by Ryosuke Niwa.
1972
1973         * BuildSlaveSupport/build.webkit.org-config/config.json:
1974
1975 2012-03-21  Johnny Ding  <jnd@chromium.org>
1976
1977         Customize layout test timeout value for different ports.
1978         https://bugs.webkit.org/show_bug.cgi?id=79859
1979
1980         Reviewed by Ojan Vafai.
1981
1982         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1983         (Manager):
1984         * Scripts/webkitpy/layout_tests/port/base.py:
1985         (Port.default_test_timeout_ms):
1986         * Scripts/webkitpy/layout_tests/port/chromium.py:
1987         (ChromiumDriver.stop):
1988         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
1989         (ChromiumAndroidPort.default_test_timeout_ms):
1990         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1991         (_set_up_derived_options):
1992
1993 2012-03-21  Mark Rowe  <mrowe@apple.com>
1994
1995         Fix the build.
1996
1997         Ensure that our test tools don't try and use RTTI unnecessarily.
1998
1999         * DumpRenderTree/mac/Configurations/Base.xcconfig:
2000         * TestWebKitAPI/Configurations/Base.xcconfig:
2001         * WebKitTestRunner/Configurations/Base.xcconfig:
2002
2003 2012-03-21  Adam Barth  <abarth@webkit.org>
2004
2005         WebKit should have some scripts to smooth interactions with GitHub
2006         https://bugs.webkit.org/show_bug.cgi?id=81842
2007
2008         Reviewed by Eric Seidel.
2009
2010         These scripts support the workflow in
2011         https://trac.webkit.org/wiki/UsingGitHub by automating the
2012         configuration and syncing operations needed for a GitHub fork of WebKit
2013         to track the main WebKit repository.
2014
2015         * Scripts/configure-github-as-upstream: Added.
2016         * Scripts/sync-master-with-upstream: Added.
2017
2018 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2019
2020         WTR - log the pid of a crashing WebProcess
2021         https://bugs.webkit.org/show_bug.cgi?id=81575
2022
2023         Reviewed by Alexey Proskuryakov.
2024
2025         This allows run-webkit-tests to find the right crash log for a
2026         given crashed WebProcess in WebKit2.
2027
2028         * WebKitTestRunner/TestController.cpp:
2029         (WTR::exceededDatabaseQuota):
2030         (WTR::TestController::createOtherPage):
2031         (WTR::TestController::initialize):
2032         (WTR::TestController::resetStateToConsistentValues):
2033         (WTR::TestController::runTest):
2034         (WTR::TestController::processDidCrash):
2035
2036 2012-03-21  Enrica Casucci  <enrica@apple.com>
2037
2038         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
2039         https://bugs.webkit.org/show_bug.cgi?id=81835
2040         <rdar://problem/11082749>
2041
2042         Reviewed by Brady Eidson.
2043
2044         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
2045         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
2046         (TestWebKitAPI::contextMenuCopyLink):
2047
2048 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
2049
2050         [CSS Shaders] Make CSS Shaders compile on Chromium
2051         https://bugs.webkit.org/show_bug.cgi?id=81435
2052
2053         Reviewed by Stephen White.
2054         
2055         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
2056
2057         * DumpRenderTree/chromium/LayoutTestController.cpp:
2058         (LayoutTestController::overridePreference):
2059         * DumpRenderTree/chromium/WebPreferences.cpp:
2060         (WebPreferences::reset):
2061         (WebPreferences::applyTo):
2062         * DumpRenderTree/chromium/WebPreferences.h:
2063         (WebPreferences):
2064
2065 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2066
2067         fix regression on gtk port introduced by r111609
2068         https://bugs.webkit.org/show_bug.cgi?id=81839
2069
2070         Reviewed by Philippe Normand.
2071
2072         r111609 should've initialized a couple of missing fields in
2073         WebKitDriver, and also should've checked for method overrides
2074         in the other ports :(.
2075
2076         * Scripts/webkitpy/layout_tests/port/gtk.py:
2077         (GtkDriver._start):
2078         * Scripts/webkitpy/layout_tests/port/webkit.py:
2079         (WebKitDriver.__init__):
2080         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2081         (MainTest.test_crash_log):
2082         (MainTest.test_web_process_crash_log):
2083
2084 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2085
2086         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
2087         https://bugs.webkit.org/show_bug.cgi?id=81603
2088
2089         Reviewed by Adam Barth.
2090
2091         Add a new _get_crash_log() overridable method on the Port
2092         object to customize how we fetch crash logs per port.
2093         Mac will now slow-spin with a timeout waiting for the crash log, rather
2094         than waiting for the ReportCrash process to exit (it appears
2095         that the mac will manage multiple crashes with a single
2096         ReportCrash process, the process waits around longer than
2097         necessary, presumably to avoid thrashing if processes are
2098         repeatedly crashing).
2099
2100         Also, add the DriverOutput should contain the crash log and other info,
2101         which is created in a port-specific manner but can then be
2102         treated generically. Previously single_test_runner would get
2103         told that something crashed and attempt to do something to get
2104         the crash log, but it didn't have the information it needed to
2105         od the right thing; better to make the driver hand back the
2106         right info.
2107
2108         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2109         (SingleTestRunner._handle_error):
2110         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2111         (write_test_result):
2112         (TestResultWriter.write_crash_log):
2113         * Scripts/webkitpy/layout_tests/port/base.py:
2114         (Port.diff_text):
2115         (Port._get_crash_log):
2116         * Scripts/webkitpy/layout_tests/port/chromium.py:
2117         (ChromiumDriver.run_test):
2118         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2119         (ChromiumDriverTest.test_crash_log):
2120         * Scripts/webkitpy/layout_tests/port/driver.py:
2121         (DriverOutput.__init__):
2122         * Scripts/webkitpy/layout_tests/port/mac.py:
2123         (MacPort.is_lion):
2124         (MacPort._get_crash_log):
2125         * Scripts/webkitpy/layout_tests/port/test.py:
2126         (TestDriver.run_test):
2127         * Scripts/webkitpy/layout_tests/port/webkit.py:
2128         (WebKitDriver.__init__):
2129         (WebKitDriver._start):
2130         (WebKitDriver.has_crashed):
2131         (WebKitDriver._check_for_driver_crash):
2132         (WebKitDriver.run_test):
2133
2134 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2135
2136         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
2137         https://bugs.webkit.org/show_bug.cgi?id=81507
2138
2139         Now that we're only loading from one directory, no need to
2140         iterate over a list of directories anymore.
2141
2142         Reviewed by Martin Robinson.
2143
2144         * DumpRenderTree/efl/CMakeLists.txt:
2145         * DumpRenderTree/efl/FontManagement.cpp:
2146         (addFontDirectory):
2147         (addFontsToEnvironment):
2148
2149 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2150
2151         webkitpy: get ServerProcess out of the reportcrash business
2152         https://bugs.webkit.org/show_bug.cgi?id=81600
2153
2154         Unreviewed, build fix.
2155
2156         Re-land r111307 and r111293 with another fix for a crash in NRWT;
2157         we need to check if the driver has crashed before attempting to
2158         read from it.
