remove support for junit-style xml output from test-webkitpy
[WebKit-https.git] / Tools / ChangeLog
1 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
2
3         remove support for junit-style xml output from test-webkitpy
4         https://bugs.webkit.org/show_bug.cgi?id=82279
5
6         Reviewed by Eric Seidel.
7
8         This was added when we were looking into integrating w/ Jenkins
9         rather than buildbot, but I believe that project got shelved, so
10         this is unused. We can always add it back in later as necessary.
11
12         * Scripts/webkitpy/test/main.py:
13         (Tester._parse_args):
14         (Tester._run_tests):
15         * Scritps/webkitpy/thirdparty/__init__.py:
16
17 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
18
19         new-run-webkit-tests: crashes when it fails to decode a stack trace
20         https://bugs.webkit.org/show_bug.cgi?id=82673
21
22         Unreviewed, build fix.
23
24         We are assuming the stdout/stderr output from the driver is utf-8
25         encoded when we get stack traces; this may not be a valid
26         assumption generally, but if we do get strings that aren't valid
27         utf-8, we would crash. Now we will ignore any decoding errors.
28
29         * Scripts/webkitpy/layout_tests/port/base.py:
30         (Port._get_crash_log):
31         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
32         (PortTestCase.test_get_crash_log):
33
34 2012-03-29  Simon Fraser  <simon.fraser@apple.com>
35
36         Scripts fail to detect when a tool crashes
37         https://bugs.webkit.org/show_bug.cgi?id=82659
38
39         Reviewed by Mark Rowe.
40         
41         Have exitStatus() detect when the process fail to exit cleanly
42         (possibly because of a crash), and return a non-zero exit code
43         in that case.
44
45         * Scripts/VCSUtils.pm:
46         (exitStatus):
47
48 2012-03-29  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
49
50         [GTK][EFL] run-javascriptcore-tests should be run through jhbuild
51         https://bugs.webkit.org/show_bug.cgi?id=82581
52
53         Reviewed by Martin Robinson.
54
55         Running Javascriptcore tests through jhbuild
56         for consistency with run-webkit-tests and in order to
57         avoid confusing libraries when facing regressions.
58
59         * Scripts/run-javascriptcore-tests:
60
61 2012-03-29  Dirk Pranke  <dpranke@chromium.org>
62
63         test-webkitpy: add --timing
64         https://bugs.webkit.org/show_bug.cgi?id=82550
65
66         Reviewed by Eric Seidel.
67
68         This patch adds a --timing option that will display the time
69         each test takes. It also removes the --silent option, since
70         probably no one ever used it, and cleans up the logging
71         option parsing code to be easier to follow.
72
73         * Scripts/webkitpy/test/main.py:
74         (Tester._parse_args):
75         (Tester._configure):
76         * Scripts/webkitpy/test/runner.py:
77         (TestRunner.write_result):
78         * Scripts/webkitpy/test/runner_unittest.py:
79         (RunnerTest.test_regular):
80         (RunnerTest.test_verbose):
81         (RunnerTest):
82         (RunnerTest.test_timing):
83
84 2012-03-29  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
85
86         HashMap<>::add should return a more descriptive object
87         https://bugs.webkit.org/show_bug.cgi?id=71063
88
89         Reviewed by Ryosuke Niwa.
90
91         Update code to use AddResult instead of a pair.
92
93         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
94         (LayoutTestController::evaluateScriptInIsolatedWorld):
95         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
96         (LayoutTestController::evaluateScriptInIsolatedWorld):
97         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
98         (WTR::LayoutTestController::evaluateScriptInIsolatedWorld):
99
100 2012-03-29  Ilya Tikhonovsky  <loislo@chromium.org>
101
102         Web Inspector: chromium: DRT --no-timeout option doesn't work.
103         https://bugs.webkit.org/show_bug.cgi?id=82608
104
105         Initial value for m_timeout was initialized in constructor and was overwritten in DRT::main.
106         This was broken by r112354 and the sequence became opposite.
107
108         Reviewed by Yury Semikhatsky.
109
110         * DumpRenderTree/chromium/TestShell.cpp:
111         (TestShell::TestShell):
112         (TestShell::initialize):
113
114 2012-03-29  Vineet Chaudhary  <rgf748@motorola.com>
115
116         Remove custom bindings form Internals.idl of attribute type Array.
117         https://bugs.webkit.org/show_bug.cgi?id=82319
118
119         Reviewed by Kentaro Hara.
120
121         Remove JSInternalsCustom.cpp and V8InternalsCustom.cpp as no longer required.
122
123         * GNUmakefile.am:
124
125 2012-03-28  Mark Rowe  <mrowe@apple.com>
126
127         Fix --no-saved-state.
128
129         Ignore the flag before Lion, rather than making it specific to Lion.
130
131         Rubber-stamped by Dan Bernstein.
132
133         * Scripts/webkitdirs.pm:
134         (argumentsForRunAndDebugMacWebKitApp):
135
136 2012-03-28  Dave Tharp  <dtharp@codeaurora.org>
137
138         Add dtharp to contributors list
139         https://bugs.webkit.org/show_bug.cgi?id=82545
140
141         Reviewed by Adam Barth.
142
143         Adding dtharp as Contributor.
144
145         * Scripts/webkitpy/common/config/committers.py:
146
147 2012-03-28  Stuart Eichert  <stuart.eichert@palm.com>
148
149         build-webkit : the word "pass" is spelled incorrectly as "passs"
150         https://bugs.webkit.org/show_bug.cgi?id=82532
151
152         Reviewed by Julien Chaffraix.
153
154         Corrected spelling of the word "pass" in build-webkit.
155
156         * Scripts/build-webkit:
157
158 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
159
160         new-run-webkit-tests doesn't log correctly after retrying failures
161         https://bugs.webkit.org/show_bug.cgi?id=82541
162
163         Unreviewed, build fix.
164
165         r112189 introduced a regression where we would accidentally
166         delete the log handler after we retried any failing tests - that
167         meant that any log messages around uploading the results got
168         dropped.
169
170         * Scripts/webkitpy/layout_tests/controllers/manager.py:
171         (Manager._run_tests):
172
173 2012-03-28  Kevin Ollivier  <kevino@theolliviers.com>
174
175         [wx] Unreviewed. Build fix, move WTF back into JSCore target
176         until issues with JSCore not linking in all WTF symbols are resolved.
177
178         * waf/build/settings.py:
179
180 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
181
182         download-built-product build step should detect 404 errors
183         https://bugs.webkit.org/show_bug.cgi?id=82491
184
185         Reviewed by Ryosuke Niwa.
186         
187         Pass --fail to curl so that it reports 404 errors.
188
189         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
190         (DownloadBuiltProduct): Pass --fail to curl so that it reports errors.
191         * BuildSlaveSupport/built-product-archive:
192         (extractBuiltProduct): Remove trailing whitespace.
193
194 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
195
196         flakiness dashboard should only keep the last 500 runs
197         https://bugs.webkit.org/show_bug.cgi?id=82530
198
199         Reviewed by Ojan Vafai.
200
201         If a bot has an overly large amount of failing tests, then
202         storing 750 runs' worth takes a lot of memory and CPU time to
203         process and we end up getting killed by AppEngine; this cases
204         data corruption and confusion down the road when we have to
205         retry the upload of the data and the tests show up as multiple
206         runs.
207
208         * TestResultServer/model/jsonresults.py:
209         (JsonResults.update_file):
210
211 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
212
213         extract-built-product build step should fail if unzipping fails
214         https://bugs.webkit.org/show_bug.cgi?id=82493
215
216         Reviewed by Simon Fraser.
217
218         Forward non-zero exit code from subprocesses so that the master can detect any failures
219         and bail out early.
220
221         * BuildSlaveSupport/built-product-archive:
222         (main):
223
224 2012-03-28  Dirk Pranke  <dpranke@chromium.org>
225
226         test-webkitpy should support files, directories, and packages as command line args
227         https://bugs.webkit.org/show_bug.cgi?id=76765
228
229         Reviewed by Adam Barth.
230
231         This patch adds support for specifying files, directories, and
232         packages to test-webkitpy along with the already existing
233         support for modules, test classes, and individual test names.
234
235         Also, fix a bug in filesystem_mock where we wouldn't normalize a
236         path containing a reference to the current directory properly,
237         for example, '/foo/./bar.py'.
238
239         * Scripts/webkitpy/common/system/filesystem_mock.py:
240         (MockFileSystem.normpath):
241         * Scripts/webkitpy/test/main.py:
242         (Tester._configure_logging):
243         (Tester._run_tests):
244         * Scripts/webkitpy/test/test_finder.py:
245         (TestDirectoryTree.find_modules):
246         (TestDirectoryTree.subpath):
247         (TestFinder.is_dotted_name):
248         (TestFinder.find_names):
249         (TestFinder):
250         (TestFinder._find_names_for_arg):
251         (TestFinder._find_in_trees):
252         (TestFinder._default_names):
253         * Scripts/webkitpy/test/test_finder_unittest.py:
254         (TestFinderTest.setUp):
255         (TestFinderTest.tearDown):
256         (TestFinderTest.check_names):
257         (TestFinderTest.test_default_names):
258         (TestFinderTest):
259         (TestFinderTest.test_paths):
260
261 2012-03-28  Simon Fraser  <simon.fraser@apple.com>
262
263         Fix inspector tests when running in internal builds.
264
265         <rdar://problem/11050694>
266
267         Reviewed by Tim Hatcher.
268
269         * DumpRenderTree/mac/DumpRenderTree.mm:
270         (resetDefaultsToConsistentValues):
271         * WebKitTestRunner/TestController.cpp:
272         (WTR::TestController::resetStateToConsistentValues):
273
274 2012-03-28  David Kilzer  <ddkilzer@apple.com>
275
276         Use xcrun to find nm and ranlib on OS X builds
277         <http://webkit.org/b/82466>
278
279         Reviewed by Dan Bernstein.
280
281         * Scripts/build-webkit: Use xcrun to find ranlib.
282         * Scripts/webkitdirs.pm:
283         (determineNmPath): Added.  Use xcrun to find nm on OS X.
284         (nmPath): Added.
285         * Scripts/webkitperl/features.pm:
286         (libraryContainsSymbol): Use nmPath().
287
288 2012-03-28  Ojan Vafai  <ojan@chromium.org>
289
290         Add history navigation to garden-o-matic
291         https://bugs.webkit.org/show_bug.cgi?id=82495
292
293         Reviewed by Dimitri Glazkov.
294
295         Also, maintain scroll offsets when returning to a tab. This makes addressing
296         expected failures considerably easier when dealing with below-the-fold tests.
297
298         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
299         -Store the scrollTop before switching to a new tab and restore the appropriate scrollTop after
300         switching.
301         -Modify window.location with the tabName in the hash. Can't use pushState because this is served
302         from a file URL and Chrome puts each different file URL in it's own origin.
303
304         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
305         Remove the href so that clicking on the accordion item does not modify window.location.hash.
306
307         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
308         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
309
310 2012-03-28  Philippe Normand  <pnormand@igalia.com>
311
312         [GTK] generate-gtk-doc doesn't cope with custom build directory
313         https://bugs.webkit.org/show_bug.cgi?id=82448
314
315         Reviewed by Martin Robinson.
316
317         Pass the product directory in generate-gtkdoc arguments from
318         build-webkit.
319
320         * Scripts/webkitdirs.pm:
321         (buildAutotoolsProject):
322         * gtk/common.py:
323         (get_build_path):
324
325 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
326
327         [GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
328         https://bugs.webkit.org/show_bug.cgi?id=82341
329
330         Reviewed by Martin Robinson.
331
332         Add --skipped=skip|ignore|only command line option.
333
334         * Scripts/run-gtk-tests:
335         (TestRunner._test_cases_to_skip): Return the list of test cases
336         to skip for the given test depending on the skipped action.
337         (TestRunner._should_run_test): Decide whether to run the test or
338         not depending on the skipped action.
339         (TestRunner.run_tests): Only show the list of skipped tests when
340         skipped action is 'skip'.
341
342 2012-03-28  Carlos Garcia Campos  <cgarcia@igalia.com>
343
344         [GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
345         https://bugs.webkit.org/show_bug.cgi?id=82333
346
347         Reviewed by Martin Robinson.
348
349         * Scripts/run-gtk-tests:
350         (TestRunner.SkippedTest): Add SkippedTest class to store more
351         information about skipped tests.
352         (TestRunner._setup_testing_environment): Use SkippedTest to add
353         TestWebKitAccessibility to the skipped list.
354         (TestRunner._find_skipped_test): Return the SkippedTest object for
355         a given test path.
356         (TestRunner._test_cases_to_skip): Returns the list of test cases
357         that should be skipped for a given test path.
358         (TestRunner._should_run_test): Helper funtion to decide whether
359         the given test path should be run or not. Tests in the skipped
360         list that don't have a list of failing test cases are not run at
361         all. Tests in the skipped list that contain a list of test cases
362         are run with gtester -s to skip the individual test cases.
363         (TestRunner._run_test): Helper funtion to run a given test.
364         (TestRunner.run_tests): Build the list of tests to run based on
365         _should_run_test() and use helper function _run_test() to run
366         every test.
367
368 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
369
370         [Qt] Couple of tests have different results on 64 bit and/or in debug mode compared to 32 bit and/or release mode
371         https://bugs.webkit.org/show_bug.cgi?id=52810
372
373         Reviewed by Simon Hausmann.
374
375         Use SSE2 floating point math on 32 bit instead of the default
376         387 to make layout test result same on 32 and on 64 bit builds.
377
378         * qmake/mkspecs/features/unix/default_post.prf:
379
380 2012-03-28  Sergio Villar Senin  <svillar@igalia.com>
381
382         [Soup] DNS prefetching spams resolver, shoots self in the foot
383         https://bugs.webkit.org/show_bug.cgi?id=41630
384
385         Reviewed by Martin Robinson.
386
387         Bump libsoup and glib dependencies.
388
389         * efl/jhbuild.modules:
390         * gtk/jhbuild.modules:
391
392 2012-03-28  Gustavo Noronha Silva  <gns@gnome.org>
393
394         [GTK] Disable introspection for jhbuild modules
395         https://bugs.webkit.org/show_bug.cgi?id=82395
396
397         Reviewed by Philippe Normand.
398
399         * efl/jhbuild.modules: add --disable-introspection to all modules
400         that support it
401         * gtk/jhbuild.modules: ditto.
402
403 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
404
405         [Qt] Unreviewed, fix API tests after adding full platforms. (r111658)
406
407         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
408         (BuildAndTestFactory.__init__):
409
410 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
411
412         [Qt] Unreviewed typo fix after r112383.
413
414         * Scripts/run-qtwebkit-tests:
415         (Main.check_crash_occurences):
416
417 2012-03-28  Ádám Kallai  <kadam@inf.u-szeged.hu>
418
419         [Qt] run-qtwebkit-tests should report crashes
420         https://bugs.webkit.org/show_bug.cgi?id=76905
421
422         Added checking of crashing tests. Now final summary includes the total number of crashing tests too.
423         Made unit test for this modification by János Badics <jbadics@inf.u-szeged.hu>.
424
425         Reviewed by Csaba Osztrogonác.
426
427         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
428         (RunQtAPITests.commandComplete):
429         (RunQtAPITests.evaluateCommand):
430         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
431         (RunQtAPITestsTest):
432         (test_failure):
433         (test_crash):
434         * Scripts/run-qtwebkit-tests:
435         (Main.announce_results_html):
436         (Main):
437         (Main.check_crash_occurences):
438         (Main.convert_to_stdout):
439         (Main.convert_to_html):
440
441 2012-03-28  Kwonjin Jeong  <gram@company100.net>
442
443         Add support for Intel compiler on Linux QT port.
444         https://bugs.webkit.org/show_bug.cgi?id=81077
445
446         Reviewed by Tor Arne Vestbø.
447
448         * qmake/mkspecs/features/default_post.prf:
449         Don't hide symbols when building with Intel compiler.
450         The symbols are needed when building QtWebKit tests.
451
452         * qmake/mkspecs/features/unix/icc.prf:
453         Added to ignore icc-specific warnings.
454         This feature is used with command-line arguments for build-webkit script.
455
456 2012-03-28  Philippe Normand  <pnormand@igalia.com>
457
458         Unreviewed, trivial update-webkit-libs-jhbuild system call fix.
459
460         * Scripts/update-webkitgtk-libs: This script is written in
461         Perl. Not Python. So use perl vocabulary to refer to command line arguments.
462
463 2012-03-28  Ryosuke Niwa  <rniwa@webkit.org>
464
465         kill-old-processes doesn't do anything on linux
466         https://bugs.webkit.org/show_bug.cgi?id=82388
467
468         Reviewed by Tony Chang.
469
470         Kill cc1plus, ld, etc... on linux. Also ps aux + grep to kill old run-webkit-tests processes.
471         This should help stabilizing Chromium linux bots.
472
473         * BuildSlaveSupport/kill-old-processes:
474
475 2012-03-28  Philippe Normand  <pnormand@igalia.com>
476
477         [GTK] jhbuild wrongly tricked into clean-up
478         https://bugs.webkit.org/show_bug.cgi?id=82216
479
480         Reviewed by Gustavo Noronha Silva.
481
482         Save jhbuild{rc,.modules}.md5sum in Dependencies instead of the
483         build tree. Also when checking if the jhbuild configuration
484         changed use absolute md5sum paths.
485
486         * Scripts/webkitdirs.pm:
487         (runAutogenForAutotoolsProjectIfNecessary):
488         (jhbuildConfigurationChanged):
489         (buildAutotoolsProject):
490
491 2012-03-28  Csaba Osztrogonác  <ossy@webkit.org>
492
493         kill-old-processes doesn't do anything on linux
494         https://bugs.webkit.org/show_bug.cgi?id=82388
495
496         One more attempt to disable kill-old-processes on the Qt bots not to kill each others.
497
498         The platform in Factory comes from config.json, accordingly platform is fullplatform,
499         for example: qt-4.8, qt-5.0-wk2. We have to check only the prefix of platform.
