505ea2ff0fe39466fe94b2f9c46c40a9744d31d6
[WebKit-https.git] / Tools / ChangeLog
1 2011-07-15  Adam Roben  <aroben@apple.com>
2
3         Rename TestFailureBugForm to FailingTestsBugForm
4
5         The new name will match better with a forthcoming FlakyTestBugForm class.
6
7         Fixes <http://webkit.org/b/64598> TestFailures page's TestFailureBugForm class has a bad
8         name
9
10         Reviewed by Daniel Bates.
11
12         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js: Renamed from Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailureBugForm.js.
13         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js: Renamed from Tools/BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailureBugForm_unittests.js.
14
15         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
16         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
17         Updated for renames.
18
19         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Ditto, and
20         reordered <script>s to put files which others depend on for parsing first.
21
22 2011-07-15  Adam Barth  <abarth@webkit.org>
23
24         NRWT stores the Chromium revision number in full_results.json
25         https://bugs.webkit.org/show_bug.cgi?id=64586
26
27         I have no earthly idea how to test this change.
28
29         * Scripts/webkitpy/layout_tests/controllers/manager.py:
30         * Scripts/webkitpy/layout_tests/port/base.py:
31
32 2011-07-14  Ojan Vafai  <ojan@chromium.org>
33
34         fix remaining style issues in the static-dashboards directory
35         https://bugs.webkit.org/show_bug.cgi?id=64561
36
37         Reviewed by Adam Barth.
38
39         * TestResultServer/static-dashboards/aggregate_results.html:
40         * TestResultServer/static-dashboards/builders.js:
41         * TestResultServer/static-dashboards/dashboard_base.js:
42         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
43         * TestResultServer/static-dashboards/timeline_explorer.html:
44         * TestResultServer/static-dashboards/treemap.html:
45
46 2011-07-14  Yuta Kitamura  <yutak@chromium.org>
47
48         WebSocket: Introduce pywebsocket-0.6b2
49         https://bugs.webkit.org/show_bug.cgi?id=64534
50
51         Reviewed by Kent Tamura.
52
53         * Scripts/webkitpy/thirdparty/__init__.py:
54
55 2011-07-14  Dimitri Glazkov  <dglazkov@chromium.org>
56
57         Introduce TestExpectationsModel, split out of TestExpectations.
58         https://bugs.webkit.org/show_bug.cgi?id=64531
59
60         This is a simple split-and-make-work refactoring, a first step among many.
61
62         Reviewed by Adam Barth.
63
64         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Moved all model-related members
65             out of TestExpectations and into TestExpectationsModel.
66
67 2011-07-14  Eric Seidel  <eric@webkit.org>
68
69         Move webkitpy off of loose mocks
70         https://bugs.webkit.org/show_bug.cgi?id=64508
71
72         Unreviewed.  Fixing a test which fails under test-webkitpy --all (but not in a normal run).
73
74         * Scripts/webkitpy/common/checkout/scm/git.py:
75
76 2011-07-14  Adam Barth  <abarth@webkit.org>
77
78         garden-o-matic should have a "rebaseline" button
79         https://bugs.webkit.org/show_bug.cgi?id=64446
80
81         Reviewed by Ojan Vafai.
82
83         This patch adds a basic Rebaseline button that copies the baselines
84         displayed in the results pane into the appropriate directory in your
85         working copy.
86
87         There are two main limitations:
88
89         1) There is no UI for actually committing the baselines.
90
91         2) The baselines are not optimized for redundancy (meaning you can have
92            identical baselines in both chromium-mac and chromium-win).
93
94         * Scripts/webkitpy/tool/commands/rebaseline.py:
95             - Turns out we need to create the directory for the baseline if it doesn't exist yet.
96         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js:
97             - Add an programatic API to call the server.
98         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
99             - Add the rebaseline button itself.
100         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
101             - Change the CSS so that the Rebaseline and Close buttons can
102               appear in the normal order in the DOM.
103         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
104             - Bind the event and translate the parameters.
105             - Hide/show the rebaseline button, as appropriate.
106         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
107             - Add some helpful utility functions for manipulating failure types.
108         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
109
110 2011-07-14  Ryosuke Niwa  <rniwa@webkit.org>
111
112         REGRESSION: webkit-patch roll-chromium-deps is broken
113         https://bugs.webkit.org/show_bug.cgi?id=64568
114
115         Reviewed by Dirk Pranke
116
117         Use _filesystem.join.
118
119         * Scripts/webkitpy/common/checkout/checkout.py:
120         * Scripts/webkitpy/common/checkout/checkout_unittests.py:
121
122 2011-07-14  Noam Rosenthal  <noam.rosenthal@nokia.com>
123
124         Adding myself to the reviewers list; No review needed.
125
126         * Scripts/webkitpy/common/config/committers.py:
127
128 2011-07-14  Adam Roben  <aroben@apple.com>
129
130         Don't use Element.prototype.classList in TestFailures
131
132         Safari 5 doesn't support it.
133
134         Fixes <http://webkit.org/b/64550> Can't expand flaky tests on TestFailures page in Safari 5
135
136         Reviewed by Daniel Bates.
137
138         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Utilities.js:
139         (Element.prototype.hasStyleClass):
140         (Element.prototype.addStyleClass):
141         (Element.prototype.removeStyleClass):
142         (Element.prototype.toggleStyleClass):
143         Added these helper functions which simulate classList functionality.
144
145         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Utilities_unittests.js:
146         Added. Tests for the above.
147
148         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
149         (ViewController.prototype._domForPossiblyFlakyTests): Changed to use
150         toggleStyleClass/hasStyleClass instead of classList.
151
152         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
153         Added Utilities_unittests.js, and reordered the tested files to be in
154         roughly dependency order (i.e., the lowest-level files are imported
155         and tested first).
156
157 2011-07-14  Eric Seidel  <eric@webkit.org>
158
159         NRWT doesn't store the svn revision in full_results.json on chromium-win
160         https://bugs.webkit.org/show_bug.cgi?id=64492
161
162         Unreviewed.  Just fixing my test-webkitpy regression.
163
164         Fix the unit tests.  Unfortunately scm does not use a filesystem
165         object so we can't control the result of detect_scm_system.
166         When detect_scm_system would fail, we would log, which would
167         cause all passing_run integration tests to fail.
168
169         * Scripts/webkitpy/common/checkout/scm/scm.py:
170         * Scripts/webkitpy/layout_tests/controllers/manager.py:
171         * Scripts/webkitpy/layout_tests/port/base.py:
172         * Scripts/webkitpy/layout_tests/port/test.py:
173         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
174
175 2011-07-14  Eric Seidel  <eric@webkit.org>
176
177         NRWT doesn't store the svn revision in full_results.json on chromium-win
178         https://bugs.webkit.org/show_bug.cgi?id=64492
179
180         Reviewed by Ojan Vafai.
181
182         This should fix the bug.  Unfortunately this code is currently
183         impossible to test since it's impossible to mock detect_scm_system at the
184         moment.  I started re-writing scm.detection.py to be mockable, but decided that
185         was best left for another day.