2159
2160
2161         * Scripts/webkitpy/layout_tests/port/server_process.py:
2162         (ServerProcess._reset):
2163         (ServerProcess._handle_possible_interrupt):
2164         (ServerProcess.write):
2165         (ServerProcess.read_stdout):
2166         (ServerProcess.has_crashed):
2167         (ServerProcess._read):
2168         (ServerProcess.stop):
2169         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2170         (TrivialMockPort.check_for_leaks):
2171         (TestServerProcess.test_broken_pipe):
2172         * Scripts/webkitpy/layout_tests/port/webkit.py:
2173         (WebKitPort._read_image_diff):
2174         (WebKitDriver.has_crashed):
2175         (WebKitDriver._check_for_driver_crash):
2176         (WebKitDriver.run_test):
2177         (WebKitDriver._read_block):
2178         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2179         (MockServerProcess.__init__):
2180         (MockServerProcess):
2181         (MockServerProcess.has_crashed):
2182
2183 2012-03-21  Zeno Albisser  <zeno@webkit.org>
2184
2185         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
2186         https://bugs.webkit.org/show_bug.cgi?id=81750
2187
2188         Building WebCore on mac currently creates an archive that is bigger than
2189         4GB. But ranlib fails to create a symbol table for such a big archive,
2190         even on 64bit machines.
2191         Therefore we start using SVGAllInOne.cpp when building debug
2192         on mac. This reduces the size of the archive by about 300MB.
2193
2194         Reviewed by Tor Arne Vestbø.
2195
2196         * qmake/mkspecs/features/mac/default_pre.prf:
2197
2198 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
2199
2200         Updated the address list of committers.py.
2201
2202         Reviewed by Csaba Osztrogonác.
2203
2204         * Scripts/webkitpy/common/config/committers.py:
2205
2206 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
2207
2208         Updating my nick in committers file.
2209
2210         Unreviewed.
2211
2212         * Scripts/webkitpy/common/config/committers.py:
2213
2214 2012-03-21  Peter Beverloo  <peter@chromium.org>
2215
2216         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
2217         https://bugs.webkit.org/show_bug.cgi?id=80861
2218
2219         Reviewed by Tony Chang.
2220
2221         Remove most Android-specific logic from webkitdirs.pm in favor of
2222         handling this during project generation time.
2223
2224         * Scripts/webkitdirs.pm:
2225         (buildChromiumMakefile):
2226
2227 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2228
2229         [GTK] Allow running javascript from location bar in MiniBrowser
2230         https://bugs.webkit.org/show_bug.cgi?id=81331
2231
2232         Reviewed by Martin Robinson.
2233
2234         * MiniBrowser/gtk/BrowserWindow.c:
2235         (activateUriEntryCallback): Use browser_window_load_uri().
2236         (browser_window_load_uri): Check whether uri starts with
2237         javascript: and run the given script, or just call
2238         webkit_web_view_load_uri() to load the given uri.
2239         * MiniBrowser/gtk/BrowserWindow.h:
2240         * MiniBrowser/gtk/main.c:
2241         (createBrowserWindow): Use browser_window_load_uri().
2242
2243 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2244
2245         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
2246         https://bugs.webkit.org/show_bug.cgi?id=75543
2247
2248         Reviewed by Martin Robinson.
2249
2250         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
2251         MiniBrowser CPP flags.
2252
2253 2012-03-20  Eric Seidel  <eric@webkit.org>
2254
2255         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
2256         https://bugs.webkit.org/show_bug.cgi?id=80911
2257
2258         Reviewed by Adam Barth.
2259
2260         Update include paths for Source/WTF.
2261
2262         * DumpRenderTree/efl/CMakeLists.txt:
2263         * DumpRenderTree/qt/DumpRenderTree.pro:
2264         * QtTestBrowser/QtTestBrowser.pro:
2265         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
2266
2267 2012-03-20  Adam Barth  <abarth@webkit.org>
2268
2269         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
2270         https://bugs.webkit.org/show_bug.cgi?id=81567
2271
2272         Reviewed by Eric Seidel.
2273
2274         This temporary measure should help us avoid disrupting the folks
2275         working on WebAudio when we move the webaudio directory to Modules.
2276
2277         * Scripts/VCSUtils.pm:
2278         (adjustPathForRecentRenamings):
2279         (parseGitDiffHeader):
2280         (parseSvnDiffHeader):
2281
2282 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2283
2284         Unreviewed, rolling out r111453.
2285         http://trac.webkit.org/changeset/111453
2286         https://bugs.webkit.org/show_bug.cgi?id=81600
2287
2288         still broken
2289
2290         * Scripts/webkitpy/layout_tests/port/server_process.py:
2291         (ServerProcess._reset):
2292         (ServerProcess.handle_interrupt):
2293         (ServerProcess.write):
2294         (ServerProcess._check_for_crash):
2295         (ServerProcess._handle_timeout):
2296         (ServerProcess._check_for_abort):
2297         (ServerProcess._read):
2298         (ServerProcess.stop):
2299         (ServerProcess):
2300         (ServerProcess.set_crashed):
2301         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2302         (TrivialMockPort.is_crash_reporter):
2303         (TestServerProcess.test_broken_pipe):
2304         * Scripts/webkitpy/layout_tests/port/webkit.py:
2305         (WebKitPort._read_image_diff):
2306         (WebKitDriver.has_crashed):
2307         (WebKitDriver._check_for_driver_crash):
2308         (WebKitDriver):
2309         (WebKitDriver._detected_crash):
2310         (WebKitDriver._subprocess_crashed):
2311         (WebKitDriver._crashed_process_name):
2312         (WebKitDriver.run_test):
2313         (WebKitDriver._read_block):
2314         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2315         (MockServerProcess.__init__):
2316
2317 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2318
2319         webkitpy: get ServerProcess out of the reportcrash business
2320         https://bugs.webkit.org/show_bug.cgi?id=81600
2321
2322         Unreviewed, build fix.
2323
2324         Re-land r111307 and r111293 with fix for crash in NRWT; I
2325         was failing to set the crashing process name.
2326
2327         * Scripts/webkitpy/layout_tests/port/server_process.py:
2328         (ServerProcess._reset):
2329         (ServerProcess._handle_possible_interrupt):
2330         (ServerProcess.write):
2331         (ServerProcess.read_stdout):
2332         (ServerProcess.has_crashed):
2333         (ServerProcess._read):
2334         (ServerProcess.stop):
2335         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2336         (TrivialMockPort.check_for_leaks):
2337         (TestServerProcess.test_broken_pipe):
2338         * Scripts/webkitpy/layout_tests/port/webkit.py:
2339         (WebKitPort._read_image_diff):
2340         (WebKitDriver.has_crashed):
2341         (WebKitDriver._check_for_driver_crash):
2342         (WebKitDriver.run_test):
2343         (WebKitDriver._read_block):
2344         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2345         (MockServerProcess.__init__):
2346         (MockServerProcess):
2347         (MockServerProcess.has_crashed):
2348
2349 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2350
2351         [mac] Restore color space switching code to run-webkit-tests
2352         https://bugs.webkit.org/show_bug.cgi?id=80571
2353         <rdar://problem/11008529>
2354
2355         Reviewed by Simon Fraser.
2356
2357         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
2358         
2359         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
2360         didn't exist until Lion.