500
501         Reviewed by Ryosuke Niwa.
502
503         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
504         (Factory.__init__):
505
506 2012-03-28  Hao Zheng  <zhenghao@chromium.org>
507
508         Make Chromium port for Android use hardware gpu path default.
509         https://bugs.webkit.org/show_bug.cgi?id=82033
510
511         Reviewed by Dirk Pranke.
512
513         Skip canvas tests, which are run in virtual gpu test suites.
514
515         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
516         (ChromiumAndroidPort.__init__):
517         (ChromiumAndroidPort.skipped_tests):
518
519 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
520
521         kill-old-processes doesn't do anything on linux
522         https://bugs.webkit.org/show_bug.cgi?id=82388
523
524         Disable kill-old-processes on the Qt bots not to kill each others.
525
526         Reviewed by Ryosuke Niwa.
527
528         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
529         (Factory.__init__):
530
531 2012-03-27  Hao Zheng  <zhenghao@chromium.org>
532
533         Handle DRT crash caused by Android OOM.
534         https://bugs.webkit.org/show_bug.cgi?id=82310
535
536         Reviewed by Tony Chang.
537
538         When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
539         is stopped silently and regarded as crashed. Re-run the test for
540         such crash.
541
542         Add some debug log to NRWT.
543
544         * Scripts/webkitpy/layout_tests/port/chromium.py:
545         (ChromiumDriver.run_test):
546         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
547         (ChromiumAndroidPort.__init__):
548         (ChromiumAndroidPort.get_last_stacktrace):
549         (ChromiumAndroidDriver.__init__):
550         (ChromiumAndroidDriver._start):
551         (ChromiumAndroidDriver.run_test):
552         (ChromiumAndroidDriver._get_drt_return_value):
553
554 2012-03-27  Ojan Vafai  <ojan@chromium.org>
555
556         Fix expected results for some unittest failures.
557
558         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
559
560 2012-03-27  Adrienne Walker  <enne@google.com>
561
562         [chromium] Fix --enabled-threaded-compositing flag in DRT
563         https://bugs.webkit.org/show_bug.cgi?id=82405
564
565         Reviewed by James Robinson.
566
567         --enable-threaded-compositing wasn't working because there was no way
568         to set the value before it was used in the TestShell constructor.
569
570         To avoid adding yet another boolean parameter to TestShell's
571         constructor, add an explicit initialize function that can operate on
572         any boolean settings that are set post-constructor.
573
574         * DumpRenderTree/chromium/DumpRenderTree.cpp:
575         (main):
576         * DumpRenderTree/chromium/TestShell.cpp:
577         (TestShell::TestShell):
578         (TestShell::initialize):
579         * DumpRenderTree/chromium/TestShell.h:
580         (TestShell):
581         (TestShell::setTestShellMode):
582
583 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
584
585         new-run-webkit-tests: suppress debug output from run-safari on results file
586         https://bugs.webkit.org/show_bug.cgi?id=82400
587
588         Reviewed by Ryosuke Niwa.
589
590         Send the run-safari output to /dev/null rather than the console.
591
592         * Scripts/webkitpy/layout_tests/port/mac.py:
593         (MacPort.show_results_html_file):
594
595 2012-03-27  Simon Fraser  <simon.fraser@apple.com>
596
597         Improve error reporting in run-javascriptcore-tests
598         https://bugs.webkit.org/show_bug.cgi?id=82379
599
600         Reviewed by Jessie Berlin.
601         
602         When the script dies, have it print out the reason for dying.
603
604         * Scripts/run-javascriptcore-tests:
605
606 2012-03-27  Eric Seidel  <eric@webkit.org>
607
608         Add Dave Barton to the Contributors list so I can more easily CC him on bugs
609         https://bugs.webkit.org/show_bug.cgi?id=82373
610
611         Reviewed by Ryosuke Niwa.
612
613         Dave's committer-status in in-progress, so until then
614         list him as a Contributor so at least I can more easily CC him on bugs.
615
616         * Scripts/webkitpy/common/config/committers.py:
617
618 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
619
620         [jhbuild] Prepend jhbuildrc's directory to sys.path instead of appending.
621         https://bugs.webkit.org/show_bug.cgi?id=82252
622
623         Reviewed by Gustavo Noronha Silva.
624
625         Appending the directory containing common.py may be problematic if
626         another path in sys.path also contains either a common.py or a
627         common/ module, so prepend to select our version instead. This is
628         the case with fail2ban, which has a common/ directory which may
629         end up being in Python's default path.
630
631         * efl/jhbuildrc:
632         * gtk/jhbuildrc:
633
634 2012-03-27  Dirk Pranke  <dpranke@chromium.org>
635
636         test-webkitpy: prepare for better test run output
637         https://bugs.webkit.org/show_bug.cgi?id=82290
638
639         Reviewed by Adam Barth.
640
641         This code basically re-implements the output of the TextTestRunner default
642         runner code from unittest, although the implementation is quite
643         different, in preparation for changing the test output to be
644         metered and possibly running in parallel.
645
646         The output is almost identical to before, except that instead of
647         logging "test_regular (webkitpy.main.RunnerTest) passed" we log
648         "webkitpy.main.RunnerTest.test_regular passed". It has always
649         annoyed me that they invert the names to be harder to read and
650         so that you can't copy & paste back to the input for
651         test-webkitpy.
652
653         This patch is provided to add a skeleton for unit tests and for
654         comparison to an upcoming patch that will actually add new
655         functionality.
656
657         * Scripts/webkitpy/test/main.py:
658         (Tester.__init__):
659         (Tester._configure_logging):
660         (Tester._run_tests):
661         * Scripts/webkitpy/test/runner.py: Added.
662         (TestRunner):
663         (TestRunner.__init__):
664         (TestRunner.test_name):
665         (TestRunner.all_test_names):
666         (TestRunner.run):
667         (TestRunner.write_result):
668         (TestRunner.write_summary):
669         * Scripts/webkitpy/test/runner_unittest.py: Added.
670         (FakeModuleSuite):
671         (FakeModuleSuite.__init__):
672         (FakeModuleSuite.__str__):
673         (FakeModuleSuite.run):
674         (FakeTopSuite):
675         (FakeTopSuite.__init__):
676         (FakeLoader):
677         (FakeLoader.__init__):
678         (FakeLoader.top_suite):
679         (FakeLoader.loadTestsFromName):
680         (RunnerTest):
681         (RunnerTest.test_regular):
682         (RunnerTest.test_verbose):
683
684 2012-03-27  Gustavo Noronha Silva  <gns@gnome.org>
685
686         [GTK] Build gnutls without p11-kit support
687         https://bugs.webkit.org/show_bug.cgi?id=82212
688
689         Reviewed by Martin Robinson.
690
691         * gtk/jhbuild.modules: pass --without-p11-kit to gnutls'
692         configure, to avoid requiring that dependency, which is not used
693         by libsoup and WebKit anyway
694
695 2012-03-27  Csaba Osztrogonác  <ossy@webkit.org>
696
697         [Qt] Make debug build work on 32 bit
698         https://bugs.webkit.org/show_bug.cgi?id=82331
699
700         Use --no-keep-memory option of ld not to cache the symbol tables of input files in
701         memory to avoid memory exhaustion during the linking phase for x86-debug builds.
702
703         Reviewed by Tor Arne Vestbø.
704
705         * qmake/mkspecs/features/unix/default_post.prf:
706         * qmake/mkspecs/features/unix/default_pre.prf:
707
708 2012-03-23  Balazs Ankes  <bank@inf.u-szeged.hu>
709
710         Add png checking to check-webkit-style
711         https://bugs.webkit.org/show_bug.cgi?id=75824
712
713         Reviewed by NOBODY Tony Chang.
714
715         * Scripts/webkitpy/style/checker.py:
716         (_all_categories):
717         (FileType):
718         (CheckerDispatcher._should_skip_file_path):
719         (CheckerDispatcher._file_type):
720         (CheckerDispatcher._create_checker):
721         * Scripts/webkitpy/style/checker_unittest.py:
722         (CheckerDispatcherDispatchTest.test_none_paths):
723         * Scripts/webkitpy/style/checkers/png.py: Added.
724         (PNGChecker):
725         (PNGChecker.__init__):
726         (PNGChecker.check):
727         (PNGChecker._config_file_path):
728         * Scripts/webkitpy/style/checkers/png_unittest.py: Added.
729         (MockSCMDetector):
730         (MockSCMDetector.__init__):
731         (MockSCMDetector.display_name):
732         (MockSCMDetector.propget):
733         (PNGCheckerTest):
734         (PNGCheckerTest.test_init):
735         (PNGCheckerTest.test_init.mock_handle_style_error):
736         (PNGCheckerTest.test_check):
737         (PNGCheckerTest.test_check.mock_handle_style_error):
738         * Scripts/webkitpy/style/patchreader.py:
739         (PatchReader.check):
740
741 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
742
743         Unreviewed. Skip GTK+ unit tests that fail in the bots
744
745         * Scripts/run-gtk-tests:
746         (TestRunner): Add failing tests to the skipped list.
747
748 2012-03-27  Carlos Garcia Campos  <cgarcia@igalia.com>
749
750         Unreviewed. Unskip all GTK+ unit test.
751
752         Currently, skipped tests contains the list of unit tests that have
753         failed in one or more bots. In this moment it's not possible to
754         skip just the test cases that fail, so the complete unit test is
755         skipped when a single test case fails. For that reason, we don't
756         know which test cases are actually failing for every skipped
757         test. We are planning to land a patch to allow skipping individual
758         test cases, but first we need to know which test cases are
759         currently failing in the bots to skip them. So, this patch simply
760         unskips all unit tests in order to watch the bots and get the list
761         of test cases failing. Failing tests will be skipped again as soon
762         as we have the new list of failing test cases.
763
764         * Scripts/run-gtk-tests:
765         (TestRunner): Reset the list of skipped tests.
766
767 2012-03-26  Scott Byer  <scottbyer@chromium.org>
768
769         Enable layout testing of the scroll animator.
770         https://bugs.webkit.org/show_bug.cgi?id=81858
771         Add a call to the InternalSettings that layout tests can use to
772         turn on scroll animation. Enable animation updates for the
773         Chromium platform DRT when scroll animation has been turned on in
774         a test. This should be a no-op for all current layout tests.
775
776         Reviewed by James Robinson.
777
778         * DumpRenderTree/chromium/WebViewHost.cpp:
779         (WebViewHost::serviceAnimation):
780         (WebViewHost::scheduleAnimation):
781         * DumpRenderTree/chromium/WebViewHost.h:
782         (WebViewHost):
783
784 2012-03-26  Dinu Jacob  <dinu.jacob@nokia.com>
785
786         [Qt][WK2] Support multi-file upload
787         https://bugs.webkit.org/show_bug.cgi?id=81589
788
789         Reviewed by Simon Hausmann.
790
791         Modified filePicker to support multi-file upload.
792
793         * MiniBrowser/qt/MiniBrowser.qrc:
794         * MiniBrowser/qt/icons/checkbox_checked.png: Added.
795         * MiniBrowser/qt/icons/checkbox_unchecked.png: Added.
796         * MiniBrowser/qt/js/MultiSelect.js: Added.
797         (values):
798         (isSelected):
799         (removeValue):
800         * MiniBrowser/qt/qml/CheckBox.qml: Added.
801         * MiniBrowser/qt/qml/FilePicker.qml:
802
803 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
804
805         Retry crashing tests serially at the end of NRWT on Apple Mac
806         https://bugs.webkit.org/show_bug.cgi?id=82233
807
808         Reviewed by Dirk Pranke.
809
810         * Scripts/webkitpy/layout_tests/controllers/manager.py:
811         (Manager.run):
812         * Scripts/webkitpy/layout_tests/port/base.py:
813         (Port.should_retry_crashes):
814         * Scripts/webkitpy/layout_tests/port/mac.py:
815         (MacPort.should_retry_crashes):
816
817 2012-03-26  Mark Hahnenberg  <mhahnenberg@apple.com>
818
819         NRWT should retry failures serially
820         https://bugs.webkit.org/show_bug.cgi?id=82241
821
822         Reviewed by Dirk Pranke.
823
824         * Scripts/webkitpy/layout_tests/controllers/manager.py:
825         (Manager._run_tests): Add another parameter to specify the number of workers to use.
826         (Manager.run): Run the failures serially.
827
828 2012-03-26  Adrienne Walker  <enne@google.com>
829
830         Allow expectations on directories in virtual test suites
831         https://bugs.webkit.org/show_bug.cgi?id=82250
832
833         Reviewed by Dirk Pranke.
834
835         lookup_virtual_test_base was only returning valid results for files
836         and not directories. Fix by falling back to the virtual test suite
837         mapping if possible.
838
839         * Scripts/webkitpy/layout_tests/port/base.py:
840         (Port.lookup_virtual_test_base):
841         * Scripts/webkitpy/layout_tests/port/test.py:
842         (TestPort.skipped_tests):
843         (TestPort.virtual_test_suites):
844
845 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
846
847         Fix duplicated ChangeLog entry from r112171.
848
849         Unreviewed, build fix.
850
851 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
852
853         nrwt: remove --worker-model flag
854         https://bugs.webkit.org/show_bug.cgi?id=82112
855
856         Reviewed by Adam Barth.
857
858         The --worker-model=inline flag is basically redundant with the
859         --child-processes=1 flag (technically, you could one run child
860         process, but this is only useful for testing); this patch
861         removes this flag, which simplifies things and enables more
862         cleanup down the road.
863
864         * Scripts/webkitpy/layout_tests/controllers/manager.py:
865         (Manager._run_tests):
866         (Manager.print_config):
867         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
868         (get):
869         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
870         (make_broker):
871         (FunctionTests.test_get__inline):
872         (FunctionTests.test_get__processes):
873         (_TestsMixin.setUp):
874         (_TestsMixin.make_broker):
875         (InlineBrokerTests.setUp):
876         (MultiProcessBrokerTests.setUp):
877         * Scripts/webkitpy/layout_tests/port/base.py:
878         (Port.default_child_processes):
879         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
880         (ChromiumAndroidPort.check_sys_deps):
881         * Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
882         (ChromiumAndroidPortTest):
883         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
884         (MockDRTPortTest.make_port):
885         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
886         (PortTestCase):
887         (PortTestCase.make_port):
888         * Scripts/webkitpy/layout_tests/port/test.py:
889         (TestPort.default_child_processes):
890         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
891         (_set_up_derived_options):
892         (parse_args):
893         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
894         (parse_args):
895         (MainTest.test_batch_size):
896         (MainTest.test_child_processes_2):
897         (MainTest.test_child_processes_min):
898         (MainTest.test_exception_raised):
899         (MainTest.test_keyboard_interrupt):
900         (MainTest.test_exit_after_n_crashes):
901         (MainTest.assert_run_order):
902         (MainTest.test_run_order__inline):
903         (MainTest.test_virtual):
904
905 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
906
907         test-webkitpy: split test-finding code into a different module
908         https://bugs.webkit.org/show_bug.cgi?id=82253
909
910         Reviewed by Adam Barth.
911
912         Per suggestion from abarth, this change splits all the
913         filesystem-crawling, test-finding code in test-webkitpy out into
914         a separate module and switches to using a FileSystem object.
915         This makes things much more testable, so we also add tests :).
916
917         We also add a realpath() method to the Filesystem object, since
918         test-webkitpy needs that in order to be able to resolve symlinks
919         properly to determine whether a file is under a particular tree
920         or not.
921
922         * Scripts/webkitpy/common/system/filesystem.py:
923         (FileSystem.realpath):
924         * Scripts/webkitpy/common/system/filesystem_mock.py:
925         (MockFileSystem.realpath):
926         * Scripts/webkitpy/test/main.py:
927         (Tester.__init__):
928         (Tester.add_tree):
929         (Tester.run):
930         (Tester._run_tests):
931         (Tester._log_exception):
932         * Scripts/webkitpy/test/test_finder.py: Added.
933         (TestDirectoryTree):
934         (TestDirectoryTree.__init__):
935         (TestDirectoryTree.find_modules):
936         (TestDirectoryTree.find_modules.file_filter):
937         (TestDirectoryTree.to_module):
938         (TestDirectoryTree.clean):
939         (TestFinder):
940         (TestFinder.__init__):
941         (TestFinder.add_tree):
942         (TestFinder.additional_paths):
943         (TestFinder.clean_trees):
944         (TestFinder.is_module):
945         (TestFinder.to_module):
946         (TestFinder.find_names):
947         (TestFinder._exclude):
948         * Scripts/webkitpy/test/test_finder_unittest.py: Added.
949         (TestFinderTest):
950         (TestFinderTest.setUp):
951         (TestFinderTest.tearDown):
952         (TestFinderTest.test_additional_system_paths):
953         (TestFinderTest.test_is_module):
954         (TestFinderTest.test_to_module):
955         (TestFinderTest.test_clean):
956         (TestFinderTest.test_find_names):
957
958 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
959
960         [jhbuild] Use $MAKE if it is defined to build jhbuild itself.
961         https://bugs.webkit.org/show_bug.cgi?id=82209
962
963         Reviewed by Martin Robinson.
964
965         Respecting the $MAKE environment variable when it is defined makes
966         building jhbuild more portable, as some platforms may have GNU
967         make as gmake, for example.
968
969         * jhbuild/jhbuild-wrapper:
970         (install_jhbuild):
971
972 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
973
974         add a Tree abstraction to test-webkitpy to better encapsulate things
975         https://bugs.webkit.org/show_bug.cgi?id=82158
976
977         Reviewed by Adam Barth.
978
979         There are no functional changes in this patch, but adding a
980         basic abstraction for the trees we are looking in for python
981         tests will allow me to add some features later on more easily
982         (like ignoring certain directories like webkitpy/thirdparty),
983         and it makes things slightly less hard-coded.
