Heap-use-after-free in WebCore::InlineFlowBox::deleteLine due to fullscreen issues.
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
2
3         new-run-webkit-tests: crashes when it fails to decode a stack trace
4         https://bugs.webkit.org/show_bug.cgi?id=82673
5
6         Unreviewed, build fix.
7
8         We are assuming the stdout/stderr output from the driver is utf-8
9         encoded when we get stack traces; this may not be a valid
10         assumption generally, but if we do get strings that aren't valid
11         utf-8, we would crash. Now we will ignore any decoding errors.
12
13         * Scripts/webkitpy/layout_tests/port/base.py:
14         (Port._get_crash_log):
15         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
16         (PortTestCase.test_get_crash_log):
17
18 2012-03-29  Simon Fraser  <simon.fraser@apple.com>
19
20         Scripts fail to detect when a tool crashes
21         https://bugs.webkit.org/show_bug.cgi?id=82659
22
23         Reviewed by Mark Rowe.
24         
25         Have exitStatus() detect when the process fail to exit cleanly
26         (possibly because of a crash), and return a non-zero exit code
27         in that case.
28
29         * Scripts/VCSUtils.pm:
30         (exitStatus):
31
32 2012-03-29  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
33
34         [GTK][EFL] run-javascriptcore-tests should be run through jhbuild
35         https://bugs.webkit.org/show_bug.cgi?id=82581
36
37         Reviewed by Martin Robinson.
38
39         Running Javascriptcore tests through jhbuild
40         for consistency with run-webkit-tests and in order to
41         avoid confusing libraries when facing regressions.
42
43         * Scripts/run-javascriptcore-tests:
44
45 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
46
47         test-webkitpy: add --timing
48         https://bugs.webkit.org/show_bug.cgi?id=82550
49
50         Reviewed by Eric Seidel.
51
52         This patch adds a --timing option that will display the time
53         each test takes. It also removes the --silent option, since
54         probably no one ever used it, and cleans up the logging
55         option parsing code to be easier to follow.
56
57         * Scripts/webkitpy/test/main.py:
58         (Tester._parse_args):
59         (Tester._configure):
60         * Scripts/webkitpy/test/runner.py:
61         (TestRunner.write_result):
62         * Scripts/webkitpy/test/runner_unittest.py:
63         (RunnerTest.test_regular):
64         (RunnerTest.test_verbose):
65         (RunnerTest):
66         (RunnerTest.test_timing):
67
68 2012-03-29  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
69
70         HashMap<>::add should return a more descriptive object
71         https://bugs.webkit.org/show_bug.cgi?id=71063
72
73         Reviewed by Ryosuke Niwa.
74
75         Update code to use AddResult instead of a pair.
76
77         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
78         (LayoutTestController::evaluateScriptInIsolatedWorld):
79         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
80         (LayoutTestController::evaluateScriptInIsolatedWorld):
81         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
82         (WTR::LayoutTestController::evaluateScriptInIsolatedWorld):
83
84 2012-03-29  Ilya Tikhonovsky  <loislo@chromium.org>
85
86         Web Inspector: chromium: DRT --no-timeout option doesn't work.
87         https://bugs.webkit.org/show_bug.cgi?id=82608
88
89         Initial value for m_timeout was initialized in constructor and was overwritten in DRT::main.
90         This was broken by r112354 and the sequence became opposite.
91
92         Reviewed by Yury Semikhatsky.
93
94         * DumpRenderTree/chromium/TestShell.cpp:
95         (TestShell::TestShell):
96         (TestShell::initialize):
97
98 2012-03-29  Vineet Chaudhary  <rgf748@motorola.com>
99
100         Remove custom bindings form Internals.idl of attribute type Array.
101         https://bugs.webkit.org/show_bug.cgi?id=82319
102
103         Reviewed by Kentaro Hara.
104
105         Remove JSInternalsCustom.cpp and V8InternalsCustom.cpp as no longer required.
106
107         * GNUmakefile.am:
108
109 2012-03-28  Mark Rowe  <mrowe@apple.com>
110
111         Fix --no-saved-state.
112
113         Ignore the flag before Lion, rather than making it specific to Lion.
114
115         Rubber-stamped by Dan Bernstein.
116
117         * Scripts/webkitdirs.pm:
118         (argumentsForRunAndDebugMacWebKitApp):
119
120 2012-03-28  Dave Tharp  <dtharp@codeaurora.org>
121
122         Add dtharp to contributors list
123         https://bugs.webkit.org/show_bug.cgi?id=82545
124
125         Reviewed by Adam Barth.
126
127         Adding dtharp as Contributor.
128
129         * Scripts/webkitpy/common/config/committers.py:
130
131 2012-03-28  Stuart Eichert  <stuart.eichert@palm.com>
132
133         build-webkit : the word "pass" is spelled incorrectly as "passs"
134         https://bugs.webkit.org/show_bug.cgi?id=82532
135
136         Reviewed by Julien Chaffraix.
137
138         Corrected spelling of the word "pass" in build-webkit.
139
140         * Scripts/build-webkit:
141
142 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
143
144         new-run-webkit-tests doesn't log correctly after retrying failures
145         https://bugs.webkit.org/show_bug.cgi?id=82541
146
147         Unreviewed, build fix.
148
149         r112189 introduced a regression where we would accidentally
150         delete the log handler after we retried any failing tests - that
151         meant that any log messages around uploading the results got
152         dropped.
153
154         * Scripts/webkitpy/layout_tests/controllers/manager.py:
155         (Manager._run_tests):
156
157 2012-03-28  Kevin Ollivier  <kevino@theolliviers.com>
158
159         [wx] Unreviewed. Build fix, move WTF back into JSCore target
160         until issues with JSCore not linking in all WTF symbols are resolved.
161
162         * waf/build/settings.py:
163
164 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
165
166         download-built-product build step should detect 404 errors
167         https://bugs.webkit.org/show_bug.cgi?id=82491
168
169         Reviewed by Ryosuke Niwa.
170         
171         Pass --fail to curl so that it reports 404 errors.
172
173         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
174         (DownloadBuiltProduct): Pass --fail to curl so that it reports errors.
175         * BuildSlaveSupport/built-product-archive:
176         (extractBuiltProduct): Remove trailing whitespace.
177
178 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
179
180         flakiness dashboard should only keep the last 500 runs
181         https://bugs.webkit.org/show_bug.cgi?id=82530
182
183         Reviewed by Ojan Vafai.
184
185         If a bot has an overly large amount of failing tests, then
186         storing 750 runs' worth takes a lot of memory and CPU time to
187         process and we end up getting killed by AppEngine; this cases
188         data corruption and confusion down the road when we have to
189         retry the upload of the data and the tests show up as multiple
190         runs.
191
192         * TestResultServer/model/jsonresults.py:
193         (JsonResults.update_file):
194
195 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
196
197         extract-built-product build step should fail if unzipping fails
198         https://bugs.webkit.org/show_bug.cgi?id=82493
199
200         Reviewed by Simon Fraser.
201
202         Forward non-zero exit code from subprocesses so that the master can detect any failures
203         and bail out early.
204
205         * BuildSlaveSupport/built-product-archive:
206         (main):
207
208 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
209
210         test-webkitpy should support files, directories, and packages as command line args
211         https://bugs.webkit.org/show_bug.cgi?id=76765
212
213         Reviewed by Adam Barth.
214
215         This patch adds support for specifying files, directories, and
216         packages to test-webkitpy along with the already existing
217         support for modules, test classes, and individual test names.
218
219         Also, fix a bug in filesystem_mock where we wouldn't normalize a
220         path containing a reference to the current directory properly,
221         for example, '/foo/./bar.py'.
222
223         * Scripts/webkitpy/common/system/filesystem_mock.py:
224         (MockFileSystem.normpath):
225         * Scripts/webkitpy/test/main.py:
226         (Tester._configure_logging):
227         (Tester._run_tests):
228         * Scripts/webkitpy/test/test_finder.py:
229         (TestDirectoryTree.find_modules):
230         (TestDirectoryTree.subpath):
231         (TestFinder.is_dotted_name):
232         (TestFinder.find_names):
233         (TestFinder):
234         (TestFinder._find_names_for_arg):
235         (TestFinder._find_in_trees):
236         (TestFinder._default_names):
237         * Scripts/webkitpy/test/test_finder_unittest.py:
238         (TestFinderTest.setUp):
239         (TestFinderTest.tearDown):
240         (TestFinderTest.check_names):
241         (TestFinderTest.test_default_names):
242         (TestFinderTest):
243         (TestFinderTest.test_paths):
244
245 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
246
247         Fix inspector tests when running in internal builds.
248
249         <rdar://problem/11050694>
250
251         Reviewed by Tim Hatcher.
252
253         * DumpRenderTree/mac/DumpRenderTree.mm:
254         (resetDefaultsToConsistentValues):
255         * WebKitTestRunner/TestController.cpp:
256         (WTR::TestController::resetStateToConsistentValues):
257
258 2012-03-28  David Kilzer  <ddkilzer@apple.com>
259
260         Use xcrun to find nm and ranlib on OS X builds
261         <http://webkit.org/b/82466>
262
263         Reviewed by Dan Bernstein.
264
265         * Scripts/build-webkit: Use xcrun to find ranlib.
266         * Scripts/webkitdirs.pm:
267         (determineNmPath): Added.  Use xcrun to find nm on OS X.
268         (nmPath): Added.
269         * Scripts/webkitperl/features.pm:
270         (libraryContainsSymbol): Use nmPath().
271
272 2012-03-28  Ojan Vafai  <ojan@chromium.org>
273
274         Add history navigation to garden-o-matic
275         https://bugs.webkit.org/show_bug.cgi?id=82495
276
277         Reviewed by Dimitri Glazkov.
278
279         Also, maintain scroll offsets when returning to a tab. This makes addressing
280         expected failures considerably easier when dealing with below-the-fold tests.
281
282         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
283         -Store the scrollTop before switching to a new tab and restore the appropriate scrollTop after
284         switching.
285         -Modify window.location with the tabName in the hash. Can't use pushState because this is served
286         from a file URL and Chrome puts each different file URL in it's own origin.
287
288         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
289         Remove the href so that clicking on the accordion item does not modify window.location.hash.
290
291         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
292         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
293
294 2012-03-28  Philippe Normand  <pnormand@igalia.com>
295
296         [GTK] generate-gtk-doc doesn't cope with custom build directory
297         https://bugs.webkit.org/show_bug.cgi?id=82448
298
299         Reviewed by Martin Robinson.
300
301         Pass the product directory in generate-gtkdoc arguments from
302         build-webkit.
303
304         * Scripts/webkitdirs.pm:
305         (buildAutotoolsProject):
306         * gtk/common.py:
307         (get_build_path):
308
309 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
310
311         [GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
312         https://bugs.webkit.org/show_bug.cgi?id=82341
313
314         Reviewed by Martin Robinson.
315
316         Add --skipped=skip|ignore|only command line option.
317
318         * Scripts/run-gtk-tests:
319         (TestRunner._test_cases_to_skip): Return the list of test cases
320         to skip for the given test depending on the skipped action.
321         (TestRunner._should_run_test): Decide whether to run the test or
322         not depending on the skipped action.
323         (TestRunner.run_tests): Only show the list of skipped tests when
324         skipped action is 'skip'.
325
326 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
327
328         [GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
329         https://bugs.webkit.org/show_bug.cgi?id=82333
330
331         Reviewed by Martin Robinson.
332
333         * Scripts/run-gtk-tests:
334         (TestRunner.SkippedTest): Add SkippedTest class to store more
335         information about skipped tests.
336         (TestRunner._setup_testing_environment): Use SkippedTest to add
337         TestWebKitAccessibility to the skipped list.
338         (TestRunner._find_skipped_test): Return the SkippedTest object for
339         a given test path.
340         (TestRunner._test_cases_to_skip): Returns the list of test cases
341         that should be skipped for a given test path.
342         (TestRunner._should_run_test): Helper funtion to decide whether
343         the given test path should be run or not. Tests in the skipped
344         list that don't have a list of failing test cases are not run at
345         all. Tests in the skipped list that contain a list of test cases
346         are run with gtester -s to skip the individual test cases.
347         (TestRunner._run_test): Helper funtion to run a given test.
348         (TestRunner.run_tests): Build the list of tests to run based on
349         _should_run_test() and use helper function _run_test() to run
350         every test.
351
352 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
353
354         [Qt] Couple of tests have different results on 64 bit and/or in debug mode compared to 32 bit and/or release mode
355         https://bugs.webkit.org/show_bug.cgi?id=52810
356
357         Reviewed by Simon Hausmann.
358
359         Use SSE2 floating point math on 32 bit instead of the default
360         387 to make layout test result same on 32 and on 64 bit builds.
361
362         * qmake/mkspecs/features/unix/default_post.prf:
363
364 2012-03-28  Sergio Villar Senin  <svillar@igalia.com>
365
366         [Soup] DNS prefetching spams resolver, shoots self in the foot
367         https://bugs.webkit.org/show_bug.cgi?id=41630
368
369         Reviewed by Martin Robinson.
370
371         Bump libsoup and glib dependencies.
372
373         * efl/jhbuild.modules:
374         * gtk/jhbuild.modules:
375
376 2012-03-28  Gustavo Noronha Silva  <gns@gnome.org>
377
378         [GTK] Disable introspection for jhbuild modules
379         https://bugs.webkit.org/show_bug.cgi?id=82395
380
381         Reviewed by Philippe Normand.
382
383         * efl/jhbuild.modules: add --disable-introspection to all modules
384         that support it
385         * gtk/jhbuild.modules: ditto.
386
387 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
388
389         [Qt] Unreviewed, fix API tests after adding full platforms. (r111658)
390
391         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
392         (BuildAndTestFactory.__init__):
393
394 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
395
396         [Qt] Unreviewed typo fix after r112383.
397
398         * Scripts/run-qtwebkit-tests:
399         (Main.check_crash_occurences):
400
401 2012-03-28  Ádám Kallai  <kadam@inf.u-szeged.hu>
402
403         [Qt] run-qtwebkit-tests should report crashes
404         https://bugs.webkit.org/show_bug.cgi?id=76905
405
406         Added checking of crashing tests. Now final summary includes the total number of crashing tests too.
407         Made unit test for this modification by János Badics <jbadics@inf.u-szeged.hu>.
408
409         Reviewed by Csaba Osztrogonác.
410
411         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
412         (RunQtAPITests.commandComplete):
413         (RunQtAPITests.evaluateCommand):
414         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
415         (RunQtAPITestsTest):
416         (test_failure):
417         (test_crash):
418         * Scripts/run-qtwebkit-tests:
419         (Main.announce_results_html):
420         (Main):
421         (Main.check_crash_occurences):
422         (Main.convert_to_stdout):
423         (Main.convert_to_html):
424
425 2012-03-28  Kwonjin Jeong  <gram@company100.net>
426
427         Add support for Intel compiler on Linux QT port.
428         https://bugs.webkit.org/show_bug.cgi?id=81077
429
430         Reviewed by Tor Arne Vestbø.
431
432         * qmake/mkspecs/features/default_post.prf:
433         Don't hide symbols when building with Intel compiler.
434         The symbols are needed when building QtWebKit tests.
435
436         * qmake/mkspecs/features/unix/icc.prf:
437         Added to ignore icc-specific warnings.
438         This feature is used with command-line arguments for build-webkit script.
439
440 2012-03-28  Philippe Normand  <pnormand@igalia.com>
441
442         Unreviewed, trivial update-webkit-libs-jhbuild system call fix.
443
444         * Scripts/update-webkitgtk-libs: This script is written in
445         Perl. Not Python. So use perl vocabulary to refer to command line arguments.
446
447 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
448
449         kill-old-processes doesn't do anything on linux
450         https://bugs.webkit.org/show_bug.cgi?id=82388
451
452         Reviewed by Tony Chang.
453
454         Kill cc1plus, ld, etc... on linux. Also ps aux + grep to kill old run-webkit-tests processes.
455         This should help stabilizing Chromium linux bots.
456
457         * BuildSlaveSupport/kill-old-processes:
458
459 2012-03-28  Philippe Normand  <pnormand@igalia.com>
460
461         [GTK] jhbuild wrongly tricked into clean-up
462         https://bugs.webkit.org/show_bug.cgi?id=82216
463
464         Reviewed by Gustavo Noronha Silva.
465
466         Save jhbuild{rc,.modules}.md5sum in Dependencies instead of the
467         build tree. Also when checking if the jhbuild configuration
468         changed use absolute md5sum paths.