2361
2362         * DumpRenderTree/mac/LayoutTestHelper.m:
2363         (installLayoutTestColorProfile):
2364         (restoreUserColorProfile):
2365
2366 2012-03-20  Adele Peterson  <adele@apple.com>
2367
2368         Update the last test to use EXPECT_WK_STREQ.
2369
2370         Reviewed by Dan Bernstein.
2371
2372         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
2373
2374 2012-03-20  Adele Peterson  <adele@apple.com>
2375
2376         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
2377         https://bugs.webkit.org/show_bug.cgi?id=81630
2378         <rdar://problem/10650660>
2379
2380         Reviewed by Alexey Proskuryakov.
2381
2382         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2383         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
2384         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
2385         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
2386         (TestWebKitAPI):
2387         (TestWebKitAPI::didFinishLoadForFrame):
2388         (TestWebKitAPI::TEST):
2389         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
2390
2391 2012-03-20  Eric Seidel  <eric@webkit.org>
2392
2393         Add link on queues.webkit.org to kov's fancy queue-depth graph
2394         https://bugs.webkit.org/show_bug.cgi?id=81690
2395
2396         Reviewed by Adam Barth.
2397
2398         * QueueStatusServer/templates/recentstatus.html:
2399
2400 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2401
2402         [mac] Restore color space switching code to run-webkit-tests
2403         https://bugs.webkit.org/show_bug.cgi?id=80571
2404         <rdar://problem/11008529>
2405
2406         Reviewed by Simon Fraser.
2407
2408         Make use of the layout test helper hooks to restore the Mac system-wide
2409         color profile switching code. This makes WkTR and DRT results match, and
2410         causes the main display's color profile to not affect results on Lion.
2411
2412         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2413         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
2414         (installLayoutTestColorProfile):
2415         (restoreUserColorProfile):
2416         (simpleSignalHandler):
2417         (main):
2418         * Scripts/webkitpy/layout_tests/port/mac.py:
2419         (MacPort.release_http_lock):
2420         (MacPort):
2421         (MacPort._path_to_helper):
2422         (MacPort.start_helper):
2423         (MacPort.stop_helper):
2424         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
2425         (WTR::PlatformWebView::PlatformWebView):
2426         (WTR::PlatformWebView::windowSnapshotImage):
2427
2428 2012-02-24  Darin Fisher  <darin@chromium.org>
2429
2430         Add more Chromium WebKit API reviewers
2431         https://bugs.webkit.org/show_bug.cgi?id=81576
2432
2433         Reviewed by Adam Barth and David Levin.
2434
2435         Simplify watchlist to lump all WebKit API into one.
2436
2437         * Scripts/webkitpy/common/config/watchlist:
2438
2439 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2440
2441         Convert hasSpellingMarker to use Internals interface.
2442         https://bugs.webkit.org/show_bug.cgi?id=81300
2443
2444         Reviewed by Ryosuke Niwa.
2445
2446         Remove hasSpellingMarker functions, because it is able to work in the
2447         cross-port way through the Internals interface.
2448
2449         * DumpRenderTree/LayoutTestController.cpp:
2450         (LayoutTestController::staticFunctions):
2451         * DumpRenderTree/LayoutTestController.h:
2452         (LayoutTestController):
2453         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2454         * DumpRenderTree/chromium/LayoutTestController.cpp:
2455         (LayoutTestController::LayoutTestController):
2456         * DumpRenderTree/chromium/LayoutTestController.h:
2457         (LayoutTestController):
2458         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2459         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2460         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2461         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2462         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2463         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2464         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2465
2466 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2467
2468         Add new Qt-WK2 performance tester bot to build.webkit.org
2469         https://bugs.webkit.org/show_bug.cgi?id=80842
2470
2471         Reviewed by Ryosuke Niwa.
2472
2473         * BuildSlaveSupport/build.webkit.org-config/config.json:
2474         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2475         (RunAndUploadPerfTestsWebKit2):
2476         (RunAndUploadPerfTestsWebKit2.start):
2477         (BuildAndPerfTestWebKit2Factory):
2478         (BuildAndPerfTestWebKit2Factory.__init__):
2479         (DownloadAndPerfTestWebKit2Factory):
2480         (DownloadAndPerfTestWebKit2Factory.__init__):
2481
2482 2012-03-20  Peter Beverloo  <peter@chromium.org>
2483
2484         [Chromium] Update to Android NDK r7b
2485         https://bugs.webkit.org/show_bug.cgi?id=81005
2486
2487         Reviewed by Adam Barth.
2488
2489         Update to version r7b of the Android NDK, release notes of which are
2490         available on the following page:
2491         http://developer.android.com/sdk/ndk/index.html
2492
2493         * Scripts/update-webkit-chromium:
2494
2495 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2496
2497         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
2498         https://bugs.webkit.org/show_bug.cgi?id=81502
2499
2500         Unreviewed.
2501
2502         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
2503
2504 2012-03-19  Jessie Berlin  <jberlin@apple.com>
2505
2506         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
2507         because they broke running the layout tests on mac.
2508
2509         * Scripts/webkitpy/layout_tests/port/webkit.py:
2510         (WebKitDriver._check_for_driver_crash):
2511
2512 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
2513
2514         [GTK] Allow running run-gtk-tests during 'make distcheck'
2515         https://bugs.webkit.org/show_bug.cgi?id=81415
2516
2517         Reviewed by Philippe Normand.
2518
2519         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
2520
2521         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
2522         main command run during 'make check.'
2523         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
2524         a requirement to run it. If jhbuild isn't around it will simply not start the
2525         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
2526         main loop and simply use the main loop in a synchronous method to start SPI
2527         daemons. Did a bunch of general cleanup around these tasks.
2528
2529 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2530
2531         jhbuild: Use the multiprocessing module to get the number of CPUs.
2532         https://bugs.webkit.org/show_bug.cgi?id=81585
2533
2534         Reviewed by Martin Robinson.
2535
2536         Now that support for Python 2.5 has been dropped, we can safely
2537         use multiprocessing.cpu_count() to obtain the current number of
2538         CPUs in the system instead of running a webkitperl script for
2539         that.
2540
2541         * efl/common.py:
2542         (top_level_path):
2543         * efl/jhbuildrc:
2544         * gtk/common.py:
2545         (build_path):
2546         * gtk/jhbuildrc:
2547
2548 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2549
2550         We should only add a live event listener once, not once per TestSelector.
2551         Adding one per TestSelector would cause accordion('option', 'active') to
2552         incorrectly be "false" if you examine multiple sets of tests.
2553
2554         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2555
2556 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2557
2558         Fix typo introduced in r111293.
2559
2560         Unreviewed, build fix.
2561
2562         * Scripts/webkitpy/layout_tests/port/webkit.py:
2563         (WebKitDriver._check_for_driver_crash):
2564
2565 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2566
2567         webkitpy: get ServerProcess out of the reportcrash business
2568         https://bugs.webkit.org/show_bug.cgi?id=81600
2569
2570         Reviewed by Adam Barth.
2571
2572         ServerProcess was half-aware that ReportCrash might run
2573         sometimes, and that the process ServerProcess was talking to
2574         might have its own crashing subprocesses; neither of these
2575         things really worked right and it made the logic convoluted, so
2576         this change makes handling crashes completely separate from the
2577         server_process code, so that it can focus on just I/O to the
2578         subprocess.
2579
2580         There should be no functional changes resulting from this patch.
2581
2582         * Scripts/webkitpy/layout_tests/port/server_process.py:
2583         (ServerProcess._reset):
2584         (ServerProcess._handle_possible_interrupt):
2585         (ServerProcess.write):
2586         (ServerProcess.read_stdout):
2587         (ServerProcess.has_crashed):
2588         (ServerProcess._read):
2589         (ServerProcess.stop):
2590         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2591         (TrivialMockPort.check_for_leaks):
2592         (TestServerProcess.test_broken_pipe):
2593         * Scripts/webkitpy/layout_tests/port/webkit.py:
2594         (WebKitPort._read_image_diff):
2595         (WebKitDriver.has_crashed):
2596         (WebKitDriver._check_for_driver_crash):
2597         (WebKitDriver.run_test):
2598         (WebKitDriver._read_block):
2599         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2600         (MockServerProcess.__init__):
2601         (MockServerProcess):
2602         (MockServerProcess.has_crashed):
2603
2604 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2605
2606         webkitpy: crashlog parsing is broken
2607         https://bugs.webkit.org/show_bug.cgi?id=81586
2608
2609         Reviewed by Adam Barth.
2610
2611         Seeking on a file opened through codecs() appears to not work
2612         correctly (at least on SL).