984
985         * Scripts/test-webkitpy:
986         * Scripts/webkitpy/test/main.py:
987         (Tester.__init__):
988         (Tester):
989         (Tester.add_tree):
990         (Tester._parse_args):
991         (Tester._configure):
992         (Tester.run):
993         (Tester._find_modules):
994         (Tester._run_tests):
995         (Tester._is_module):
996         (Tester._log_exception):
997         (TestDirectoryTree):
998         (TestDirectoryTree.__init__):
999         (TestDirectoryTree.find_modules):
1000         (TestDirectoryTree.clean):
1001
1002 2012-03-26  Ojan Vafai  <ojan@chromium.org>
1003
1004         garden-o-matic should special case reftests
1005         https://bugs.webkit.org/show_bug.cgi?id=81812
1006
1007         Reviewed by Adam Barth.
1008
1009         On the examine page, don't show a rebaseline button for reftests.
1010         We still show it on the failure summary page because tests are grouped
1011         together. Long-term, we should make sure webkit-patch rebaseline-test
1012         never tries to rebaseline reftests.
1013
1014         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1015         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1016         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1017
1018 2012-03-26  Dirk Pranke  <dpranke@chromium.org>
1019
1020         clean up NRWT logging/metering, be less verbose
1021         https://bugs.webkit.org/show_bug.cgi?id=73846
1022
1023         Reviewed by Ojan Vafai.
1024
1025         Re-land r112014 with a build fix for Python 2.6 compatibility.
1026
1027         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1028         (ManagerTest.test_http_locking):
1029         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1030         (AbstractWorker.run):
1031         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1032         (Worker.__init__):
1033         (Worker._set_up_logging):
1034         (Worker):
1035         (Worker._set_up_host_and_port):
1036         (Worker.run):
1037         (Worker.cleanup):
1038         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
1039         (WorkerTest.test_default_platform_in_worker):
1040         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1041         (run):
1042         (main):
1043         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1044         (MeteredStream):
1045         (MeteredStream.implements):
1046         (MeteredStream._erasure):
1047         (MeteredStream._ensure_newline):
1048         (MeteredStream.__init__):
1049         (MeteredStream.__del__):
1050         (MeteredStream.cleanup):
1051         (MeteredStream.write_throttled_update):
1052         (MeteredStream.write_update):
1053         (MeteredStream.write):
1054         (MeteredStream.writeln):
1055         (MeteredStream._erase_last_partial_line):
1056         (_LogHandler):
1057         (_LogHandler.__init__):
1058         (_LogHandler.emit):
1059         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1060         (RegularTest):
1061         (RegularTest.setUp):
1062         (RegularTest.tearDown):
1063         (RegularTest.time_fn):
1064         (RegularTest.test_logging_not_included):
1065         (RegularTest._basic):
1066         (RegularTest.test_basic):
1067         (RegularTest._log_after_update):
1068         (RegularTest.test_log_after_update):
1069         (TtyTest):
1070         (TtyTest.test_basic):
1071         (TtyTest.test_log_after_update):
1072         (VerboseTest):
1073         (VerboseTest.test_basic):
1074         (VerboseTest.test_log_after_update):
1075         * Scripts/webkitpy/layout_tests/views/printing.py:
1076         (parse_print_options):
1077         (Printer.__init__):
1078         (Printer.cleanup):
1079         (Printer.print_progress):
1080         (Printer.print_update):
1081         (Printer._write):
1082         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1083         (Testprinter.get_printer):
1084         (Testprinter.test_print_progress):
1085         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1086         (PerfTestsRunner.__init__):
1087         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1088         (test_run_test_pause_before_testing):
1089
1090 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1091
1092         [jhbuild] Revert $MAKE environment hack introduced in r101929.
1093         https://bugs.webkit.org/show_bug.cgi?id=82234
1094
1095         Reviewed by Martin Robinson.
1096
1097         The problem it was supposedly fixing was fixed in jhbuild itself
1098         in git revision da8944d8ca987ca48c91b350257a530050406092, so we
1099         can remove the workaround added to the script.
1100
1101         * jhbuild/jhbuild-wrapper:
1102         (install_jhbuild):
1103
1104 2012-03-26  Leo Yang  <leo.yang@torchmobile.com.cn>
1105
1106         [BlackBerry] Skeleton code of AsyncFileSystemBlackBerry.cpp
1107         https://bugs.webkit.org/show_bug.cgi?id=82157
1108
1109         Reviewed by Rob Buis.
1110
1111         * Scripts/build-webkit:
1112         Make ENABLE_FILE_SYSTEM default on for blackberry.
1113
1114 2012-03-26  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1115
1116         [jhbuild] Explicitly disable DTrace when building glib.
1117         https://bugs.webkit.org/show_bug.cgi?id=82145
1118
1119         Reviewed by Gustavo Noronha Silva.
1120
1121         DTrace support is not really needed by WebKit, so disabling it
1122         helps avoid some build errors on more exotic platforms (such as
1123         FreeBSD).
1124
1125         * efl/jhbuild.modules:
1126         * gtk/jhbuild.modules:
1127
1128 2012-03-26  Sheriff Bot  <webkit.review.bot@gmail.com>
1129
1130         Unreviewed, rolling out r111993.
1131         http://trac.webkit.org/changeset/111993
1132         https://bugs.webkit.org/show_bug.cgi?id=82184
1133
1134         It broke layout and API tests and made WTR crash (Requested by
1135         Ossy on #webkit).
1136
1137         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1138         (WTR::PlatformWebView::PlatformWebView):
1139
1140 2012-03-25  Kevin Ollivier  <kevino@theolliviers.com>
1141
1142         [wx] Unreviewed build fix. Move WTF to its own static lib build.
1143
1144         * waf/build/settings.py:
1145         (common_configure):
1146
1147 2012-03-24  Chris Fleizach  <cfleizach@apple.com>
1148
1149         AX: Support solution to handle invalid ax text marker
1150         https://bugs.webkit.org/show_bug.cgi?id=82023
1151
1152         Reviewed by Oliver Hunt.
1153
1154         * DumpRenderTree/AccessibilityUIElement.cpp:
1155         (indexForTextMarkerCallback):
1156         (isTextMarkerValidCallback):
1157         (textMarkerForIndexCallback):
1158         (AccessibilityUIElement::indexForTextMarker):
1159         (AccessibilityUIElement::isTextMarkerValid):
1160         (AccessibilityUIElement::textMarkerForIndex):
1161         (AccessibilityUIElement::getJSClass):
1162         * DumpRenderTree/AccessibilityUIElement.h:
1163         (AccessibilityUIElement):
1164         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1165         (AccessibilityUIElement::indexForTextMarker):
1166         (AccessibilityUIElement::textMarkerForIndex):
1167         (AccessibilityUIElement::isTextMarkerValid):
1168         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
1169         (WTR::AccessibilityUIElement::indexForTextMarker):
1170         (WTR::AccessibilityUIElement::isTextMarkerValid):
1171         (WTR::AccessibilityUIElement::textMarkerForIndex):
1172         (WTR):
1173         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
1174         (AccessibilityUIElement):
1175         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
1176         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
1177         (WTR):
1178         (WTR::AccessibilityUIElement::indexForTextMarker):
1179         (WTR::AccessibilityUIElement::isTextMarkerValid):
1180         (WTR::AccessibilityUIElement::textMarkerForIndex):
1181
1182 2012-03-24  Sheriff Bot  <webkit.review.bot@gmail.com>
1183
1184         Unreviewed, rolling out r112014.
1185         http://trac.webkit.org/changeset/112014
1186         https://bugs.webkit.org/show_bug.cgi?id=82138
1187
1188         Broke SnowLeopard, Chromium and GTK+ bots (Requested by kov on
1189         #webkit).
1190
1191         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1192         (ManagerTest.test_http_locking):
1193         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1194         (AbstractWorker.run):
1195         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1196         (Worker.__init__):
1197         (Worker.safe_init):
1198         (Worker.run):
1199         (Worker.cleanup):
1200         * Scripts/webkitpy/layout_tests/controllers/worker_unittest.py:
1201         (WorkerTest.test_default_platform_in_worker):
1202         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1203         (run):
1204         (main):
1205         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1206         (MeteredStream):
1207         (MeteredStream.is):
1208         (MeteredStream.__init__):
1209         (MeteredStream.write):
1210         (MeteredStream.update):
1211         (MeteredStream._overwrite):
1212         (MeteredStream._reset):
1213         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1214         (TestMeteredStream):
1215         (TestMeteredStream.test_regular):
1216         * Scripts/webkitpy/layout_tests/views/printing.py:
1217         (_configure_logging):
1218         (_restore_logging):
1219         (Printer.__init__):
1220         (Printer.cleanup):
1221         (Printer.print_progress):
1222         (Printer.print_update):
1223         (Printer._write):
1224         (Printer):
1225         (Printer._update):
1226         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1227         (TestUtilityFunctions.assertEmpty):
1228         (TestUtilityFunctions):
1229         (TestUtilityFunctions.assertNotEmpty):
1230         (TestUtilityFunctions.assertWritten):
1231         (TestUtilityFunctions.test_configure_logging):
1232         (Testprinter.get_printer):
1233         (Testprinter.test_print_progress):
1234         (Testprinter.test_print_progress.mock_time):
1235         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1236         (PerfTestsRunner.__init__):
1237         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1238         (test_run_test_pause_before_testing):
1239
1240 2012-03-24  Kevin Ollivier  <kevino@theolliviers.com>
1241
1242         [wx] Unreviewed. Fix the build after WTF move.
1243
1244         * waf/build/build_utils.py:
1245         (get_excludes):
1246         (get_excludes_in_dirs):
1247         * waf/build/settings.py:
1248         (get_port_excludes):
1249         (common_configure):
1250
1251 2012-03-24  Dirk Pranke  <dpranke@chromium.org>
1252
1253         clean up NRWT logging/metering, be less verbose
1254         https://bugs.webkit.org/show_bug.cgi?id=73846
1255
1256         Reviewed by Ojan Vafai.
1257
1258         This patch totally rewrites the metered_stream module and
1259         changes the way it and the printing module now interact with the
1260         logging subsystem.
1261
1262         The printing module becomes a dumb layout_tests-specific wrapper around
1263         the metered stream. All of the intelligence of how and when to
1264         actually write (or erase) lines and how things interact with
1265         logging moves into the metered stream.
1266
1267         As of now, anything written to the metered stream *bypasses the
1268         logging system*. If a logger is passed to the meter constructor,
1269         the metered stream will also attach a log handler so that all
1270         messages fed to the logger will also be written to the metered
1271         stream. This detangles the logic that used to exist where
1272         sometimes the meter would write to the logger and sometimes the
1273         logger would write to the meter.
1274
1275         All of this makes the logic generic and reusable for
1276         test-webkitpy (coming in another patch). It also helps clarify
1277         what the mysterious "configure_logging" parameter to the printer
1278         constructor used to do (that parameter itself is gone now, and
1279         by default the printer and metered_stream will not ever touch
1280         the logging system; this makes a bunch of tests less fragile).
1281
1282         All of the above should produce no user-visible changes.
1283
1284         However, to also fix the bug in question, we change the message
1285         format in --verbose mode to %H:%M:%S.%frac %pid %message,
1286         eliminating the date, file:lineno, and level parameters. I am
1287         making this change at the same time because it would be
1288         difficult to write things consistently between the logs and the
1289         metered stream if we were still logging the file:lineno.
1290
1291         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1292           Remove the configure_logging parameter
1293         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1294           Use metered stream to configure the logging (we don't need a printer)
1295           and be careful about when and how we manipulate the log level
1296           and handlers on the root logger.
1297         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1298           Pass in the root logger so that log messages are formatted
1299           properly, and set the log level outside of any routines called
1300           by unit tests so as to not interfere.
1301         * Scripts/webkitpy/layout_tests/views/manager_worker_broker.py:
1302           remove a duplicate log message.
1303         * Scripts/webkitpy/layout_tests/views/metered_stream.py:
1304           complete rewrite :).
1305         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py:
1306           complete rewrite :).
1307         * Scripts/webkitpy/layout_tests/views/printing.py:
1308           move the logging configuration into metered_stream, and also
1309           move the "when should I log progress messages" into metered stream.
1310         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
1311           remove tests that are no longer necessary.
1312         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1313           remove the configure_logging parameter.
1314         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1315           update the test; since the printer no longer writes to the log,
1316           we need to check what is written to the regular_output instead.
1317           (test_run_test_pause_before_testing):
1318
1319 2012-03-24  Gustavo Noronha Silva  <gns@gnome.org>
1320
1321         Add myself and Martin Robinson to:
1322           * GStreamerGraphics
1323           * GtkWebKit2PublicAPI
1324           * SoupNetwork
1325         and to the newly created WebKitGTKTranslations.
1326
1327         Rubber-stamped by Martin Robinson.
1328
1329         * Scripts/webkitpy/common/config/watchlist:
1330
1331 2012-03-24  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
1332
1333         [Qt][WK2] Implement PageClient::isViewWindowActive()
1334         https://bugs.webkit.org/show_bug.cgi?id=81143
1335
1336         Reviewed by Kenneth Rohde Christiansen.
1337
1338         Fix the WrapperWindow from our PlatformWebView
1339         so it correctly creates the platform related
1340         stuff for QWindow, which is not created unless
1341         QWindow::setVisible() or QWindow::show() are called.
1342
1343         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1344         (WTR::PlatformWebView::PlatformWebView):
1345
1346 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1347
1348         nrwt: don't wait for safari to exit before exiting after showing the results file
1349         https://bugs.webkit.org/show_bug.cgi?id=81845
1350
1351         Reviewed by Ryosuke Niwa.
1352
1353         Prior to this change, if you ran new-run-webkit-tests and
1354         displayed the HTML results file at the end, we would block
1355         waiting for the user to quit the browser. There doesn't seem to
1356         be a need for that, and the Chromium ports don't do this.
1357
1358         Also, update the mac tests to capture the output and be quiet :).
1359
1360         * Scripts/webkitpy/common/system/executive_mock.py:
1361           Make popen() testable, implement should_log for it.
1362         * Scripts/webkitpy/layout_tests/port/mac.py:
1363         (MacPort.show_results_html_file):
1364         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1365           Silence some of the tests via outputcapture.
1366
1367 2012-03-23  Gustavo Noronha Silva  <gns@gnome.org>
1368
1369         [GTK] libgcrypt and p11-kit should not be in jhbuild modules
1370         https://bugs.webkit.org/show_bug.cgi?id=82073
1371
1372         Reviewed by Martin Robinson.
1373
1374         * Scripts/webkitdirs.pm:
1375         (jhbuildConfigurationChanged): new function that isolates the
1376         jhbuild configuration change checking from the autogen logic.
1377         (mustReRunAutogen): code for the new function lived here.
1378         (buildAutotoolsProject): now calls jhbuildConfigurationChanged
1379         itself, and cleans up the jhbuild root if configuration has
1380         changed; it also forces autogen to be run in that case, to
1381         make sure the new libraries are used.
1382         * gtk/jhbuild.modules: removed libgcrypt and p11-kit.
1383         * jhbuild/jhbuild-wrapper:
1384         (ensure_jhbuild): this might be run inside a jhbuild environment,
1385         in which case aclocal fails when trying to use the now deleted
1386         aclocal directory in the jhbuild prefix, so work around that.
1387
1388 2012-03-23  Ojan Vafai  <ojan@chromium.org>
1389
1390         garden-o-matic scrolls to the top of the page when you click an accordion item
1391         https://bugs.webkit.org/show_bug.cgi?id=82088
1392
1393         Reviewed by Adam Barth.
1394
1395         Check the scroll position and make sure that the clicked accordion item is visible.
1396
1397         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
1398
1399 2012-03-23  Dirk Pranke  <dpranke@chromium.org>
1400
1401         LayoutTestHelper could get torn down earlier (mostly to reset color profile)
1402         https://bugs.webkit.org/show_bug.cgi?id=80567
1403
1404         Reviewed by Ryosuke Niwa.
1405
1406         Move clean_up_run to occur right after we finish testing and
1407         before we start processing the results. Also makes set_up_run()
1408         and clean_up_run() private since there's no real need for
1409         callers to be aware of them.
1410
1411         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1412         (Manager._set_up_run):
1413         (Manager.run):
1414         (Manager._clean_up_run):
1415         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1416         (ManagerTest.test_http_locking):
1417         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1418         (run):
1419
1420 2012-03-23  Dean Jackson  <dino@apple.com>
1421
1422         Disable CSS_SHADERS in Apple builds
1423         https://bugs.webkit.org/show_bug.cgi?id=81996
1424
1425         Reviewed by Simon Fraser.
1426
1427         Default enableCSSShaders to be 0 on all builds that use
1428         this script. Use the flag to turn it back on.
1429
1430         * Scripts/build-webkit:
1431
1432 2012-03-23  Tony Chang  <tony@chromium.org>
1433
1434         [chromium] rename newwtf target back to wtf
1435         https://bugs.webkit.org/show_bug.cgi?id=82064
1436
1437         Reviewed by Adam Barth.
1438
1439         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1440
1441 2012-03-23  Jessie Berlin  <jberlin@apple.com>
1442
1443         Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on
1444         the Lion Intel WebKit2 Debug bots.
1445         https://bugs.webkit.org/show_bug.cgi?id=81012.
1446
1447         Rubber-stamped by Dan Bernstein.
1448
1449         Add a step to kill the pasteboard server while we investigate what is causing these
1450         assertion failures (to get the bots green).