186
187         * Scripts/webkitpy/layout_tests/controllers/manager.py:
188         * Scripts/webkitpy/layout_tests/port/base.py:
189
190 2011-07-14  Ojan Vafai  <ojan@chromium.org>
191
192         fix coding style of dashboard_base.js
193         https://bugs.webkit.org/show_bug.cgi?id=64545
194
195         Reviewed by Adam Roben.
196
197         The code changes to the files other than dashboard_base.js
198         are just fallout from renaming globals in dashboard_base.js.
199
200         * TestResultServer/static-dashboards/aggregate_results.html:
201         * TestResultServer/static-dashboards/builders.js:
202         * TestResultServer/static-dashboards/dashboard_base.js:
203         * TestResultServer/static-dashboards/flakiness_dashboard.html:
204         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
205         * TestResultServer/static-dashboards/timeline_explorer.html:
206         * TestResultServer/static-dashboards/treemap.html:
207
208 2011-07-14  Carlos Garcia Campos  <cgarcia@igalia.com>
209
210         Unreviewed. Fix WebKit2 GTK build after r90953.
211
212         * WebKitTestRunner/GNUmakefile.am:
213
214 2011-07-14  Adam Barth  <abarth@webkit.org>
215
216         garden-o-matic results pane should be more discoverable
217         https://bugs.webkit.org/show_bug.cgi?id=64513
218
219         Reviewed by Eric Seidel.
220
221         This patch causes us to trigger the results pane on mouse clicks
222         instead of mousein.  Also, handle the case of no results details more
223         elegantly.
224
225         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
226         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
227             - Change the binding to "click".
228         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
229             - Handle the "no results URLs" case explicitly instead of never
230               calling the callback.
231         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
232         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
233             - Show a message when there are no result details to show.
234         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
235
236 2011-07-14  Adam Barth  <abarth@webkit.org>
237
238         garden-o-matic should badge TIMEOUT failures
239         https://bugs.webkit.org/show_bug.cgi?id=64435
240
241         Reviewed by Eric Seidel.
242
243         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
244         (.regression .what a[draggable].TIMEOUT::before):
245             - Add TIMEOUT.
246
247 2011-07-14  Adam Barth  <abarth@webkit.org>
248
249         Fix Python exception blocking the commit-queue.
250
251         * Scripts/webkitpy/common/checkout/scm/git.py:
252
253 2011-07-13  Eric Seidel  <eric@webkit.org>
254
255         Move webkitpy off of loose mocks
256         https://bugs.webkit.org/show_bug.cgi?id=64508
257
258         Reviewed by Adam Barth.
259
260         Using Mock has caused us more pain than help.
261         It's possible that there was a cleaner way to use it
262         (maybe Mock(class) instead of inheriting from it?).
263         But for now, I've removed all uses of Mock from mocktool.py.
264
265         I also moved run_command into the only 3 files which call it
266         instead of leaving the deprecated method in executive.py.
267
268         I changed various direct calls to os.* to use filesystem instead.
269
270         * Scripts/webkitpy/common/checkout/checkout.py:
271         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
272         * Scripts/webkitpy/common/checkout/scm/git.py:
273         * Scripts/webkitpy/common/checkout/scm/scm.py:
274         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
275         * Scripts/webkitpy/common/checkout/scm/svn.py:
276         * Scripts/webkitpy/common/system/executive.py:
277         * Scripts/webkitpy/common/system/executive_unittest.py:
278         * Scripts/webkitpy/tool/mocktool.py:
279         * Scripts/webkitpy/tool/steps/cleanworkingdirectory_unittest.py:
280         * Scripts/webkitpy/tool/steps/ensurelocalcommitifneeded.py:
281
282 2011-07-13  Eric Seidel  <eric@webkit.org>
283
284         NRWT doesn't store the svn revision in full_results.json on chromium-win
285         https://bugs.webkit.org/show_bug.cgi?id=64492
286
287         Reviewed by Adam Barth.
288
289         Add a warning message in the revision='' case to help us diagnose
290         what's going wrong on the bot.
291
292         * Scripts/webkitpy/layout_tests/controllers/manager.py:
293
294 2011-07-13  Dirk Pranke  <dpranke@chromium.org>
295
296         new-run-webkit-tests: print baseline search path as part of config output
297         https://bugs.webkit.org/show_bug.cgi?id=64499
298
299         Reviewed by Eric Seidel.
300
301         * Scripts/webkitpy/layout_tests/controllers/manager.py:
302         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
303
304 2011-07-13  Adam Barth  <abarth@webkit.org>
305
306         gardening server should have an API for parsing changelogs
307         https://bugs.webkit.org/show_bug.cgi?id=64495
308
309         Reviewed by Eric Seidel.
310
311         This patch exposes much of the same information from CommitInfo in a
312         dictionary form, which is easier to send over-the-wire as JSON to the
313         frontend.
314
315         * Scripts/webkitpy/common/checkout/checkout.py:
316         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
317         * Scripts/webkitpy/tool/mocktool.py:
318         * Scripts/webkitpy/tool/servers/gardeningserver.py:
319         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
320         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
321
322 2011-07-13  Eric Seidel  <eric@webkit.org>
323
324         REGRESSION: GitTestWithMock.test_create_patch fails
325         https://bugs.webkit.org/show_bug.cgi?id=62945
326
327         Reviewed by Daniel Bates.
328
329         I was not able to reproduce the exact failure seen in the bug,
330         however this test was failing on my machine for other reasons.
331
332         I went through and did an audit of our run_command usage, it's
333         entirely in scm classes after this change.  (Not surprising given
334         that scm.py was the second file ever created in webkit.py.)
335
336         The real bug I'm fixing here is that we were setting executive.should_log
337         when the value had changed to executive._should_log.  Now we set the right one
338         and the test works again.
339
340         * Scripts/webkitpy/common/checkout/checkout.py:
341         * Scripts/webkitpy/common/checkout/scm/git.py:
342         * Scripts/webkitpy/common/checkout/scm/scm.py:
343         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
344         * Scripts/webkitpy/common/checkout/scm/svn.py:
345
346 2011-07-13  Ilya Sherman  <isherman@chromium.org>
347
348         Fix WTF header guard style check
349         https://bugs.webkit.org/show_bug.cgi?id=64488
350
351         Reviewed by David Levin.
352
353         * Scripts/webkitpy/style/checkers/cpp.py:
354         * Scripts/webkitpy/style/checkers/cpp_unittest.py: Verify that we don't *always* suggest "WTF_" as a prefix
355
356 2011-07-13  Ojan Vafai  <ojan@chromium.org>
357
358         bring flakiness_dashboard.html closer to webkit style
359         https://bugs.webkit.org/show_bug.cgi?id=64477
360
361         Reviewed by Adam Barth.
362
363         * TestResultServer/static-dashboards/dashboard_base.js:
364         Fix bad variable name that would hit only when running the tests.
365         * TestResultServer/static-dashboards/flakiness_dashboard.html:
366         No code/logic changes. All moving/removing brackets, indents and moving things
367         to one line.