2613         
2614         The code was more complicated than it needed to be, so I have
2615         revamped it to just read a full crashlog at once and then look
2616         for matches. 
2617         
2618         I have also added the ability to optionally propagate
2619         errors back (which can be helpful to debug races when ReportCrash is
2620         still running and you need to see that you couldn't open some
2621         files, rather than just ignoring them).
2622
2623         * Scripts/webkitpy/common/system/crashlogs.py:
2624         (CrashLogs.find_newest_log):
2625         (CrashLogs._find_newest_log_darwin):
2626         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2627         (CrashLogsTest.test_find_log_darwin):
2628
2629 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2630
2631         webkitpy: clean up some port._filesystem references to not reference private members
2632         https://bugs.webkit.org/show_bug.cgi?id=81595
2633
2634         Reviewed by Eric Seidel.
2635
2636         There should be no functional changes in this patch, just some
2637         refactoring. Also, fixes a minor nit in write_image_diff_files(()
2638
2639         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2640         (SingleTestRunner.__init__):
2641         (SingleTestRunner._run_compare_test):
2642         (SingleTestRunner._run_rebaseline):
2643         (SingleTestRunner._save_baseline_data):
2644         (SingleTestRunner._handle_error):
2645         (SingleTestRunner._run_reftest):
2646         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2647         (write_test_result):
2648         (TestResultWriter.__init__):
2649         (TestResultWriter._make_output_directory):
2650         (TestResultWriter.output_filename):
2651         (TestResultWriter._output_testname):
2652         (TestResultWriter.write_output_files):
2653         (TestResultWriter.write_stderr):
2654         (TestResultWriter.write_crash_report):
2655         (TestResultWriter.create_text_diff_and_write_result):
2656         (TestResultWriter.write_image_diff_files):
2657         (copy_file):
2658         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
2659         (TestResultWriterTest.test_reftest_diff_image):
2660
2661 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2662
2663         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
2664         https://bugs.webkit.org/show_bug.cgi?id=81598
2665
2666         Reviewed by Eric Seidel.
2667
2668         "DumpRenderTree" shouldn't be hard-coded here, since some other
2669         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
2670         
2671         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2672         (TestExpectations):
2673
2674 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2675
2676         webkitpy: allow instance data in TestFailure objects
2677         https://bugs.webkit.org/show_bug.cgi?id=81580
2678
2679         Reviewed by Tony Chang.
2680
2681         Current TestFailures are basically enums, but it would be nice,
2682         especially for crashes, if they contained the name and pid of
2683         the process that crashed.
2684
2685         * Scripts/webkitpy/layout_tests/models/test_failures.py:
2686         (TestFailure.message):
2687         (FailureTimeout.message):
2688         (FailureCrash):
2689         (FailureCrash.__init__):
2690         (FailureCrash.message):
2691         (FailureMissingResult.message):
2692         (FailureTextMismatch.message):
2693         (FailureMissingImageHash.message):
2694         (FailureMissingImage.message):
2695         (FailureImageHashMismatch.message):
2696         (FailureImageHashIncorrect.message):
2697         (FailureReftestMismatch.message):
2698         (FailureReftestMismatchDidNotOccur.message):
2699         (FailureReftestNoImagesGenerated.message):
2700         (FailureMissingAudio.message):
2701         (FailureAudioMismatch.message):
2702         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
2703         (TestFailuresTest.test_equals):
2704         (TestFailuresTest):
2705         (TestFailuresTest.test_crashes):
2706
2707 2012-03-19  Brady Eidson  <beidson@apple.com>
2708
2709         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
2710         Crash in 3rd party WebKit apps under XHR/Cache code
2711
2712         Reviewed by Antti Koivisto.
2713
2714         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2715         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
2716         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
2717         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
2718         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
2719         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
2720         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
2721         (TestWebKitAPI):
2722         (TestWebKitAPI::TEST):
2723
2724 2012-03-19  Brady Eidson  <beidson@apple.com>
2725
2726         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
2727         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
2728
2729         Reviewed by Enrica Casucci.
2730
2731         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
2732         are in the correct format.
2733
2734         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2735         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
2736         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
2737         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
2738         (TestWebKitAPI::contextMenuCopyLink):
2739         (TestWebKitAPI::TEST):
2740
2741 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2742
2743         Make the selected item in the accordion a link to the flakiness dashboard
2744         https://bugs.webkit.org/show_bug.cgi?id=81571
2745
2746         Reviewed by Dimitri Glazkov.
2747
2748         Also, make it so that you can select the text of the link, instead of the current behavior
2749         of acting as if you had clicked on the accordion header.
2750
2751         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2752         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2753
2754 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2755
2756         NRWT runs some tests that are skipped with -i command line option
2757         https://bugs.webkit.org/show_bug.cgi?id=81535
2758
2759         Reviewed by Ojan Vafai.
2760
2761         This change modifies the interaction of Skipped files and
2762         test_expectations files so that entries in Skipped files (and
2763         the -i command line) override *everything* in the expectations
2764         file.
2765
2766         Specifically, a directory in a Skipped file will cause all of
2767         the tests in the dir to be skipped even if individual tests in
2768         the dir are listed in the test_expectations.txt.
2769
2770         Skipped files also override anything in an overrides files.
2771
2772         This seems to make more intuitive sense, since if you list
2773         something in the Skips file (and even more specify it on the
2774         command line) you probably want it to be universally applied.
2775
2776         Theoretically we could add more precedence levels and have full
2777         paths in an expectations file override dirs in a Skipped file
2778         (but not the command line), but I don't know yet that that level
2779         of complexity is justified.
2780         
2781         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2782         (TestExpectations.__init__):
2783         (TestExpectations._add_skipped_tests):
2784         (TestExpectations._add_test):
2785         (TestExpectations._add_expectations):
2786         (TestExpectations._add_expectation_line):
2787         (TestExpectations._already_seen_better_match):
2788         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2789         (test_more_specific_override_resets_skip):
2790         (SkippedTests):
2791         (SkippedTests.get_exp):
2792         (SkippedTests.assert_exp):
2793         (SkippedTests.test_skipped_tests_work):
2794         (SkippedTests.test_duplicate_skipped_test_fails_lint):
2795         (SkippedTests.test_skipped_file_overrides_expectations):
2796         (SkippedTests.test_skipped_dir_overrides_expectations):
2797         (SkippedTests.test_skipped_file_overrides_overrides):
2798         (SkippedTests.test_skipped_dir_overrides_overrides):