1451
1452         * BuildSlaveSupport/kill-old-processes:
1453
1454 2012-03-22  Martin Robinson  <mrobinson@igalia.com>
1455
1456         [check-webkit-style] Alphabetical sorting errors in headers are reported for the line after the first out of order header
1457         https://bugs.webkit.org/show_bug.cgi?id=81986
1458
1459         Reviewed by David Levin.
1460
1461         Track whether or not errors are filtered in error handlers. If an
1462         alphabetical sorting error is reported for a header and filtered, try
1463         reporting it for the other potential source of the error. This ensures
1464         that the style bot can still find new sorting errors in both potential
1465         situations:
1466
1467             #include <foo.h> <-- 1. edited line
1468             #include <baz.h> <-- 2. edited line
1469
1470         * Scripts/webkitpy/style/checkers/cpp.py:
1471         (check_include_line): Potentially report the error for both lines,
1472         if the first error is filtered.
1473         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Implement a new style of
1474         test that allows only reporting errors for certain lines. This verifies that
1475         when one line of an ordering error is filtered, the error is reported on the
1476         other line.
1477         (ErrorCollector.__init__):
1478         (ErrorCollector.__call__):
1479         (CppStyleTestBase.perform_lint):
1480         (CppStyleTestBase.perform_language_rules_check):
1481         (CppStyleTestBase.assert_language_rules_check):
1482         (OrderOfIncludesTest.test_check_alphabetical_include_order_errors_reported_for_both_lines):
1483         Added a new test that verifies that when one line of the two lines of an ordering
1484         error occur, the error is reported on the other line.
1485         * Scripts/webkitpy/style/checkers/jsonchecker_unittest.py:
1486         (MockErrorHandler.__call__): Report True because the error is handled.
1487         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1488         (ErrorCollector.__call__): Ditto.
1489         * Scripts/webkitpy/style/checkers/watchlist_unittest.py:
1490         (MockErrorHandler.__call__): Ditto.
1491         * Scripts/webkitpy/style/checkers/xcodeproj_unittest.py:
1492         (TestErrorHandler.__call__): Ditto.
1493         * Scripts/webkitpy/style/checkers/xml_unittest.py:
1494         (MockErrorHandler.__call__): Ditto.
1495         * Scripts/webkitpy/style/error_handlers.py:
1496         (DefaultStyleErrorHandler.__call__): Report True if the error is handled
1497         and False if it is filtered.
1498
1499 2012-03-23  Patrick Gansterer  <paroga@webkit.org>
1500
1501         Build fix for WinCE after r111778.
1502
1503         * WinCELauncher/CMakeLists.txt:
1504
1505 2012-03-23  Peter Beverloo  <peter@chromium.org>
1506
1507         Unreviewed buildfix for the Chromium Android builder.
1508
1509         Disable the ImageDiff target for now. Android needs to build this for
1510         host rather than for target (as other configurations do), but since the
1511         WTF move the dependency chain got significantly more complicated.
1512
1513         I'll work out a proper fix in the following bug, but since several
1514         things are broken due to this, this buildfix should be landed first.
1515         https://bugs.webkit.org/show_bug.cgi?id=82039
1516
1517         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1518
1519 2012-03-23  Grzegorz Czajkowski  <g.czajkowski@samsung.com>
1520
1521         [EFL][DRT] Implement LayoutTestController's methods related with editing.
1522         https://bugs.webkit.org/show_bug.cgi?id=80136
1523
1524         Reviewed by Ryosuke Niwa.
1525
1526         Implements EFL's LayoutTestController methods e.i. setSmartInsertDeleteEnabled,
1527         setSelectTrailingWhitespaceEnabled.
1528
1529         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1530         (LayoutTestController::setSmartInsertDeleteEnabled):
1531         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
1532
1533 2012-03-22  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
1534
1535         Convert hasGrammarMarker to use Internals interface
1536         https://bugs.webkit.org/show_bug.cgi?id=82004
1537
1538         Reviewed by Ryosuke Niwa.
1539
1540         Remove hasGrammarMarker functions, because it is able to work in the
1541         cross-port way through the Internals interface.
1542
1543         * DumpRenderTree/LayoutTestController.cpp:
1544         (LayoutTestController::staticFunctions):
1545         * DumpRenderTree/LayoutTestController.h:
1546         (LayoutTestController):
1547         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
1548         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1549         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
1550         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
1551         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1552         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
1553
1554 2012-03-22  Adam Barth  <abarth@webkit.org>
1555
1556         sync-master-with-upstream creates non-linear git history
1557         https://bugs.webkit.org/show_bug.cgi?id=82006
1558
1559         Reviewed by Julien Chaffraix.
1560
1561         Previously this script would create a merge commit at the HEAD of
1562         origin/master, causing your fork of WebKit to diverge from upstream.
1563         The new version of this script pushes directly from upstream to origin
1564         (without round-tripping through any local branches), avoiding this
1565         problem.
1566
1567         * Scripts/sync-master-with-upstream:
1568
1569 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1570
1571         Unreviewed build fix. Add knowledge of the WTF project to
1572         the autotools build system scripts so that build-jsc works
1573         correctly for GTK+.
1574
1575         * Scripts/webkitdirs.pm:
1576         (buildAutotoolsProject): handle the WTF project, build the
1577         libWTF.la target.
1578         (buildGtkProject): add WTF to the list of supported projects.
1579
1580 2012-03-22  Eric Seidel  <eric@webkit.org>
1581
1582         Make svn-apply re-write JavaScriptCore/wtf to WTF/wtf
1583         https://bugs.webkit.org/show_bug.cgi?id=81984
1584
1585         Reviewed by Adam Barth.
1586
1587         * Scripts/VCSUtils.pm:
1588         (adjustPathForRecentRenamings):
1589
1590 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1591
1592         [CMake] Fix build-jsc after r111778 and r111797.
1593         https://bugs.webkit.org/show_bug.cgi?id=82001
1594
1595         Reviewed by Filip Pizlo.
1596
1597         Contrary to the comment in build-jsc, buildCMakeProjectOrExit()
1598         does return if no error occurred, and build-jsc failed with
1599         "building not defined for this platform".
1600
1601         * Scripts/build-jsc: Always exit after caling
1602         buildCMakeProjectOrExit().
1603         * Scripts/webkitdirs.pm:
1604         (buildCMakeProjectOrExit): Return 0 if no error occurred.
1605
1606 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1607
1608         [CMake] Unreviewed build fix after r111778.
1609
1610         * DumpRenderTree/efl/CMakeLists.txt: Replace
1611         ${JAVASCRIPTCORE_DIR}/wtf with ${WTF_DIR}/wtf in the include
1612         paths.
1613         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Use the right
1614         wtf include.
1615         * DumpRenderTree/efl/WorkQueueItemEfl.cpp: Ditto.
1616         * WinCELauncher/CMakeLists.txt: Replace ${JAVASCRIPTCORE_DIR}/wtf
1617         with ${WTF_DIR}/wtf in the include paths.
1618
1619 2012-03-22  Filip Pizlo  <fpizlo@apple.com>
1620
1621         Tools/Scripts/build-jsc doesn't work after WTF move
1622         https://bugs.webkit.org/show_bug.cgi?id=81999
1623
1624         Reviewed by Eric Seidel.
1625         
1626         For ports that require WTF and JSC to be built as two projects and where
1627         the machinery to build both does not already exist, teach the tool to do
1628         so for us.
1629
1630         * Scripts/build-jsc:
1631         (buildMyProject):
1632
1633 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1634
1635         Fix accidental missing line in r111793.
1636
1637         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1638         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
1639         Remove base.values. It was unused and duplicates the builtin Object.values.
1640         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1641         Accidentally deleted this line before committing r111793.
1642
1643 2012-03-22  Ojan Vafai  <ojan@chromium.org>
1644
1645         Move garden-o-matic progress feedback from non-modal dialogs to a status console anchored to the bottom of the page
1646         https://bugs.webkit.org/show_bug.cgi?id=81983
1647
1648         Reviewed by Adam Barth.
1649
1650         The dialogs get in the way if you are doing multiple rebaselines. Also,
1651         a number of people didn't realize that you could do multiple rebaselines
1652         in parallel. A non-modal dialog is just confusing.
1653
1654         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
1655         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
1656         If we're only rebaselining 1 test, show the test name in the initial message. Otherwise,
1657         show the number of tests being rebaselined.
1658
1659         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
1660         Turn MessageBox into StatusArea. StatusArea is a singleton and messages in the StatusArea are grouped
1661         by ID. That way, the results of each UI action (e.g. clicking rebaseline) are grouped together
1662         into a single area.
1663
1664         Once addFinalMessage has been called for all IDs, we show the close button. Clicking close
1665         also serves to clear all teh content in the StatusArea. 
1666
1667         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
1668         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
1669
1670 2012-03-22  Tony Chang  <tony@chromium.org>
1671
1672         Unreviewed, fix chromium build after wtf move.
1673
1674         Only use newwtf, remove references to wtf.
1675
1676         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1677
1678 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1679
1680         Actually move WTF files to their new home
1681         https://bugs.webkit.org/show_bug.cgi?id=81844
1682
1683         [Qt] Unreviewed buildfix after r111778.
1684
1685         * DumpRenderTree/qt/DumpRenderTree.pro:
1686
1687 2012-03-22  Dan Bernstein  <mitz@apple.com>
1688
1689         Font fallback in WebKitTestRunner depends on user settings
1690         https://bugs.webkit.org/show_bug.cgi?id=81978
1691
1692         Reviewed by Jessie Berlin.
1693
1694         * WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:
1695         (WTR::InjectedBundle::platformInitialize): Added an AppleLanguages key
1696         with a value consisting of the array ("en") to the override defaults
1697         dictionary. This is the same value used in DumpRenderTree, and it
1698         ensures that font fallback follows a consistent path that is not
1699         affected by the language preferences of the user running the test.
1700
1701 2012-03-22  Andy Estes  <aestes@apple.com>
1702
1703         check-for-inappropriate-objc-class-names should ignore the class defined by libarclite.
1704         https://bugs.webkit.org/show_bug.cgi?id=81971
1705
1706         Reviewed by Dan Bernstein.
1707
1708         Xcode will automatically link ObjC binaries against libarclite in some
1709         cases, which defines a class called __ARCLite__.
1710         check-for-inappropriate-objc-class-names shouldn't complain about this
1711         class name.
1712
1713         * Scripts/check-for-inappropriate-objc-class-names:
1714
1715 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1716
1717         new-run-webkit-tests blames the wrong test for crashing
1718         https://bugs.webkit.org/show_bug.cgi?id=81951
1719
1720         Reviewed by Adam Barth.
1721
1722         NRWT wasn't extracting the WebProcess pid correct; should've
1723         used re.search instead of re.match :(. Add more tests!
1724
1725         * Scripts/webkitpy/layout_tests/port/mac.py:
1726         (MacPort._get_crash_log):
1727         * Scripts/webkitpy/layout_tests/port/webkit.py:
1728         (WebKitDriver.has_crashed):
1729         (WebKitDriver._check_for_driver_crash):
1730         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1731         (WebKitDriverTest.test_no_timeout):
1732         (WebKitDriverTest):
1733         (WebKitDriverTest.test_check_for_driver_crash):
1734         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess):
1735         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.__init__):
1736         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.pid):
1737         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.name):
1738         (WebKitDriverTest.test_check_for_driver_crash.FakeServerProcess.has_crashed):
1739         (WebKitDriverTest.test_check_for_driver_crash.assert_crash):
1740
1741 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1742
1743         [EFL] [DRT] Implement LayoutTestController::markerTextForListItem()
1744         https://bugs.webkit.org/show_bug.cgi?id=81512
1745
1746         Adding missing implementation markerTextForListItem to EFL's
1747         LayoutTestController so that we can unskip related tests from
1748         the skip list.
1749
1750         Reviewed by Antonio Gomes.
1751
1752         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1753         (LayoutTestController::markerTextForListItem): Implemented.
1754
1755 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1756
1757         [EFL] Enable view mode media feature layout tests
1758         https://bugs.webkit.org/show_bug.cgi?id=81205
1759
1760         Adding missing implementation setViewModeMediaFeature to EFL's 
1761         LayoutTestController so that we can unskip view mode 
1762         media feature layout tests.
1763
1764         Reviewed by Antonio Gomes.
1765
1766         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1767         (LayoutTestController::setViewModeMediaFeature): Implemented.
1768
1769 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1770
1771         nrwt: crash while stopping layout test helper on apple mac lion
1772         https://bugs.webkit.org/show_bug.cgi?id=81949
1773
1774         Reviewed by Eric Seidel.
1775
1776         stop_helper() didn't work right if the helper process wasn't
1777         still running or didn't respond properly to being shut down. Fix
1778         this and add some tests. Also add code to MockExecutive() to be
1779         able to mock executive.popen() properly.
1780
1781         * Scripts/webkitpy/common/system/executive_mock.py:
1782         (MockProcess.__init__):
1783         (MockProcess.wait):
1784         (MockExecutive.__init__):
1785         (MockExecutive.popen):
1786         * Scripts/webkitpy/common/system/systemhost_mock.py:
1787         (MockSystemHost.__init__):
1788         * Scripts/webkitpy/layout_tests/port/mac.py:
1789         (MacPort.start_helper):
1790         (MacPort.stop_helper):
1791         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1792         (test_get_crash_log):
1793         (test_helper_starts):
1794         (test_helper_fails_to_start):
1795         (test_helper_fails_to_stop):
1796         (test_helper_fails_to_stop.bad_waiter):
1797
1798 2012-03-22  Gustavo Noronha Silva  <gns@gnome.org>
1799
1800         [GTK] Add make to the jhbuild moduleset
1801         https://bugs.webkit.org/show_bug.cgi?id=81769
1802
1803         Reviewed by Martin Robinson.
1804
1805         Mostly based on a patch by Carlos Garcia Campos <cgarcia@igalia.com>.
1806         Add GNU make with custom patches to fix several issues:
1807
1808         - make-3.82-arg-list-length.patch: Patch by Ralf Wildenhues to
1809           fix the argument list limit of GNU make when running make
1810           distcheck.
1811         - make-3.82-parallel-build.patch: Patch by Kamil Mierzejewski
1812           and Jeremy Devenport to fix parallel build issues present in
1813           GNU make 3.82
1814
1815         * gtk/jhbuild.modules: Added make with patches.
1816         * gtk/patches/make-3.82-arg-list-length.patch: Added.
1817         * gtk/patches/make-3.82-parallel-build.patch: Added.
1818
1819 2012-03-22  Kevin Ollivier  <kevino@theolliviers.com>
1820
1821         [wx] Unreviewed. Adding Source/WTF to the build and updating
1822         waf port list.
1823
1824         * DumpRenderTree/wscript:
1825         * waf/build/settings.py:
1826
1827 2012-03-22  Dirk Pranke  <dpranke@chromium.org>
1828
1829         Re-land fix for r81603 yet again (!) with latest build fix.
1830         https://bugs.webkit.org/show_bug.cgi?id=81603
1831
1832         Unreviewed, build fix.
1833
1834         The latest problem was a bad sprintf() in
1835         BasePort._get_crash_log(). I've reworked that routine and added
1836         unit tests for coverage.
1837
1838         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1839         (SingleTestRunner._handle_error):
1840         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1841         (write_test_result):
1842         (TestResultWriter.write_crash_log):
1843         * Scripts/webkitpy/layout_tests/port/base.py:
1844         (Port.diff_text):
1845         (Port._get_crash_log):
1846         * Scripts/webkitpy/layout_tests/port/chromium.py:
1847         (ChromiumDriver.run_test):
1848         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1849         (ChromiumDriverTest.test_crash_log):
1850         * Scripts/webkitpy/layout_tests/port/driver.py:
1851         (DriverOutput.__init__):
1852         * Scripts/webkitpy/layout_tests/port/mac.py:
1853         (MacPort.is_lion):
1854         (MacPort._get_crash_log):
1855         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1856         (test_default_child_processes):
1857         (test_get_crash_log):
1858         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1859         (MockDRTPortTest.test_get_crash_log):
1860         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1861         (PortTestCase.test_get_crash_log):
1862         * Scripts/webkitpy/layout_tests/port/test.py:
1863         (TestDriver.run_test):
1864         * Scripts/webkitpy/layout_tests/port/webkit.py:
1865         (WebKitDriver.__init__):
1866         (WebKitDriver._start):
1867         (WebKitDriver.has_crashed):
1868         (WebKitDriver._check_for_driver_crash):
1869         (WebKitDriver.run_test):
1870
1871 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1872
1873         [Qt] Enable support for handling database quota's in the MiniBrowser
1874         https://bugs.webkit.org/show_bug.cgi?id=81928
1875
1876         Reviewed by Simon Hausmann.
1877
1878         * MiniBrowser/qt/qml/BrowserWindow.qml:
1879
1880 2012-03-22  Sudarsana Nagineni  <sudarsana.nagineni@linux.intel.com>
1881
1882         [EFL] Implement LayoutTestController::setEditingBehavior
1883         https://bugs.webkit.org/show_bug.cgi?id=81124
1884
1885         Adding missing implementation setEditingBehavior to EFL's
1886         LayoutTestController. This implementation allows us to
1887         unskip some tests from the skip list.
1888
1889         Reviewed by Martin Robinson.
1890
1891         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1892         (LayoutTestController::setEditingBehavior): Implemented.
1893
1894 2012-03-22  Carlos Garcia Campos  <cgarcia@bb-webkit-rel-64.local.igalia.com>
1895
1896         [GTK] Use the angle-bracket form to include wtf headers
1897         https://bugs.webkit.org/show_bug.cgi?id=81884
1898
1899         Reviewed by Eric Seidel.
1900
1901         Use #include <wtf/foo> instead of #include "foo".
1902
1903         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1904         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1905         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1906         * DumpRenderTree/gtk/EditingCallbacks.cpp:
1907         * DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
1908         * WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:
1909
1910 2012-03-22  Alexander Færøy  <alexander.faeroy@nokia.com>
1911
1912         Unreviewed, add Michael Brüning and Allan Jensen to committers.py.
1913
1914         * Scripts/webkitpy/common/config/committers.py:
1915
1916 2012-03-22  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
1917
1918         [Qt] Enable FAST_MOBILE_SCROLLING when scrolling is delegated.