469
470         * Scripts/webkitdirs.pm:
471         (runAutogenForAutotoolsProjectIfNecessary):
472         (jhbuildConfigurationChanged):
473         (buildAutotoolsProject):
474
475 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
476
477         kill-old-processes doesn't do anything on linux
478         https://bugs.webkit.org/show_bug.cgi?id=82388
479
480         One more attempt to disable kill-old-processes on the Qt bots not to kill each others.
481
482         The platform in Factory comes from config.json, accordingly platform is fullplatform,
483         for example: qt-4.8, qt-5.0-wk2. We have to check only the prefix of platform.
484
485         Reviewed by Ryosuke Niwa.
486
487         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
488         (Factory.__init__):
489
490 2012-03-28  Hao Zheng  <zhenghao@chromium.org>
491
492         Make Chromium port for Android use hardware gpu path default.
493         https://bugs.webkit.org/show_bug.cgi?id=82033
494
495         Reviewed by Dirk Pranke.
496
497         Skip canvas tests, which are run in virtual gpu test suites.
498
499         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
500         (ChromiumAndroidPort.__init__):
501         (ChromiumAndroidPort.skipped_tests):
502
503 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
504
505         kill-old-processes doesn't do anything on linux
506         https://bugs.webkit.org/show_bug.cgi?id=82388
507
508         Disable kill-old-processes on the Qt bots not to kill each others.
509
510         Reviewed by Ryosuke Niwa.
511
512         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
513         (Factory.__init__):
514
515 2012-03-27  Hao Zheng  <zhenghao@chromium.org>
516
517         Handle DRT crash caused by Android OOM.
518         https://bugs.webkit.org/show_bug.cgi?id=82310
519
520         Reviewed by Tony Chang.
521
522         When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
523         is stopped silently and regarded as crashed. Re-run the test for
524         such crash.
525
526         Add some debug log to NRWT.
527
528         * Scripts/webkitpy/layout_tests/port/chromium.py:
529         (ChromiumDriver.run_test):
530         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
531         (ChromiumAndroidPort.__init__):
532         (ChromiumAndroidPort.get_last_stacktrace):
533         (ChromiumAndroidDriver.__init__):
534         (ChromiumAndroidDriver._start):
535         (ChromiumAndroidDriver.run_test):
536         (ChromiumAndroidDriver._get_drt_return_value):
537
538 2012-03-27  Ojan Vafai  <ojan@chromium.org>
539
540         Fix expected results for some unittest failures.
541
542         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
543
544 2012-03-27  Adrienne Walker  <enne@google.com>
545
546         [chromium] Fix --enabled-threaded-compositing flag in DRT
547         https://bugs.webkit.org/show_bug.cgi?id=82405
548
549         Reviewed by James Robinson.
550
551         --enable-threaded-compositing wasn't working because there was no way
552         to set the value before it was used in the TestShell constructor.
553
554         To avoid adding yet another boolean parameter to TestShell's
555         constructor, add an explicit initialize function that can operate on
556         any boolean settings that are set post-constructor.
557
558         * DumpRenderTree/chromium/DumpRenderTree.cpp:
559         (main):
560         * DumpRenderTree/chromium/TestShell.cpp:
561         (TestShell::TestShell):
562         (TestShell::initialize):
563         * DumpRenderTree/chromium/TestShell.h:
564         (TestShell):
565         (TestShell::setTestShellMode):
566
567 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
568
569         new-run-webkit-tests: suppress debug output from run-safari on results file
570         https://bugs.webkit.org/show_bug.cgi?id=82400
571
572         Reviewed by Ryosuke Niwa.
573
574         Send the run-safari output to /dev/null rather than the console.
575
576         * Scripts/webkitpy/layout_tests/port/mac.py:
577         (MacPort.show_results_html_file):
578
579 2012-03-27  Simon Fraser  <simon.fraser@apple.com>
580
581         Improve error reporting in run-javascriptcore-tests
582         https://bugs.webkit.org/show_bug.cgi?id=82379
583
584         Reviewed by Jessie Berlin.
585         
586         When the script dies, have it print out the reason for dying.
587
588         * Scripts/run-javascriptcore-tests:
589
590 2012-03-27  Eric Seidel  <eric@webkit.org>
591
592         Add Dave Barton to the Contributors list so I can more easily CC him on bugs
593         https://bugs.webkit.org/show_bug.cgi?id=82373
594
595         Reviewed by Ryosuke Niwa.
596
597         Dave's committer-status in in-progress, so until then
598         list him as a Contributor so at least I can more easily CC him on bugs.
599
600         * Scripts/webkitpy/common/config/committers.py:
601
602 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
603
604         [jhbuild] Prepend jhbuildrc's directory to sys.path instead of appending.
605         https://bugs.webkit.org/show_bug.cgi?id=82252
606
607         Reviewed by Gustavo Noronha Silva.
608
609         Appending the directory containing common.py may be problematic if
610         another path in sys.path also contains either a common.py or a
611         common/ module, so prepend to select our version instead. This is
612         the case with fail2ban, which has a common/ directory which may
613         end up being in Python's default path.
614
615         * efl/jhbuildrc:
616         * gtk/jhbuildrc:
617
618 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
619
620         test-webkitpy: prepare for better test run output
621         https://bugs.webkit.org/show_bug.cgi?id=82290
622
623         Reviewed by Adam Barth.
624
625         This code basically re-implements the output of the TextTestRunner default
626         runner code from unittest, although the implementation is quite
627         different, in preparation for changing the test output to be
628         metered and possibly running in parallel.
629
630         The output is almost identical to before, except that instead of
631         logging "test_regular (webkitpy.main.RunnerTest) passed" we log
632         "webkitpy.main.RunnerTest.test_regular passed". It has always
633         annoyed me that they invert the names to be harder to read and
634         so that you can't copy & paste back to the input for
635         test-webkitpy.
636
637         This patch is provided to add a skeleton for unit tests and for
638         comparison to an upcoming patch that will actually add new
639         functionality.
640
641         * Scripts/webkitpy/test/main.py:
642         (Tester.__init__):
643         (Tester._configure_logging):
644         (Tester._run_tests):
645         * Scripts/webkitpy/test/runner.py: Added.
646         (TestRunner):
647         (TestRunner.__init__):
648         (TestRunner.test_name):
649         (TestRunner.all_test_names):
650         (TestRunner.run):
651         (TestRunner.write_result):
652         (TestRunner.write_summary):
653         * Scripts/webkitpy/test/runner_unittest.py: Added.
654         (FakeModuleSuite):
655         (FakeModuleSuite.__init__):
656         (FakeModuleSuite.__str__):
657         (FakeModuleSuite.run):
658         (FakeTopSuite):
659         (FakeTopSuite.__init__):
660         (FakeLoader):
661         (FakeLoader.__init__):
662         (FakeLoader.top_suite):
663         (FakeLoader.loadTestsFromName):
664         (RunnerTest):
665         (RunnerTest.test_regular):
666         (RunnerTest.test_verbose):
667
668 2012-03-27  Gustavo Noronha Silva  <gns@gnome.org>
669
670         [GTK] Build gnutls without p11-kit support
671         https://bugs.webkit.org/show_bug.cgi?id=82212
672
673         Reviewed by Martin Robinson.
674
675         * gtk/jhbuild.modules: pass --without-p11-kit to gnutls'
676         configure, to avoid requiring that dependency, which is not used
677         by libsoup and WebKit anyway
678
679 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
680
681         [Qt] Make debug build work on 32 bit
682         https://bugs.webkit.org/show_bug.cgi?id=82331
683
684         Use --no-keep-memory option of ld not to cache the symbol tables of input files in
685         memory to avoid memory exhaustion during the linking phase for x86-debug builds.
686
687         Reviewed by Tor Arne Vestbø.
688
689         * qmake/mkspecs/features/unix/default_post.prf:
690         * qmake/mkspecs/features/unix/default_pre.prf:
691
692 2012-03-23  Balazs Ankes  <bank@inf.u-szeged.hu>
693
694         Add png checking to check-webkit-style
695         https://bugs.webkit.org/show_bug.cgi?id=75824
696
697         Reviewed by NOBODY Tony Chang.
698
699         * Scripts/webkitpy/style/checker.py:
700         (_all_categories):
701         (FileType):
702         (CheckerDispatcher._should_skip_file_path):
703         (CheckerDispatcher._file_type):
704         (CheckerDispatcher._create_checker):
705         * Scripts/webkitpy/style/checker_unittest.py:
706         (CheckerDispatcherDispatchTest.test_none_paths):
707         * Scripts/webkitpy/style/checkers/png.py: Added.
708         (PNGChecker):
709         (PNGChecker.__init__):
710         (PNGChecker.check):
711         (PNGChecker._config_file_path):
712         * Scripts/webkitpy/style/checkers/png_unittest.py: Added.
713         (MockSCMDetector):
714         (MockSCMDetector.__init__):
715         (MockSCMDetector.display_name):
716         (MockSCMDetector.propget):
717         (PNGCheckerTest):
718         (PNGCheckerTest.test_init):
719         (PNGCheckerTest.test_init.mock_handle_style_error):
720         (PNGCheckerTest.test_check):
721         (PNGCheckerTest.test_check.mock_handle_style_error):
722         * Scripts/webkitpy/style/patchreader.py:
723         (PatchReader.check):
724
725 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
726
727         Unreviewed. Skip GTK+ unit tests that fail in the bots
728
729         * Scripts/run-gtk-tests:
730         (TestRunner): Add failing tests to the skipped list.
731
732 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
733
734         Unreviewed. Unskip all GTK+ unit test.
735
736         Currently, skipped tests contains the list of unit tests that have
737         failed in one or more bots. In this moment it's not possible to
738         skip just the test cases that fail, so the complete unit test is
739         skipped when a single test case fails. For that reason, we don't
740         know which test cases are actually failing for every skipped
741         test. We are planning to land a patch to allow skipping individual
742         test cases, but first we need to know which test cases are
743         currently failing in the bots to skip them. So, this patch simply
744         unskips all unit tests in order to watch the bots and get the list
745         of test cases failing. Failing tests will be skipped again as soon
746         as we have the new list of failing test cases.
747
748         * Scripts/run-gtk-tests:
749         (TestRunner): Reset the list of skipped tests.
750
751 2012-03-26  Scott Byer  <scottbyer@chromium.org>
752
753         Enable layout testing of the scroll animator.
754         https://bugs.webkit.org/show_bug.cgi?id=81858
755         Add a call to the InternalSettings that layout tests can use to
756         turn on scroll animation. Enable animation updates for the
757         Chromium platform DRT when scroll animation has been turned on in
758         a test. This should be a no-op for all current layout tests.
759
760         Reviewed by James Robinson.
761
762         * DumpRenderTree/chromium/WebViewHost.cpp:
763         (WebViewHost::serviceAnimation):
764         (WebViewHost::scheduleAnimation):
765         * DumpRenderTree/chromium/WebViewHost.h:
766         (WebViewHost):
767
768 2012-03-26  Dinu Jacob  <dinu.jacob@nokia.com>
769
770         [Qt][WK2] Support multi-file upload
771         https://bugs.webkit.org/show_bug.cgi?id=81589
772
773         Reviewed by Simon Hausmann.
774
775         Modified filePicker to support multi-file upload.
776
777         * MiniBrowser/qt/MiniBrowser.qrc:
778         * MiniBrowser/qt/icons/checkbox_checked.png: Added.
779         * MiniBrowser/qt/icons/checkbox_unchecked.png: Added.
780         * MiniBrowser/qt/js/MultiSelect.js: Added.
781         (values):
782         (isSelected):
783         (removeValue):
784         * MiniBrowser/qt/qml/CheckBox.qml: Added.
785         * MiniBrowser/qt/qml/FilePicker.qml:
786
787 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
788
789         Retry crashing tests serially at the end of NRWT on Apple Mac
790         https://bugs.webkit.org/show_bug.cgi?id=82233
791
792         Reviewed by Dirk Pranke.
793
794         * Scripts/webkitpy/layout_tests/controllers/manager.py:
795         (Manager.run):
796         * Scripts/webkitpy/layout_tests/port/base.py:
797         (Port.should_retry_crashes):
798         * Scripts/webkitpy/layout_tests/port/mac.py:
799         (MacPort.should_retry_crashes):
800
801 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
802
803         NRWT should retry failures serially
804         https://bugs.webkit.org/show_bug.cgi?id=82241
805
806         Reviewed by Dirk Pranke.
807
808         * Scripts/webkitpy/layout_tests/controllers/manager.py:
809         (Manager._run_tests): Add another parameter to specify the number of workers to use.
810         (Manager.run): Run the failures serially.
811
812 2012-03-26  Adrienne Walker  <enne@google.com>
813
814         Allow expectations on directories in virtual test suites
815         https://bugs.webkit.org/show_bug.cgi?id=82250
816
817         Reviewed by Dirk Pranke.
818
819         lookup_virtual_test_base was only returning valid results for files
820         and not directories. Fix by falling back to the virtual test suite
821         mapping if possible.
822
823         * Scripts/webkitpy/layout_tests/port/base.py:
824         (Port.lookup_virtual_test_base):
825         * Scripts/webkitpy/layout_tests/port/test.py:
826         (TestPort.skipped_tests):
827         (TestPort.virtual_test_suites):
828
829 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
830
831         Fix duplicated ChangeLog entry from r112171.
832
833         Unreviewed, build fix.
834
835 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
836
837         nrwt: remove --worker-model flag
838         https://bugs.webkit.org/show_bug.cgi?id=82112
839
840         Reviewed by Adam Barth.
841
842         The --worker-model=inline flag is basically redundant with the
843         --child-processes=1 flag (technically, you could one run child
844         process, but this is only useful for testing); this patch
845         removes this flag, which simplifies things and enables more
846         cleanup down the road.
847
848         * Scripts/webkitpy/layout_tests/controllers/manager.py:
849         (Manager._run_tests):
850         (Manager.print_config):
851         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
852         (get):
853         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
854         (make_broker):
855         (FunctionTests.test_get__inline):
856         (FunctionTests.test_get__processes):
857         (_TestsMixin.setUp):
858         (_TestsMixin.make_broker):
859         (InlineBrokerTests.setUp):
860         (MultiProcessBrokerTests.setUp):
861         * Scripts/webkitpy/layout_tests/port/base.py:
862         (Port.default_child_processes):
863         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
864         (ChromiumAndroidPort.check_sys_deps):
865         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
866         (ChromiumAndroidPortTest):
867         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
868         (MockDRTPortTest.make_port):
869         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
870         (PortTestCase):
871         (PortTestCase.make_port):
872         * Scripts/webkitpy/layout_tests/port/test.py:
873         (TestPort.default_child_processes):
874         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
875         (_set_up_derived_options):
876         (parse_args):
877         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
878         (parse_args):
879         (MainTest.test_batch_size):
880         (MainTest.test_child_processes_2):
881         (MainTest.test_child_processes_min):
882         (MainTest.test_exception_raised):
883         (MainTest.test_keyboard_interrupt):
884         (MainTest.test_exit_after_n_crashes):
885         (MainTest.assert_run_order):
886         (MainTest.test_run_order__inline):
887         (MainTest.test_virtual):
888
889 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
890
891         test-webkitpy: split test-finding code into a different module
892         https://bugs.webkit.org/show_bug.cgi?id=82253
893
894         Reviewed by Adam Barth.
895
896         Per suggestion from abarth, this change splits all the
897         filesystem-crawling, test-finding code in test-webkitpy out into
898         a separate module and switches to using a FileSystem object.