368
369 2011-07-13  Xan Lopez  <xlopez@igalia.com>
370
371         [GTK] Fix distcheck
372
373         Reviewed by Martin Robinson.
374
375         * WebKitTestRunner/GNUmakefile.am: mark built sources as nodist.
376
377 2011-07-13  Joseph Pecoraro  <joepeck@webkit.org>
378
379         Unreviewed fix. Add a "\n" to a printf that somehow got lost.
380
381         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
382         (WebCore::DumpRenderTree::dumpApplicationCacheQuota):
383
384 2011-07-13  Joseph Pecoraro  <joepeck@webkit.org>
385
386         ApplicationCache Quota Output is Flakey
387         https://bugs.webkit.org/show_bug.cgi?id=64410
388
389         Reviewed by Alexey Proskuryakov.
390
391         Unify the delegate logging for reaching application cache
392         quotas. Truncate the space needed to the nearest 10000
393         for less flakey test output. This also better supports
394         printing NSUInteger on non-64 bit systems by casting to
395         an unsigned long.
396
397         * DumpRenderTree/mac/UIDelegate.mm:
398         (-[UIDelegate webView:exceededApplicationCacheOriginQuotaForSecurityOrigin:totalSpaceNeeded:]):
399         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
400         (WebCore::DumpRenderTree::dumpApplicationCacheQuota):
401
402 2011-07-13  Dimitri Glazkov  <dglazkov@chromium.org>
403
404         Remove unused TestExpectations._overrides.
405         https://bugs.webkit.org/show_bug.cgi?id=64470
406
407         Reviewed by Adam Barth.
408
409         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Removed unused member.
410
411 2011-07-13  Dimitri Glazkov  <dglazkov@chromium.org>
412
413         Introduce TestExpectationSerializer.list_to_string.
414         https://bugs.webkit.org/show_bug.cgi?id=64462
415
416         Reviewed by Adam Barth.
417
418         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added list_to_string and change the relevant callsite to use it.
419         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests for it.
420
421 2011-07-13  Dimitri Glazkov  <dglazkov@chromium.org>
422
423         Consolidate expectations parsing code.
424         https://bugs.webkit.org/show_bug.cgi?id=64460
425
426         Reviewed by Adam Barth.
427
428         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Folded TestExpectationParser._split_expectation_string into TestExpectationParser.parse.
429
430 2011-07-12  Brent Fulgham  <bfulgham@webkit.org>
431
432         Standardize WinCairo conditionalized code under PLATFORM macro.
433         https://bugs.webkit.org/show_bug.cgi?id=64377
434
435         Reviewed by Maciej Stachowiak.
436
437         Update compile-time conditions that had previously used #ifdef
438         of WIN_CAIRO to consistently use PLATFORM(WIN_CAIRO).
439
440         * DumpRenderTree/config.h: Switch to PLATFORM(WIN_CAIRO)
441         * DumpRenderTree/win/DumpRenderTreeCairo.vsprops: Remove redundant
442           define of WIN_CAIRO=1.
443         * DumpRenderTree/win/DumpRenderTreeDebugCairoCFLite.vsprops: Add
444           WinCairo.vsprops to set (to match Release build).
445
446 2011-07-13  Dimitri Glazkov  <dglazkov@chromium.org>
447
448         Eliminate TestExpectationsFile.
449         https://bugs.webkit.org/show_bug.cgi?id=64458
450
451         Turns out, we can just use a Python list.
452
453         Reviewed by Adam Barth.
454
455         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Folded TestExpectationsFile.append into TestExpectationParser.parse_list,
456             removed TestExpectationsFile.
457         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Moved tests to better reflect new names, removed iterator test,
458             since there's no more custom iterator machinery.
459
460 2011-07-13  Philippe Normand  <pnormand@igalia.com>
461
462         Unreviewed, added my other email addresses.
463
464         * Scripts/webkitpy/common/config/committers.py:
465
466 2011-07-13  Adam Barth  <abarth@webkit.org>
467
468         Folks should only be listed once.
469
470         * Scripts/webkitpy/common/config/committers.py:
471
472 2011-07-13  Philippe Normand  <pnormand@igalia.com>
473
474         Unreviewed, adding myself as Reviewer.
475
476         * Scripts/webkitpy/common/config/committers.py:
477
478 2011-07-13  Adam Barth  <abarth@webkit.org>
479
480         Fix flickering bug introduced by my previous patch.  I forgot to change
481         the name of the class everywhere.
482
483         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
484
485 2011-07-12  Philippe Normand  <pnormand@igalia.com>
486
487         [GTK] media/media-blocked-by-willsendrequest.html fails
488         https://bugs.webkit.org/show_bug.cgi?id=63699
489
490         Reviewed by Martin Robinson.
491
492         * DumpRenderTree/gtk/DumpRenderTree.cpp:
493         (willSendRequestCallback): Abort the request if explicitely asked
494         by the LayoutTestController's willSendRequestReturnsNull() function.
495
496 2011-07-13  Adam Barth  <abarth@webkit.org>
497
498         Remove "Dismiss" button from garden-o-matic butterbar
499         https://bugs.webkit.org/show_bug.cgi?id=64443
500
501         Reviewed by Dimitri Glazkov.
502
503         We don't have any persistent butter bar messages yet, so the dismiss
504         button is premature (and fairly heavy, visually).
505
506         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
507         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
508
509 2011-07-13  Adam Barth  <abarth@webkit.org>
510
511         Refine garden-o-matic status pane
512         https://bugs.webkit.org/show_bug.cgi?id=64442
513
514         Reviewed by Dimitri Glazkov.
515
516         This patch makes two improvements to the status pane:
517
518         1) We only query the server for failure types that we're expecting.
519            This dramatically reduces the number of HTTP requests, making
520            loading the status pane faster.
521
522         2) The status pane now displays which test and which builder it is
523            showing results for.
524
525         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js:
526         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
527         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
528         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
529         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
530         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
531         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
532
533 2011-07-13  Adam Barth  <abarth@webkit.org>
534
535         garden-o-matic should show test results
536         https://bugs.webkit.org/show_bug.cgi?id=64440
537
538         Reviewed by Adam Roben.
539
540         This patch introduces the results detail pane, which appears at the
541         bottom of the window and contains test results from the bots.
542         Currently, you can activate the pane by mousing over one of the builder
543         names associated with a failing test.