1919         https://bugs.webkit.org/show_bug.cgi?id=81889
1920
1921         Reviewed by Kenneth Rohde Christiansen.
1922
1923         Enable FAST_MOBILE_SCROLLING feature.
1924
1925         * qmake/mkspecs/features/features.prf:
1926
1927 2012-03-22  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
1928
1929         [EFL] DRT: Do not pass a relative path to DOWNLOADED_FONTS_DIR.
1930         https://bugs.webkit.org/show_bug.cgi?id=81900
1931
1932         Reviewed by Gustavo Noronha Silva.
1933
1934         r111598 added the DOWNLOADED_FONTS_DIR define, but it used a relative
1935         path which assumed a certain directory structure for the build
1936         directory which may noy be true.
1937
1938         Instead, hardcode the path to the downloaded fonts starting from the
1939         top-level source directory. While this solution is also far from
1940         perfect, it makes it more consistent and reliable, and easier to
1941         tackle at once with the rest of bug 81475.
1942
1943         * DumpRenderTree/efl/CMakeLists.txt:
1944
1945 2012-03-22  Carlos Garcia Campos  <cgarcia@igalia.com>
1946
1947         [GTK] Allow to run any jhbuild command with jhbuild-wrapper script
1948         https://bugs.webkit.org/show_bug.cgi?id=81888
1949
1950         Reviewed by Philippe Normand.
1951
1952         * Scripts/update-webkit-libs-jhbuild:
1953         (runJhbuild): Helper function to run jhbuild with a given command.
1954         Instead of running jhbuild directly, use the jhbuild-wrapper, so
1955         that we don't need to run it first to check whether jhbuild is
1956         installed or not.
1957         * efl/run-with-jhbuild: Call jhbuild-wrapper with run command.
1958         * gtk/run-with-jhbuild: Ditto.
1959         * jhbuild/jhbuild-wrapper: Allow to execute any jhbuild command,
1960         not only run.
1961
1962 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1963
1964         Unreviewed, test-webkitpy fix after r111661.
1965
1966         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1967         (MainTest.test_crash_log):
1968         (MainTest.test_web_process_crash_log):
1969
1970 2012-03-22  Philippe Normand  <pnormand@igalia.com>
1971
1972         Unreviewed, test-webkitpy fix after r111642.
1973
1974         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1975         (ChromiumDriverTest.setUp):
1976         (ChromiumDriverTest.test_two_drivers):
1977
1978 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
1979
1980         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
1981         https://bugs.webkit.org/show_bug.cgi?id=81603
1982
1983         Unreviewed rolling out r111609 and part of r111615,
1984         because it broke NRWT on Qt-WK2 platform.
1985
1986         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
1987         (SingleTestRunner._handle_error):
1988         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
1989         (write_test_result):
1990         (TestResultWriter.write_crash_report):
1991         * Scripts/webkitpy/layout_tests/port/base.py:
1992         (Port.is_crash_reporter):
1993         (Port._driver_class):
1994         * Scripts/webkitpy/layout_tests/port/chromium.py:
1995         (ChromiumDriver.run_test):
1996         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
1997         (ChromiumDriverTest.test_crashed_process_name):
1998         * Scripts/webkitpy/layout_tests/port/driver.py:
1999         (DriverOutput.__init__):
2000         * Scripts/webkitpy/layout_tests/port/mac.py:
2001         (MacPort):
2002         (MacPort.is_crash_reporter):
2003         (MacPort.release_http_lock):
2004         * Scripts/webkitpy/layout_tests/port/test.py:
2005         (TestDriver.run_test):
2006         * Scripts/webkitpy/layout_tests/port/webkit.py:
2007         (WebKitDriver.__init__):
2008         (WebKitDriver._start):
2009         (WebKitDriver.has_crashed):
2010         (WebKitDriver._check_for_driver_crash):
2011         (WebKitDriver.run_test):
2012
2013 2012-03-22  Csaba Osztrogonác  <ossy@webkit.org>
2014
2015         [Qt] Add full platforms to Qt buildslaves
2016         https://bugs.webkit.org/show_bug.cgi?id=81877
2017
2018         Reviewed by Ryosuke Niwa.
2019
2020         * BuildSlaveSupport/build.webkit.org-config/config.json:
2021
2022 2012-03-21  Johnny Ding  <jnd@chromium.org>
2023
2024         Customize layout test timeout value for different ports.
2025         https://bugs.webkit.org/show_bug.cgi?id=79859
2026
2027         Reviewed by Ojan Vafai.
2028
2029         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2030         (Manager):
2031         * Scripts/webkitpy/layout_tests/port/base.py:
2032         (Port.default_test_timeout_ms):
2033         * Scripts/webkitpy/layout_tests/port/chromium.py:
2034         (ChromiumDriver.stop):
2035         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
2036         (ChromiumAndroidPort.default_test_timeout_ms):
2037         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2038         (_set_up_derived_options):
2039
2040 2012-03-21  Mark Rowe  <mrowe@apple.com>
2041
2042         Fix the build.
2043
2044         Ensure that our test tools don't try and use RTTI unnecessarily.
2045
2046         * DumpRenderTree/mac/Configurations/Base.xcconfig:
2047         * TestWebKitAPI/Configurations/Base.xcconfig:
2048         * WebKitTestRunner/Configurations/Base.xcconfig:
2049
2050 2012-03-21  Adam Barth  <abarth@webkit.org>
2051
2052         WebKit should have some scripts to smooth interactions with GitHub
2053         https://bugs.webkit.org/show_bug.cgi?id=81842
2054
2055         Reviewed by Eric Seidel.
2056
2057         These scripts support the workflow in
2058         https://trac.webkit.org/wiki/UsingGitHub by automating the
2059         configuration and syncing operations needed for a GitHub fork of WebKit
2060         to track the main WebKit repository.
2061
2062         * Scripts/configure-github-as-upstream: Added.
2063         * Scripts/sync-master-with-upstream: Added.
2064
2065 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2066
2067         WTR - log the pid of a crashing WebProcess
2068         https://bugs.webkit.org/show_bug.cgi?id=81575
2069
2070         Reviewed by Alexey Proskuryakov.
2071
2072         This allows run-webkit-tests to find the right crash log for a
2073         given crashed WebProcess in WebKit2.
2074
2075         * WebKitTestRunner/TestController.cpp:
2076         (WTR::exceededDatabaseQuota):
2077         (WTR::TestController::createOtherPage):
2078         (WTR::TestController::initialize):
2079         (WTR::TestController::resetStateToConsistentValues):
2080         (WTR::TestController::runTest):
2081         (WTR::TestController::processDidCrash):
2082
2083 2012-03-21  Enrica Casucci  <enrica@apple.com>
2084
2085         WebKitURLWithTitles pasteboard format should support URLs containing Emoji characters.
2086         https://bugs.webkit.org/show_bug.cgi?id=81835
2087         <rdar://problem/11082749>
2088
2089         Reviewed by Brady Eidson.
2090
2091         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
2092         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
2093         (TestWebKitAPI::contextMenuCopyLink):
2094
2095 2012-03-21  Alexandru Chiculita  <achicu@adobe.com>
2096
2097         [CSS Shaders] Make CSS Shaders compile on Chromium
2098         https://bugs.webkit.org/show_bug.cgi?id=81435
2099
2100         Reviewed by Stephen White.
2101         
2102         I've added WebKitCSSCustomFilterEnabled support for the layoutTestController.overridePreference function.
2103
2104         * DumpRenderTree/chromium/LayoutTestController.cpp:
2105         (LayoutTestController::overridePreference):
2106         * DumpRenderTree/chromium/WebPreferences.cpp:
2107         (WebPreferences::reset):
2108         (WebPreferences::applyTo):
2109         * DumpRenderTree/chromium/WebPreferences.h:
2110         (WebPreferences):
2111
2112 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2113
2114         fix regression on gtk port introduced by r111609
2115         https://bugs.webkit.org/show_bug.cgi?id=81839
2116
2117         Reviewed by Philippe Normand.
2118
2119         r111609 should've initialized a couple of missing fields in
2120         WebKitDriver, and also should've checked for method overrides
2121         in the other ports :(.
2122
2123         * Scripts/webkitpy/layout_tests/port/gtk.py:
2124         (GtkDriver._start):
2125         * Scripts/webkitpy/layout_tests/port/webkit.py:
2126         (WebKitDriver.__init__):
2127         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2128         (MainTest.test_crash_log):
2129         (MainTest.test_web_process_crash_log):
2130
2131 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2132
2133         webkitpy: clean up actually getting crash logs from DRT/WTR crashes
2134         https://bugs.webkit.org/show_bug.cgi?id=81603
2135
2136         Reviewed by Adam Barth.
2137
2138         Add a new _get_crash_log() overridable method on the Port
2139         object to customize how we fetch crash logs per port.
2140         Mac will now slow-spin with a timeout waiting for the crash log, rather
2141         than waiting for the ReportCrash process to exit (it appears
2142         that the mac will manage multiple crashes with a single
2143         ReportCrash process, the process waits around longer than
2144         necessary, presumably to avoid thrashing if processes are
2145         repeatedly crashing).
2146
2147         Also, add the DriverOutput should contain the crash log and other info,
2148         which is created in a port-specific manner but can then be
2149         treated generically. Previously single_test_runner would get
2150         told that something crashed and attempt to do something to get
2151         the crash log, but it didn't have the information it needed to
2152         od the right thing; better to make the driver hand back the
2153         right info.
2154
2155         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2156         (SingleTestRunner._handle_error):
2157         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2158         (write_test_result):
2159         (TestResultWriter.write_crash_log):
2160         * Scripts/webkitpy/layout_tests/port/base.py:
2161         (Port.diff_text):
2162         (Port._get_crash_log):
2163         * Scripts/webkitpy/layout_tests/port/chromium.py:
2164         (ChromiumDriver.run_test):
2165         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2166         (ChromiumDriverTest.test_crash_log):
2167         * Scripts/webkitpy/layout_tests/port/driver.py:
2168         (DriverOutput.__init__):
2169         * Scripts/webkitpy/layout_tests/port/mac.py:
2170         (MacPort.is_lion):
2171         (MacPort._get_crash_log):
2172         * Scripts/webkitpy/layout_tests/port/test.py:
2173         (TestDriver.run_test):
2174         * Scripts/webkitpy/layout_tests/port/webkit.py:
2175         (WebKitDriver.__init__):
2176         (WebKitDriver._start):
2177         (WebKitDriver.has_crashed):
2178         (WebKitDriver._check_for_driver_crash):
2179         (WebKitDriver.run_test):
2180
2181 2012-03-21  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
2182
2183         [EFL] Use jhbuild downloaded fonts instead of hardcoded system font paths
2184         https://bugs.webkit.org/show_bug.cgi?id=81507
2185
2186         Now that we're only loading from one directory, no need to
2187         iterate over a list of directories anymore.
2188
2189         Reviewed by Martin Robinson.
2190
2191         * DumpRenderTree/efl/CMakeLists.txt:
2192         * DumpRenderTree/efl/FontManagement.cpp:
2193         (addFontDirectory):
2194         (addFontsToEnvironment):
2195
2196 2012-03-21  Dirk Pranke  <dpranke@chromium.org>
2197
2198         webkitpy: get ServerProcess out of the reportcrash business
2199         https://bugs.webkit.org/show_bug.cgi?id=81600
2200
2201         Unreviewed, build fix.
2202
2203         Re-land r111307 and r111293 with another fix for a crash in NRWT;
2204         we need to check if the driver has crashed before attempting to
2205         read from it.
2206
2207
2208         * Scripts/webkitpy/layout_tests/port/server_process.py:
2209         (ServerProcess._reset):
2210         (ServerProcess._handle_possible_interrupt):
2211         (ServerProcess.write):
2212         (ServerProcess.read_stdout):
2213         (ServerProcess.has_crashed):
2214         (ServerProcess._read):
2215         (ServerProcess.stop):
2216         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2217         (TrivialMockPort.check_for_leaks):
2218         (TestServerProcess.test_broken_pipe):
2219         * Scripts/webkitpy/layout_tests/port/webkit.py:
2220         (WebKitPort._read_image_diff):
2221         (WebKitDriver.has_crashed):
2222         (WebKitDriver._check_for_driver_crash):
2223         (WebKitDriver.run_test):
2224         (WebKitDriver._read_block):
2225         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2226         (MockServerProcess.__init__):
2227         (MockServerProcess):
2228         (MockServerProcess.has_crashed):
2229
2230 2012-03-21  Zeno Albisser  <zeno@webkit.org>
2231
2232         [Qt][Mac] ranlib segfaults when creating symbol tables for libWebCore.a.
2233         https://bugs.webkit.org/show_bug.cgi?id=81750
2234
2235         Building WebCore on mac currently creates an archive that is bigger than
2236         4GB. But ranlib fails to create a symbol table for such a big archive,
2237         even on 64bit machines.
2238         Therefore we start using SVGAllInOne.cpp when building debug
2239         on mac. This reduces the size of the archive by about 300MB.
2240
2241         Reviewed by Tor Arne Vestbø.
2242
2243         * qmake/mkspecs/features/mac/default_pre.prf:
2244
2245 2012-03-21  Szilard Ledan  <szledan@inf.u-szeged.hu>
2246
2247         Updated the address list of committers.py.
2248
2249         Reviewed by Csaba Osztrogonác.
2250
2251         * Scripts/webkitpy/common/config/committers.py:
2252
2253 2012-03-21  W. James MacLean  <wjmaclean@chromium.org>
2254
2255         Updating my nick in committers file.
2256
2257         Unreviewed.
2258
2259         * Scripts/webkitpy/common/config/committers.py:
2260
2261 2012-03-21  Peter Beverloo  <peter@chromium.org>
2262
2263         [Chromium] Remove Android build-fix when the proper fix rolled into WebKit
2264         https://bugs.webkit.org/show_bug.cgi?id=80861
2265
2266         Reviewed by Tony Chang.
2267
2268         Remove most Android-specific logic from webkitdirs.pm in favor of
2269         handling this during project generation time.
2270
2271         * Scripts/webkitdirs.pm:
2272         (buildChromiumMakefile):
2273
2274 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2275
2276         [GTK] Allow running javascript from location bar in MiniBrowser
2277         https://bugs.webkit.org/show_bug.cgi?id=81331
2278
2279         Reviewed by Martin Robinson.
2280
2281         * MiniBrowser/gtk/BrowserWindow.c:
2282         (activateUriEntryCallback): Use browser_window_load_uri().
2283         (browser_window_load_uri): Check whether uri starts with
2284         javascript: and run the given script, or just call
2285         webkit_web_view_load_uri() to load the given uri.
2286         * MiniBrowser/gtk/BrowserWindow.h:
2287         * MiniBrowser/gtk/main.c:
2288         (createBrowserWindow): Use browser_window_load_uri().
2289
2290 2012-03-21  Carlos Garcia Campos  <cgarcia@igalia.com>
2291
2292         [GTK] Add webkit_web_view_run_javascript() to WebKit2 GTK+
2293         https://bugs.webkit.org/show_bug.cgi?id=75543
2294
2295         Reviewed by Martin Robinson.
2296
2297         * MiniBrowser/gtk/GNUmakefile.am: Add javascriptcore_cppflags to
2298         MiniBrowser CPP flags.
2299
2300 2012-03-20  Eric Seidel  <eric@webkit.org>
2301
2302         Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf
2303         https://bugs.webkit.org/show_bug.cgi?id=80911
2304
2305         Reviewed by Adam Barth.
2306
2307         Update include paths for Source/WTF.
2308
2309         * DumpRenderTree/efl/CMakeLists.txt:
2310         * DumpRenderTree/qt/DumpRenderTree.pro:
2311         * QtTestBrowser/QtTestBrowser.pro:
2312         * Scripts/build-webkit: Build Source/WTF on AppleWin as well.
2313
2314 2012-03-20  Adam Barth  <abarth@webkit.org>
2315
2316         svn-apply should move inflight patches from WebCore/webaudio to WebCore/Modules/webaudio
2317         https://bugs.webkit.org/show_bug.cgi?id=81567
2318
2319         Reviewed by Eric Seidel.
2320
2321         This temporary measure should help us avoid disrupting the folks
2322         working on WebAudio when we move the webaudio directory to Modules.
2323
2324         * Scripts/VCSUtils.pm:
2325         (adjustPathForRecentRenamings):
2326         (parseGitDiffHeader):
2327         (parseSvnDiffHeader):
2328
2329 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2330
2331         Unreviewed, rolling out r111453.
2332         http://trac.webkit.org/changeset/111453
2333         https://bugs.webkit.org/show_bug.cgi?id=81600
2334
2335         still broken
2336
2337         * Scripts/webkitpy/layout_tests/port/server_process.py:
2338         (ServerProcess._reset):
2339         (ServerProcess.handle_interrupt):
2340         (ServerProcess.write):
2341         (ServerProcess._check_for_crash):
2342         (ServerProcess._handle_timeout):
2343         (ServerProcess._check_for_abort):
2344         (ServerProcess._read):
2345         (ServerProcess.stop):
2346         (ServerProcess):
2347         (ServerProcess.set_crashed):
2348         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2349         (TrivialMockPort.is_crash_reporter):
2350         (TestServerProcess.test_broken_pipe):
2351         * Scripts/webkitpy/layout_tests/port/webkit.py:
2352         (WebKitPort._read_image_diff):
2353         (WebKitDriver.has_crashed):
2354         (WebKitDriver._check_for_driver_crash):
2355         (WebKitDriver):
2356         (WebKitDriver._detected_crash):
2357         (WebKitDriver._subprocess_crashed):
2358         (WebKitDriver._crashed_process_name):
2359         (WebKitDriver.run_test):
2360         (WebKitDriver._read_block):
2361         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2362         (MockServerProcess.__init__):
2363
2364 2012-03-20  Dirk Pranke  <dpranke@chromium.org>
2365
2366         webkitpy: get ServerProcess out of the reportcrash business
2367         https://bugs.webkit.org/show_bug.cgi?id=81600
2368
2369         Unreviewed, build fix.
2370
2371         Re-land r111307 and r111293 with fix for crash in NRWT; I
2372         was failing to set the crashing process name.