2799
2800 2012-03-19  Eric Seidel  <eric@webkit.org>
2801
2802         Fix WTF header include discipline in Chromium WebKit
2803         https://bugs.webkit.org/show_bug.cgi?id=81281
2804
2805         Reviewed by James Robinson.
2806
2807         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2808         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
2809
2810 2012-03-19  Andy Chen  <andchen@rim.com>
2811
2812         [BlackBerry] Add new layout test for text match marker
2813         https://bugs.webkit.org/show_bug.cgi?id=81531
2814
2815         Make the LayoutTestControllerBlackBerry::findString to test the
2816         BlackBerry port implementation.
2817
2818         Reviewed by Rob Buis.
2819
2820         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2821         (LayoutTestController::findString):
2822
2823 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2824
2825         Don't show WONTFIX tests in garden-o-matic
2826         https://bugs.webkit.org/show_bug.cgi?id=81554
2827
2828         Reviewed by Dimitri Glazkov.
2829
2830         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2831         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2832
2833 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
2834
2835         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
2836         https://bugs.webkit.org/show_bug.cgi?id=81530
2837
2838         Reviewed by Jessie Berlin.
2839
2840         Add rowAtIndex to WebKitTestRunner.
2841
2842         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
2843         (WTR::AccessibilityUIElement::rowAtIndex):
2844         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
2845         (AccessibilityUIElement):
2846         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
2847         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2848         (WTR):
2849         (WTR::AccessibilityUIElement::rowAtIndex):
2850
2851 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2852
2853         Change garden-o-matic All Failures tab to Expected Failures
2854         https://bugs.webkit.org/show_bug.cgi?id=81540
2855
2856         Reviewed by Dimitri Glazkov.
2857
2858         Also, fix some broken unittests that had just gotten stale.
2859
2860         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
2861         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
2862         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
2863         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2864         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2865         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2866         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2867         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
2868         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
2869         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2870
2871 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
2872
2873         [chromium] DRT crashes at shutdown.
2874         https://bugs.webkit.org/show_bug.cgi?id=81504
2875
2876         Reviewed by James Robinson.
2877
2878         WebCompositor must be destroyed after WebView is properly destroyed,
2879         or some compositor objects like CCLayerTreeHost may not be destroyed.
2880         However, it is really hard to make the destruction order correct
2881         explicitly in ~TestShell. So move it out to DumpRenderTree.
2882
2883         * DumpRenderTree/chromium/DumpRenderTree.cpp:
2884         (main):
2885         * DumpRenderTree/chromium/TestShell.cpp:
2886         (TestShell::~TestShell):
2887
2888 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2889
2890         add dpranke to webkitpy watchlist
2891         https://bugs.webkit.org/show_bug.cgi?id=81433
2892
2893         Reviewed by Benjamin Poulain.
2894
2895         * Scripts/webkitpy/common/config/watchlist:
2896
2897 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2898
2899         test-webkitpy is failing on SL/Python 2.6.1
2900         https://bugs.webkit.org/show_bug.cgi?id=81533
2901
2902         Reviewed by Ojan Vafai.
2903
2904         It appears that itertools.combinations() is buggy in 2.6.1; this
2905         backports the implementation from 2.7.
2906
2907         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2908         (TestConfigurationConverter.collapse_macros):
2909         (TestConfigurationConverter):
2910         (TestConfigurationConverter.combinations):
2911
2912 2012-03-19  David Michael Barr  <davidbarr@google.com>
2913
2914         Optimize the cleanup sequence in the start-queue.sh script
2915         https://github.com/abarth/webkit/pull/1
2916
2917         Reviewed by Adam Barth.
2918
2919         Updating the working copy is an expensive operation
2920         for a repository the size of WebKit.
2921         Attempt to reset only to the target revision.
2922
2923         * EWSTools/start-queue.sh:
2924
2925 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
2926
2927         Adjust architecture type of the EFL build bot.
2928         https://bugs.webkit.org/show_bug.cgi?id=81452
2929
2930         Reviewed by Tony Chang.
2931
2932         * BuildSlaveSupport/build.webkit.org-config/config.json: The
2933         buildslave is a 64-bit machine, therefore describe it as being
2934         x86_64, not i386.
2935
2936 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2937
2938         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
2939         https://bugs.webkit.org/show_bug.cgi?id=81514
2940
2941         Reviewed by Simon Hausmann.
2942
2943         * MiniBrowser/qt/qml/BrowserWindow.qml:
2944
2945 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2946
2947         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
2948         https://bugs.webkit.org/show_bug.cgi?id=81491
2949
2950         Reviewed by Philippe Normand.
2951
2952         For the find_path(), find_library() etc CMake calls to find the
2953         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
2954         set, otherwise CMake cannot know the dependencies exist and will only
2955         look in the standard directories.
2956
2957         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
2958         the jhbuild Root directory.
2959
2960 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2961
2962         [EFL] Add XDG env vars to jhbuildrc
2963         https://bugs.webkit.org/show_bug.cgi?id=81510
2964
2965         Needed to ensure correct glib behaviour in jhbuild environment.
2966
2967         Reviewed by Philippe Normand.
2968
2969         * efl/jhbuildrc:
2970
2971 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2972
2973         Select best target for tap gesture.
2974         https://bugs.webkit.org/show_bug.cgi?id=78801
2975
2976         Reviewed by Kenneth Rohde Christiansen.
2977
2978         Add TOUCH_ADJUSTMENT to enabled features.
2979
2980         * qmake/mkspecs/features/features.prf:
2981
2982 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
2983
2984         [chromium] synthesize wheel events for fling on main thread
2985         https://bugs.webkit.org/show_bug.cgi?id=81458
2986
2987         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
2988
2989         Reviewed by James Robinson.
2990
2991         * DumpRenderTree/chromium/EventSender.cpp:
2992         (EventSender::EventSender):
2993         (EventSender::gestureFlingCancel):
2994         (EventSender::gestureFlingStart):
2995         * DumpRenderTree/chromium/EventSender.h:
2996         (EventSender):
2997
2998 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2999
3000         [Qt] Add command line option to set the user agent in the MiniBrowser
3001         https://bugs.webkit.org/show_bug.cgi?id=81508
3002
3003         Reviewed by Simon Hausmann.
3004
3005         * MiniBrowser/qt/BrowserWindow.cpp:
3006         (BrowserWindow::BrowserWindow):
3007         (BrowserWindow::webViewExperimental):
3008         * MiniBrowser/qt/BrowserWindow.h:
3009         (BrowserWindow):
3010         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3011         (printHelp):
3012         (MiniBrowserApplication::handleUserOptions):
3013         * MiniBrowser/qt/MiniBrowserApplication.h:
3014         (WindowOptions::WindowOptions):
3015         (WindowOptions::setUserAgent):
3016         (WindowOptions::userAgent):
3017         (WindowOptions):
3018
3019 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
3020
3021         [Qt] Remove dead user agent code from the MiniBrowser.
3022
3023         Reviewed by Simon Hausmann.
3024
3025         * MiniBrowser/qt/BrowserWindow.cpp:
3026         * MiniBrowser/qt/BrowserWindow.h:
3027         (BrowserWindow):
3028         * MiniBrowser/qt/MiniBrowser.qrc:
3029         * MiniBrowser/qt/useragentlist.txt: Removed.
3030
3031 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3032
3033         jhbuild: Call jhbuild's autogen.sh via bash.
3034         https://bugs.webkit.org/show_bug.cgi?id=81485
3035
3036         Reviewed by Philippe Normand.
3037
3038         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
3039         this is usually true only on Linux systems, and even then bash for
3040         some reason might not be available in that location. It is safer to
3041         call bash directly and pass autogen.sh to it instead.