544
545         This is just a first iteration of the UI.  There's no way to resize or
546         zoom in on elements of the details pane, and images likely aren't sized
547         correctly, but it's a place to start.
548
549         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js:
550             - Add constants for our data attributes so we don't typo them!
551         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
552             - Add DOM structure for the details pane.
553         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
554         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
555             - Bind events for showing and hiding the details pane.
556         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
557             - Reduce the number of result types that we query for to avoid
558               overfilling the details pane.  We'll probably need another
559               solution here in the long-term.  For example, we could use tabs
560               to pack more results into the pane.
561         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
562         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html:
563             - Now that we're using the config package during testing, we need
564               to include it in the testing HTML.
565         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
566         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
567             - Change the DOM structure of the results detail to make it fit
568               nicely in the details pane (rather than flowing freely in the
569               body, as it did before).
570
571 2011-07-13  Adam Roben  <aroben@apple.com>
572
573         Make TestFailures show every time a possibly-flaky test failed, but hide it by default
574
575         It's useful to be able to see every time a flaky test failed to see whether it failed the
576         same way every time. But doing so takes a lot of space, so the list of failures is now
577         collapsed by default and can be revealed using a disclosure triangle.
578
579         Fixes <http://webkit.org/b/64455> TestFailures page doesn't show as much information for
580         flaky tests as I would like, even though the page is already so long
581
582         Reviewed by Dimitri Glazkov.
583
584         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyLayoutTestDetector.js:
585         (FlakyLayoutTestDetector.prototype.allFailures): Replaced flakinessExamples with this
586         function. Now returns all failures for the given test.
587
588         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyLayoutTestDetector_unittests.js:
589         Added. This just contains some simple tests of the FlakyLayoutTestDetector class. We'll add
590         more over time.
591
592         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestHistoryAnalyzer.js:
593         (LayoutTestHistoryAnalyzer.prototype.start): Updated the documentation comment to reflect
594         that we no longer return passing builds for possibly-flaky tests.
595
596         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
597         (.existing-bugs-list, .suspect-revisions-list, .flakiness-examples-list): Make the list of
598         flakiness examples small, too, since it can get quite long.
599
600         (.disclosure-triangle):
601         (.expanded > .disclosure-triangle):
602         Simple styles for the disclosure triangle.
603
604         (.flakiness-examples-list): Collapse the list by default.
605
606         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
607         (ViewController.prototype._displayBuilder): Pass the total number of builds analyzed to
608         _domForPossiblyFlakyTests.
609         (ViewController.prototype._domForPossiblyFlakyTests): Put a disclosure triangle to the left
610         of each test name, and the number of failures to the right. When the disclosure triangle is
611         clicked for the first time, we build up the list of failures and expand the element. After
612         that we just collapse or expand the element on subsequent clicks.
613
614         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
615         Pulled in new tests.
616
617 2011-07-13  Adam Roben  <aroben@apple.com>
618         
619         Teach TestFailures to understand NRWT's output when it exits early due to too many failures
620
621         Fixes <http://webkit.org/b/64456> TestFailures page reports way too many failures when NRWT
622         exits early
623
624         Reviewed by Dimitri Glazkov.
625
626         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
627         (Builder.prototype.getNumberOfFailingTests): Relaxed the "Exiting early" test not to require
628         it to be at the beginning of the line, since NRWT prints a bunch of junk earlier in the
629         line. Tightened up the regex that's used to parse the number of failing tests to require the
630         leading number to be followed by whitespace so that we won't parse the "2011" in
631         "2011-07-13" as a number of failures.
632
633         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
634         Added a test for the above.
635
636 2011-07-12  Dimitri Glazkov  <dglazkov@chromium.org>
637
638         Extract model-like TestExpectationLine and TestExpectationFile from TestExpectations.
639         https://bugs.webkit.org/show_bug.cgi?id=64386
640
641         This is the first step in converting TestExpectations to a real model.
642         * TestExpectationsLine represents a line in the test_expectations.txt file, and
643         * TestExpectationsFile represents the file, which is a collection of lines.
644
645         Reviewed by Adam Barth.
646
647         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
648         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
649         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
650
651 2011-07-13  Xan Lopez  <xlopez@igalia.com>
652
653         [GTK] Do not grab focus too early in DRT.
654
655         Reviewed by Gustavo Noronha.
656
657         It causes a layout to happen and a progress signal to be emitted
658         since r90900, but at this point we don't have a
659         LayoutTestController object and we'll eventually crash. Since we
660         already grab focus at the beginning of runTest() this is
661         redundant, so get rid of it to fix the crash.
662
663         * DumpRenderTree/gtk/DumpRenderTree.cpp:
664         (main): remove call to grab_focus
665
666 2011-07-13  Sheriff Bot  <webkit.review.bot@gmail.com>
667
668         Unreviewed, rolling out r90893 and r90894.
669         http://trac.webkit.org/changeset/90893
670         http://trac.webkit.org/changeset/90894
671         https://bugs.webkit.org/show_bug.cgi?id=64441
672
673         NRWT still doesn't work on qt-mac platform (Requested by Ossy
674         on #webkit).
675
676         * Scripts/run-webkit-tests:
677         (useNewRunWebKitTests):
678
679 2011-07-13  Csaba Osztrogonác  <ossy@webkit.org>
680
681         [Qt] NRWT should pick up the right httpd config file
682         https://bugs.webkit.org/show_bug.cgi?id=64086
683
684         * Scripts/run-webkit-tests: Enable NRWT on qt-mac platform after r90810.
685         (useNewRunWebKitTests):
686
687 2011-07-12  Mark Rowe  <mrowe@apple.com>
688
689         Fix the 32-bit build.
690
691         * DumpRenderTree/mac/UIDelegate.mm:
692         (-[UIDelegate webView:exceededApplicationCacheOriginQuotaForSecurityOrigin:totalSpaceNeeded:]):
693         Cast the NSUInteger value to unsigned long to match the format specifier.
694
695 2011-07-12  Adam Barth  <abarth@webkit.org>
696
697         Improve garden-o-matic UI when the bots fail to report revision numbers
698         https://bugs.webkit.org/show_bug.cgi?id=64427
699
700         Reviewed by Dimitri Glazkov.
701
702         At least Win (dbg)(2) seems to fail to report the SVN revision number
703         in full_results.json.  This patch makes garden-o-matic more robust to
704         missing revision numbers.
705
706         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
707         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
708         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
709         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
710         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
711
712 2011-07-12  Adam Barth  <abarth@webkit.org>
713
714         Adjust garden-o-matic layout to use fewer lines
715         https://bugs.webkit.org/show_bug.cgi?id=64422
716
717         Reviewed by Ojan Vafai.
718
719         This patch effectively merges the test name line with the list of
720         builders on which the test fails.  Each test failure now occupies two
721         lines instead of three.