2373
2374         * Scripts/webkitpy/layout_tests/port/server_process.py:
2375         (ServerProcess._reset):
2376         (ServerProcess._handle_possible_interrupt):
2377         (ServerProcess.write):
2378         (ServerProcess.read_stdout):
2379         (ServerProcess.has_crashed):
2380         (ServerProcess._read):
2381         (ServerProcess.stop):
2382         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2383         (TrivialMockPort.check_for_leaks):
2384         (TestServerProcess.test_broken_pipe):
2385         * Scripts/webkitpy/layout_tests/port/webkit.py:
2386         (WebKitPort._read_image_diff):
2387         (WebKitDriver.has_crashed):
2388         (WebKitDriver._check_for_driver_crash):
2389         (WebKitDriver.run_test):
2390         (WebKitDriver._read_block):
2391         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2392         (MockServerProcess.__init__):
2393         (MockServerProcess):
2394         (MockServerProcess.has_crashed):
2395
2396 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2397
2398         [mac] Restore color space switching code to run-webkit-tests
2399         https://bugs.webkit.org/show_bug.cgi?id=80571
2400         <rdar://problem/11008529>
2401
2402         Reviewed by Simon Fraser.
2403
2404         http://trac.webkit.org/changeset/111429 broke tools build on Snow Leopard.
2405         
2406         Use the old CM* API on Snow Leopard, since CGDisplayCreateUUIDFromDisplayID
2407         didn't exist until Lion.
2408
2409         * DumpRenderTree/mac/LayoutTestHelper.m:
2410         (installLayoutTestColorProfile):
2411         (restoreUserColorProfile):
2412
2413 2012-03-20  Adele Peterson  <adele@apple.com>
2414
2415         Update the last test to use EXPECT_WK_STREQ.
2416
2417         Reviewed by Dan Bernstein.
2418
2419         * TestWebKitAPI/Tests/mac/AttributedString.mm: (TestWebKitAPI::TEST):
2420
2421 2012-03-20  Adele Peterson  <adele@apple.com>
2422
2423         "Attempt to insert nil value " exception when calling attributed string APIs on content with a custom font
2424         https://bugs.webkit.org/show_bug.cgi?id=81630
2425         <rdar://problem/10650660>
2426
2427         Reviewed by Alexey Proskuryakov.
2428
2429         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2430         * TestWebKitAPI/Tests/mac/Ahem.ttf: Added.
2431         * TestWebKitAPI/Tests/mac/AttributedString.mm: Added.
2432         (-[AttributedStringTest webView:didFinishLoadForFrame:]):
2433         (TestWebKitAPI):
2434         (TestWebKitAPI::didFinishLoadForFrame):
2435         (TestWebKitAPI::TEST):
2436         * TestWebKitAPI/Tests/mac/attributedStringCustomFont.html: Added.
2437
2438 2012-03-20  Eric Seidel  <eric@webkit.org>
2439
2440         Add link on queues.webkit.org to kov's fancy queue-depth graph
2441         https://bugs.webkit.org/show_bug.cgi?id=81690
2442
2443         Reviewed by Adam Barth.
2444
2445         * QueueStatusServer/templates/recentstatus.html:
2446
2447 2012-03-20  Tim Horton  <timothy_horton@apple.com>
2448
2449         [mac] Restore color space switching code to run-webkit-tests
2450         https://bugs.webkit.org/show_bug.cgi?id=80571
2451         <rdar://problem/11008529>
2452
2453         Reviewed by Simon Fraser.
2454
2455         Make use of the layout test helper hooks to restore the Mac system-wide
2456         color profile switching code. This makes WkTR and DRT results match, and
2457         causes the main display's color profile to not affect results on Lion.
2458
2459         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
2460         * DumpRenderTree/mac/LayoutTestHelper.m: Added.
2461         (installLayoutTestColorProfile):
2462         (restoreUserColorProfile):
2463         (simpleSignalHandler):
2464         (main):
2465         * Scripts/webkitpy/layout_tests/port/mac.py:
2466         (MacPort.release_http_lock):
2467         (MacPort):
2468         (MacPort._path_to_helper):
2469         (MacPort.start_helper):
2470         (MacPort.stop_helper):
2471         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
2472         (WTR::PlatformWebView::PlatformWebView):
2473         (WTR::PlatformWebView::windowSnapshotImage):
2474
2475 2012-02-24  Darin Fisher  <darin@chromium.org>
2476
2477         Add more Chromium WebKit API reviewers
2478         https://bugs.webkit.org/show_bug.cgi?id=81576
2479
2480         Reviewed by Adam Barth and David Levin.
2481
2482         Simplify watchlist to lump all WebKit API into one.
2483
2484         * Scripts/webkitpy/common/config/watchlist:
2485
2486 2012-03-20  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2487
2488         Convert hasSpellingMarker to use Internals interface.
2489         https://bugs.webkit.org/show_bug.cgi?id=81300
2490
2491         Reviewed by Ryosuke Niwa.
2492
2493         Remove hasSpellingMarker functions, because it is able to work in the
2494         cross-port way through the Internals interface.
2495
2496         * DumpRenderTree/LayoutTestController.cpp:
2497         (LayoutTestController::staticFunctions):
2498         * DumpRenderTree/LayoutTestController.h:
2499         (LayoutTestController):
2500         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2501         * DumpRenderTree/chromium/LayoutTestController.cpp:
2502         (LayoutTestController::LayoutTestController):
2503         * DumpRenderTree/chromium/LayoutTestController.h:
2504         (LayoutTestController):
2505         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
2506         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
2507         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2508         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2509         * DumpRenderTree/qt/LayoutTestControllerQt.h:
2510         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
2511         * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
2512
2513 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2514
2515         Add new Qt-WK2 performance tester bot to build.webkit.org
2516         https://bugs.webkit.org/show_bug.cgi?id=80842
2517
2518         Reviewed by Ryosuke Niwa.
2519
2520         * BuildSlaveSupport/build.webkit.org-config/config.json:
2521         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2522         (RunAndUploadPerfTestsWebKit2):
2523         (RunAndUploadPerfTestsWebKit2.start):
2524         (BuildAndPerfTestWebKit2Factory):
2525         (BuildAndPerfTestWebKit2Factory.__init__):
2526         (DownloadAndPerfTestWebKit2Factory):
2527         (DownloadAndPerfTestWebKit2Factory.__init__):
2528
2529 2012-03-20  Peter Beverloo  <peter@chromium.org>
2530
2531         [Chromium] Update to Android NDK r7b
2532         https://bugs.webkit.org/show_bug.cgi?id=81005
2533
2534         Reviewed by Adam Barth.
2535
2536         Update to version r7b of the Android NDK, release notes of which are
2537         available on the following page:
2538         http://developer.android.com/sdk/ndk/index.html
2539
2540         * Scripts/update-webkit-chromium:
2541
2542 2012-03-20  Csaba Osztrogonác  <ossy@webkit.org>
2543
2544         x86 GCC bug triggers a compilation error in generated copy constructor of CollapsedBorderValue
2545         https://bugs.webkit.org/show_bug.cgi?id=81502
2546
2547         Unreviewed.
2548
2549         * qmake/mkspecs/features/unix/default_post.prf: Reenable Werror after r111171.
2550
2551 2012-03-19  Jessie Berlin  <jberlin@apple.com>
2552
2553         Unreviewed; roll out http://trac.webkit.org/changeset/111307 and http://trac.webkit.org/changeset/111293
2554         because they broke running the layout tests on mac.
2555
2556         * Scripts/webkitpy/layout_tests/port/webkit.py:
2557         (WebKitDriver._check_for_driver_crash):
2558
2559 2012-03-16  Martin Robinson  <mrobinson@igalia.com>
2560
2561         [GTK] Allow running run-gtk-tests during 'make distcheck'
2562         https://bugs.webkit.org/show_bug.cgi?id=81415
2563
2564         Reviewed by Philippe Normand.
2565
2566         Modify the build and run-gtk-tests so that it can run during 'make distcheck'
2567
2568         * GNUmakefile.am: Distribute run-gtk-tests during 'make dist' and make it the
2569         main command run during 'make check.'
2570         * Scripts/run-gtk-tests: Refactored run-gtk-tests so that jhbuild is no longer
2571         a requirement to run it. If jhbuild isn't around it will simply not start the
2572         SPI daemons and skip the WebKit2 accessibility test. Remove test runs from the
2573         main loop and simply use the main loop in a synchronous method to start SPI
2574         daemons. Did a bunch of general cleanup around these tasks.
2575
2576 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2577
2578         jhbuild: Use the multiprocessing module to get the number of CPUs.
2579         https://bugs.webkit.org/show_bug.cgi?id=81585
2580
2581         Reviewed by Martin Robinson.
2582
2583         Now that support for Python 2.5 has been dropped, we can safely
2584         use multiprocessing.cpu_count() to obtain the current number of
2585         CPUs in the system instead of running a webkitperl script for
2586         that.
2587
2588         * efl/common.py:
2589         (top_level_path):
2590         * efl/jhbuildrc:
2591         * gtk/common.py:
2592         (build_path):
2593         * gtk/jhbuildrc:
2594
2595 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2596
2597         We should only add a live event listener once, not once per TestSelector.
2598         Adding one per TestSelector would cause accordion('option', 'active') to
2599         incorrectly be "false" if you examine multiple sets of tests.
2600
2601         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2602
2603 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2604
2605         Fix typo introduced in r111293.
2606
2607         Unreviewed, build fix.
2608
2609         * Scripts/webkitpy/layout_tests/port/webkit.py:
2610         (WebKitDriver._check_for_driver_crash):
2611
2612 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2613
2614         webkitpy: get ServerProcess out of the reportcrash business
2615         https://bugs.webkit.org/show_bug.cgi?id=81600
2616
2617         Reviewed by Adam Barth.
2618
2619         ServerProcess was half-aware that ReportCrash might run
2620         sometimes, and that the process ServerProcess was talking to
2621         might have its own crashing subprocesses; neither of these
2622         things really worked right and it made the logic convoluted, so
2623         this change makes handling crashes completely separate from the
2624         server_process code, so that it can focus on just I/O to the
2625         subprocess.
2626
2627         There should be no functional changes resulting from this patch.
2628
2629         * Scripts/webkitpy/layout_tests/port/server_process.py:
2630         (ServerProcess._reset):
2631         (ServerProcess._handle_possible_interrupt):
2632         (ServerProcess.write):
2633         (ServerProcess.read_stdout):
2634         (ServerProcess.has_crashed):
2635         (ServerProcess._read):
2636         (ServerProcess.stop):
2637         * Scripts/webkitpy/layout_tests/port/server_process_unittest.py:
2638         (TrivialMockPort.check_for_leaks):
2639         (TestServerProcess.test_broken_pipe):
2640         * Scripts/webkitpy/layout_tests/port/webkit.py:
2641         (WebKitPort._read_image_diff):
2642         (WebKitDriver.has_crashed):
2643         (WebKitDriver._check_for_driver_crash):
2644         (WebKitDriver.run_test):
2645         (WebKitDriver._read_block):
2646         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2647         (MockServerProcess.__init__):
2648         (MockServerProcess):
2649         (MockServerProcess.has_crashed):
2650
2651 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2652
2653         webkitpy: crashlog parsing is broken
2654         https://bugs.webkit.org/show_bug.cgi?id=81586
2655
2656         Reviewed by Adam Barth.
2657
2658         Seeking on a file opened through codecs() appears to not work
2659         correctly (at least on SL).
2660         
2661         The code was more complicated than it needed to be, so I have
2662         revamped it to just read a full crashlog at once and then look
2663         for matches. 
2664         
2665         I have also added the ability to optionally propagate
2666         errors back (which can be helpful to debug races when ReportCrash is
2667         still running and you need to see that you couldn't open some
2668         files, rather than just ignoring them).
2669
2670         * Scripts/webkitpy/common/system/crashlogs.py:
2671         (CrashLogs.find_newest_log):
2672         (CrashLogs._find_newest_log_darwin):
2673         * Scripts/webkitpy/common/system/crashlogs_unittest.py:
2674         (CrashLogsTest.test_find_log_darwin):
2675
2676 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2677
2678         webkitpy: clean up some port._filesystem references to not reference private members
2679         https://bugs.webkit.org/show_bug.cgi?id=81595
2680
2681         Reviewed by Eric Seidel.
2682
2683         There should be no functional changes in this patch, just some
2684         refactoring. Also, fixes a minor nit in write_image_diff_files(()
2685
2686         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2687         (SingleTestRunner.__init__):
2688         (SingleTestRunner._run_compare_test):
2689         (SingleTestRunner._run_rebaseline):
2690         (SingleTestRunner._save_baseline_data):
2691         (SingleTestRunner._handle_error):
2692         (SingleTestRunner._run_reftest):
2693         * Scripts/webkitpy/layout_tests/controllers/test_result_writer.py:
2694         (write_test_result):
2695         (TestResultWriter.__init__):
2696         (TestResultWriter._make_output_directory):
2697         (TestResultWriter.output_filename):
2698         (TestResultWriter._output_testname):
2699         (TestResultWriter.write_output_files):
2700         (TestResultWriter.write_stderr):
2701         (TestResultWriter.write_crash_report):
2702         (TestResultWriter.create_text_diff_and_write_result):
2703         (TestResultWriter.write_image_diff_files):
2704         (copy_file):
2705         * Scripts/webkitpy/layout_tests/controllers/test_result_writer_unittest.py:
2706         (TestResultWriterTest.test_reftest_diff_image):
2707
2708 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2709
2710         webkitpy: fix nit - incorrect reference to DRT in test_expectations.py
2711         https://bugs.webkit.org/show_bug.cgi?id=81598
2712
2713         Reviewed by Eric Seidel.
2714
2715         "DumpRenderTree" shouldn't be hard-coded here, since some other
2716         process may actually be crashing (WebKitTestRunner, WebProcess, etc.)
2717         
2718         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2719         (TestExpectations):
2720
2721 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2722
2723         webkitpy: allow instance data in TestFailure objects
2724         https://bugs.webkit.org/show_bug.cgi?id=81580
2725
2726         Reviewed by Tony Chang.
2727
2728         Current TestFailures are basically enums, but it would be nice,
2729         especially for crashes, if they contained the name and pid of
2730         the process that crashed.
2731
2732         * Scripts/webkitpy/layout_tests/models/test_failures.py:
2733         (TestFailure.message):
2734         (FailureTimeout.message):
2735         (FailureCrash):
2736         (FailureCrash.__init__):
2737         (FailureCrash.message):
2738         (FailureMissingResult.message):
2739         (FailureTextMismatch.message):
2740         (FailureMissingImageHash.message):
2741         (FailureMissingImage.message):
2742         (FailureImageHashMismatch.message):
2743         (FailureImageHashIncorrect.message):
2744         (FailureReftestMismatch.message):
2745         (FailureReftestMismatchDidNotOccur.message):
2746         (FailureReftestNoImagesGenerated.message):
2747         (FailureMissingAudio.message):
2748         (FailureAudioMismatch.message):
2749         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py:
2750         (TestFailuresTest.test_equals):
2751         (TestFailuresTest):
2752         (TestFailuresTest.test_crashes):
2753
2754 2012-03-19  Brady Eidson  <beidson@apple.com>
2755
2756         <rdar://problem/10917120> and https://bugs.webkit.org/show_bug.cgi?id=81066
2757         Crash in 3rd party WebKit apps under XHR/Cache code
2758
2759         Reviewed by Antti Koivisto.
2760
2761         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2762         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.html: Added.
2763         * TestWebKitAPI/Tests/mac/MemoryCachePruneWithinResourceLoadDelegate.mm: Added.
2764         (-[MemoryCachePruneTestResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):
2765         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:willSendRequest:redirectResponse:fromDataSource:]):
2766         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFinishLoadingFromDataSource:]):
2767         (-[MemoryCachePruneTestResourceLoadDelegate webView:resource:didFailLoadingWithError:fromDataSource:]):
2768         (TestWebKitAPI):
2769         (TestWebKitAPI::TEST):
2770
2771 2012-03-19  Brady Eidson  <beidson@apple.com>
2772
2773         <rdar://problem/10848575> and https://bugs.webkit.org/show_bug.cgi?id=81516
2774         REGRESSION (r107435) URLs copied from WebKit apps aren't in the right pasteboard format
2775
2776         Reviewed by Enrica Casucci.
2777
2778         Bring up a context menu on a link, copy the link, then verify the pasteboard contents
2779         are in the correct format.
2780
2781         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
2782         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html: Added.
2783         * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm: Added.
2784         (-[ContextMenuCanCopyURLDelegate webView:didFinishLoadForFrame:]):
2785         (TestWebKitAPI::contextMenuCopyLink):
2786         (TestWebKitAPI::TEST):
2787
2788 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2789
2790         Make the selected item in the accordion a link to the flakiness dashboard
2791         https://bugs.webkit.org/show_bug.cgi?id=81571
2792
2793         Reviewed by Dimitri Glazkov.
2794
2795         Also, make it so that you can select the text of the link, instead of the current behavior
2796         of acting as if you had clicked on the accordion header.