3042
3043         * jhbuild/jhbuild-wrapper:
3044         (install_jhbuild):
3045
3046 2012-03-19  Philippe Normand  <pnormand@igalia.com>
3047
3048         [GTK] Prefix run-launcher with run-with-jhbuild
3049         https://bugs.webkit.org/show_bug.cgi?id=81503
3050
3051         Reviewed by Hajime Morita.
3052
3053         * Scripts/run-launcher:
3054
3055 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3056
3057         [EFL] Prefix run-launcher with run-with-jhbuild
3058         https://bugs.webkit.org/show_bug.cgi?id=80026
3059
3060         Reviewed by Hajime Morita.
3061
3062         * Scripts/run-launcher:
3063
3064 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
3065
3066         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
3067         https://bugs.webkit.org/show_bug.cgi?id=81498
3068
3069         * qmake/mkspecs/features/unix/default_post.prf:
3070
3071 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3072
3073         Unreviewed; switch to using my FreeBSD.org email address, the
3074         other one is going to disappear soon.
3075
3076         * Scripts/webkitpy/common/config/committers.py:
3077         * Scripts/webkitpy/common/config/watchlist:
3078
3079 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
3080
3081         Unreviewed; fix webkitpy unit test failures after r111116.
3082
3083         It turns out that cmd_line in the patch in bug 80025 had the right
3084         signature when it was submitted, but r109242 added a few more
3085         required parameters, and since the method (and EflDriver) did not
3086         exist at the time we ended up committing a broken method.
3087
3088         * Scripts/webkitpy/layout_tests/port/efl.py:
3089         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
3090         arguments.
3091
3092 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
3093
3094         [BlackBerry] Enable Web Timing for performance profiling and improvement.
3095         https://bugs.webkit.org/show_bug.cgi?id=81085
3096
3097         Reviewed by Rob Buis.
3098
3099         * Scripts/build-webkit:
3100
3101 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3102
3103         Add a dependencies installation step for EFL to buildbot master.cfg
3104         https://bugs.webkit.org/show_bug.cgi?id=81337
3105
3106         Reviewed by Gustavo Noronha Silva.
3107
3108         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3109         (InstallEflDependencies):
3110         (Factory.__init__):
3111
3112 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3113
3114         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
3115         https://bugs.webkit.org/show_bug.cgi?id=80025
3116
3117         Reviewed by Gustavo Noronha Silva.
3118
3119         * Scripts/webkitpy/layout_tests/port/efl.py:
3120         (EflDriver):
3121         (EflDriver.cmd_line):
3122         (EflPort._driver_class):
3123
3124 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3125
3126         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
3127         https://bugs.webkit.org/show_bug.cgi?id=79904
3128
3129         build-webkit --efl will use jhbuild for fetching and building dependencies.
3130
3131         Reviewed by Gustavo Noronha Silva.
3132
3133         * Scripts/update-webkitefl-libs: Added.
3134         * Scripts/webkitdirs.pm:
3135         (jhbuildWrapperPrefixIfNeeded):
3136         (generateBuildSystemFromCMakeProject):
3137         (buildCMakeGeneratedProject):
3138         * efl/common.py: Added.
3139         (script_path):
3140         (top_level_path):
3141         (number_of_cpus):
3142         * efl/jhbuild.modules: Added.
3143         * efl/jhbuildrc: Added.
3144         * efl/run-with-jhbuild: Added.
3145
3146 2012-03-16  Stephanie Lewis  <slewis@apple.com>
3147
3148         https://bugs.webkit.org/show_bug.cgi?id=81065
3149         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
3150         Fix crash preventing WebKitTestRunner from working in Release.  
3151         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
3152         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
3153         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
3154
3155         I removed ostringstream in favor of StringBuilder.  
3156
3157         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
3158
3159         Reviewed by Geoff Garen.
3160
3161         Add DEBUG_DEFINES to set NDEBUG on release builds.
3162         * WebKitTestRunner/Configurations/Base.xcconfig:
3163         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
3164
3165         Remove ostringstream member in favor of StringBuilder.
3166         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3167         (WTR::InjectedBundle::initialize):
3168         (WTR::InjectedBundle::beginTesting):
3169         (WTR::InjectedBundle::done):
3170         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3171         (WTF):
3172         (WTR::InjectedBundle::stringBuilder):
3173         (InjectedBundle):
3174
3175         Append output to StringBuilder.
3176         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
3177         (WTR::hasPrefix):
3178         (WTR::dumpPath):
3179         (WTR::rangeToStr):
3180         (WTR::styleDecToStr):
3181         (WTR::frameToStr):
3182         (WTR::dumpFrameScrollPosition):
3183         (WTR::dumpFrameText):
3184         (WTR::dumpDescendantFramesText):
3185         (WTR::InjectedBundlePage::dump):
3186         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
3187         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
3188         (WTR::InjectedBundlePage::willSendRequestForFrame):
3189         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
3190         (WTR::lastFileURLPathComponent):
3191         (WTR::InjectedBundlePage::willAddMessageToConsole):
3192         (WTR::InjectedBundlePage::willSetStatusbarText):
3193         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
3194         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
3195         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
3196         (WTR::InjectedBundlePage::shouldBeginEditing):
3197         (WTR::InjectedBundlePage::shouldEndEditing):
3198         (WTR::InjectedBundlePage::shouldInsertNode):
3199         (WTR::InjectedBundlePage::shouldInsertText):
3200         (WTR::InjectedBundlePage::shouldDeleteRange):
3201         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
3202         (WTR::InjectedBundlePage::shouldApplyStyle):
3203         (WTR::InjectedBundlePage::didBeginEditing):
3204         (WTR::InjectedBundlePage::didEndEditing):
3205         (WTR::InjectedBundlePage::didChange):
3206         (WTR::InjectedBundlePage::didChangeSelection):
3207         (WTR::InjectedBundlePage::supportsFullScreen):
3208         (WTR::InjectedBundlePage::enterFullScreenForElement):
3209         (WTR::InjectedBundlePage::exitFullScreenForElement):
3210         (WTR::InjectedBundlePage::beganEnterFullScreen):
3211         (WTR::InjectedBundlePage::beganExitFullScreen):
3212         (WTR::dumpBackForwardListItem):
3213         (WTR::InjectedBundlePage::dumpBackForwardList):
3214         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3215         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
3216         (WTR::LayoutTestController::dumpConfigurationForViewport):
3217         
3218         Convert between WTF::String and WKStringRef.
3219         * WebKitTestRunner/StringFunctions.h:
3220         (WTR::toWTFString):
3221         (WTR):
3222         * WebKitTestRunner/TestInvocation.cpp:
3223         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
3224
3225 2012-03-16  Eric Seidel  <eric@webkit.org>
3226
3227         Add a land-from-url command
3228         https://bugs.webkit.org/show_bug.cgi?id=81411
3229
3230         Reviewed by Adam Barth.
3231
3232         Very basic so far.  The goal is to make it landing smarter
3233         so that users don't have to bother to parse out the ids out of the urls.
3234
3235         * Scripts/webkitpy/common/config/urls.py:
3236         (parse_bug_id):
3237         (parse_attachment_id):
3238         * Scripts/webkitpy/common/config/urls_unittest.py:
3239         (URLsTest.test_parse_bug_id):
3240         (URLsTest):
3241         (URLsTest.test_parse_attachment_id):
3242         * Scripts/webkitpy/tool/commands/download.py:
3243         (ProcessURLsMixin):
3244         (ProcessURLsMixin._fetch_list_of_patches_to_process):
3245         (LandFromURL):
3246         * Scripts/webkitpy/tool/commands/download_unittest.py:
3247         (test_land_from_bug):
3248         (test_land_from_url):
3249
3250 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
3251
3252         build-webkit launcher instructions use wrong relative path for Qt
3253         https://bugs.webkit.org/show_bug.cgi?id=81357
3254
3255         Reviewed by Eric Seidel.
3256
3257         Ensure we are at the proper directory before generating run-launcher
3258         instructions.