899         This makes things much more testable, so we also add tests :).
900
901         We also add a realpath() method to the Filesystem object, since
902         test-webkitpy needs that in order to be able to resolve symlinks
903         properly to determine whether a file is under a particular tree
904         or not.
905
906         * Scripts/webkitpy/common/system/filesystem.py:
907         (FileSystem.realpath):
908         * Scripts/webkitpy/common/system/filesystem_mock.py:
909         (MockFileSystem.realpath):
910         * Scripts/webkitpy/test/main.py:
911         (Tester.__init__):
912         (Tester.add_tree):
913         (Tester.run):
914         (Tester._run_tests):
915         (Tester._log_exception):
916         * Scripts/webkitpy/test/test_finder.py: Added.
917         (TestDirectoryTree):
918         (TestDirectoryTree.__init__):
919         (TestDirectoryTree.find_modules):
920         (TestDirectoryTree.find_modules.file_filter):
921         (TestDirectoryTree.to_module):
922         (TestDirectoryTree.clean):
923         (TestFinder):
924         (TestFinder.__init__):
925         (TestFinder.add_tree):
926         (TestFinder.additional_paths):
927         (TestFinder.clean_trees):
928         (TestFinder.is_module):
929         (TestFinder.to_module):
930         (TestFinder.find_names):
931         (TestFinder._exclude):
932         * Scripts/webkitpy/test/test_finder_unittest.py: Added.
933         (TestFinderTest):
934         (TestFinderTest.setUp):
935         (TestFinderTest.tearDown):
936         (TestFinderTest.test_additional_system_paths):
937         (TestFinderTest.test_is_module):
938         (TestFinderTest.test_to_module):
939         (TestFinderTest.test_clean):
940         (TestFinderTest.test_find_names):
941
942 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
943
944         [jhbuild] Use $MAKE if it is defined to build jhbuild itself.
945         https://bugs.webkit.org/show_bug.cgi?id=82209
946
947         Reviewed by Martin Robinson.
948
949         Respecting the $MAKE environment variable when it is defined makes
950         building jhbuild more portable, as some platforms may have GNU
951         make as gmake, for example.
952
953         * jhbuild/jhbuild-wrapper:
954         (install_jhbuild):
955
956 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
957
958         add a Tree abstraction to test-webkitpy to better encapsulate things
959         https://bugs.webkit.org/show_bug.cgi?id=82158
960
961         Reviewed by Adam Barth.
962
963         There are no functional changes in this patch, but adding a
964         basic abstraction for the trees we are looking in for python
965         tests will allow me to add some features later on more easily
966         (like ignoring certain directories like webkitpy/thirdparty),
967         and it makes things slightly less hard-coded.
968
969         * Scripts/test-webkitpy:
970         * Scripts/webkitpy/test/main.py:
971         (Tester.__init__):
972         (Tester):
973         (Tester.add_tree):
974         (Tester._parse_args):
975         (Tester._configure):
976         (Tester.run):
977         (Tester._find_modules):
978         (Tester._run_tests):
979         (Tester._is_module):
980         (Tester._log_exception):
981         (TestDirectoryTree):
982         (TestDirectoryTree.__init__):
983         (TestDirectoryTree.find_modules):
984         (TestDirectoryTree.clean):
985
986 2012-03-26  Ojan Vafai  <ojan@chromium.org>
987
988         garden-o-matic should special case reftests
989         https://bugs.webkit.org/show_bug.cgi?id=81812
990
991         Reviewed by Adam Barth.
992
993         On the examine page, don't show a rebaseline button for reftests.
994         We still show it on the failure summary page because tests are grouped
995         together. Long-term, we should make sure webkit-patch rebaseline-test
996         never tries to rebaseline reftests.
997
998         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
999         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1000         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1001
1002 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1003
1004         clean up NRWT logging/metering, be less verbose
1005         https://bugs.webkit.org/show_bug.cgi?id=73846
1006
1007         Reviewed by Ojan Vafai.
1008
1009         Re-land r112014 with a build fix for Python 2.6 compatibility.
1010
1011         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1012         (ManagerTest.test_http_locking):
1013         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1014         (AbstractWorker.run):
1015         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1016         (Worker.__init__):
1017         (Worker._set_up_logging):
1018         (Worker):
1019         (Worker._set_up_host_and_port):
1020         (Worker.run):
1021         (Worker.cleanup):
1022         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
1023         (WorkerTest.test_default_platform_in_worker):
1024         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1025         (run):
1026         (main):
1027         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1028         (MeteredStream):
1029         (MeteredStream.implements):
1030         (MeteredStream._erasure):
1031         (MeteredStream._ensure_newline):
1032         (MeteredStream.__init__):
1033         (MeteredStream.__del__):
1034         (MeteredStream.cleanup):
1035         (MeteredStream.write_throttled_update):
1036         (MeteredStream.write_update):
1037         (MeteredStream.write):
1038         (MeteredStream.writeln):
1039         (MeteredStream._erase_last_partial_line):
1040         (_LogHandler):
1041         (_LogHandler.__init__):
1042         (_LogHandler.emit):
1043         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1044         (RegularTest):
1045         (RegularTest.setUp):
1046         (RegularTest.tearDown):
1047         (RegularTest.time_fn):
1048         (RegularTest.test_logging_not_included):
1049         (RegularTest._basic):
1050         (RegularTest.test_basic):
1051         (RegularTest._log_after_update):
1052         (RegularTest.test_log_after_update):
1053         (TtyTest):
1054         (TtyTest.test_basic):
1055         (TtyTest.test_log_after_update):
1056         (VerboseTest):
1057         (VerboseTest.test_basic):
1058         (VerboseTest.test_log_after_update):
1059         * Scripts/webkitpy/layout_tests/views/printing.py:
1060         (parse_print_options):
1061         (Printer.__init__):
1062         (Printer.cleanup):
1063         (Printer.print_progress):
1064         (Printer.print_update):
1065         (Printer._write):
1066         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1067         (Testprinter.get_printer):
1068         (Testprinter.test_print_progress):
1069         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1070         (PerfTestsRunner.__init__):
1071         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1072         (test_run_test_pause_before_testing):
1073
1074 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1075
1076         [jhbuild] Revert $MAKE environment hack introduced in r101929.
1077         https://bugs.webkit.org/show_bug.cgi?id=82234
1078
1079         Reviewed by Martin Robinson.
1080
1081         The problem it was supposedly fixing was fixed in jhbuild itself
1082         in git revision da8944d8ca987ca48c91b350257a530050406092, so we
1083         can remove the workaround added to the script.
1084
1085         * jhbuild/jhbuild-wrapper:
1086         (install_jhbuild):
1087
1088 2012-03-26  Leo Yang  <leo.yang@torchmobile.com.cn>
1089
1090         [BlackBerry] Skeleton code of AsyncFileSystemBlackBerry.cpp
1091         https://bugs.webkit.org/show_bug.cgi?id=82157
1092
1093         Reviewed by Rob Buis.
1094
1095         * Scripts/build-webkit:
1096         Make ENABLE_FILE_SYSTEM default on for blackberry.
1097
1098 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1099
1100         [jhbuild] Explicitly disable DTrace when building glib.
1101         https://bugs.webkit.org/show_bug.cgi?id=82145
1102
1103         Reviewed by Gustavo Noronha Silva.
1104
1105         DTrace support is not really needed by WebKit, so disabling it
1106         helps avoid some build errors on more exotic platforms (such as
1107         FreeBSD).
1108
1109         * efl/jhbuild.modules:
1110         * gtk/jhbuild.modules:
1111
1112 2012-03-26  Sheriff Bot  <webkit.review.bot@gmail.com>
1113
1114         Unreviewed, rolling out r111993.
1115         http://trac.webkit.org/changeset/111993
1116         https://bugs.webkit.org/show_bug.cgi?id=82184
1117
1118         It broke layout and API tests and made WTR crash (Requested by
1119         Ossy on #webkit).
1120
1121         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1122         (WTR::PlatformWebView::PlatformWebView):
1123
1124 2012-03-25  Kevin Ollivier  <kevino@theolliviers.com>
1125
1126         [wx] Unreviewed build fix. Move WTF to its own static lib build.
1127
1128         * waf/build/settings.py:
1129         (common_configure):
1130
1131 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
1132
1133         AX: Support solution to handle invalid ax text marker
1134         https://bugs.webkit.org/show_bug.cgi?id=82023
1135
1136         Reviewed by Oliver Hunt.
1137
1138         * DumpRenderTree/AccessibilityUIElement.cpp:
1139         (indexForTextMarkerCallback):
1140         (isTextMarkerValidCallback):
1141         (textMarkerForIndexCallback):
1142         (AccessibilityUIElement::indexForTextMarker):
1143         (AccessibilityUIElement::isTextMarkerValid):
1144         (AccessibilityUIElement::textMarkerForIndex):
1145         (AccessibilityUIElement::getJSClass):
1146         * DumpRenderTree/AccessibilityUIElement.h:
1147         (AccessibilityUIElement):
1148         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1149         (AccessibilityUIElement::indexForTextMarker):
1150         (AccessibilityUIElement::textMarkerForIndex):
1151         (AccessibilityUIElement::isTextMarkerValid):
1152         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1153         (WTR::AccessibilityUIElement::indexForTextMarker):
1154         (WTR::AccessibilityUIElement::isTextMarkerValid):
1155         (WTR::AccessibilityUIElement::textMarkerForIndex):
1156         (WTR):
1157         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1158         (AccessibilityUIElement):
1159         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1160         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1161         (WTR):
1162         (WTR::AccessibilityUIElement::indexForTextMarker):
1163         (WTR::AccessibilityUIElement::isTextMarkerValid):
1164         (WTR::AccessibilityUIElement::textMarkerForIndex):
1165
1166 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
1167
1168         Unreviewed, rolling out r112014.
1169         http://trac.webkit.org/changeset/112014
1170         https://bugs.webkit.org/show_bug.cgi?id=82138
1171
1172         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
1173         #webkit).
1174
1175         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1176         (ManagerTest.test_http_locking):
1177         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1178         (AbstractWorker.run):
1179         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1180         (Worker.__init__):
1181         (Worker.safe_init):
1182         (Worker.run):
1183         (Worker.cleanup):
1184         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
1185         (WorkerTest.test_default_platform_in_worker):
1186         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1187         (run):
1188         (main):
1189         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1190         (MeteredStream):
1191         (MeteredStream.is):
1192         (MeteredStream.__init__):
1193         (MeteredStream.write):
1194         (MeteredStream.update):
1195         (MeteredStream._overwrite):
1196         (MeteredStream._reset):
1197         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1198         (TestMeteredStream):
1199         (TestMeteredStream.test_regular):
1200         * Scripts/webkitpy/layout_tests/views/printing.py:
1201         (_configure_logging):
1202         (_restore_logging):
1203         (Printer.__init__):
1204         (Printer.cleanup):
1205         (Printer.print_progress):
1206         (Printer.print_update):
1207         (Printer._write):
1208         (Printer):
1209         (Printer._update):
1210         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1211         (TestUtilityFunctions.assertEmpty):
1212         (TestUtilityFunctions):
1213         (TestUtilityFunctions.assertNotEmpty):
1214         (TestUtilityFunctions.assertWritten):
1215         (TestUtilityFunctions.test_configure_logging):
1216         (Testprinter.get_printer):
1217         (Testprinter.test_print_progress):
1218         (Testprinter.test_print_progress.mock_time):
1219         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1220         (PerfTestsRunner.__init__):
1221         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1222         (test_run_test_pause_before_testing):
1223
1224 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
1225
1226         [wx] Unreviewed. Fix the build after WTF move.
1227
1228         * waf/build/build_utils.py:
1229         (get_excludes):
1230         (get_excludes_in_dirs):
1231         * waf/build/settings.py:
1232         (get_port_excludes):
1233         (common_configure):
1234
1235 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
1236
1237         clean up NRWT logging/metering, be less verbose
1238         https://bugs.webkit.org/show_bug.cgi?id=73846
1239
1240         Reviewed by Ojan Vafai.
1241
1242         This patch totally rewrites the metered_stream module and
1243         changes the way it and the printing module now interact with the
1244         logging subsystem.
1245
1246         The printing module becomes a dumb layout_tests-specific wrapper around
1247         the metered stream. All of the intelligence of how and when to
1248         actually write (or erase) lines and how things interact with
1249         logging moves into the metered stream.
1250
1251         As of now, anything written to the metered stream *bypasses the
1252         logging system*. If a logger is passed to the meter constructor,
1253         the metered stream will also attach a log handler so that all
1254         messages fed to the logger will also be written to the metered
1255         stream. This detangles the logic that used to exist where
1256         sometimes the meter would write to the logger and sometimes the
1257         logger would write to the meter.
1258
1259         All of this makes the logic generic and reusable for
1260         test-webkitpy (coming in another patch). It also helps clarify
1261         what the mysterious "configure_logging" parameter to the printer
1262         constructor used to do (that parameter itself is gone now, and
1263         by default the printer and metered_stream will not ever touch
1264         the logging system; this makes a bunch of tests less fragile).
1265
1266         All of the above should produce no user-visible changes.
1267
1268         However, to also fix the bug in question, we change the message
1269         format in --verbose mode to %H:%M:%S.%frac %pid %message,
1270         eliminating the date, file:lineno, and level parameters. I am
1271         making this change at the same time because it would be
1272         difficult to write things consistently between the logs and the
1273         metered stream if we were still logging the file:lineno.
1274
1275         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1276           Remove the configure_logging parameter
1277         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1278           Use metered stream to configure the logging (we don't need a printer)
1279           and be careful about when and how we manipulate the log level
1280           and handlers on the root logger.
1281         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1282           Pass in the root logger so that log messages are formatted
1283           properly, and set the log level outside of any routines called
1284           by unit tests so as to not interfere.
1285         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
1286           remove a duplicate log message.
1287         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1288           complete rewrite :).
1289         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1290           complete rewrite :).
1291         * Scripts/webkitpy/layout_tests/views/printing.py:
1292           move the logging configuration into metered_stream, and also
1293           move the "when should I log progress messages" into metered stream.
1294         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1295           remove tests that are no longer necessary.
1296         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1297           remove the configure_logging parameter.
1298         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1299           update the test; since the printer no longer writes to the log,
1300           we need to check what is written to the regular_output instead.
1301           (test_run_test_pause_before_testing):
1302
1303 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
1304
1305         Add myself and Martin Robinson to:
1306           * GStreamerGraphics
1307           * GtkWebKit2PublicAPI
1308           * SoupNetwork
1309         and to the newly created WebKitGTKTranslations.
1310
1311         Rubber-stamped by Martin Robinson.
1312
1313         * Scripts/webkitpy/common/config/watchlist:
1314
1315 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
1316
1317         [Qt][WK2] Implement PageClient::isViewWindowActive()
1318         https://bugs.webkit.org/show_bug.cgi?id=81143
1319
1320         Reviewed by Kenneth Rohde Christiansen.
1321
1322         Fix the WrapperWindow from our PlatformWebView
1323         so it correctly creates the platform related
1324         stuff for QWindow, which is not created unless
1325         QWindow::setVisible() or QWindow::show() are called.
1326
1327         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1328         (WTR::PlatformWebView::PlatformWebView):
1329
1330 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1331
1332         nrwt: don't wait for safari to exit before exiting after showing the results file
1333         https://bugs.webkit.org/show_bug.cgi?id=81845
1334
1335         Reviewed by Ryosuke Niwa.
1336
1337         Prior to this change, if you ran new-run-webkit-tests and
1338         displayed the HTML results file at the end, we would block
1339         waiting for the user to quit the browser. There doesn't seem to
1340         be a need for that, and the Chromium ports don't do this.