722
723         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
724         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
725         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
726
727 2011-07-12  Adam Barth  <abarth@webkit.org>
728
729         garden-o-matic should label tests failures of new tests as such
730         https://bugs.webkit.org/show_bug.cgi?id=64421
731
732         Reviewed by Ojan Vafai.
733
734         Also, we shouldn't dim failures of new tests, even if we've only seen
735         them once because they're likely to be real problems that need
736         attention.
737
738         * Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js:
739         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js: Added.
740         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout_unittests.js: Added.
741         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
742         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
743         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
744         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
745         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html:
746         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
747         * Scripts/webkitpy/tool/servers/gardeningserver.py:
748
749 2011-07-12  Adam Barth  <abarth@webkit.org>
750
751         garden-o-matic should display how many times we've seen a failure
752         https://bugs.webkit.org/show_bug.cgi?id=64417
753
754         Reviewed by Ojan Vafai.
755
756         This patch adds some UI to display how many times we've seen a given
757         failure, which can be helpful for determining whether that failure is a
758         real failure or a flaky test.
759
760         When a failure has only been seen once (i.e., only a single run on a
761         single bot), we set the opacity of to 50% to avoid distracting the
762         gardener.
763
764         This patch also refactors the failure walker to have a simpler API
765         internally by moving from an object-oriented paradigm to a functional
766         paradigm.
767
768         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
769         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
770         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
771         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
772         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
773         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
774
775 2011-07-12  Chris Rogers  <crogers@google.com>
776
777         Enable Web Audio for chromium DRT
778         https://bugs.webkit.org/show_bug.cgi?id=64409
779
780         Reviewed by James Robinson.
781
782         * DumpRenderTree/chromium/TestShell.cpp:
783         (TestShell::TestShell):
784
785 2011-07-12  Adam Barth  <abarth@webkit.org>
786
787         Tweak some UI in garden-o-matic now that the tree actually has a
788         failure and I can see what this all looks like.
789
790         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
791         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
792         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
793
794 2011-07-12  Joseph Pecoraro  <joepeck@webkit.org>
795
796         Unreviewed. Skipping a few tests which fail due to differing output
797         in recent ApplicationCache quota tests. Reenabling the tests is
798         tracked by <http://webkit.org/b/64410>.
799
800         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
801         (WebCore::DumpRenderTree::dumpApplicationCacheQuota): This should
802         have been a space, not a newline.
803
804 2011-07-12  Adam Barth  <abarth@webkit.org>
805
806         garden-o-matic should display regression ranges
807         https://bugs.webkit.org/show_bug.cgi?id=64407
808
809         Reviewed by Dimitri Glazkov.
810
811         This patch computes a regression range for a failure by intersecting
812         the regression ranges seen by the various bots.  We make the underlying
813         assumption that a test is only failing due to one revision at any given
814         moment.  If that's not true, this code probably explodes.
815
816         The regression ranges appear asynchronously, which might be a jarring
817         UI.  We'll have to experiment to see.
818
819         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
820         * Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js:
821         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
822         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
823         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
824         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
825         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
826
827 2011-07-12  Adam Barth  <abarth@webkit.org>
828
829         cr-linux-ews should run pixel tests
830         https://bugs.webkit.org/show_bug.cgi?id=64394
831
832         Reviewed by Ojan Vafai.
833
834         This patch enabled pixel test on the cr-linux-ews.  Running a quick
835         test on the bots, it looks like we have five pixel failures, related to
836         some missing fonts.  I've also made the bots skip failing tests to
837         avoid uploading huge zip files to bugs.webkit.org (with all the exected
838         IMAGE failures).
839
840         * Scripts/webkitpy/common/config/ports.py:
841         * Scripts/webkitpy/common/config/ports_unittest.py:
842
843 2011-07-12  Joseph Pecoraro  <joepeck@webkit.org>
844
845         ApplicationCache update should not immediately fail when reaching per-origin quota
846         https://bugs.webkit.org/show_bug.cgi?id=64177
847
848         Reviewed by Alexey Proskuryakov.
849
850         Add layoutTestController.disallowIncreaseForApplicationCacheQuota()
851         to disable the default behavior of raising the default per-origin
852         quota to 5MB when reached. This allows us to test what happens
853         when the quota is reached and not increased.
854
855         * DumpRenderTree/LayoutTestController.cpp:
856         (LayoutTestController::LayoutTestController):
857         (disallowIncreaseForApplicationCacheQuotaCallback):
858         (LayoutTestController::staticFunctions):
859         * DumpRenderTree/LayoutTestController.h:
860         (LayoutTestController::disallowIncreaseForApplicationCacheQuota):
861         (LayoutTestController::setDisallowIncreaseForApplicationCacheQuota):
862         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
863         (LayoutTestController::reset):
864         * DumpRenderTree/qt/LayoutTestControllerQt.h:
865         (LayoutTestController::disallowIncreaseForApplicationCacheQuota):
866         (LayoutTestController::dumpApplicationCacheDelegateCallbacks):
867         Expose a function to prevent automatically increasing the per-origin
868         quota when the quota is reached. This allows us to test what happens
869         when a user would "disallow" a quota increase.
870
871         * DumpRenderTree/mac/UIDelegate.mm:
872         (-[UIDelegate webView:exceededApplicationCacheOriginQuotaForSecurityOrigin:spaceNeeded:]):
873         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
874         (WebCore::DumpRenderTree::dumpApplicationCacheQuota):
875         Respect the new disallowIncrease flag and dump the spaceNeeded values
876         for ports that implement application cache quotas.
877
878 2011-07-12  Ojan Vafai  <ojan@chromium.org>
879
880         fix flakiness dashboard to work with new crash log filenames
881         https://bugs.webkit.org/show_bug.cgi?id=64393
882
883         Reviewed by Adam Barth.
884
885         * TestResultServer/static-dashboards/flakiness_dashboard.html:
886
887 2011-07-12  Adam Barth  <abarth@webkit.org>
888
889         Add a happy animation to garden-o-matic when there are no failures
890         https://bugs.webkit.org/show_bug.cgi?id=64382
891
892         Reviewed by Ojan Vafai.
893
894         As requested by Ojan.
895
896         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
897         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
898         * Scripts/webkitpy/tool/servers/data/gardeningserver/partytime.gif: Added.