3259
3260         * Scripts/build-webkit:
3261
3262 2012-03-16  Brady Eidson  <beidson@apple.com>
3263
3264         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
3265         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
3266
3267         Reviewed by Geoff Garen.
3268
3269         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3270         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
3271
3272 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3273
3274         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
3275         https://bugs.webkit.org/show_bug.cgi?id=81407
3276
3277         Reviewed by Adam Barth.
3278
3279         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3280         (summarize_results):
3281         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
3282         (ResultSummaryTest.summarized_results):
3283         (ResultSummaryTest.test_no_svn_revision):
3284         (ResultSummaryTest.test_svn_revision):
3285         (ResultSummaryTest):
3286         (ResultSummaryTest.test_summarized_results_wontfix):
3287
3288 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3289
3290         Have webkit-patch rebaseline-test update test_expectations.txt
3291         https://bugs.webkit.org/show_bug.cgi?id=81054
3292
3293         Recommit http://trac.webkit.org/changeset/110783 now that
3294         http://trac.webkit.org/changeset/111034 has landed.
3295
3296         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3297         (TestExpectations.remove_configuration_from_test):
3298         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3299         (RemoveConfigurationsTest):
3300         (RemoveConfigurationsTest.test_remove):
3301         (test_remove_line):
3302         * Scripts/webkitpy/tool/commands/rebaseline.py:
3303         (RebaselineTest._update_expectations_file):
3304         (RebaselineTest._rebaseline_test_and_update_expectations):
3305         (RebaselineTest):
3306         (RebaselineTest.execute):
3307         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3308         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3309         (test_rebaseline_updates_expectations_file):
3310
3311 2012-03-15  Ojan Vafai  <ojan@chromium.org>
3312
3313         Specifier collapsing when writing test expectations lines gets a number of cases wrong
3314         https://bugs.webkit.org/show_bug.cgi?id=81309
3315
3316         Reviewed by Dimitri Glazkov.
3317
3318         I've run this over all the lines in the current Chromium test_expectations.txt file,
3319         so I'm relatively confident we now cover all the cases.
3320
3321         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3322         (TestConfigurationConverter.__init__):
3323         (TestConfigurationConverter.collapse_macros):
3324         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
3325         (TestConfigurationConverter):
3326         (TestConfigurationConverter.intersect_combination):
3327         (TestConfigurationConverter.symmetric_difference):
3328         (TestConfigurationConverter.to_specifiers_list):
3329         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
3330         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
3331         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3332         (make_mock_all_test_configurations_set):
3333         (TestConfigurationConverterTest.test_symmetric_difference):
3334         (TestConfigurationConverterTest.test_to_config_set):
3335         (TestConfigurationConverterTest.test_macro_expansion):
3336         (TestConfigurationConverterTest.test_to_specifier_lists):
3337         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3338         * Scripts/webkitpy/layout_tests/port/chromium.py:
3339         (ChromiumPort):
3340         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3341         (ChromiumAndroidPort.__init__):
3342         The android port uses "arm" as it's architecture, which is technically correct,
3343         but considerably complicates making collapsing work. We probably should kill
3344         the concept of architecture entirely. The benefits are not worth the code
3345         complexity.
3346
3347         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3348         (ChromiumPortTest.test_all_test_configurations):
3349         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3350         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3351         (test_rebaseline_updates_expectations_file):
3352
3353 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
3354
3355         [Qt][Wk2] Assertion Failure and crash on file upload
3356         https://bugs.webkit.org/show_bug.cgi?id=80854
3357
3358         Reviewed by Simon Hausmann.
3359
3360         Added filePicker to WebView using experimental API.
3361
3362         * MiniBrowser/qt/MiniBrowser.qrc:
3363         * MiniBrowser/qt/icons/folder.png: Added.
3364         * MiniBrowser/qt/icons/titlebar.png: Added.
3365         * MiniBrowser/qt/icons/up.png: Added.
3366         * MiniBrowser/qt/qml/BrowserWindow.qml:
3367         * MiniBrowser/qt/qml/FilePicker.qml: Added.
3368
3369 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
3370
3371         Updating email for committer.py script.
3372
3373         * Scripts/webkitpy/common/config/committers.py:
3374
3375 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
3376
3377         [chromium] DRT: Add support for sending gesture scrolls without a begin
3378         https://bugs.webkit.org/show_bug.cgi?id=81292
3379
3380         Reviewed by Kent Tamura.
3381
3382         * DumpRenderTree/chromium/EventSender.cpp:
3383         (EventSender::EventSender):
3384         (EventSender::gestureScrollFirstPoint):
3385         * DumpRenderTree/chromium/EventSender.h:
3386         (EventSender):
3387
3388 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
3389
3390         [GTK] WebKitIconDatabase doesn't keep icons cached
3391         https://bugs.webkit.org/show_bug.cgi?id=56200
3392
3393         Reviewed by Martin Robinson.
3394
3395         Add support for favicons in GtkLauncher.
3396
3397         * GtkLauncher/main.c:
3398         (activateUriEntryCb):
3399         (iconLoadedCb):
3400         (createBrowser):
3401
3402 2012-03-16  Adam Barth  <abarth@webkit.org>
3403
3404         Remove sheriff-bot's last-green-revision command
3405         https://bugs.webkit.org/show_bug.cgi?id=81314
3406
3407         Reviewed by Eric Seidel.
3408
3409         This command is a cruel joke.  WebKit doesn't really ever have a
3410         last-green-revision.
3411
3412         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3413         (BuildBot._find_green_revision):
3414         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3415         (MockBuildBot.builder_statuses):
3416         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3417         (test_find_green_revision):
3418         * Scripts/webkitpy/tool/bot/irc_command.py:
3419         (IRCCommand.execute):
3420         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
3421         (SheriffIRCBotTest.test_help):
3422         * Scripts/webkitpy/tool/commands/queries.py:
3423         (PatchesToReview.execute):
3424
3425 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3426
3427         [Qt] Make gccdepends test aware of broken icecream deps generation
3428
3429         Icecream seems to build the files without passing -o to gcc with
3430         the original destination path, which means the target name in the
3431         dependency file is wrong.
3432
3433         Also, echo something to the final file instead of just touching it,
3434         so it's not deleted on every build.
3435
3436         Reviewed by Simon Hausmann.
3437
3438         * qmake/config.tests/gccdepends/gccdepends.pro:
3439
3440 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
3441
3442         Support for Battery Status API
3443         https://bugs.webkit.org/show_bug.cgi?id=62698
3444
3445         Add include directory and file.
3446         http://www.w3.org/TR/battery-status/
3447
3448         Reviewed by Adam Barth.
3449
3450         * Scripts/build-webkit:
3451
3452 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
3453
3454         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
3455         https://bugs.webkit.org/show_bug.cgi?id=81007
3456
3457         Reviewed by Adam Barth.
3458
3459         A dedidcated build slave is available for efl on build.webkit.org. 
3460         This change add's an entry for the elf port in list of available port builders.
3461         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
3462
3463         * Scripts/webkitpy/layout_tests/port/builders.py:
3464
3465 2012-03-16  Adam Barth  <abarth@webkit.org>
3466
3467         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
3468         https://bugs.webkit.org/show_bug.cgi?id=81302
3469
3470         Reviewed by Eric Seidel.
3471
3472         This is a trivial patch, but the variable name turned up in a grep of
3473         webkitpy and looks like copy pasta.