1341
1342         Also, update the mac tests to capture the output and be quiet :).
1343
1344         * Scripts/webkitpy/common/system/executive_mock.py:
1345           Make popen() testable, implement should_log for it.
1346         * Scripts/webkitpy/layout_tests/port/mac.py:
1347         (MacPort.show_results_html_file):
1348         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1349           Silence some of the tests via outputcapture.
1350
1351 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
1352
1353         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
1354         https://bugs.webkit.org/show_bug.cgi?id=82073
1355
1356         Reviewed by Martin Robinson.
1357
1358         * Scripts/webkitdirs.pm:
1359         (jhbuildConfigurationChanged): new function that isolates the
1360         jhbuild configuration change checking from the autogen logic.
1361         (mustReRunAutogen): code for the new function lived here.
1362         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
1363         itself, and cleans up the jhbuild root if configuration has
1364         changed; it also forces autogen to be run in that case, to
1365         make sure the new libraries are used.
1366         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
1367         * jhbuild/jhbuild-wrapper:
1368         (ensure_jhbuild): this might be run inside a jhbuild environment,
1369         in which case aclocal fails when trying to use the now deleted
1370         aclocal directory in the jhbuild prefix, so work around that.
1371
1372 2012-03-23  Ojan Vafai  <ojan@chromium.org>
1373
1374         garden-o-matic scrolls to the top of the page when you click an accordion item
1375         https://bugs.webkit.org/show_bug.cgi?id=82088
1376
1377         Reviewed by Adam Barth.
1378
1379         Check the scroll position and make sure that the clicked accordion item is visible.
1380
1381         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1382
1383 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1384
1385         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
1386         https://bugs.webkit.org/show_bug.cgi?id=80567
1387
1388         Reviewed by Ryosuke Niwa.
1389
1390         Move clean_up_run to occur right after we finish testing and
1391         before we start processing the results. Also makes set_up_run()
1392         and clean_up_run() private since there's no real need for
1393         callers to be aware of them.
1394
1395         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1396         (Manager._set_up_run):
1397         (Manager.run):
1398         (Manager._clean_up_run):
1399         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1400         (ManagerTest.test_http_locking):
1401         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1402         (run):
1403
1404 2012-03-23  Dean Jackson  <dino@apple.com>
1405
1406         Disable CSS_SHADERS in Apple builds
1407         https://bugs.webkit.org/show_bug.cgi?id=81996
1408
1409         Reviewed by Simon Fraser.
1410
1411         Default enableCSSShaders to be 0 on all builds that use
1412         this script. Use the flag to turn it back on.
1413
1414         * Scripts/build-webkit:
1415
1416 2012-03-23  Tony Chang  <tony@chromium.org>
1417
1418         [chromium] rename newwtf target back to wtf
1419         https://bugs.webkit.org/show_bug.cgi?id=82064
1420
1421         Reviewed by Adam Barth.
1422
1423         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1424
1425 2012-03-23  Jessie Berlin  <jberlin@apple.com>
1426
1427         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
1428         the Lion Intel WebKit2 Debug bots.
1429         https://bugs.webkit.org/show_bug.cgi?id=81012.
1430
1431         Rubber-stamped by Dan Bernstein.
1432
1433         Add a step to kill the pasteboard server while we investigate what is causing these
1434         assertion failures (to get the bots green).
1435
1436         * BuildSlaveSupport/kill-old-processes:
1437
1438 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
1439
1440         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
1441         https://bugs.webkit.org/show_bug.cgi?id=81986
1442
1443         Reviewed by David Levin.
1444
1445         Track whether or not errors are filtered in error handlers. If an
1446         alphabetical sorting error is reported for a header and filtered, try
1447         reporting it for the other potential source of the error. This ensures
1448         that the style bot can still find new sorting errors in both potential
1449         situations:
1450
1451             #include <foo.h> <-- 1. edited line
1452             #include <baz.h> <-- 2. edited line
1453
1454         * Scripts/webkitpy/style/checkers/cpp.py:
1455         (check_include_line): Potentially report the error for both lines,
1456         if the first error is filtered.
1457         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
1458         test that allows only reporting errors for certain lines. This verifies that
1459         when one line of an ordering error is filtered, the error is reported on the
1460         other line.
1461         (ErrorCollector.__init__):
1462         (ErrorCollector.__call__):
1463         (CppStyleTestBase.perform_lint):
1464         (CppStyleTestBase.perform_language_rules_check):
1465         (CppStyleTestBase.assert_language_rules_check):
1466         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
1467         Added a new test that verifies that when one line of the two lines of an ordering
1468         error occur, the error is reported on the other line.
1469         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
1470         (MockErrorHandler.__call__): Report True because the error is handled.
1471         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1472         (ErrorCollector.__call__): Ditto.
1473         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
1474         (MockErrorHandler.__call__): Ditto.
1475         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
1476         (TestErrorHandler.__call__): Ditto.
1477         * Scripts/webkitpy/style/checkers/xml_unittest.py:
1478         (MockErrorHandler.__call__): Ditto.
1479         * Scripts/webkitpy/style/error_handlers.py:
1480         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
1481         and False if it is filtered.
1482
1483 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
1484
1485         Build fix for WinCE after r111778.
1486
1487         * WinCELauncher/CMakeLists.txt:
1488
1489 2012-03-23  Peter Beverloo  <peter@chromium.org>
1490
1491         Unreviewed buildfix for the Chromium Android builder.
1492
1493         Disable the ImageDiff target for now. Android needs to build this for
1494         host rather than for target (as other configurations do), but since the
1495         WTF move the dependency chain got significantly more complicated.
1496
1497         I'll work out a proper fix in the following bug, but since several
1498         things are broken due to this, this buildfix should be landed first.
1499         https://bugs.webkit.org/show_bug.cgi?id=82039
1500
1501         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1502
1503 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
1504
1505         [EFL][DRT] Implement LayoutTestController's methods related with editing.
1506         https://bugs.webkit.org/show_bug.cgi?id=80136
1507
1508         Reviewed by Ryosuke Niwa.
1509
1510         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
1511         setSelectTrailingWhitespaceEnabled.
1512
1513         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1514         (LayoutTestController::setSmartInsertDeleteEnabled):
1515         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
1516
1517 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1518
1519         Convert hasGrammarMarker to use Internals interface
1520         https://bugs.webkit.org/show_bug.cgi?id=82004
1521
1522         Reviewed by Ryosuke Niwa.
1523
1524         Remove hasGrammarMarker functions, because it is able to work in the
1525         cross-port way through the Internals interface.
1526
1527         * DumpRenderTree/LayoutTestController.cpp:
1528         (LayoutTestController::staticFunctions):
1529         * DumpRenderTree/LayoutTestController.h:
1530         (LayoutTestController):
1531         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1532         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1533         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1534         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1535         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1536         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1537
1538 2012-03-22  Adam Barth  <abarth@webkit.org>
1539
1540         sync-master-with-upstream creates non-linear git history
1541         https://bugs.webkit.org/show_bug.cgi?id=82006
1542
1543         Reviewed by Julien Chaffraix.
1544
1545         Previously this script would create a merge commit at the HEAD of
1546         origin/master, causing your fork of WebKit to diverge from upstream.
1547         The new version of this script pushes directly from upstream to origin
1548         (without round-tripping through any local branches), avoiding this
1549         problem.
1550
1551         * Scripts/sync-master-with-upstream:
1552
1553 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1554
1555         Unreviewed build fix. Add knowledge of the WTF project to
1556         the autotools build system scripts so that build-jsc works
1557         correctly for GTK+.
1558
1559         * Scripts/webkitdirs.pm:
1560         (buildAutotoolsProject): handle the WTF project, build the
1561         libWTF.la target.
1562         (buildGtkProject): add WTF to the list of supported projects.
1563
1564 2012-03-22  Eric Seidel  <eric@webkit.org>
1565
1566         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
1567         https://bugs.webkit.org/show_bug.cgi?id=81984
1568
1569         Reviewed by Adam Barth.
1570
1571         * Scripts/VCSUtils.pm:
1572         (adjustPathForRecentRenamings):
1573
1574 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1575
1576         [CMake] Fix build-jsc after r111778 and r111797.
1577         https://bugs.webkit.org/show_bug.cgi?id=82001
1578
1579         Reviewed by Filip Pizlo.
1580
1581         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
1582         does return if no error occurred, and build-jsc failed with
1583         "building not defined for this platform".
1584
1585         * Scripts/build-jsc: Always exit after caling
1586         buildCMakeProjectOrExit().
1587         * Scripts/webkitdirs.pm:
1588         (buildCMakeProjectOrExit): Return 0 if no error occurred.
1589
1590 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1591
1592         [CMake] Unreviewed build fix after r111778.
1593
1594         * DumpRenderTree/efl/CMakeLists.txt: Replace
1595         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
1596         paths.
1597         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
1598         wtf include.
1599         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
1600         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
1601         with ${WTF_DIR}/wtf in the include paths.
1602
1603 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
1604
1605         Tools/Scripts/build-jsc doesn't work after WTF move
1606         https://bugs.webkit.org/show_bug.cgi?id=81999
1607
1608         Reviewed by Eric Seidel.
1609         
1610         For ports that require WTF and JSC to be built as two projects and where
1611         the machinery to build both does not already exist, teach the tool to do
1612         so for us.
1613
1614         * Scripts/build-jsc:
1615         (buildMyProject):
1616
1617 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1618
1619         Fix accidental missing line in r111793.
1620
1621         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1622         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
1623         Remove base.values. It was unused and duplicates the builtin Object.values.
1624         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1625         Accidentally deleted this line before committing r111793.
1626
1627 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1628
1629         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
1630         https://bugs.webkit.org/show_bug.cgi?id=81983
1631
1632         Reviewed by Adam Barth.
1633
1634         The dialogs get in the way if you are doing multiple rebaselines. Also,
1635         a number of people didn't realize that you could do multiple rebaselines
1636         in parallel. A non-modal dialog is just confusing.
1637
1638         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1639         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1640         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
1641         show the number of tests being rebaselined.
1642
1643         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1644         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
1645         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
1646         into a single area.
1647
1648         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
1649         also serves to clear all teh content in the StatusArea. 
1650
1651         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1652         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1653
1654 2012-03-22  Tony Chang  <tony@chromium.org>
1655
1656         Unreviewed, fix chromium build after wtf move.
1657
1658         Only use newwtf, remove references to wtf.
1659
1660         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1661
1662 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1663
1664         Actually move WTF files to their new home
1665         https://bugs.webkit.org/show_bug.cgi?id=81844
1666
1667         [Qt] Unreviewed buildfix after r111778.
1668
1669         * DumpRenderTree/qt/DumpRenderTree.pro:
1670
1671 2012-03-22  Dan Bernstein  <mitz@apple.com>
1672
1673         Font fallback in WebKitTestRunner depends on user settings
1674         https://bugs.webkit.org/show_bug.cgi?id=81978
1675
1676         Reviewed by Jessie Berlin.
1677
1678         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
1679         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
1680         with a value consisting of the array ("en") to the override defaults
1681         dictionary. This is the same value used in DumpRenderTree, and it
1682         ensures that font fallback follows a consistent path that is not
1683         affected by the language preferences of the user running the test.
1684
1685 2012-03-22  Andy Estes  <aestes@apple.com>
1686
1687         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
1688         https://bugs.webkit.org/show_bug.cgi?id=81971
1689
1690         Reviewed by Dan Bernstein.
1691
1692         Xcode will automatically link ObjC binaries against libarclite in some
1693         cases, which defines a class called __ARCLite__.
1694         check-for-inappropriate-objc-class-names shouldn't complain about this
1695         class name.
1696
1697         * Scripts/check-for-inappropriate-objc-class-names:
1698
1699 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1700
1701         new-run-webkit-tests blames the wrong test for crashing
1702         https://bugs.webkit.org/show_bug.cgi?id=81951
1703
1704         Reviewed by Adam Barth.
1705
1706         NRWT wasn't extracting the WebProcess pid correct; should've
1707         used re.search instead of re.match :(. Add more tests!
1708
1709         * Scripts/webkitpy/layout_tests/port/mac.py:
1710         (MacPort._get_crash_log):
1711         * Scripts/webkitpy/layout_tests/port/webkit.py:
1712         (WebKitDriver.has_crashed):
1713         (WebKitDriver._check_for_driver_crash):
1714         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1715         (WebKitDriverTest.test_no_timeout):
1716         (WebKitDriverTest):
1717         (WebKitDriverTest.test_check_for_driver_crash):
1718         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
1719         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
1720         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
1721         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
1722         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
1723         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
1724
1725 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1726
1727         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
1728         https://bugs.webkit.org/show_bug.cgi?id=81512
1729
1730         Adding missing implementation markerTextForListItem to EFL's
1731         LayoutTestController so that we can unskip related tests from
1732         the skip list.
1733
1734         Reviewed by Antonio Gomes.
1735
1736         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1737         (LayoutTestController::markerTextForListItem): Implemented.
1738
1739 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1740
1741         [EFL] Enable view mode media feature layout tests
1742         https://bugs.webkit.org/show_bug.cgi?id=81205
1743
1744         Adding missing implementation setViewModeMediaFeature to EFL's 
1745         LayoutTestController so that we can unskip view mode 
1746         media feature layout tests.
1747
1748         Reviewed by Antonio Gomes.
1749
1750         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1751         (LayoutTestController::setViewModeMediaFeature): Implemented.
1752
1753 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1754
1755         nrwt: crash while stopping layout test helper on apple mac lion
1756         https://bugs.webkit.org/show_bug.cgi?id=81949
1757
1758         Reviewed by Eric Seidel.
1759
1760         stop_helper() didn't work right if the helper process wasn't
1761         still running or didn't respond properly to being shut down. Fix
1762         this and add some tests. Also add code to MockExecutive() to be
1763         able to mock executive.popen() properly.
1764
1765         * Scripts/webkitpy/common/system/executive_mock.py:
1766         (MockProcess.__init__):
1767         (MockProcess.wait):
1768         (MockExecutive.__init__):
1769         (MockExecutive.popen):
1770         * Scripts/webkitpy/common/system/systemhost_mock.py:
1771         (MockSystemHost.__init__):
1772         * Scripts/webkitpy/layout_tests/port/mac.py:
1773         (MacPort.start_helper):
1774         (MacPort.stop_helper):
1775         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1776         (test_get_crash_log):
1777         (test_helper_starts):
1778         (test_helper_fails_to_start):
1779         (test_helper_fails_to_stop):
1780         (test_helper_fails_to_stop.bad_waiter):
1781
1782 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1783
1784         [GTK] Add make to the jhbuild moduleset
1785         https://bugs.webkit.org/show_bug.cgi?id=81769
1786
1787         Reviewed by Martin Robinson.
1788
1789         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
1790         Add GNU make with custom patches to fix several issues:
1791
1792         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
1793           fix the argument list limit of GNU make when running make
1794           distcheck.
1795         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
1796           and Jeremy Devenport to fix parallel build issues present in
1797           GNU make 3.82
1798
1799         * gtk/jhbuild.modules: Added make with patches.
1800         * gtk/patches/make-3.82-arg-list-length.patch: Added.
1801         * gtk/patches/make-3.82-parallel-build.patch: Added.
1802
1803 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
1804
1805         [wx] Unreviewed. Adding Source/WTF to the build and updating
1806         waf port list.
1807
1808         * DumpRenderTree/wscript:
1809         * waf/build/settings.py:
1810
1811 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1812
1813         Re-land fix for r81603 yet again (!) with latest build fix.
1814         https://bugs.webkit.org/show_bug.cgi?id=81603
1815
1816         Unreviewed, build fix.
1817
1818         The latest problem was a bad sprintf() in
1819         BasePort._get_crash_log(). I've reworked that routine and added
1820         unit tests for coverage.