899         * Scripts/webkitpy/tool/servers/gardeningserver.py:
900
901 2011-07-12  Adam Barth  <abarth@webkit.org>
902
903         sheriffbot can't perform rollouts
904         https://bugs.webkit.org/show_bug.cgi?id=64370
905
906         Reviewed by Eric Seidel.
907
908         Now that we're not updating the working copy before processing every
909         IRC message, the rollout command needs to update it manually.
910
911         * Scripts/webkitpy/tool/bot/irc_command.py:
912         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
913
914 2011-07-12  Adam Barth  <abarth@webkit.org>
915
916         Fix the grammar in the FIXME comment.
917
918         * Scripts/webkitpy/common/net/resultsjsonparser.py:
919
920 2011-07-12  Eric Seidel  <eric@webkit.org>
921
922         NRWT should open test results page with Safari trunk, not the system provided one on Mac
923         https://bugs.webkit.org/show_bug.cgi?id=64346
924
925         Reviewed by Adam Barth.
926
927         To fix this I implemented Port.show_results_html_file in Mac, Gtk and Qt ports with
928         implementations (mostly) matching those found in old-run-webkit-tests.
929         There are still some minor differences for Qt which Qt hackers may wish to tweak.
930
931         I had to add a WebKitPort._port_flag_for_scripts method (similar to flag() in common.config.ports.py)
932         for the Qt/Gtk ports which always require a flag passed to scripts.
933
934         While trying to test this, I found that FactoryTest.test_chromium_gpu_linux
935         was using a real filesystem (due to assert_platform) which due to
936         global variables in config.py was causing set-webkit-configuration to have
937         an affect on unit test results!  So I fixed this by making FactoryTest.assert_port
938         pass a mock file system whenever calling factory.get.
939
940         Unfortunately TestPort was depending on always being passed a None filesystem
941         and asserting filesystem._tests (only true for unit_test_filesystem()).
942         So I just removed the FactoryTest.test_test and FactoryTest.test_dryrun tests
943         deciding that they were pretty much useless anyway.  If others feel strongly
944         I'm happy to fix this in a different way.
945
946         * Scripts/webkitpy/common/system/executive.py:
947          - default arguments in python are screwy.  They use a single shared
948            instance, so it's better to use argument=None and then argument = argument or Default()
949            if you have any chance of mutating (or returning) the default argument.
950         * Scripts/webkitpy/layout_tests/port/config.py:
951          - This code is wrong.  We don't need to use a global variable here (as far as I can tell).
952            I'm not fixing it in this patch, but I've marked it with a FIXME and we can convert to
953            storing the results of the read on the Config object (which should only be created once during normal operation).
954            Unit tests shouldn't be hitting the disk anyway.  It's possible Config should move off of Port and onto Tool/Host directly.
955         * Scripts/webkitpy/layout_tests/port/factory.py:
956         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
957         * Scripts/webkitpy/layout_tests/port/gtk.py:
958         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py: Added.
959         * Scripts/webkitpy/layout_tests/port/mac.py:
960         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
961         * Scripts/webkitpy/layout_tests/port/qt.py:
962         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
963         * Scripts/webkitpy/layout_tests/port/webkit.py:
964
965 2011-07-12  Adam Roben  <aroben@apple.com>
966
967         Teach TestFailures to recognize when run-webkit-tests gets killed by buildbot
968
969         Fixes <http://webkit.org/b/64358> TestFailures page thinks all tests passed in
970         http://build.webkit.org/builders/Windows%207%20Release%20(Tests)/builds/14672
971
972         Reviewed by Daniel Bates.
973
974         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
975         (Builder.prototype.getNumberOfFailingTests): If run-webkit-tests exited with a non-zero
976         exit status but we didn't find any failure counts, assume that there was some error that
977         caused run-webkit-tests to die early (like being killed by buildbot due to a timeout).
978
979         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
980         Added a new test that shows that we get a failingTestCount of -1 when run-webkit-tests dies
981         early.
982         (runGetNumberOfFailingTestsTest): Moved most code here from the only pre-existing test in this
983         file.
984
985         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:
986         (LayoutTestResultsLoader.prototype.start): Bump the cache number so old cached data that was
987         tainted by the bug fixed in this patch will be evicted.
988
989 2011-07-12  Adam Barth  <abarth@webkit.org>
990
991         cr-linux-ews complains about tests that aren't actually failing
992         https://bugs.webkit.org/show_bug.cgi?id=64367
993
994         The underlying problem here is that full_results.json doesn't have
995         enough information to determine whether a given test result was
996         expected because whether an actual result is expected depends on
997         whether full_results.json was generated duing a test run that included
998         pixel tests.
999
1000         The right long-term solution is to make full_results.json a complete
1001         description of what happened durning a test run.  In this patch
1002         however, to stop the spam, this patch makes the jsonresultsparser
1003         ignore pixel failures.
1004
1005         (I'm landing this patch unreviewed in order to stop the bots from
1006         spamming.  I'll happily address any review comments in a follow-up
1007         patch.)
1008
1009         * Scripts/webkitpy/common/net/resultsjsonparser.py:
1010         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
1011
1012 2011-07-12  Adam Barth  <abarth@webkit.org>
1013
1014         Re-work garden-o-matic UI to begin to look like the mocks
1015         https://bugs.webkit.org/show_bug.cgi?id=64334
1016
1017         Reviewed by Dimitri Glazkov.
1018
1019         This patch changes how we display failures to be more like the mock the
1020         dglazkov made. We now can display the same information much more
1021         compactly.
1022
1023         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
1024         * Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js:
1025         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
1026         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css: Added.
1027         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
1028         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
1029         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
1030         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
1031         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
1032         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1033
1034 2011-07-11  Adam Roben  <aroben@apple.com>
1035
1036         Extract TestFailures's bug-filing code into two new classes
1037
1038         Fixes <http://webkit.org/b/64300> TestFailures page's new-bug-filing code is a mess!
1039
1040         Reviewed by Darin Adler and Adam Barth.
1041
1042         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Buildbot.js:
1043         (Buildbot.prototype.resultsDirectoryURL): Changed to return a URI-encoded URL. Otherwise the
1044         URL isn't valid (and it's harder to mock this function).
1045
1046         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Buildbot_unittests.js:
1047         Added. Just tests the above change (for now).
1048
1049         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
1050         Added a license header, enclosed everything in a closure to avoid polluting the global
1051         namespace, changed the test name to actually describe the passing condition, and replaced
1052         uses of equals() with equal(). (The latter is the actual name of the function, and matches
1053         deepEqual, etc.)
1054
1055         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NewBugForm.js: Added.
1056         (NewBugForm): This class knows how to construct a <form> used for filing new bugs in
1057         Bugzilla based on some parameters.
1058         (NewBugForm.prototype.domElement): Creates and returns the <form> element. Code came from
1059         ViewController.prototype._domForNewAndExistingBugs.
1060
1061         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NewBugForm_unittests.js:
1062         Added. Tests the above code.
1063
1064         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailureBugForm.js: Added.
1065         (TestFailureBugForm): This class knows how to construct a <form> element used for filing new
1066         bugs specifically about test failures. Code came from
1067         ViewController.prototype._domForNewAndExistingBugs.
1068         (TestFailureBugForm.prototype.domElement): Slightly customizes the <form> element returned
1069         by NewBugForm.