3474
3475         * Scripts/webkitpy/common/net/statusserver.py:
3476         (StatusServer.next_work_item):
3477
3478 2012-03-16  Andrew Lo  <anlo@rim.com>
3479
3480         [BlackBerry] Enable PAGE_VISIBILITY_API
3481         https://bugs.webkit.org/show_bug.cgi?id=81257
3482
3483         Enable Page Visibility API for BlackBerry.
3484
3485         Reviewed by Rob Buis.
3486
3487         * Scripts/build-webkit:
3488
3489 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
3490
3491         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
3492         https://bugs.webkit.org/show_bug.cgi?id=81232
3493
3494         Reviewed by Philippe Normand.
3495
3496         Use the newly-added helper method in DumpRenderTreeSupportGtk.
3497
3498         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3499         (LayoutTestController::computedStyleIncludingVisitedInfo):
3500
3501 2012-03-15  Eric Seidel  <eric@webkit.org>
3502
3503         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
3504         https://bugs.webkit.org/show_bug.cgi?id=81313
3505
3506         Reviewed by Adam Barth.
3507
3508         * Scripts/webkitpy/common/checkout/changelog.py:
3509         (parse_bug_id_from_changelog):
3510         * Scripts/webkitpy/common/config/urls.py:
3511         (parse_bug_id):
3512         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
3513         (URLsTest):
3514         (URLsTest.test_parse_bug_id):
3515         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
3516         (test_parse_bug_id):
3517         * Scripts/webkitpy/tool/bot/irc_command.py:
3518         (_post_error_and_check_for_bug_url):
3519         * Scripts/webkitpy/tool/bot/sheriff.py:
3520         (Sheriff.post_rollout_patch):
3521         (Sheriff.post_chromium_deps_roll):
3522
3523 2012-03-15  Adam Barth  <abarth@webkit.org>
3524
3525         The commit-queue should fast-track patches that have already passed the testing EWS bots
3526         https://bugs.webkit.org/show_bug.cgi?id=81305
3527
3528         Reviewed by Eric Seidel.
3529
3530         If a patch has already passed the testing EWS bots (currently only the
3531         chromium-ews bot), we should be able to land the patch without running
3532         the tests again.  This patch is somewhat risky in that the tests might
3533         have been broken between when the EWS bots ran and when the
3534         commit-queue is processing the patch.  If that turns out to be a
3535         problem in practice, we'll tighten up this fast-track in some way,
3536         likely by making sure that the pass status is recent.
3537
3538         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
3539         (CommitQueueTaskDelegate.did_pass_testing_ews):
3540         (CommitQueueTask._did_pass_tests_recently):
3541         (CommitQueueTask.run):
3542         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3543         (MockCommitQueue.did_pass_testing_ews):
3544         (test_fast_success_case):
3545         * Scripts/webkitpy/tool/commands/queues.py:
3546         (CommitQueue.did_pass_testing_ews):
3547         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3548         (test_did_pass_testing_ews):
3549
3550 2012-03-15  Kentaro Hara  <haraken@chromium.org>
3551
3552         [PerformanceTests] run-perf-tests should output correct units
3553         https://bugs.webkit.org/show_bug.cgi?id=78303
3554
3555         Reviewed by Ryosuke Niwa.
3556
3557         Previously run-perf-tests just supported "ms" for units.
3558         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
3559         This patch fixes run-perf-tests so that they can support custom units.
3560         The default unit is "ms".
3561
3562         The test result looks like this:
3563
3564         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
3565         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
3566         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
3567
3568         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3569         (PerfTestsRunner._process_parser_test_result):
3570
3571 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
3572
3573         [WinCairo] Unreviewed build change after wtf path changes.
3574
3575         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
3576         directories to look at the private header directory to find 'wtf'.
3577
3578 2012-03-15  Kent Tamura  <tkent@chromium.org>
3579
3580         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
3581         https://bugs.webkit.org/show_bug.cgi?id=81293
3582
3583         Reviewed by Adam Barth.
3584
3585         r110595 introduced new format of the author field of ChangeLog entry:
3586         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
3587         Updte chanelog.py to support it.
3588
3589         * Scripts/webkitpy/common/checkout/changelog.py:
3590         (ChangeLogEntry._split_contributor_names):
3591         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
3592         (test_parse_authors):
3593
3594 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
3595
3596         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
3597
3598         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
3599
3600         Reviewed by Jessie Berlin.
3601
3602         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3603         (LayoutTestController::setValueForUser):
3604         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3605         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3606         (WTR::LayoutTestController::setValueForUser):
3607         (WTR):
3608         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3609         (LayoutTestController):
3610
3611 2012-03-15  Adam Barth  <abarth@webkit.org>
3612
3613         Update the start-queue.sh script to match what we're using on the bots
3614         https://bugs.webkit.org/show_bug.cgi?id=81277
3615
3616         Reviewed by Eric Seidel.
3617
3618         Over time we've added more magic git commands to clean up the local repo after
3619         it has been borked in various ways. The idea behind this sequence is to clear
3620         out any changes in the working copy and then to rebuild the master branch from
3621         origin/master.
3622
3623         This patch also removes the watchdog timer that we've stopped using on the
3624         bots.
3625
3626         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
3627         Note: Eric and I are experimenting with GitHub to see if it might be
3628         useful for WebKit.  If you have thoughts on this topic, please feel
3629         free to grab Eric or me on #webkit to chat.  I assume that he and I are
3630         the only ones who care about this file so using this alternative
3631         process for this patch shouldn't cause any trouble.
3632
3633         * EWSTools/start-queue.sh:
3634
3635 2012-03-15  Adam Barth  <abarth@webkit.org>
3636
3637         The commit-queue can hang when the test_expectations.txt style check fails during commit
3638         https://bugs.webkit.org/show_bug.cgi?id=81251
3639
3640         Reviewed by Eric Seidel.
3641
3642         We forgot to check the --non-interactive flag, so we end up prompting
3643         the "user" who doesn't exist on the bot.  We might consider teaching
3644         the User class about non-interactive if we have this sort of bug again.
3645
3646         * Scripts/webkitpy/tool/steps/commit.py:
3647         (Commit.options):
3648         (Commit._check_test_expectations):
3649         (Commit.run):
3650         * Scripts/webkitpy/tool/steps/commit_unittest.py:
3651         (CommitTest.test_check_test_expectations):
3652
3653 2012-03-15  Jessie Berlin  <jberlin@apple.com>
3654
3655         fast/dom/Window/window-properties.html and fast/dom/prototype-inheritance-2.html failing on
3656         Mac WK2 because WebAudio is not enabled
3657         https://bugs.webkit.org/show_bug.cgi?id=81259
3658
3659         Reviewed by Dan Bernstein.
3660
3661         Enable WebAudio so that it appears in the window object in the tests.
3662         The webaudio tests are all still skipped in the Mac Skipped list.
3663
3664         * WebKitTestRunner/TestController.cpp:
3665         (WTR::TestController::resetStateToConsistentValues):
3666
3667 2012-03-15  David Levin  <levin@chromium.org>
3668
3669         Various autoresize fixes.
3670         https://bugs.webkit.org/show_bug.cgi?id=81199
3671
3672         Reviewed by Darin Fisher.
3673