1821
1822         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1823         (SingleTestRunner._handle_error):
1824         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1825         (write_test_result):
1826         (TestResultWriter.write_crash_log):
1827         * Scripts/webkitpy/layout_tests/port/base.py:
1828         (Port.diff_text):
1829         (Port._get_crash_log):
1830         * Scripts/webkitpy/layout_tests/port/chromium.py:
1831         (ChromiumDriver.run_test):
1832         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1833         (ChromiumDriverTest.test_crash_log):
1834         * Scripts/webkitpy/layout_tests/port/driver.py:
1835         (DriverOutput.__init__):
1836         * Scripts/webkitpy/layout_tests/port/mac.py:
1837         (MacPort.is_lion):
1838         (MacPort._get_crash_log):
1839         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1840         (test_default_child_processes):
1841         (test_get_crash_log):
1842         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1843         (MockDRTPortTest.test_get_crash_log):
1844         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1845         (PortTestCase.test_get_crash_log):
1846         * Scripts/webkitpy/layout_tests/port/test.py:
1847         (TestDriver.run_test):
1848         * Scripts/webkitpy/layout_tests/port/webkit.py:
1849         (WebKitDriver.__init__):
1850         (WebKitDriver._start):
1851         (WebKitDriver.has_crashed):
1852         (WebKitDriver._check_for_driver_crash):
1853         (WebKitDriver.run_test):
1854
1855 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1856
1857         [Qt] Enable support for handling database quota's in the MiniBrowser
1858         https://bugs.webkit.org/show_bug.cgi?id=81928
1859
1860         Reviewed by Simon Hausmann.
1861
1862         * MiniBrowser/qt/qml/BrowserWindow.qml:
1863
1864 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1865
1866         [EFL] Implement LayoutTestController::setEditingBehavior
1867         https://bugs.webkit.org/show_bug.cgi?id=81124
1868
1869         Adding missing implementation setEditingBehavior to EFL's
1870         LayoutTestController. This implementation allows us to
1871         unskip some tests from the skip list.
1872
1873         Reviewed by Martin Robinson.
1874
1875         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1876         (LayoutTestController::setEditingBehavior): Implemented.
1877
1878 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
1879
1880         [GTK] Use the angle-bracket form to include wtf headers
1881         https://bugs.webkit.org/show_bug.cgi?id=81884
1882
1883         Reviewed by Eric Seidel.
1884
1885         Use #include <wtf/foo> instead of #include "foo".
1886
1887         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1888         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1889         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1890         * DumpRenderTree/gtk/EditingCallbacks.cpp:
1891         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
1892         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
1893
1894 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1895
1896         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
1897
1898         * Scripts/webkitpy/common/config/committers.py:
1899
1900 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1901
1902         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
1903         https://bugs.webkit.org/show_bug.cgi?id=81889
1904
1905         Reviewed by Kenneth Rohde Christiansen.
1906
1907         Enable FAST_MOBILE_SCROLLING feature.
1908
1909         * qmake/mkspecs/features/features.prf:
1910
1911 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1912
1913         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
1914         https://bugs.webkit.org/show_bug.cgi?id=81900
1915
1916         Reviewed by Gustavo Noronha Silva.
1917
1918         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
1919         path which assumed a certain directory structure for the build
1920         directory which may noy be true.
1921
1922         Instead, hardcode the path to the downloaded fonts starting from the
1923         top-level source directory. While this solution is also far from
1924         perfect, it makes it more consistent and reliable, and easier to
1925         tackle at once with the rest of bug 81475.
1926
1927         * DumpRenderTree/efl/CMakeLists.txt:
1928
1929 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
1930
1931         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
1932         https://bugs.webkit.org/show_bug.cgi?id=81888
1933
1934         Reviewed by Philippe Normand.
1935
1936         * Scripts/update-webkit-libs-jhbuild:
1937         (runJhbuild): Helper function to run jhbuild with a given command.
1938         Instead of running jhbuild directly, use the jhbuild-wrapper, so
1939         that we don't need to run it first to check whether jhbuild is
1940         installed or not.
1941         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
1942         * gtk/run-with-jhbuild: Ditto.
1943         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
1944         not only run.
1945
1946 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1947
1948         Unreviewed, test-webkitpy fix after r111661.
1949
1950         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1951         (MainTest.test_crash_log):
1952         (MainTest.test_web_process_crash_log):
1953
1954 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1955
1956         Unreviewed, test-webkitpy fix after r111642.
1957
1958         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1959         (ChromiumDriverTest.setUp):
1960         (ChromiumDriverTest.test_two_drivers):
1961
1962 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1963
1964         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1965         https://bugs.webkit.org/show_bug.cgi?id=81603
1966
1967         Unreviewed rolling out r111609 and part of r111615,
1968         because it broke NRWT on Qt-WK2 platform.
1969
1970         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1971         (SingleTestRunner._handle_error):
1972         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1973         (write_test_result):
1974         (TestResultWriter.write_crash_report):
1975         * Scripts/webkitpy/layout_tests/port/base.py:
1976         (Port.is_crash_reporter):
1977         (Port._driver_class):
1978         * Scripts/webkitpy/layout_tests/port/chromium.py:
1979         (ChromiumDriver.run_test):
1980         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1981         (ChromiumDriverTest.test_crashed_process_name):
1982         * Scripts/webkitpy/layout_tests/port/driver.py:
1983         (DriverOutput.__init__):
1984         * Scripts/webkitpy/layout_tests/port/mac.py:
1985         (MacPort):
1986         (MacPort.is_crash_reporter):
1987         (MacPort.release_http_lock):
1988         * Scripts/webkitpy/layout_tests/port/test.py:
1989         (TestDriver.run_test):
1990         * Scripts/webkitpy/layout_tests/port/webkit.py:
1991         (WebKitDriver.__init__):
1992         (WebKitDriver._start):
1993         (WebKitDriver.has_crashed):
1994         (WebKitDriver._check_for_driver_crash):
1995         (WebKitDriver.run_test):
1996
1997 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1998
1999         [Qt] Add full platforms to Qt buildslaves
2000         https://bugs.webkit.org/show_bug.cgi?id=81877
2001
2002         Reviewed by Ryosuke Niwa.
2003
2004         * BuildSlaveSupport/build.webkit.org-config/config.json:
2005
2006 2012-03-21  Johnny Ding  <jnd@chromium.org>
2007
2008         Customize layout test timeout value for different ports.
2009         https://bugs.webkit.org/show_bug.cgi?id=79859
2010
2011         Reviewed by Ojan Vafai.
2012
2013         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2014         (Manager):
2015         * Scripts/webkitpy/layout_tests/port/base.py:
2016         (Port.default_test_timeout_ms):
2017         * Scripts/webkitpy/layout_tests/port/chromium.py:
2018         (ChromiumDriver.stop):
2019         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2020         (ChromiumAndroidPort.default_test_timeout_ms):
2021         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2022         (_set_up_derived_options):
2023
2024 2012-03-21  Mark Rowe  <mrowe@apple.com>
2025
2026         Fix the build.
2027
2028         Ensure that our test tools don't try and use RTTI unnecessarily.
2029
2030         * DumpRenderTree/mac/Configurations/Base.xcconfig:
2031         * TestWebKitAPI/Configurations/Base.xcconfig:
2032         * WebKitTestRunner/Configurations/Base.xcconfig:
2033
2034 2012-03-21  Adam Barth  <abarth@webkit.org>
2035
2036         WebKit should have some scripts to smooth interactions with GitHub
2037         https://bugs.webkit.org/show_bug.cgi?id=81842
2038
2039         Reviewed by Eric Seidel.
2040
2041         These scripts support the workflow in
2042         https://trac.webkit.org/wiki/UsingGitHub by automating the
2043         configuration and syncing operations needed for a GitHub fork of WebKit
2044         to track the main WebKit repository.
2045
2046         * Scripts/configure-github-as-upstream: Added.
2047         * Scripts/sync-master-with-upstream: Added.
2048
2049 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2050
2051         WTR - log the pid of a crashing WebProcess
2052         https://bugs.webkit.org/show_bug.cgi?id=81575
2053
2054         Reviewed by Alexey Proskuryakov.
2055
2056         This allows run-webkit-tests to find the right crash log for a
2057         given crashed WebProcess in WebKit2.
2058
2059         * WebKitTestRunner/TestController.cpp:
2060         (WTR::exceededDatabaseQuota):
2061         (WTR::TestController::createOtherPage):
2062         (WTR::TestController::initialize):
2063         (WTR::TestController::resetStateToConsistentValues):
2064         (WTR::TestController::runTest):
2065         (WTR::TestController::processDidCrash):
2066
2067 2012-03-21  Enrica Casucci  <enrica@apple.com>
2068
2069         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
2070         https://bugs.webkit.org/show_bug.cgi?id=81835
2071         <rdar://problem/11082749>
2072
2073         Reviewed by Brady Eidson.
2074
2075         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
2076         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
2077         (TestWebKitAPI::contextMenuCopyLink):
2078
2079 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
2080
2081         [CSS Shaders] Make CSS Shaders compile on Chromium
2082         https://bugs.webkit.org/show_bug.cgi?id=81435
2083
2084         Reviewed by Stephen White.
2085         
2086         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
2087
2088         * DumpRenderTree/chromium/LayoutTestController.cpp:
2089         (LayoutTestController::overridePreference):
2090         * DumpRenderTree/chromium/WebPreferences.cpp:
2091         (WebPreferences::reset):
2092         (WebPreferences::applyTo):
2093         * DumpRenderTree/chromium/WebPreferences.h:
2094         (WebPreferences):
2095
2096 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2097
2098         fix regression on gtk port introduced by r111609
2099         https://bugs.webkit.org/show_bug.cgi?id=81839
2100
2101         Reviewed by Philippe Normand.
2102
2103         r111609 should've initialized a couple of missing fields in
2104         WebKitDriver, and also should've checked for method overrides
2105         in the other ports :(.
2106
2107         * Scripts/webkitpy/layout_tests/port/gtk.py:
2108         (GtkDriver._start):
2109         * Scripts/webkitpy/layout_tests/port/webkit.py:
2110         (WebKitDriver.__init__):
2111         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2112         (MainTest.test_crash_log):
2113         (MainTest.test_web_process_crash_log):
2114
2115 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2116
2117         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
2118         https://bugs.webkit.org/show_bug.cgi?id=81603
2119
2120         Reviewed by Adam Barth.
2121
2122         Add a new _get_crash_log() overridable method on the Port
2123         object to customize how we fetch crash logs per port.
2124         Mac will now slow-spin with a timeout waiting for the crash log, rather
2125         than waiting for the ReportCrash process to exit (it appears
2126         that the mac will manage multiple crashes with a single
2127         ReportCrash process, the process waits around longer than
2128         necessary, presumably to avoid thrashing if processes are
2129         repeatedly crashing).
2130
2131         Also, add the DriverOutput should contain the crash log and other info,
2132         which is created in a port-specific manner but can then be
2133         treated generically. Previously single_test_runner would get
2134         told that something crashed and attempt to do something to get
2135         the crash log, but it didn't have the information it needed to
2136         od the right thing; better to make the driver hand back the
2137         right info.
2138
2139         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2140         (SingleTestRunner._handle_error):
2141         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2142         (write_test_result):
2143         (TestResultWriter.write_crash_log):
2144         * Scripts/webkitpy/layout_tests/port/base.py:
2145         (Port.diff_text):
2146         (Port._get_crash_log):
2147         * Scripts/webkitpy/layout_tests/port/chromium.py:
2148         (ChromiumDriver.run_test):
2149         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2150         (ChromiumDriverTest.test_crash_log):
2151         * Scripts/webkitpy/layout_tests/port/driver.py:
2152         (DriverOutput.__init__):
2153         * Scripts/webkitpy/layout_tests/port/mac.py:
2154         (MacPort.is_lion):
2155         (MacPort._get_crash_log):
2156         * Scripts/webkitpy/layout_tests/port/test.py:
2157         (TestDriver.run_test):
2158         * Scripts/webkitpy/layout_tests/port/webkit.py:
2159         (WebKitDriver.__init__):
2160         (WebKitDriver._start):
2161         (WebKitDriver.has_crashed):
2162         (WebKitDriver._check_for_driver_crash):
2163         (WebKitDriver.run_test):
2164
2165 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2166
2167         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
2168         https://bugs.webkit.org/show_bug.cgi?id=81507
2169
2170         Now that we're only loading from one directory, no need to
2171         iterate over a list of directories anymore.
2172
2173         Reviewed by Martin Robinson.
2174
2175         * DumpRenderTree/efl/CMakeLists.txt:
2176         * DumpRenderTree/efl/FontManagement.cpp:
2177         (addFontDirectory):
2178         (addFontsToEnvironment):
2179
2180 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2181
2182         webkitpy: get ServerProcess out of the reportcrash business
2183         https://bugs.webkit.org/show_bug.cgi?id=81600
2184
2185         Unreviewed, build fix.
2186
2187         Re-land r111307 and r111293 with another fix for a crash in NRWT;
2188         we need to check if the driver has crashed before attempting to
2189         read from it.
2190
2191
2192         * Scripts/webkitpy/layout_tests/port/server_process.py:
2193         (ServerProcess._reset):
2194         (ServerProcess._handle_possible_interrupt):
2195         (ServerProcess.write):
2196         (ServerProcess.read_stdout):
2197         (ServerProcess.has_crashed):
2198         (ServerProcess._read):
2199         (ServerProcess.stop):
2200         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2201         (TrivialMockPort.check_for_leaks):
2202         (TestServerProcess.test_broken_pipe):
2203         * Scripts/webkitpy/layout_tests/port/webkit.py:
2204         (WebKitPort._read_image_diff):
2205         (WebKitDriver.has_crashed):
2206         (WebKitDriver._check_for_driver_crash):
2207         (WebKitDriver.run_test):
2208         (WebKitDriver._read_block):
2209         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2210         (MockServerProcess.__init__):
2211         (MockServerProcess):
2212         (MockServerProcess.has_crashed):
2213
2214 2012-03-21  Zeno Albisser  <zeno@webkit.org>
2215
2216         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
2217         https://bugs.webkit.org/show_bug.cgi?id=81750
2218
2219         Building WebCore on mac currently creates an archive that is bigger than
2220         4GB. But ranlib fails to create a symbol table for such a big archive,
2221         even on 64bit machines.
2222         Therefore we start using SVGAllInOne.cpp when building debug
2223         on mac. This reduces the size of the archive by about 300MB.
2224
2225         Reviewed by Tor Arne Vestbø.
2226
2227         * qmake/mkspecs/features/mac/default_pre.prf:
2228
2229 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
2230
2231         Updated the address list of committers.py.
2232
2233         Reviewed by Csaba Osztrogonác.
2234
2235         * Scripts/webkitpy/common/config/committers.py:
2236
2237 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
2238
2239         Updating my nick in committers file.
2240
2241         Unreviewed.
2242
2243         * Scripts/webkitpy/common/config/committers.py:
2244
2245 2012-03-21  Peter Beverloo  <peter@chromium.org>
2246
2247         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
2248         https://bugs.webkit.org/show_bug.cgi?id=80861
2249
2250         Reviewed by Tony Chang.
2251
2252         Remove most Android-specific logic from webkitdirs.pm in favor of
2253         handling this during project generation time.
2254
2255         * Scripts/webkitdirs.pm:
2256         (buildChromiumMakefile):
2257
2258 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2259
2260         [GTK] Allow running javascript from location bar in MiniBrowser
2261         https://bugs.webkit.org/show_bug.cgi?id=81331
2262
2263         Reviewed by Martin Robinson.
2264
2265         * MiniBrowser/gtk/BrowserWindow.c:
2266         (activateUriEntryCallback): Use browser_window_load_uri().
2267         (browser_window_load_uri): Check whether uri starts with
2268         javascript: and run the given script, or just call
2269         webkit_web_view_load_uri() to load the given uri.