1070         (TestFailureBugForm.prototype._computeOperatingSystem):
1071         (TestFailureBugForm.prototype._computePlatform):
1072         (TestFailureBugForm.prototype._createBugTitle):
1073         (TestFailureBugForm.prototype._failingResultsHTMLURL):
1074         (TestFailureBugForm.prototype._failingRevision):
1075         (TestFailureBugForm.prototype._passingRevision):
1076         (TestFailureBugForm.prototype._regressionRangeString):
1077         Code came from ViewController.prototype._domForNewAndExistingBugs. I broke it out into
1078         separate functions to break up the rat's nest a bit.
1079
1080         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailureBugForm_unittests.js:
1081         Added. Tests the above code.
1082
1083         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1084         (ViewController.prototype._domForNewAndExistingBugs): Moved code from here to
1085         TestFailureBugForm/NewBugForm, and changed this code to use a TestFailureBugForm.
1086
1087         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
1088         NewBugForm/TestFailureBugForm.js.
1089
1090         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
1091         Added new tests and required files.
1092
1093 2011-07-12  Adam Roben  <aroben@apple.com>
1094
1095         Test that no intermediate WTF::Strings are created when concatenating with string literals
1096
1097         Test for <http://webkit.org/b/63330> Concatenating string literals and WTF::Strings using
1098         operator+ is suboptimal
1099
1100         Reviewed by Darin Adler.
1101
1102         * TestWebKitAPI/Tests/WTF/StringOperators.cpp: Added.
1103         (TestWebKitAPI::TEST): Test that a bunch of different string concatenation expressions don't
1104         create any intermediate WTF::Strings while they're being evaluated.
1105
1106         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1107         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
1108         Added new file.
1109
1110 2011-07-12  Eric Seidel  <eric@webkit.org>
1111
1112         [Qt] NRWT should pick up the right httpd config file
1113         https://bugs.webkit.org/show_bug.cgi?id=64086
1114
1115         Reviewed by Adam Barth.
1116
1117         This is more fall-out from the Port class inappropriately encapsulating
1118         both platform and port knowledge.  We need to split out some of this
1119         platform knowledge into a separate class which can be better shared
1120         between ports.
1121
1122         The fix was to move all the _path_to_apache_config_file logic down
1123         into the WebKitPort baseclass so that all of the ports can find the
1124         right apache config file, regardless of what platform they may be running on.
1125
1126         I did not move this down into base.Port (even though I considered it).
1127         Chromium duplicates some of this logic, but since they have separate
1128         subclasses for each port-OS combination (e.g. ChromiumMac, ChromiumWin)
1129         they wouldn't notice the change.  Eventually we'll move this logic
1130         out of Port entirely, and then it will be shared by all ports.
1131
1132         I also cleaned up the http-config logic for ORWT while I was there,
1133         although since we're killing that code, I'm happy to revert that part
1134         of the change if changing it is viewed as needlessly risky.
1135
1136         During this cleanup, I noticed that no ports use the "default" httpd.conf
1137         which is left over from Mac OS X Tiger (and old linux distros) which used
1138         Apache 1.3.  I've removed httpd.conf (and associated support in ORWT) since
1139         we no longer support any ports which use this httpd.conf.
1140
1141         * Scripts/webkitperl/httpd.pm:
1142         (getHTTPDConfigPathForTestDirectory):
1143         * Scripts/webkitpy/layout_tests/port/base.py:
1144         * Scripts/webkitpy/layout_tests/port/gtk.py:
1145         * Scripts/webkitpy/layout_tests/port/mac.py:
1146         * Scripts/webkitpy/layout_tests/port/qt.py:
1147         * Scripts/webkitpy/layout_tests/port/webkit.py:
1148         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1149         * Scripts/webkitpy/layout_tests/port/win.py:
1150         * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
1151
1152 2011-07-11  Hironori Bono  <hbono@chromium.org>
1153
1154         Reviewed by Adam Roben.
1155
1156         Implement layoutTestController.setTextDirection for WebKit2 and Windows.
1157         https://bugs.webkit.org/show_bug.cgi?id=61931
1158
1159         This change implements layoutTextController.setTextDirection for WebKit2
1160         and Windows so we can run a layout test added by r87770 on them. Each
1161         implementation adds a wrapper function for Editor::setBaseWritingDirection()
1162         so LayoutTestController can call it.
1163
1164         * DumpRenderTree/win/LayoutTestControllerWin.cpp:
1165         (LayoutTestController::setTextDirection): Implemented the binding function
1166         for layoutTestController.setTextDirection (Windows).
1167         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
1168         Added a binding function for layoutTestController.setTextDirection (WebKit2).
1169         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp: ditto.
1170         (WTR::LayoutTestController::setTextDirection):
1171         * WebKitTestRunner/InjectedBundle/LayoutTestController.h: ditto.
1172
1173 2011-07-11  Raphael Kubo da Costa  <kubo@profusion.mobi>
1174
1175         [EFL] ImageDiff: Make sure gEcoreEvas is destroyed at the right time.
1176         https://bugs.webkit.org/show_bug.cgi?id=64293
1177
1178         Reviewed by Kent Tamura.
1179
1180         gEcoreEvas was being destroyed after shutdownEfl() was called, so the
1181         call to ecore_evas_free() failed.
1182
1183         We now shut it down manually before shutdownEfl() is called.
1184
1185         * DumpRenderTree/efl/ImageDiff.cpp:
1186         (main): Destroy gEcoreEvas before shutting down the EFL.
1187
1188 2011-07-11  Ryosuke Niwa  <rniwa@webkit.org>
1189
1190         Make Chromium Mac and Linux tests bot core builders
1191         https://bugs.webkit.org/show_bug.cgi?id=63196
1192
1193         Reviewed by Adam Barth.
1194
1195         Added Chromium Mac and Linux Release Tests bots core builders.
1196
1197         Also rearranged the order of bots so that test bots show up right next to
1198         respective builders.
1199
1200         * BuildSlaveSupport/build.webkit.org-config/config.json:
1201         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
1202         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
1203
1204 2011-07-11  Raphael Kubo da Costa  <kubo@profusion.mobi>
1205
1206         [EFL] DRT: Add code related to font management.
1207         https://bugs.webkit.org/show_bug.cgi?id=63989
1208
1209         Reviewed by Kent Tamura.
1210
1211         The added files are responsible for managing fontconfig and adding the
1212         fonts required by some layout tests to fontconfig's path.
1213
1214         * DumpRenderTree/efl/FontManagement.cpp: Added.
1215         (getFontDirectories):
1216         (getFontFiles):
1217         (addFontDirectories):
1218         (addFontFiles):
1219         (addFontsToEnvironment):
1220         * DumpRenderTree/efl/FontManagement.h: Added.