2797
2798         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
2799         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:
2800
2801 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2802
2803         NRWT runs some tests that are skipped with -i command line option
2804         https://bugs.webkit.org/show_bug.cgi?id=81535
2805
2806         Reviewed by Ojan Vafai.
2807
2808         This change modifies the interaction of Skipped files and
2809         test_expectations files so that entries in Skipped files (and
2810         the -i command line) override *everything* in the expectations
2811         file.
2812
2813         Specifically, a directory in a Skipped file will cause all of
2814         the tests in the dir to be skipped even if individual tests in
2815         the dir are listed in the test_expectations.txt.
2816
2817         Skipped files also override anything in an overrides files.
2818
2819         This seems to make more intuitive sense, since if you list
2820         something in the Skips file (and even more specify it on the
2821         command line) you probably want it to be universally applied.
2822
2823         Theoretically we could add more precedence levels and have full
2824         paths in an expectations file override dirs in a Skipped file
2825         (but not the command line), but I don't know yet that that level
2826         of complexity is justified.
2827         
2828         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2829         (TestExpectations.__init__):
2830         (TestExpectations._add_skipped_tests):
2831         (TestExpectations._add_test):
2832         (TestExpectations._add_expectations):
2833         (TestExpectations._add_expectation_line):
2834         (TestExpectations._already_seen_better_match):
2835         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2836         (test_more_specific_override_resets_skip):
2837         (SkippedTests):
2838         (SkippedTests.get_exp):
2839         (SkippedTests.assert_exp):
2840         (SkippedTests.test_skipped_tests_work):
2841         (SkippedTests.test_duplicate_skipped_test_fails_lint):
2842         (SkippedTests.test_skipped_file_overrides_expectations):
2843         (SkippedTests.test_skipped_dir_overrides_expectations):
2844         (SkippedTests.test_skipped_file_overrides_overrides):
2845         (SkippedTests.test_skipped_dir_overrides_overrides):
2846
2847 2012-03-19  Eric Seidel  <eric@webkit.org>
2848
2849         Fix WTF header include discipline in Chromium WebKit
2850         https://bugs.webkit.org/show_bug.cgi?id=81281
2851
2852         Reviewed by James Robinson.
2853
2854         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
2855         * Scripts/update-webkit-chromium: Made it possible to run this from a directory other than the root.
2856
2857 2012-03-19  Andy Chen  <andchen@rim.com>
2858
2859         [BlackBerry] Add new layout test for text match marker
2860         https://bugs.webkit.org/show_bug.cgi?id=81531
2861
2862         Make the LayoutTestControllerBlackBerry::findString to test the
2863         BlackBerry port implementation.
2864
2865         Reviewed by Rob Buis.
2866
2867         * DumpRenderTree/blackberry/LayoutTestControllerBlackBerry.cpp:
2868         (LayoutTestController::findString):
2869
2870 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2871
2872         Don't show WONTFIX tests in garden-o-matic
2873         https://bugs.webkit.org/show_bug.cgi?id=81554
2874
2875         Reviewed by Dimitri Glazkov.
2876
2877         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2878         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2879
2880 2012-03-19  Chris Fleizach  <cfleizach@apple.com>
2881
2882         platform/mac/accessibility/selected-rows-table.html failing on Lion Intel Debug WebKit2 testers
2883         https://bugs.webkit.org/show_bug.cgi?id=81530
2884
2885         Reviewed by Jessie Berlin.
2886
2887         Add rowAtIndex to WebKitTestRunner.
2888
2889         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
2890         (WTR::AccessibilityUIElement::rowAtIndex):
2891         * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
2892         (AccessibilityUIElement):
2893         * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
2894         * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
2895         (WTR):
2896         (WTR::AccessibilityUIElement::rowAtIndex):
2897
2898 2012-03-19  Ojan Vafai  <ojan@chromium.org>
2899
2900         Change garden-o-matic All Failures tab to Expected Failures
2901         https://bugs.webkit.org/show_bug.cgi?id=81540
2902
2903         Reviewed by Dimitri Glazkov.
2904
2905         Also, fix some broken unittests that had just gotten stale.
2906
2907         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
2908         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers/rebaseline.js:
2909         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
2910         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
2911         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
2912         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
2913         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
2914         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
2915         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
2916         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
2917
2918 2012-03-19  Hao Zheng  <zhenghao@chromium.org>
2919
2920         [chromium] DRT crashes at shutdown.
2921         https://bugs.webkit.org/show_bug.cgi?id=81504
2922
2923         Reviewed by James Robinson.
2924
2925         WebCompositor must be destroyed after WebView is properly destroyed,
2926         or some compositor objects like CCLayerTreeHost may not be destroyed.
2927         However, it is really hard to make the destruction order correct
2928         explicitly in ~TestShell. So move it out to DumpRenderTree.
2929
2930         * DumpRenderTree/chromium/DumpRenderTree.cpp:
2931         (main):
2932         * DumpRenderTree/chromium/TestShell.cpp:
2933         (TestShell::~TestShell):
2934
2935 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2936
2937         add dpranke to webkitpy watchlist
2938         https://bugs.webkit.org/show_bug.cgi?id=81433
2939
2940         Reviewed by Benjamin Poulain.
2941
2942         * Scripts/webkitpy/common/config/watchlist:
2943
2944 2012-03-19  Dirk Pranke  <dpranke@chromium.org>
2945
2946         test-webkitpy is failing on SL/Python 2.6.1
2947         https://bugs.webkit.org/show_bug.cgi?id=81533
2948
2949         Reviewed by Ojan Vafai.
2950
2951         It appears that itertools.combinations() is buggy in 2.6.1; this
2952         backports the implementation from 2.7.
2953
2954         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2955         (TestConfigurationConverter.collapse_macros):
2956         (TestConfigurationConverter):
2957         (TestConfigurationConverter.combinations):
2958
2959 2012-03-19  David Michael Barr  <davidbarr@google.com>
2960
2961         Optimize the cleanup sequence in the start-queue.sh script
2962         https://github.com/abarth/webkit/pull/1
2963
2964         Reviewed by Adam Barth.
2965
2966         Updating the working copy is an expensive operation
2967         for a repository the size of WebKit.
2968         Attempt to reset only to the target revision.
2969
2970         * EWSTools/start-queue.sh:
2971
2972 2012-03-19  Raphael Kubo da Costa  <kubo@profusion.mobi>
2973
2974         Adjust architecture type of the EFL build bot.
2975         https://bugs.webkit.org/show_bug.cgi?id=81452
2976
2977         Reviewed by Tony Chang.
2978
2979         * BuildSlaveSupport/build.webkit.org-config/config.json: The
2980         buildslave is a 64-bit machine, therefore describe it as being
2981         x86_64, not i386.
2982
2983 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
2984
2985         [Qt] printLoadedUrls prints "Object" instead of the actual URL.
2986         https://bugs.webkit.org/show_bug.cgi?id=81514
2987
2988         Reviewed by Simon Hausmann.
2989
2990         * MiniBrowser/qt/qml/BrowserWindow.qml:
2991
2992 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
2993
2994         [EFL] jhbuild: Set CMAKE_PREFIX_PATH in jhbuildrc.
2995         https://bugs.webkit.org/show_bug.cgi?id=81491
2996
2997         Reviewed by Philippe Normand.
2998
2999         For the find_path(), find_library() etc CMake calls to find the
3000         dependencies built with jhbuild, CMAKE_PREFIX_PATH must be properly
3001         set, otherwise CMake cannot know the dependencies exist and will only
3002         look in the standard directories.
3003
3004         * efl/jhbuildrc: Set the CMAKE_PREFIX_PATH environment variable with
3005         the jhbuild Root directory.
3006
3007 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3008
3009         [EFL] Add XDG env vars to jhbuildrc
3010         https://bugs.webkit.org/show_bug.cgi?id=81510
3011
3012         Needed to ensure correct glib behaviour in jhbuild environment.
3013
3014         Reviewed by Philippe Normand.
3015
3016         * efl/jhbuildrc:
3017
3018 2012-03-19  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
3019
3020         Select best target for tap gesture.
3021         https://bugs.webkit.org/show_bug.cgi?id=78801
3022
3023         Reviewed by Kenneth Rohde Christiansen.
3024
3025         Add TOUCH_ADJUSTMENT to enabled features.
3026
3027         * qmake/mkspecs/features/features.prf:
3028
3029 2012-03-19  Robert Kroeger  <rjkroege@chromium.org>
3030
3031         [chromium] synthesize wheel events for fling on main thread
3032         https://bugs.webkit.org/show_bug.cgi?id=81458
3033
3034         Added the ability to dispatch GestureFlingStart and GestureFlingCancel from EventSender.
3035
3036         Reviewed by James Robinson.
3037
3038         * DumpRenderTree/chromium/EventSender.cpp:
3039         (EventSender::EventSender):
3040         (EventSender::gestureFlingCancel):
3041         (EventSender::gestureFlingStart):
3042         * DumpRenderTree/chromium/EventSender.h:
3043         (EventSender):
3044
3045 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
3046
3047         [Qt] Add command line option to set the user agent in the MiniBrowser
3048         https://bugs.webkit.org/show_bug.cgi?id=81508
3049
3050         Reviewed by Simon Hausmann.
3051
3052         * MiniBrowser/qt/BrowserWindow.cpp:
3053         (BrowserWindow::BrowserWindow):
3054         (BrowserWindow::webViewExperimental):
3055         * MiniBrowser/qt/BrowserWindow.h:
3056         (BrowserWindow):
3057         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3058         (printHelp):
3059         (MiniBrowserApplication::handleUserOptions):
3060         * MiniBrowser/qt/MiniBrowserApplication.h:
3061         (WindowOptions::WindowOptions):
3062         (WindowOptions::setUserAgent):
3063         (WindowOptions::userAgent):
3064         (WindowOptions):
3065
3066 2012-03-19  Alexander Færøy  <alexander.faeroy@nokia.com>
3067
3068         [Qt] Remove dead user agent code from the MiniBrowser.
3069
3070         Reviewed by Simon Hausmann.
3071
3072         * MiniBrowser/qt/BrowserWindow.cpp:
3073         * MiniBrowser/qt/BrowserWindow.h:
3074         (BrowserWindow):
3075         * MiniBrowser/qt/MiniBrowser.qrc:
3076         * MiniBrowser/qt/useragentlist.txt: Removed.
3077
3078 2012-03-19  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3079
3080         jhbuild: Call jhbuild's autogen.sh via bash.
3081         https://bugs.webkit.org/show_bug.cgi?id=81485
3082
3083         Reviewed by Philippe Normand.
3084
3085         jhbuild's autogen.sh uses "/bin/bash" in the shebang line, however
3086         this is usually true only on Linux systems, and even then bash for
3087         some reason might not be available in that location. It is safer to
3088         call bash directly and pass autogen.sh to it instead.
3089
3090         * jhbuild/jhbuild-wrapper:
3091         (install_jhbuild):
3092
3093 2012-03-19  Philippe Normand  <pnormand@igalia.com>
3094
3095         [GTK] Prefix run-launcher with run-with-jhbuild
3096         https://bugs.webkit.org/show_bug.cgi?id=81503
3097
3098         Reviewed by Hajime Morita.
3099
3100         * Scripts/run-launcher:
3101
3102 2012-03-19  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3103
3104         [EFL] Prefix run-launcher with run-with-jhbuild
3105         https://bugs.webkit.org/show_bug.cgi?id=80026
3106
3107         Reviewed by Hajime Morita.
3108
3109         * Scripts/run-launcher:
3110
3111 2012-03-18  Csaba Osztrogonác  <ossy@webkit.org>
3112
3113         [Qt] Unreviewed. Disable -Werror on x86 until proper fix.
3114         https://bugs.webkit.org/show_bug.cgi?id=81498
3115
3116         * qmake/mkspecs/features/unix/default_post.prf:
3117
3118 2012-03-17  Raphael Kubo da Costa  <rakuco@FreeBSD.org>
3119
3120         Unreviewed; switch to using my FreeBSD.org email address, the
3121         other one is going to disappear soon.
3122
3123         * Scripts/webkitpy/common/config/committers.py:
3124         * Scripts/webkitpy/common/config/watchlist:
3125
3126 2012-03-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
3127
3128         Unreviewed; fix webkitpy unit test failures after r111116.
3129
3130         It turns out that cmd_line in the patch in bug 80025 had the right
3131         signature when it was submitted, but r109242 added a few more
3132         required parameters, and since the method (and EflDriver) did not
3133         exist at the time we ended up committing a broken method.
3134
3135         * Scripts/webkitpy/layout_tests/port/efl.py:
3136         (EflDriver.cmd_line): Add the `pixel_tests' and `per_test_args'
3137         arguments.
3138
3139 2012-03-17  Charles Wei  <charles.wei@torchmobile.com.cn>
3140
3141         [BlackBerry] Enable Web Timing for performance profiling and improvement.
3142         https://bugs.webkit.org/show_bug.cgi?id=81085
3143
3144         Reviewed by Rob Buis.
3145
3146         * Scripts/build-webkit:
3147
3148 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3149
3150         Add a dependencies installation step for EFL to buildbot master.cfg
3151         https://bugs.webkit.org/show_bug.cgi?id=81337
3152
3153         Reviewed by Gustavo Noronha Silva.
3154
3155         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3156         (InstallEflDependencies):
3157         (Factory.__init__):
3158
3159 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3160
3161         [EFL] Prefixing run-webkit-tests with run-with-jhbuild
3162         https://bugs.webkit.org/show_bug.cgi?id=80025
3163
3164         Reviewed by Gustavo Noronha Silva.
3165
3166         * Scripts/webkitpy/layout_tests/port/efl.py:
3167         (EflDriver):
3168         (EflDriver.cmd_line):
3169         (EflPort._driver_class):
3170
3171 2012-03-17  Dominik Röttsches  <dominik.rottsches@linux.intel.com>
3172
3173         [EFL] Add and use run-with-jhbuild and update-webkitefl-libs scripts for EFL
3174         https://bugs.webkit.org/show_bug.cgi?id=79904
3175
3176         build-webkit --efl will use jhbuild for fetching and building dependencies.
3177
3178         Reviewed by Gustavo Noronha Silva.
3179
3180         * Scripts/update-webkitefl-libs: Added.
3181         * Scripts/webkitdirs.pm:
3182         (jhbuildWrapperPrefixIfNeeded):
3183         (generateBuildSystemFromCMakeProject):
3184         (buildCMakeGeneratedProject):
3185         * efl/common.py: Added.
3186         (script_path):
3187         (top_level_path):
3188         (number_of_cpus):
3189         * efl/jhbuild.modules: Added.
3190         * efl/jhbuildrc: Added.
3191         * efl/run-with-jhbuild: Added.
3192
3193 2012-03-16  Stephanie Lewis  <slewis@apple.com>
3194
3195         https://bugs.webkit.org/show_bug.cgi?id=81065
3196         <rdar://problem/10944309> CrashTracer: [USER] 111 crashes in WebProcess at WebKitTestRunnerInjectedBundle: WTR::InjectedBundle::done + 142
3197         Fix crash preventing WebKitTestRunner from working in Release.  
3198         WebKitTestRunner used ostringstream to collect output from the test.  ostringstream is incompatible with our
3199         FastMalloc overrides.  When copying strings out ostringstream a new string is allocated with FastMalloc because
3200         the allocation is inlined.  When that string is freed is uses the system malloc because that call is not inlined.
3201
3202         I removed ostringstream in favor of StringBuilder.  
3203
3204         Properly set NDEBUG on Release builds so we don't crash when when we free the strings StringBuilder created.
3205
3206         Reviewed by Geoff Garen.
3207
3208         Add DEBUG_DEFINES to set NDEBUG on release builds.
3209         * WebKitTestRunner/Configurations/Base.xcconfig:
3210         * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
3211
3212         Remove ostringstream member in favor of StringBuilder.
3213         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
3214         (WTR::InjectedBundle::initialize):
3215         (WTR::InjectedBundle::beginTesting):
3216         (WTR::InjectedBundle::done):
3217         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
3218         (WTF):
3219         (WTR::InjectedBundle::stringBuilder):
3220         (InjectedBundle):
3221
3222         Append output to StringBuilder.
3223         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
3224         (WTR::hasPrefix):
3225         (WTR::dumpPath):
3226         (WTR::rangeToStr):
3227         (WTR::styleDecToStr):
3228         (WTR::frameToStr):
3229         (WTR::dumpFrameScrollPosition):
3230         (WTR::dumpFrameText):
3231         (WTR::dumpDescendantFramesText):
3232         (WTR::InjectedBundlePage::dump):
3233         (WTR::InjectedBundlePage::didReceiveTitleForFrame):
3234         (WTR::InjectedBundlePage::didFinishDocumentLoadForFrame):
3235         (WTR::InjectedBundlePage::willSendRequestForFrame):
3236         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
3237         (WTR::lastFileURLPathComponent):
3238         (WTR::InjectedBundlePage::willAddMessageToConsole):
3239         (WTR::InjectedBundlePage::willSetStatusbarText):
3240         (WTR::InjectedBundlePage::willRunJavaScriptAlert):
3241         (WTR::InjectedBundlePage::willRunJavaScriptConfirm):
3242         (WTR::InjectedBundlePage::willRunJavaScriptPrompt):
3243         (WTR::InjectedBundlePage::shouldBeginEditing):
3244         (WTR::InjectedBundlePage::shouldEndEditing):
3245         (WTR::InjectedBundlePage::shouldInsertNode):
3246         (WTR::InjectedBundlePage::shouldInsertText):
3247         (WTR::InjectedBundlePage::shouldDeleteRange):
3248         (WTR::InjectedBundlePage::shouldChangeSelectedRange):
3249         (WTR::InjectedBundlePage::shouldApplyStyle):
3250         (WTR::InjectedBundlePage::didBeginEditing):
3251         (WTR::InjectedBundlePage::didEndEditing):
3252         (WTR::InjectedBundlePage::didChange):
3253         (WTR::InjectedBundlePage::didChangeSelection):
3254         (WTR::InjectedBundlePage::supportsFullScreen):
3255         (WTR::InjectedBundlePage::enterFullScreenForElement):
3256         (WTR::InjectedBundlePage::exitFullScreenForElement):
3257         (WTR::InjectedBundlePage::beganEnterFullScreen):
3258         (WTR::InjectedBundlePage::beganExitFullScreen):
3259         (WTR::dumpBackForwardListItem):
3260         (WTR::InjectedBundlePage::dumpBackForwardList):
3261         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3262         (WTR::LayoutTestController::waitToDumpWatchdogTimerFired):
3263         (WTR::LayoutTestController::dumpConfigurationForViewport):
3264         
3265         Convert between WTF::String and WKStringRef.
3266         * WebKitTestRunner/StringFunctions.h:
3267         (WTR::toWTFString):
3268         (WTR):
3269         * WebKitTestRunner/TestInvocation.cpp:
3270         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
3271
3272 2012-03-16  Eric Seidel  <eric@webkit.org>
3273
3274         Add a land-from-url command
3275         https://bugs.webkit.org/show_bug.cgi?id=81411
3276
3277         Reviewed by Adam Barth.
3278
3279         Very basic so far.  The goal is to make it landing smarter
3280         so that users don't have to bother to parse out the ids out of the urls.