2270         * MiniBrowser/gtk/BrowserWindow.h:
2271         * MiniBrowser/gtk/main.c:
2272         (createBrowserWindow): Use browser_window_load_uri().
2273
2274 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2275
2276         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
2277         https://bugs.webkit.org/show_bug.cgi?id=75543
2278
2279         Reviewed by Martin Robinson.
2280
2281         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
2282         MiniBrowser CPP flags.
2283
2284 2012-03-20  Eric Seidel  <eric@webkit.org>
2285
2286         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
2287         https://bugs.webkit.org/show_bug.cgi?id=80911
2288
2289         Reviewed by Adam Barth.
2290
2291         Update include paths for Source/WTF.
2292
2293         * DumpRenderTree/efl/CMakeLists.txt:
2294         * DumpRenderTree/qt/DumpRenderTree.pro:
2295         * QtTestBrowser/QtTestBrowser.pro:
2296         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
2297
2298 2012-03-20  Adam Barth  <abarth@webkit.org>
2299
2300         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
2301         https://bugs.webkit.org/show_bug.cgi?id=81567
2302
2303         Reviewed by Eric Seidel.
2304
2305         This temporary measure should help us avoid disrupting the folks
2306         working on WebAudio when we move the webaudio directory to Modules.
2307
2308         * Scripts/VCSUtils.pm:
2309         (adjustPathForRecentRenamings):
2310         (parseGitDiffHeader):
2311         (parseSvnDiffHeader):
2312
2313 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2314
2315         Unreviewed, rolling out r111453.
2316         http://trac.webkit.org/changeset/111453
2317         https://bugs.webkit.org/show_bug.cgi?id=81600
2318
2319         still broken
2320
2321         * Scripts/webkitpy/layout_tests/port/server_process.py:
2322         (ServerProcess._reset):
2323         (ServerProcess.handle_interrupt):
2324         (ServerProcess.write):
2325         (ServerProcess._check_for_crash):
2326         (ServerProcess._handle_timeout):
2327         (ServerProcess._check_for_abort):
2328         (ServerProcess._read):
2329         (ServerProcess.stop):
2330         (ServerProcess):
2331         (ServerProcess.set_crashed):
2332         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2333         (TrivialMockPort.is_crash_reporter):
2334         (TestServerProcess.test_broken_pipe):
2335         * Scripts/webkitpy/layout_tests/port/webkit.py:
2336         (WebKitPort._read_image_diff):
2337         (WebKitDriver.has_crashed):
2338         (WebKitDriver._check_for_driver_crash):
2339         (WebKitDriver):
2340         (WebKitDriver._detected_crash):
2341         (WebKitDriver._subprocess_crashed):
2342         (WebKitDriver._crashed_process_name):
2343         (WebKitDriver.run_test):
2344         (WebKitDriver._read_block):
2345         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2346         (MockServerProcess.__init__):
2347
2348 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2349
2350         webkitpy: get ServerProcess out of the reportcrash business
2351         https://bugs.webkit.org/show_bug.cgi?id=81600
2352
2353         Unreviewed, build fix.
2354
2355         Re-land r111307 and r111293 with fix for crash in NRWT; I
2356         was failing to set the crashing process name.
2357
2358         * Scripts/webkitpy/layout_tests/port/server_process.py:
2359         (ServerProcess._reset):
2360         (ServerProcess._handle_possible_interrupt):
2361         (ServerProcess.write):
2362         (ServerProcess.read_stdout):
2363         (ServerProcess.has_crashed):
2364         (ServerProcess._read):
2365         (ServerProcess.stop):
2366         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2367         (TrivialMockPort.check_for_leaks):
2368         (TestServerProcess.test_broken_pipe):
2369         * Scripts/webkitpy/layout_tests/port/webkit.py:
2370         (WebKitPort._read_image_diff):
2371         (WebKitDriver.has_crashed):
2372         (WebKitDriver._check_for_driver_crash):
2373         (WebKitDriver.run_test):
2374         (WebKitDriver._read_block):
2375         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2376         (MockServerProcess.__init__):
2377         (MockServerProcess):
2378         (MockServerProcess.has_crashed):
2379
2380 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2381
2382         [mac] Restore color space switching code to run-webkit-tests
2383         https://bugs.webkit.org/show_bug.cgi?id=80571
2384         <rdar://problem/11008529>
2385
2386         Reviewed by Simon Fraser.
2387
2388         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
2389         
2390         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
2391         didn't exist until Lion.
2392
2393         * DumpRenderTree/mac/LayoutTestHelper.m:
2394         (installLayoutTestColorProfile):
2395         (restoreUserColorProfile):
2396
2397 2012-03-20  Adele Peterson  <adele@apple.com>
2398
2399         Update the last test to use EXPECT_WK_STREQ.
2400
2401         Reviewed by Dan Bernstein.
2402
2403         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
2404
2405 2012-03-20  Adele Peterson  <adele@apple.com>
2406
2407         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
2408         https://bugs.webkit.org/show_bug.cgi?id=81630
2409         <rdar://problem/10650660>
2410
2411         Reviewed by Alexey Proskuryakov.
2412
2413         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2414         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
2415         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
2416         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
2417         (TestWebKitAPI):
2418         (TestWebKitAPI::didFinishLoadForFrame):
2419         (TestWebKitAPI::TEST):
2420         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
2421
2422 2012-03-20  Eric Seidel  <eric@webkit.org>
2423
2424         Add link on queues.webkit.org to kov's fancy queue-depth graph
2425         https://bugs.webkit.org/show_bug.cgi?id=81690
2426
2427         Reviewed by Adam Barth.
2428
2429         * QueueStatusServer/templates/recentstatus.html:
2430
2431 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2432
2433         [mac] Restore color space switching code to run-webkit-tests
2434         https://bugs.webkit.org/show_bug.cgi?id=80571
2435         <rdar://problem/11008529>
2436
2437         Reviewed by Simon Fraser.
2438
2439         Make use of the layout test helper hooks to restore the Mac system-wide
2440         color profile switching code. This makes WkTR and DRT results match, and
2441         causes the main display's color profile to not affect results on Lion.
2442
2443         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2444         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
2445         (installLayoutTestColorProfile):
2446         (restoreUserColorProfile):
2447         (simpleSignalHandler):
2448         (main):
2449         * Scripts/webkitpy/layout_tests/port/mac.py:
2450         (MacPort.release_http_lock):
2451         (MacPort):
2452         (MacPort._path_to_helper):
2453         (MacPort.start_helper):
2454         (MacPort.stop_helper):
2455         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
2456         (WTR::PlatformWebView::PlatformWebView):
2457         (WTR::PlatformWebView::windowSnapshotImage):
2458
2459 2012-02-24  Darin Fisher  <darin@chromium.org>
2460
2461         Add more Chromium WebKit API reviewers
2462         https://bugs.webkit.org/show_bug.cgi?id=81576
2463
2464         Reviewed by Adam Barth and David Levin.
2465
2466         Simplify watchlist to lump all WebKit API into one.
2467
2468         * Scripts/webkitpy/common/config/watchlist:
2469
2470 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2471
2472         Convert hasSpellingMarker to use Internals interface.
2473         https://bugs.webkit.org/show_bug.cgi?id=81300
2474
2475         Reviewed by Ryosuke Niwa.
2476
2477         Remove hasSpellingMarker functions, because it is able to work in the
2478         cross-port way through the Internals interface.
2479
2480         * DumpRenderTree/LayoutTestController.cpp:
2481         (LayoutTestController::staticFunctions):
2482         * DumpRenderTree/LayoutTestController.h:
2483         (LayoutTestController):
2484         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2485         * DumpRenderTree/chromium/LayoutTestController.cpp:
2486         (LayoutTestController::LayoutTestController):
2487         * DumpRenderTree/chromium/LayoutTestController.h:
2488         (LayoutTestController):
2489         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2490         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2491         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2492         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2493         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2494         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2495         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2496
2497 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2498
2499         Add new Qt-WK2 performance tester bot to build.webkit.org
2500         https://bugs.webkit.org/show_bug.cgi?id=80842
2501
2502         Reviewed by Ryosuke Niwa.
2503
2504         * BuildSlaveSupport/build.webkit.org-config/config.json:
2505         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2506         (RunAndUploadPerfTestsWebKit2):
2507         (RunAndUploadPerfTestsWebKit2.start):
2508         (BuildAndPerfTestWebKit2Factory):
2509         (BuildAndPerfTestWebKit2Factory.__init__):
2510         (DownloadAndPerfTestWebKit2Factory):
2511         (DownloadAndPerfTestWebKit2Factory.__init__):
2512
2513 2012-03-20  Peter Beverloo  <peter@chromium.org>
2514
2515         [Chromium] Update to Android NDK r7b
2516         https://bugs.webkit.org/show_bug.cgi?id=81005
2517
2518         Reviewed by Adam Barth.
2519
2520         Update to version r7b of the Android NDK, release notes of which are
2521         available on the following page:
2522         http://developer.android.com/sdk/ndk/index.html
2523
2524         * Scripts/update-webkit-chromium:
2525
2526 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2527
2528         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
2529         https://bugs.webkit.org/show_bug.cgi?id=81502
2530
2531         Unreviewed.
2532
2533         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
2534
2535 2012-03-19  Jessie Berlin  <jberlin@apple.com>
2536
2537         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
2538         because they broke running the layout tests on mac.
2539
2540         * Scripts/webkitpy/layout_tests/port/webkit.py:
2541         (WebKitDriver._check_for_driver_crash):
2542
2543 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
2544
2545         [GTK] Allow running run-gtk-tests during 'make distcheck'
2546         https://bugs.webkit.org/show_bug.cgi?id=81415
2547
2548         Reviewed by Philippe Normand.
2549
2550         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
2551
2552         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
2553         main command run during 'make check.'
2554         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
2555         a requirement to run it. If jhbuild isn't around it will simply not start the
2556         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
2557         main loop and simply use the main loop in a synchronous method to start SPI
2558         daemons. Did a bunch of general cleanup around these tasks.
2559
2560 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2561
2562         jhbuild: Use the multiprocessing module to get the number of CPUs.
2563         https://bugs.webkit.org/show_bug.cgi?id=81585
2564
2565         Reviewed by Martin Robinson.
2566
2567         Now that support for Python 2.5 has been dropped, we can safely
2568         use multiprocessing.cpu_count() to obtain the current number of
2569         CPUs in the system instead of running a webkitperl script for
2570         that.
2571
2572         * efl/common.py:
2573         (top_level_path):
2574         * efl/jhbuildrc:
2575         * gtk/common.py:
2576         (build_path):
2577         * gtk/jhbuildrc:
2578
2579 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2580
2581         We should only add a live event listener once, not once per TestSelector.
2582         Adding one per TestSelector would cause accordion('option', 'active') to
2583         incorrectly be "false" if you examine multiple sets of tests.
2584
2585         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2586
2587 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2588
2589         Fix typo introduced in r111293.
2590
2591         Unreviewed, build fix.
2592
2593         * Scripts/webkitpy/layout_tests/port/webkit.py:
2594         (WebKitDriver._check_for_driver_crash):
2595
2596 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2597
2598         webkitpy: get ServerProcess out of the reportcrash business
2599         https://bugs.webkit.org/show_bug.cgi?id=81600
2600
2601         Reviewed by Adam Barth.
2602
2603         ServerProcess was half-aware that ReportCrash might run
2604         sometimes, and that the process ServerProcess was talking to
2605         might have its own crashing subprocesses; neither of these
2606         things really worked right and it made the logic convoluted, so
2607         this change makes handling crashes completely separate from the
2608         server_process code, so that it can focus on just I/O to the
2609         subprocess.
2610
2611         There should be no functional changes resulting from this patch.
2612
2613         * Scripts/webkitpy/layout_tests/port/server_process.py:
2614         (ServerProcess._reset):
2615         (ServerProcess._handle_possible_interrupt):
2616         (ServerProcess.write):
2617         (ServerProcess.read_stdout):
2618         (ServerProcess.has_crashed):
2619         (ServerProcess._read):
2620         (ServerProcess.stop):
2621         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2622         (TrivialMockPort.check_for_leaks):
2623         (TestServerProcess.test_broken_pipe):
2624         * Scripts/webkitpy/layout_tests/port/webkit.py:
2625         (WebKitPort._read_image_diff):
2626         (WebKitDriver.has_crashed):
2627         (WebKitDriver._check_for_driver_crash):
2628         (WebKitDriver.run_test):
2629         (WebKitDriver._read_block):
2630         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2631         (MockServerProcess.__init__):
2632         (MockServerProcess):
2633         (MockServerProcess.has_crashed):
2634
2635 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2636
2637         webkitpy: crashlog parsing is broken
2638         https://bugs.webkit.org/show_bug.cgi?id=81586
2639
2640         Reviewed by Adam Barth.
2641
2642         Seeking on a file opened through codecs() appears to not work
2643         correctly (at least on SL).
2644         
2645         The code was more complicated than it needed to be, so I have
2646         revamped it to just read a full crashlog at once and then look
2647         for matches. 
2648         
2649         I have also added the ability to optionally propagate
2650         errors back (which can be helpful to debug races when ReportCrash is
2651         still running and you need to see that you couldn't open some
2652         files, rather than just ignoring them).
2653
2654         * Scripts/webkitpy/common/system/crashlogs.py:
2655         (CrashLogs.find_newest_log):
2656         (CrashLogs._find_newest_log_darwin):
2657         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2658         (CrashLogsTest.test_find_log_darwin):
2659
2660 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2661
2662         webkitpy: clean up some port._filesystem references to not reference private members
2663         https://bugs.webkit.org/show_bug.cgi?id=81595
2664
2665         Reviewed by Eric Seidel.
2666
2667         There should be no functional changes in this patch, just some
2668         refactoring. Also, fixes a minor nit in write_image_diff_files(()
2669
2670         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2671         (SingleTestRunner.__init__):
2672         (SingleTestRunner._run_compare_test):
2673         (SingleTestRunner._run_rebaseline):
2674         (SingleTestRunner._save_baseline_data):
2675         (SingleTestRunner._handle_error):
2676         (SingleTestRunner._run_reftest):
2677         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2678         (write_test_result):
2679         (TestResultWriter.__init__):
2680         (TestResultWriter._make_output_directory):
2681         (TestResultWriter.output_filename):
2682         (TestResultWriter._output_testname):
2683         (TestResultWriter.write_output_files):
2684         (TestResultWriter.write_stderr):
2685         (TestResultWriter.write_crash_report):
2686         (TestResultWriter.create_text_diff_and_write_result):
2687         (TestResultWriter.write_image_diff_files):
2688         (copy_file):
2689         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
2690         (TestResultWriterTest.test_reftest_diff_image):
2691
2692 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2693
2694         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
2695         https://bugs.webkit.org/show_bug.cgi?id=81598
2696
2697         Reviewed by Eric Seidel.
2698
2699         "DumpRenderTree" shouldn't be hard-coded here, since some other
2700         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
2701         
2702         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2703         (TestExpectations):
2704
2705 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2706
2707         webkitpy: allow instance data in TestFailure objects
2708         https://bugs.webkit.org/show_bug.cgi?id=81580
2709
2710         Reviewed by Tony Chang.
2711
2712         Current TestFailures are basically enums, but it would be nice,
2713         especially for crashes, if they contained the name and pid of
2714         the process that crashed.
2715
2716         * Scripts/webkitpy/layout_tests/models/test_failures.py:
2717         (TestFailure.message):
2718         (FailureTimeout.message):
2719         (FailureCrash):
2720         (FailureCrash.__init__):
2721         (FailureCrash.message):
2722         (FailureMissingResult.message):
2723         (FailureTextMismatch.message):
2724         (FailureMissingImageHash.message):
2725         (FailureMissingImage.message):
2726         (FailureImageHashMismatch.message):
2727         (FailureImageHashIncorrect.message):
2728         (FailureReftestMismatch.message):
2729         (FailureReftestMismatchDidNotOccur.message):
2730         (FailureReftestNoImagesGenerated.message):
2731         (FailureMissingAudio.message):
2732         (FailureAudioMismatch.message):
2733         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
2734         (TestFailuresTest.test_equals):
2735         (TestFailuresTest):
2736         (TestFailuresTest.test_crashes):
2737
2738 2012-03-19  Brady Eidson  <beidson@apple.com>
2739
2740         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
2741         Crash in 3rd party WebKit apps under XHR/Cache code
2742
2743         Reviewed by Antti Koivisto.
2744
2745         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2746         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
2747         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
2748         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
2749         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
2750         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
2751         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
2752         (TestWebKitAPI):
2753         (TestWebKitAPI::TEST):
2754
2755 2012-03-19  Brady Eidson  <beidson@apple.com>
2756
2757         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
2758         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
2759
2760         Reviewed by Enrica Casucci.
2761
2762         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
2763         are in the correct format.
2764
2765         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2766         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
2767         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
2768         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
2769         (TestWebKitAPI::contextMenuCopyLink):
2770         (TestWebKitAPI::TEST):
2771
2772 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2773
2774         Make the selected item in the accordion a link to the flakiness dashboard
2775         https://bugs.webkit.org/show_bug.cgi?id=81571
2776
2777         Reviewed by Dimitri Glazkov.
2778
2779         Also, make it so that you can select the text of the link, instead of the current behavior
2780         of acting as if you had clicked on the accordion header.
2781
2782         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2783         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2784
2785 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2786
2787         NRWT runs some tests that are skipped with -i command line option
2788         https://bugs.webkit.org/show_bug.cgi?id=81535
2789
2790         Reviewed by Ojan Vafai.
2791
2792         This change modifies the interaction of Skipped files and
2793         test_expectations files so that entries in Skipped files (and
2794         the -i command line) override *everything* in the expectations
2795         file.
2796
2797         Specifically, a directory in a Skipped file will cause all of
2798         the tests in the dir to be skipped even if individual tests in
2799         the dir are listed in the test_expectations.txt.
2800
2801         Skipped files also override anything in an overrides files.
2802
2803         This seems to make more intuitive sense, since if you list
2804         something in the Skips file (and even more specify it on the
2805         command line) you probably want it to be universally applied.
2806
2807         Theoretically we could add more precedence levels and have full
2808         paths in an expectations file override dirs in a Skipped file
2809         (but not the command line), but I don't know yet that that level
2810         of complexity is justified.
2811         
2812         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2813         (TestExpectations.__init__):
2814         (TestExpectations._add_skipped_tests):
2815         (TestExpectations._add_test):
2816         (TestExpectations._add_expectations):
2817         (TestExpectations._add_expectation_line):
2818         (TestExpectations._already_seen_better_match):
2819         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2820         (test_more_specific_override_resets_skip):
2821         (SkippedTests):
2822         (SkippedTests.get_exp):
2823         (SkippedTests.assert_exp):
2824         (SkippedTests.test_skipped_tests_work):
2825         (SkippedTests.test_duplicate_skipped_test_fails_lint):
2826         (SkippedTests.test_skipped_file_overrides_expectations):
2827         (SkippedTests.test_skipped_dir_overrides_expectations):
2828         (SkippedTests.test_skipped_file_overrides_overrides):
2829         (SkippedTests.test_skipped_dir_overrides_overrides):
2830
2831 2012-03-19  Eric Seidel  <eric@webkit.org>
2832
2833         Fix WTF header include discipline in Chromium WebKit
2834         https://bugs.webkit.org/show_bug.cgi?id=81281
2835
2836         Reviewed by James Robinson.
2837
2838         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2839         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
2840
2841 2012-03-19  Andy Chen  <andchen@rim.com>
2842
2843         [BlackBerry] Add new layout test for text match marker
2844         https://bugs.webkit.org/show_bug.cgi?id=81531
2845
2846         Make the LayoutTestControllerBlackBerry::findString to test the
2847         BlackBerry port implementation.
2848
2849         Reviewed by Rob Buis.
2850
2851         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2852         (LayoutTestController::findString):
2853
2854 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2855
2856         Don't show WONTFIX tests in garden-o-matic
2857         https://bugs.webkit.org/show_bug.cgi?id=81554
2858
2859         Reviewed by Dimitri Glazkov.
2860
2861         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2862         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2863
2864 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
2865
2866         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
2867         https://bugs.webkit.org/show_bug.cgi?id=81530
2868
2869         Reviewed by Jessie Berlin.
2870
2871         Add rowAtIndex to WebKitTestRunner.
2872
2873         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
2874         (WTR::AccessibilityUIElement::rowAtIndex):
2875         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
2876         (AccessibilityUIElement):
2877         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
2878         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2879         (WTR):
2880         (WTR::AccessibilityUIElement::rowAtIndex):
2881
2882 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2883
2884         Change garden-o-matic All Failures tab to Expected Failures
2885         https://bugs.webkit.org/show_bug.cgi?id=81540
2886
2887         Reviewed by Dimitri Glazkov.
2888
2889         Also, fix some broken unittests that had just gotten stale.
2890
2891         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
2892         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
2893         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
2894         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2895         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2896         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2897         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2898         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
2899         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
2900         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2901
2902 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
2903
2904         [chromium] DRT crashes at shutdown.
2905         https://bugs.webkit.org/show_bug.cgi?id=81504
2906
2907         Reviewed by James Robinson.
2908
2909         WebCompositor must be destroyed after WebView is properly destroyed,
2910         or some compositor objects like CCLayerTreeHost may not be destroyed.
2911         However, it is really hard to make the destruction order correct
2912         explicitly in ~TestShell. So move it out to DumpRenderTree.
2913
2914         * DumpRenderTree/chromium/DumpRenderTree.cpp:
2915         (main):
2916         * DumpRenderTree/chromium/TestShell.cpp:
2917         (TestShell::~TestShell):
2918
2919 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2920
2921         add dpranke to webkitpy watchlist
2922         https://bugs.webkit.org/show_bug.cgi?id=81433
2923
2924         Reviewed by Benjamin Poulain.
2925
2926         * Scripts/webkitpy/common/config/watchlist:
2927
2928 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2929
2930         test-webkitpy is failing on SL/Python 2.6.1
2931         https://bugs.webkit.org/show_bug.cgi?id=81533
2932
2933         Reviewed by Ojan Vafai.
2934
2935         It appears that itertools.combinations() is buggy in 2.6.1; this
2936         backports the implementation from 2.7.
2937
2938         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2939         (TestConfigurationConverter.collapse_macros):
2940         (TestConfigurationConverter):
2941         (TestConfigurationConverter.combinations):
2942
2943 2012-03-19  David Michael Barr  <davidbarr@google.com>
2944
2945         Optimize the cleanup sequence in the start-queue.sh script
2946         https://github.com/abarth/webkit/pull/1
2947
2948         Reviewed by Adam Barth.
2949
2950         Updating the working copy is an expensive operation
2951         for a repository the size of WebKit.
2952         Attempt to reset only to the target revision.
2953
2954         * EWSTools/start-queue.sh:
2955
2956 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
2957
2958         Adjust architecture type of the EFL build bot.
2959         https://bugs.webkit.org/show_bug.cgi?id=81452
2960
2961         Reviewed by Tony Chang.
2962
2963         * BuildSlaveSupport/build.webkit.org-config/config.json: The
2964         buildslave is a 64-bit machine, therefore describe it as being
2965         x86_64, not i386.
2966
2967 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2968
2969         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
2970         https://bugs.webkit.org/show_bug.cgi?id=81514
2971
2972         Reviewed by Simon Hausmann.
2973
2974         * MiniBrowser/qt/qml/BrowserWindow.qml:
2975
2976 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2977
2978         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
2979         https://bugs.webkit.org/show_bug.cgi?id=81491
2980
2981         Reviewed by Philippe Normand.
2982
2983         For the find_path(), find_library() etc CMake calls to find the
2984         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
2985         set, otherwise CMake cannot know the dependencies exist and will only
2986         look in the standard directories.
2987
2988         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
2989         the jhbuild Root directory.
2990
2991 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2992
2993         [EFL] Add XDG env vars to jhbuildrc
2994         https://bugs.webkit.org/show_bug.cgi?id=81510
2995
2996         Needed to ensure correct glib behaviour in jhbuild environment.
2997
2998         Reviewed by Philippe Normand.
2999
3000         * efl/jhbuildrc:
3001
3002 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
3003
3004         Select best target for tap gesture.
3005         https://bugs.webkit.org/show_bug.cgi?id=78801
3006
3007         Reviewed by Kenneth Rohde Christiansen.
3008
3009         Add TOUCH_ADJUSTMENT to enabled features.
3010
3011         * qmake/mkspecs/features/features.prf:
3012
3013 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
3014
3015         [chromium] synthesize wheel events for fling on main thread
3016         https://bugs.webkit.org/show_bug.cgi?id=81458
3017
3018         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
3019
3020         Reviewed by James Robinson.
3021
3022         * DumpRenderTree/chromium/EventSender.cpp:
3023         (EventSender::EventSender):
3024         (EventSender::gestureFlingCancel):
3025         (EventSender::gestureFlingStart):
3026         * DumpRenderTree/chromium/EventSender.h:
3027         (EventSender):
3028
3029 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
3030
3031         [Qt] Add command line option to set the user agent in the MiniBrowser
3032         https://bugs.webkit.org/show_bug.cgi?id=81508
3033
3034         Reviewed by Simon Hausmann.
3035
3036         * MiniBrowser/qt/BrowserWindow.cpp:
3037         (BrowserWindow::BrowserWindow):
3038         (BrowserWindow::webViewExperimental):
3039         * MiniBrowser/qt/BrowserWindow.h:
3040         (BrowserWindow):
3041         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3042         (printHelp):
3043         (MiniBrowserApplication::handleUserOptions):
3044         * MiniBrowser/qt/MiniBrowserApplication.h:
3045         (WindowOptions::WindowOptions):
3046         (WindowOptions::setUserAgent):
3047         (WindowOptions::userAgent):
3048         (WindowOptions):
3049
3050 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
3051
3052         [Qt] Remove dead user agent code from the MiniBrowser.
3053
3054         Reviewed by Simon Hausmann.
3055
3056         * MiniBrowser/qt/BrowserWindow.cpp:
3057         * MiniBrowser/qt/BrowserWindow.h:
3058         (BrowserWindow):
3059         * MiniBrowser/qt/MiniBrowser.qrc:
3060         * MiniBrowser/qt/useragentlist.txt: Removed.
3061
3062 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3063
3064         jhbuild: Call jhbuild's autogen.sh via bash.
3065         https://bugs.webkit.org/show_bug.cgi?id=81485
3066
3067         Reviewed by Philippe Normand.
3068
3069         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
3070         this is usually true only on Linux systems, and even then bash for
3071         some reason might not be available in that location. It is safer to
3072         call bash directly and pass autogen.sh to it instead.
3073
3074         * jhbuild/jhbuild-wrapper:
3075         (install_jhbuild):
3076
3077 2012-03-19  Philippe Normand  <pnormand@igalia.com>
3078
3079         [GTK] Prefix run-launcher with run-with-jhbuild
3080         https://bugs.webkit.org/show_bug.cgi?id=81503
3081
3082         Reviewed by Hajime Morita.
3083
3084         * Scripts/run-launcher:
3085
3086 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3087
3088         [EFL] Prefix run-launcher with run-with-jhbuild
3089         https://bugs.webkit.org/show_bug.cgi?id=80026
3090
3091         Reviewed by Hajime Morita.
3092
3093         * Scripts/run-launcher:
3094
3095 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
3096
3097         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
3098         https://bugs.webkit.org/show_bug.cgi?id=81498
3099
3100         * qmake/mkspecs/features/unix/default_post.prf:
3101
3102 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3103
3104         Unreviewed; switch to using my FreeBSD.org email address, the
3105         other one is going to disappear soon.
3106
3107         * Scripts/webkitpy/common/config/committers.py:
3108         * Scripts/webkitpy/common/config/watchlist:
3109
3110 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
3111
3112         Unreviewed; fix webkitpy unit test failures after r111116.
3113
3114         It turns out that cmd_line in the patch in bug 80025 had the right
3115         signature when it was submitted, but r109242 added a few more
3116         required parameters, and since the method (and EflDriver) did not
3117         exist at the time we ended up committing a broken method.
3118
3119         * Scripts/webkitpy/layout_tests/port/efl.py:
3120         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
3121         arguments.
3122
3123 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
3124
3125         [BlackBerry] Enable Web Timing for performance profiling and improvement.
3126         https://bugs.webkit.org/show_bug.cgi?id=81085
3127
3128         Reviewed by Rob Buis.
3129
3130         * Scripts/build-webkit:
3131
3132 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3133
3134         Add a dependencies installation step for EFL to buildbot master.cfg
3135         https://bugs.webkit.org/show_bug.cgi?id=81337
3136
3137         Reviewed by Gustavo Noronha Silva.
3138
3139         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3140         (InstallEflDependencies):
3141         (Factory.__init__):
3142
3143 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3144
3145         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
3146         https://bugs.webkit.org/show_bug.cgi?id=80025
3147
3148         Reviewed by Gustavo Noronha Silva.
3149
3150         * Scripts/webkitpy/layout_tests/port/efl.py:
3151         (EflDriver):
3152         (EflDriver.cmd_line):
3153         (EflPort._driver_class):
3154
3155 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3156
3157         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
3158         https://bugs.webkit.org/show_bug.cgi?id=79904
3159
3160         build-webkit --efl will use jhbuild for fetching and building dependencies.
3161
3162         Reviewed by Gustavo Noronha Silva.
3163
3164         * Scripts/update-webkitefl-libs: Added.
3165         * Scripts/webkitdirs.pm:
3166         (jhbuildWrapperPrefixIfNeeded):
3167         (generateBuildSystemFromCMakeProject):
3168         (buildCMakeGeneratedProject):
3169         * efl/common.py: Added.
3170         (script_path):
3171         (top_level_path):
3172         (number_of_cpus):
3173         * efl/jhbuild.modules: Added.
3174         * efl/jhbuildrc: Added.
3175         * efl/run-with-jhbuild: Added.