3281
3282         * Scripts/webkitpy/common/config/urls.py:
3283         (parse_bug_id):
3284         (parse_attachment_id):
3285         * Scripts/webkitpy/common/config/urls_unittest.py:
3286         (URLsTest.test_parse_bug_id):
3287         (URLsTest):
3288         (URLsTest.test_parse_attachment_id):
3289         * Scripts/webkitpy/tool/commands/download.py:
3290         (ProcessURLsMixin):
3291         (ProcessURLsMixin._fetch_list_of_patches_to_process):
3292         (LandFromURL):
3293         * Scripts/webkitpy/tool/commands/download_unittest.py:
3294         (test_land_from_bug):
3295         (test_land_from_url):
3296
3297 2012-03-16  Dave Tharp  <dtharp@codeaurora.org>
3298
3299         build-webkit launcher instructions use wrong relative path for Qt
3300         https://bugs.webkit.org/show_bug.cgi?id=81357
3301
3302         Reviewed by Eric Seidel.
3303
3304         Ensure we are at the proper directory before generating run-launcher
3305         instructions.
3306
3307         * Scripts/build-webkit:
3308
3309 2012-03-16  Brady Eidson  <beidson@apple.com>
3310
3311         <rdar://problem/11027997> and https://bugs.webkit.org/show_bug.cgi?id=81412
3312         REGRESSION (r107435) Copy a link and paste to Mail: Nothing is pasted
3313
3314         Reviewed by Geoff Garen.
3315
3316         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
3317         * TestWebKitAPI/Tests/mac/WebViewCanPasteURL.mm: Added.
3318
3319 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3320
3321         Mark wontfix tests in full_results.json so we can ignore them in garden-o-matic
3322         https://bugs.webkit.org/show_bug.cgi?id=81407
3323
3324         Reviewed by Adam Barth.
3325
3326         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3327         (summarize_results):
3328         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
3329         (ResultSummaryTest.summarized_results):
3330         (ResultSummaryTest.test_no_svn_revision):
3331         (ResultSummaryTest.test_svn_revision):
3332         (ResultSummaryTest):
3333         (ResultSummaryTest.test_summarized_results_wontfix):
3334
3335 2012-03-16  Ojan Vafai  <ojan@chromium.org>
3336
3337         Have webkit-patch rebaseline-test update test_expectations.txt
3338         https://bugs.webkit.org/show_bug.cgi?id=81054
3339
3340         Recommit http://trac.webkit.org/changeset/110783 now that
3341         http://trac.webkit.org/changeset/111034 has landed.
3342
3343         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3344         (TestExpectations.remove_configuration_from_test):
3345         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3346         (RemoveConfigurationsTest):
3347         (RemoveConfigurationsTest.test_remove):
3348         (test_remove_line):
3349         * Scripts/webkitpy/tool/commands/rebaseline.py:
3350         (RebaselineTest._update_expectations_file):
3351         (RebaselineTest._rebaseline_test_and_update_expectations):
3352         (RebaselineTest):
3353         (RebaselineTest.execute):
3354         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3355         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3356         (test_rebaseline_updates_expectations_file):
3357
3358 2012-03-15  Ojan Vafai  <ojan@chromium.org>
3359
3360         Specifier collapsing when writing test expectations lines gets a number of cases wrong
3361         https://bugs.webkit.org/show_bug.cgi?id=81309
3362
3363         Reviewed by Dimitri Glazkov.
3364
3365         I've run this over all the lines in the current Chromium test_expectations.txt file,
3366         so I'm relatively confident we now cover all the cases.
3367
3368         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
3369         (TestConfigurationConverter.__init__):
3370         (TestConfigurationConverter.collapse_macros):
3371         (TestConfigurationConverter.collapse_macros.collapse_individual_specifier_set):
3372         (TestConfigurationConverter):
3373         (TestConfigurationConverter.intersect_combination):
3374         (TestConfigurationConverter.symmetric_difference):
3375         (TestConfigurationConverter.to_specifiers_list):
3376         (TestConfigurationConverter.to_specifiers_list.try_collapsing):
3377         (TestConfigurationConverter.to_specifiers_list.try_abbreviating):
3378         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
3379         (make_mock_all_test_configurations_set):
3380         (TestConfigurationConverterTest.test_symmetric_difference):
3381         (TestConfigurationConverterTest.test_to_config_set):
3382         (TestConfigurationConverterTest.test_macro_expansion):
3383         (TestConfigurationConverterTest.test_to_specifier_lists):
3384         (TestConfigurationConverterTest.test_converter_macro_collapsing):
3385         * Scripts/webkitpy/layout_tests/port/chromium.py:
3386         (ChromiumPort):
3387         * Scripts/webkitpy/layout_tests/port/chromium_android.py:
3388         (ChromiumAndroidPort.__init__):
3389         The android port uses "arm" as it's architecture, which is technically correct,
3390         but considerably complicates making collapsing work. We probably should kill
3391         the concept of architecture entirely. The benefits are not worth the code
3392         complexity.
3393
3394         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3395         (ChromiumPortTest.test_all_test_configurations):
3396         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3397         (TestRebaseline.test_rebaseline_updates_expectations_file_noop):
3398         (test_rebaseline_updates_expectations_file):
3399
3400 2012-03-16  Dinu Jacob  <dinu.jacob@nokia.com>
3401
3402         [Qt][Wk2] Assertion Failure and crash on file upload
3403         https://bugs.webkit.org/show_bug.cgi?id=80854
3404
3405         Reviewed by Simon Hausmann.
3406
3407         Added filePicker to WebView using experimental API.
3408
3409         * MiniBrowser/qt/MiniBrowser.qrc:
3410         * MiniBrowser/qt/icons/folder.png: Added.
3411         * MiniBrowser/qt/icons/titlebar.png: Added.
3412         * MiniBrowser/qt/icons/up.png: Added.
3413         * MiniBrowser/qt/qml/BrowserWindow.qml:
3414         * MiniBrowser/qt/qml/FilePicker.qml: Added.
3415
3416 2012-03-16  Mahesh Kulkarni  <mahesh.kulkarni@nokia.com>
3417
3418         Updating email for committer.py script.
3419
3420         * Scripts/webkitpy/common/config/committers.py:
3421
3422 2012-03-16  Robert Kroeger  <rjkroege@chromium.org>
3423
3424         [chromium] DRT: Add support for sending gesture scrolls without a begin
3425         https://bugs.webkit.org/show_bug.cgi?id=81292
3426
3427         Reviewed by Kent Tamura.
3428
3429         * DumpRenderTree/chromium/EventSender.cpp:
3430         (EventSender::EventSender):
3431         (EventSender::gestureScrollFirstPoint):
3432         * DumpRenderTree/chromium/EventSender.h:
3433         (EventSender):
3434
3435 2012-03-02  Carlos Garcia Campos  <cgarcia@igalia.com> and Sergio Villar Senin  <svillar@igalia.com>
3436
3437         [GTK] WebKitIconDatabase doesn't keep icons cached
3438         https://bugs.webkit.org/show_bug.cgi?id=56200
3439
3440         Reviewed by Martin Robinson.
3441
3442         Add support for favicons in GtkLauncher.
3443
3444         * GtkLauncher/main.c:
3445         (activateUriEntryCb):
3446         (iconLoadedCb):
3447         (createBrowser):
3448
3449 2012-03-16  Adam Barth  <abarth@webkit.org>
3450
3451         Remove sheriff-bot's last-green-revision command
3452         https://bugs.webkit.org/show_bug.cgi?id=81314
3453
3454         Reviewed by Eric Seidel.
3455
3456         This command is a cruel joke.  WebKit doesn't really ever have a
3457         last-green-revision.
3458
3459         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3460         (BuildBot._find_green_revision):
3461         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3462         (MockBuildBot.builder_statuses):
3463         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3464         (test_find_green_revision):
3465         * Scripts/webkitpy/tool/bot/irc_command.py:
3466         (IRCCommand.execute):
3467         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
3468         (SheriffIRCBotTest.test_help):
3469         * Scripts/webkitpy/tool/commands/queries.py:
3470         (PatchesToReview.execute):
3471
3472 2012-03-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
3473
3474         [Qt] Make gccdepends test aware of broken icecream deps generation
3475
3476         Icecream seems to build the files without passing -o to gcc with
3477         the original destination path, which means the target name in the
3478         dependency file is wrong.
3479
3480         Also, echo something to the final file instead of just touching it,
3481         so it's not deleted on every build.
3482
3483         Reviewed by Simon Hausmann.
3484
3485         * qmake/config.tests/gccdepends/gccdepends.pro:
3486
3487 2012-03-16  Kihong Kwon  <kihong.kwon@samsung.com>
3488
3489         Support for Battery Status API
3490         https://bugs.webkit.org/show_bug.cgi?id=62698
3491
3492         Add include directory and file.
3493         http://www.w3.org/TR/battery-status/
3494
3495         Reviewed by Adam Barth.
3496
3497         * Scripts/build-webkit:
3498
3499 2012-03-16  Kishore Bolisetty  <kbolisetty@innominds.com>
3500
3501         check-webkit-style fails on LayoutTests/platform/efl/test_expectations.txt
3502         https://bugs.webkit.org/show_bug.cgi?id=81007
3503
3504         Reviewed by Adam Barth.
3505
3506         A dedidcated build slave is available for efl on build.webkit.org. 
3507         This change add's an entry for the elf port in list of available port builders.
3508         This is required because test_expectations.py when triggered, looks for a valid port object for the specified port. 
3509
3510         * Scripts/webkitpy/layout_tests/port/builders.py:
3511
3512 2012-03-16  Adam Barth  <abarth@webkit.org>
3513
3514         [webkitpy] StatusServer.next_work_item has an incorrectly named local variable
3515         https://bugs.webkit.org/show_bug.cgi?id=81302
3516
3517         Reviewed by Eric Seidel.
3518
3519         This is a trivial patch, but the variable name turned up in a grep of
3520         webkitpy and looks like copy pasta.
3521
3522         * Scripts/webkitpy/common/net/statusserver.py:
3523         (StatusServer.next_work_item):
3524
3525 2012-03-16  Andrew Lo  <anlo@rim.com>
3526
3527         [BlackBerry] Enable PAGE_VISIBILITY_API
3528         https://bugs.webkit.org/show_bug.cgi?id=81257
3529
3530         Enable Page Visibility API for BlackBerry.
3531
3532         Reviewed by Rob Buis.
3533
3534         * Scripts/build-webkit:
3535
3536 2012-03-16  Zan Dobersek  <zandobersek@gmail.com>
3537
3538         [Gtk] Add computedStyleIncludingVisitedInfo to DumpRenderTreeSupport
3539         https://bugs.webkit.org/show_bug.cgi?id=81232
3540
3541         Reviewed by Philippe Normand.
3542
3543         Use the newly-added helper method in DumpRenderTreeSupportGtk.
3544
3545         * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
3546         (LayoutTestController::computedStyleIncludingVisitedInfo):
3547
3548 2012-03-15  Eric Seidel  <eric@webkit.org>
3549
3550         Move parse_bug_id into config/urls in preparation for re-use elsewhere.
3551         https://bugs.webkit.org/show_bug.cgi?id=81313
3552
3553         Reviewed by Adam Barth.
3554
3555         * Scripts/webkitpy/common/checkout/changelog.py:
3556         (parse_bug_id_from_changelog):
3557         * Scripts/webkitpy/common/config/urls.py:
3558         (parse_bug_id):
3559         * Scripts/webkitpy/common/config/urls_unittest.py: Copied from Tools/Scripts/webkitpy/common/config/urls.py.
3560         (URLsTest):
3561         (URLsTest.test_parse_bug_id):
3562         * Scripts/webkitpy/common/net/bugzilla/bugzilla_unittest.py:
3563         (test_parse_bug_id):
3564         * Scripts/webkitpy/tool/bot/irc_command.py:
3565         (_post_error_and_check_for_bug_url):
3566         * Scripts/webkitpy/tool/bot/sheriff.py:
3567         (Sheriff.post_rollout_patch):
3568         (Sheriff.post_chromium_deps_roll):
3569
3570 2012-03-15  Adam Barth  <abarth@webkit.org>
3571
3572         The commit-queue should fast-track patches that have already passed the testing EWS bots
3573         https://bugs.webkit.org/show_bug.cgi?id=81305
3574
3575         Reviewed by Eric Seidel.
3576
3577         If a patch has already passed the testing EWS bots (currently only the
3578         chromium-ews bot), we should be able to land the patch without running
3579         the tests again.  This patch is somewhat risky in that the tests might
3580         have been broken between when the EWS bots ran and when the
3581         commit-queue is processing the patch.  If that turns out to be a
3582         problem in practice, we'll tighten up this fast-track in some way,
3583         likely by making sure that the pass status is recent.
3584
3585         * Scripts/webkitpy/tool/bot/commitqueuetask.py:
3586         (CommitQueueTaskDelegate.did_pass_testing_ews):
3587         (CommitQueueTask._did_pass_tests_recently):
3588         (CommitQueueTask.run):
3589         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
3590         (MockCommitQueue.did_pass_testing_ews):
3591         (test_fast_success_case):
3592         * Scripts/webkitpy/tool/commands/queues.py:
3593         (CommitQueue.did_pass_testing_ews):
3594         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3595         (test_did_pass_testing_ews):
3596
3597 2012-03-15  Kentaro Hara  <haraken@chromium.org>
3598
3599         [PerformanceTests] run-perf-tests should output correct units
3600         https://bugs.webkit.org/show_bug.cgi?id=78303
3601
3602         Reviewed by Ryosuke Niwa.
3603
3604         Previously run-perf-tests just supported "ms" for units.
3605         Consequently, Dromaeo perf tests had been reporting runs/s as "ms".
3606         This patch fixes run-perf-tests so that they can support custom units.
3607         The default unit is "ms".
3608
3609         The test result looks like this:
3610
3611         $ ./Tools/Scripts/run-perf-tests Dromaeo/dom-attr.html
3612         RESULT Dromaeo: dom-attr= 6465.9525483 runs/s
3613         median= 0.0 runs/s, stdev= 24.2983433436 runs/s, min= 6435.87649402 runs/s, max= 6515.63693392 runs/s
3614
3615         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
3616         (PerfTestsRunner._process_parser_test_result):
3617
3618 2012-03-15  Brent Fulgham  <bfulgham@webkit.org>
3619
3620         [WinCairo] Unreviewed build change after wtf path changes.
3621
3622         * DumpRenderTree/win/ImageDiffWinCairoCommon.vsprops: Updated include
3623         directories to look at the private header directory to find 'wtf'.
3624
3625 2012-03-15  Kent Tamura  <tkent@chromium.org>
3626
3627         REGRESSION(r110595): webkit-patch suggest-reviewers doesn't work
3628         https://bugs.webkit.org/show_bug.cgi?id=81293
3629
3630         Reviewed by Adam Barth.
3631
3632         r110595 introduced new format of the author field of ChangeLog entry:
3633         "Adam Barth  <abarth@webkit.org> && Benjamin Poulain  <bpoulain@apple.com>"
3634         Updte chanelog.py to support it.
3635
3636         * Scripts/webkitpy/common/checkout/changelog.py:
3637         (ChangeLogEntry._split_contributor_names):
3638         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
3639         (test_parse_authors):
3640
3641 2012-03-15  Lynn Neir  <lynn.neir@skype.net>
3642
3643         Implement LayoutTestController::setValueForUser on Windows https://bugs.webkit.org/show_bug.cgi?id=55834
3644
3645         Implemented setValueForUser in both DumpRenderTree and WebKitTestRunner.
3646
3647         Reviewed by Jessie Berlin.
3648
3649         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
3650         (LayoutTestController::setValueForUser):
3651         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
3652         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
3653         (WTR::LayoutTestController::setValueForUser):
3654         (WTR):
3655         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
3656         (LayoutTestController):
3657
3658 2012-03-15  Adam Barth  <abarth@webkit.org>
3659
3660         Update the start-queue.sh script to match what we're using on the bots
3661         https://bugs.webkit.org/show_bug.cgi?id=81277
3662
3663         Reviewed by Eric Seidel.
3664
3665         Over time we've added more magic git commands to clean up the local repo after
3666         it has been borked in various ways. The idea behind this sequence is to clear
3667         out any changes in the working copy and then to rebuild the master branch from
3668         origin/master.
3669
3670         This patch also removes the watchdog timer that we've stopped using on the
3671         bots.
3672
3673         This patch was reviewed on <https://github.com/eseidel/webkit/pull/1>.
3674         Note: Eric and I are experi