3176
3177 2012-03-16  Stephanie Lewis  <slewis@apple.com>
3178
3179         https://bugs.webkit.org/show_bug.cgi?id=81065
3180         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
3181         Fix crash preventing WebKitTestRunner from working in Release.  
3182         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
3183         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
3184         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
3185
3186         I removed ostringstream in favor of StringBuilder.  
3187
3188         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
3189
3190         Reviewed by Geoff Garen.
3191
3192         Add DEBUG_DEFINES to set NDEBUG on release builds.
3193         * WebKitTestRunner/Configurations/Base.xcconfig:
3194         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
3195
3196         Remove ostringstream member in favor of StringBuilder.
3197         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3198         (WTR::InjectedBundle::initialize):
3199         (WTR::InjectedBundle::beginTesting):
3200         (WTR::InjectedBundle::done):
3201         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3202         (WTF):
3203         (WTR::InjectedBundle::stringBuilder):
3204         (InjectedBundle):
3205
3206         Append output to StringBuilder.
3207         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
3208         (WTR::hasPrefix):
3209         (WTR::dumpPath):
3210         (WTR::rangeToStr):
3211         (WTR::styleDecToStr):
3212         (WTR::frameToStr):
3213         (WTR::dumpFrameScrollPosition):
3214         (WTR::dumpFrameText):
3215         (WTR::dumpDescendantFramesText):
3216         (WTR::InjectedBundlePage::dump):
3217         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
3218         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
3219         (WTR::InjectedBundlePage::willSendRequestForFrame):
3220         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
3221         (WTR::lastFileURLPathComponent):
3222         (WTR::InjectedBundlePage::willAddMessageToConsole):
3223         (WTR::InjectedBundlePage::willSetStatusbarText):
3224         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
3225         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
3226         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
3227         (WTR::InjectedBundlePage::shouldBeginEditing):
3228         (WTR::InjectedBundlePage::shouldEndEditing):
3229         (WTR::InjectedBundlePage::shouldInsertNode):
3230         (WTR::InjectedBundlePage::shouldInsertText):
3231         (WTR::InjectedBundlePage::shouldDeleteRange):
3232         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
3233         (WTR::InjectedBundlePage::shouldApplyStyle):
3234         (WTR::InjectedBundlePage::didBeginEditing):
3235         (WTR::InjectedBundlePage::didEndEditing):
3236         (WTR::InjectedBundlePage::didChange):
3237         (WTR::InjectedBundlePage::didChangeSelection):
3238         (WTR::InjectedBundlePage::supportsFullScreen):
3239         (WTR::InjectedBundlePage::enterFullScreenForElement):
3240         (WTR::InjectedBundlePage::exitFullScreenForElement):
3241         (WTR::InjectedBundlePage::beganEnterFullScreen):
3242         (WTR::InjectedBundlePage::beganExitFullScreen):
3243         (WTR::dumpBackForwardListItem):
3244         (WTR::InjectedBundlePage::dumpBackForwardList):
3245         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3246         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
3247         (WTR::LayoutTestController::dumpConfigurationForViewport):
3248         
3249         Convert between WTF::String and WKStringRef.
3250         * WebKitTestRunner/StringFunctions.h:
3251         (WTR::toWTFString):
3252         (WTR):
3253         * WebKitTestRunner/TestInvocation.cpp:
3254         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
3255
3256 2012-03-16  Eric Seidel  <eric@webkit.org>
3257
3258         Add a land-from-url command
3259         https://bugs.webkit.org/show_bug.cgi?id=81411
3260
3261         Reviewed by Adam Barth.
3262
3263         Very basic so far.  The goal is to make it landing smarter
3264         so that users don't have to bother to parse out the ids out of the urls.
3265
3266         * Scripts/webkitpy/common/config/urls.py:
3267         (parse_bug_id):
3268         (parse_attachment_id):
3269         * Scripts/webkitpy/common/config/urls_unittest.py:
3270         (URLsTest.test_parse_bug_id):
3271         (URLsTest):
3272         (URLsTest.test_parse_attachment_id):
3273         * Scripts/webkitpy/tool/commands/download.py:
3274         (ProcessURLsMixin):
3275         (ProcessURLsMixin._fetch_list_of_patches_to_process):
3276         (LandFromURL):
3277         * Scripts/webkitpy/tool/commands/download_unittest.py:
3278         (test_land_from_bug):
3279         (test_land_from_url):
3280
3281 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
3282
3283         build-webkit launcher instructions use wrong relative path for Qt
3284         https://bugs.webkit.org/show_bug.cgi?id=81357
3285
3286         Reviewed by Eric Seidel.
3287
3288         Ensure we are at the proper directory before generating run-launcher
3289         instructions.
3290
3291         * Scripts/build-webkit:
3292
3293 2012-03-16  Brady Eidson  <beidson@apple.com>
3294
3295         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
3296         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
3297
3298         Reviewed by Geoff Garen.
3299
3300         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3301         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
3302
3303 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3304
3305         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
3306         https://bugs.webkit.org/show_bug.cgi?id=81407
3307
3308         Reviewed by Adam Barth.
3309
3310         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3311         (summarize_results):
3312         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
3313         (ResultSummaryTest.summarized_results):
3314         (ResultSummaryTest.test_no_svn_revision):
3315         (ResultSummaryTest.test_svn_revision):
3316         (ResultSummaryTest):
3317         (ResultSummaryTest.test_summarized_results_wontfix):
3318
3319 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3320
3321         Have webkit-patch rebaseline-test update test_expectations.txt
3322         https://bugs.webkit.org/show_bug.cgi?id=81054
3323
3324         Recommit http://trac.webkit.org/changeset/110783 now that
3325         http://trac.webkit.org/changeset/111034 has landed.
3326
3327         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3328         (TestExpectations.remove_configuration_from_test):
3329         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3330         (RemoveConfigurationsTest):
3331         (RemoveConfigurationsTest.test_remove):
3332         (test_remove_line):
3333         * Scripts/webkitpy/tool/commands/rebaseline.py:
3334         (RebaselineTest._update_expectations_file):
3335         (RebaselineTest._rebaseline_test_and_update_expectations):
3336         (RebaselineTest):
3337         (RebaselineTest.execute):
3338         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3339         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3340         (test_rebaseline_updates_expectations_file):
3341
3342 2012-03-15  Ojan Vafai  <ojan@chromium.org>
3343
3344         Specifier collapsing when writing test expectations lines gets a number of cases wrong
3345         https://bugs.webkit.org/show_bug.cgi?id=81309
3346
3347         Reviewed by Dimitri Glazkov.
3348
3349         I've run this over all the lines in the current Chromium test_expectations.txt file,
3350         so I'm relatively confident we now cover all the cases.
3351
3352         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3353         (TestConfigurationConverter.__init__):
3354         (TestConfigurationConverter.collapse_macros):
3355         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
3356         (TestConfigurationConverter):
3357         (TestConfigurationConverter.intersect_combination):
3358         (TestConfigurationConverter.symmetric_difference):
3359         (TestConfigurationConverter.to_specifiers_list):
3360         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
3361         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
3362         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3363         (make_mock_all_test_configurations_set):
3364         (TestConfigurationConverterTest.test_symmetric_difference):
3365         (TestConfigurationConverterTest.test_to_config_set):
3366         (TestConfigurationConverterTest.test_macro_expansion):
3367         (TestConfigurationConverterTest.test_to_specifier_lists):
3368         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3369         * Scripts/webkitpy/layout_tests/port/chromium.py:
3370         (ChromiumPort):
3371         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3372         (ChromiumAndroidPort.__init__):
3373         The android port uses "arm" as it's architecture, which is technically correct,
3374         but considerably complicates making collapsing work. We probably should kill
3375         the concept of architecture entirely. The benefits are not worth the code
3376         complexity.
3377
3378         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3379         (ChromiumPortTest.test_all_test_configurations):
3380         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3381         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3382         (test_rebaseline_updates_expectations_file):
3383
3384 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
3385
3386         [Qt][Wk2] Assertion Failure and crash on file upload
3387         https://bugs.webkit.org/show_bug.cgi?id=80854
3388
3389         Reviewed by Simon Hausmann.
3390
3391         Added filePicker to WebView using experimental API.
3392
3393         * MiniBrowser/qt/MiniBrowser.qrc:
3394         * MiniBrowser/qt/icons/folder.png: Added.
3395         * MiniBrowser/qt/icons/titlebar.png: Added.
3396         * MiniBrowser/qt/icons/up.png: Added.
3397         * MiniBrowser/qt/qml/BrowserWindow.qml:
3398         * MiniBrowser/qt/qml/FilePicker.qml: Added.
3399
3400 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
3401
3402         Updating email for committer.py script.
3403
3404         * Scripts/webkitpy/common/config/committers.py:
3405
3406 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
3407
3408         [chromium] DRT: Add support for sending gesture scrolls without a begin
3409         https://bugs.webkit.org/show_bug.cgi?id=81292
3410
3411         Reviewed by Kent Tamura.
3412
3413         * DumpRenderTree/chromium/EventSender.cpp:
3414         (EventSender::EventSender):
3415         (EventSender::gestureScrollFirstPoint):
3416         * DumpRenderTree/chromium/EventSender.h:
3417         (EventSender):
3418
3419 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
3420
3421         [GTK] WebKitIconDatabase doesn't keep icons cached
3422         https://bugs.webkit.org/show_bug.cgi?id=56200
3423
3424         Reviewed by Martin Robinson.
3425
3426         Add support for favicons in GtkLauncher.
3427
3428         * GtkLauncher/main.c:
3429         (activateUriEntryCb):
3430         (iconLoadedCb):
3431         (createBrowser):
3432
3433 2012-03-16  Adam Barth  <abarth@webkit.org>
3434
3435         Remove sheriff-bot's last-green-revision command
3436         https://bugs.webkit.org/show_bug.cgi?id=81314
3437
3438         Reviewed by Eric Seidel.
3439
3440         This command is a cruel joke.  WebKit doesn't really ever have a
3441         last-green-revision.
3442
3443         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3444         (BuildBot._find_green_revision):
3445         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3446         (MockBuildBot.builder_statuses):
3447         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3448         (test_find_green_revision):
3449         * Scripts/webkitpy/tool/bot/irc_command.py:
3450         (IRCCommand.execute):
3451         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
3452         (SheriffIRCBotTest.test_help):
3453         * Scripts/webkitpy/tool/commands/queries.py:
3454         (PatchesToReview.execute):
3455
3456 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3457
3458         [Qt] Make gccdepends test aware of broken icecream deps generation
3459
3460         Icecream seems to build the files without passing -o to gcc with
3461         the original destination path, which means the target name in the
3462         dependency file is wrong.
3463
3464         Also, echo something to the final file instead of just touching it,
3465         so it's not deleted on every build.
3466
3467         Reviewed by Simon Hausmann.
3468
3469         * qmake/config.tests/gccdepends/gccdepends.pro:
3470
3471 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
3472
3473         Support for Battery Status API
3474         https://bugs.webkit.org/show_bug.cgi?id=62698
3475
3476         Add include directory and file.
3477         http://www.w3.org/TR/battery-status/
3478
3479         Reviewed by Adam Barth.
3480
3481         * Scripts/build-webkit:
3482
3483 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
3484
3485         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
3486         https://bugs.webkit.org/show_bug.cgi?id=81007
3487
3488         Reviewed by Adam Barth.
3489
3490         A dedidcated build slave is available for efl on build.webkit.org. 
3491         This change add's an entry for the elf port in list of available port builders.
3492         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
3493
3494         * Scripts/webkitpy/layout_tests/port/builders.py:
3495
3496 2012-03-16  Adam Barth  <abarth@webkit.org>
3497
3498         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
3499         https://bugs.webkit.org/show_bug.cgi?id=81302
3500
3501         Reviewed by Eric Seidel.
3502
3503         This is a trivial patch, but the variable name turned up in a grep of
3504         webkitpy and looks like copy pasta.
3505
3506         * Scripts/webkitpy/common/net/statusserver.py:
3507         (StatusServer.next_work_item):
3508
3509 2012-03-16  Andrew Lo  <anlo@rim.com>
3510
3511         [BlackBerry] Enable PAGE_VISIBILITY_API
3512         https://bugs.webkit.org/show_bug.cgi?id=81257
3513
3514         Enable Page Visibility API for BlackBerry.
3515
3516         Reviewed by Rob Buis.
3517
3518         * Scripts/build-webkit:
3519
3520 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
3521
3522         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
3523         https://bugs.webkit.org/show_bug.cgi?id=81232
3524
3525         Reviewed by Philippe Normand.
3526
3527         Use the newly-added helper method in DumpRenderTreeSupportGtk.
3528
3529         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3530         (LayoutTestController::computedStyleIncludingVisitedInfo):
3531
3532 2012-03-15  Eric Seidel  <eric@webkit.org>
3533
3534         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
3535         https://bugs.webkit.org/show_bug.cgi?id=81313
3536
3537         Reviewed by Adam Barth.
3538
3539         * Scripts/webkitpy/common/checkout/changelog.py:
3540         (parse_bug_id_from_changelog):
3541         * Scripts/webkitpy/common/config/urls.py:
3542         (parse_bug_id):
3543         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
3544         (URLsTest):
3545         (URLsTest.test_parse_bug_id):
3546         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
3547         (test_parse_bug_id):
3548         * Scripts/webkitpy/tool/bot/irc_command.py:
3549         (_post_error_and_check_for_bug_url):
3550         * Scripts/webkitpy/tool/bot/sheriff.py:
3551         (Sheriff.post_rollout_patch):
3552         (Sheriff.post_chromium_deps_roll):
3553
3554 2012-03-15  Adam Barth  <abarth@webkit.org>
3555
3556         The commit-queue should fast-track patches that have already passed the testing EWS bots
3557         https://bugs.webkit.org/show_bug.cgi?id=81305
3558
3559         Reviewed by Eric Seidel.
3560
3561         If a patch has already passed the testing EWS bots (currently only the
3562         chromium-ews bot), we should be able to land the patch without running
3563         the tests again.  This patch is somewhat risky in that the tests might
3564         have been broken between when the EWS bots ran and when the
3565         commit-queue is processing the patch.  If that turns out to be a
3566         problem in practice, we'll tighten up this fast-track in some way,
3567         likely by making sure that the pass status is recent.
3568
3569         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
3570         (CommitQueueTaskDelegate.did_pass_testing_ews):
3571         (CommitQueueTask._did_pass_tests_recently):
3572         (CommitQueueTask.run):
3573         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3574         (MockCommitQueue.did_pass_testing_ews):
3575         (test_fast_success_case):
3576         * Scripts/webkitpy/tool/commands/queues.py:
3577         (CommitQueue.did_pass_testing_ews):
3578         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3579         (test_did_pass_testing_ews):
3580
3581 2012-03-15  Kentaro Hara  <haraken@chromium.org>
3582
3583         [PerformanceTests] run-perf-tests should output correct units
3584         https://bugs.webkit.org/show_bug.cgi?id=78303
3585
3586         Reviewed by Ryosuke Niwa.
3587
3588         Previously run-perf-tests just supported "ms" for units.
3589         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
3590         This patch fixes run-perf-tests so that they can support custom units.
3591         The default unit is "ms".
3592
3593         The test result looks like this:
3594
3595         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
3596         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
3597         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
3598
3599         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3600         (PerfTestsRunner._process_parser_test_result):
3601
3602 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
3603
3604         [WinCairo] Unreviewed build change after wtf path changes.
3605
3606         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
3607         directories to look at the private header directory to find 'wtf'.
3608
3609 2012-03-15  Kent Tamura  <tkent@chromium.org>
3610
3611         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
3612         https://bugs.webkit.org/show_bug.cgi?id=81293
3613
3614         Reviewed by Adam Barth.
3615
3616         r110595 introduced new format of the author field of ChangeLog entry:
3617         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
3618         Updte chanelog.py to support it.
3619
3620         * Scripts/webkitpy/common/checkout/changelog.py:
3621         (ChangeLogEntry._split_contributor_names):
3622         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
3623         (test_parse_authors):
3624
3625 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
3626
3627         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
3628
3629         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
3630
3631         Reviewed by Jessie Berlin.
3632
3633         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3634         (LayoutTestController::setValueForUser):
3635         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3636         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3637         (WTR::LayoutTestController::setValueForUser):
3638         (WTR):
3639         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3640         (LayoutTestController):
3641
3642 2012-03-15  Adam Barth  <abarth@webkit.org>
3643
3644         Update the start-queue.sh script to match what we're using on the bots
3645         https://bugs.webkit.org/show_bug.cgi?id=81277
3646
3647         Reviewed by Eric Seidel.
3648
3649         Over time we've added more magic git commands to clean up the local repo after
3650         it has been borked in various ways. The idea behind this sequence is to clear
3651         out any changes in the working copy and then to rebuild the master branch from
3652         origin/master.
3653
3654         This patch also removes the watchdog timer that we've stopped using on the
3655         bots.
3656
3657         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
3658         Note: Eric and I are experimenting with GitHub to see if it might be
3659         useful for WebKit.  If you have thoughts on this topic, please feel
3660         free to grab Eric or me on #webkit to chat.  I assume that he and I are
3661         the only ones who care about this file so using this alternative
3662         process for this patch shouldn't cause any trouble.
3663
3664         * EWSTools/start-queue.sh:
3665
3666 2012-03-15  Adam Barth  <abarth@webkit.org>
3667
3668         The commit-queue can hang when the test_expectations.txt style check fails during commit
3669         https://bugs.webkit.org/show_bug.cgi?id=81251
3670
3671         Reviewed by Eric Seidel.
3672
3673         We forgot to check the --non-interactive