1221
1222 2011-07-08  Dirk Pranke  <dpranke@chromium.org>
1223
1224         nrwt: linting fixes
1225         https://bugs.webkit.org/show_bug.cgi?id=64225
1226
1227         Reviewed by Eric Siedel.
1228
1229         Miscellaneous linting fixes. The most notable change is that
1230         we add public attributes for user, executive, filesystem, and
1231         options on the Port object, so we don't have to refer to the
1232         "protected" versions all over the place".
1233
1234         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1235         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1236         * Scripts/webkitpy/layout_tests/controllers/message_broker.py:
1237         * Scripts/webkitpy/layout_tests/controllers/message_broker_unittest.py:
1238         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1239         * Scripts/webkitpy/layout_tests/port/base.py:
1240         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
1241
1242 2011-07-11  Ryosuke Niwa  <rniwa@webkit.org>
1243
1244         webkit-patch roll-chromium-deps no longer works
1245         https://bugs.webkit.org/show_bug.cgi?id=64324
1246
1247         Reviewed by James Robinson.
1248
1249         Lower the minimum commit message length to 10.
1250
1251         * Scripts/webkitpy/tool/steps/commit.py:
1252
1253 2011-07-11  Alice Boxhall  <aboxhall@chromium.org>
1254
1255         Convert json_results_generator.py to output version 4 JSON.
1256         https://bugs.webkit.org/show_bug.cgi?id=60869
1257
1258         Reviewed by Ojan Vafai.
1259
1260         * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
1261         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
1262         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1263
1264 2011-07-11  Martin Robinson  <mrobinson@igalia.com>
1265
1266         [GTK] [NRWT] Pixel tests do not work
1267         https://bugs.webkit.org/show_bug.cgi?id=64091
1268
1269         Reviewed by Eric Seidel.
1270
1271         Fix pixel test runs for GTK+ new-run-webkit-tests by exposing the
1272         path to the ImageDiff binary.
1273
1274         * Scripts/webkitpy/layout_tests/port/gtk.py: Implement _path_to_image_diff.
1275
1276 2011-07-11  Dirk Pranke  <dpranke@chromium.org>
1277
1278         Fix hang on win32 in a multiprocessing test that shouldn't have been running.
1279
1280         Unreviewed, build fix.
1281
1282         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1283
1284 2011-07-11  Adam Roben  <aroben@apple.com>
1285
1286         Don't count leaks as test failures on TestFailures's front page
1287
1288         As a bonus, this patch adds our first unit test.
1289
1290         Fixes <http://webkit.org/b/64303> REGRESSION (r90489): TestFailures page says far more tests
1291         are failing on the Leaks bot than actually are
1292
1293         Reviewed by Adam Barth.
1294
1295         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
1296         (Builder.prototype.getNumberOfFailingTests): Exclude lines containing "leak" when summing
1297         failure counts.
1298
1299         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
1300         Added. Contains a simple test to show that leaks aren't counted as test failures.
1301
1302         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
1303         Added. This is the test harness.
1304
1305 2011-06-14  Mike Stegeman  <mrstegeman@gmail.com>
1306
1307         [GTK] Add API to allow setting local storage database path
1308         https://bugs.webkit.org/show_bug.cgi?id=62091
1309
1310         Reviewed by Martin Robinson and Gustavo Noronha.
1311
1312         Expose the path of the localStorage databases through a setting
1313         to allow HTML5 localStorage to be persistent. New setting is:
1314         html5-local-storage-database-path
1315
1316         * DumpRenderTree/gtk/DumpRenderTree.cpp:
1317         (resetDefaultsToConsistentValues):
1318
1319 2011-07-11  Adam Barth  <abarth@webkit.org>
1320
1321         garden-o-matic should be able to rebaseline tests
1322         https://bugs.webkit.org/show_bug.cgi?id=64186
1323
1324         Reviewed by Eric Seidel.
1325
1326         This patch contains a simple HTTP server binding for the
1327         rebaseline-test command.  The frontend will appear in another patch.
1328
1329         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1330         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
1331         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
1332
1333 2011-07-11  Adam Barth  <abarth@webkit.org>
1334
1335         Add some basic UI for showing regression ranges
1336         https://bugs.webkit.org/show_bug.cgi?id=64243
1337
1338         Reviewed by Eric Seidel.
1339
1340         This UI is all temporary.  It exists mostly just to have some buttons
1341         to click to exercise the code.
1342
1343         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
1344         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
1345         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
1346         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
1347         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
1348
1349 2011-07-11  Adam Barth  <abarth@webkit.org>
1350
1351         garden-o-matic should be able to determine which revisions caused a given failure
1352         https://bugs.webkit.org/show_bug.cgi?id=64189
1353
1354         Reviewed by Adam Roben.
1355
1356         Walking the failure history looking for failures turns out to be
1357         slightly tricky because the network requests are asynchronous.
1358         Currently we do all the fetches serially and our cache is unbounded.
1359         We'll probably optimize both those parameters eventually.
1360
1361         This patch also generalizes some functionality in the unit testing
1362         framework to make testing this sort of code easier.
1363
1364         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
1365         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
1366         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
1367
1368 2011-07-11  Adam Barth  <abarth@webkit.org>
1369
1370         Add a webkit-patch command for rebaselining an individual test
1371         https://bugs.webkit.org/show_bug.cgi?id=64246
1372
1373         Reviewed by Eric Seidel.
1374
1375         This patch introduces a command that's able to rebaseline a single
1376         test.  Currently, the command works only with the build.chromium.org
1377         buildbots, but extending it to work with the build.webkit.org bots
1378         shouldn't be that hard.
1379
1380         A complete rebaseling tool should also include an "optimize baselines"
1381         command (which moves/deletes baselines in order to reduce the number of
1382         expected results files), but that will come in a future patch.
1383
1384         Really BuilderToPort should be merged into builders.py, but I'm going
1385         to save that for a future patch as well.  (We need to stop shaving yaks
1386         at some point.)
1387
1388         * Scripts/webkitpy/tool/commands/rebaseline.py:
1389
1390 2011-07-11  Adam Barth  <abarth@webkit.org>
1391
1392         garden-o-matic should be able to roll out patches
1393         https://bugs.webkit.org/show_bug.cgi?id=64185
1394
1395         Reviewed by Eric Seidel.
1396
1397         This gardening server API is simply a binding to the rollout machinery
1398         already present in webkit-patch.
1399
1400         * Scripts/webkitpy/tool/commands/gardenomatic.py:
1401         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1402         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added.
1403
1404 2011-07-11  Csaba Osztrogonác  <ossy@webkit.org>
1405
1406         [Qt][Mac] Unreviewed fix after r90746.
1407
1408         * Scripts/run-webkit-tests:
1409         (useNewRunWebKitTests): Really disable NRWT for qt-mac platform.
1410
1411 2011-07-11  Adam Roben  <aroben@apple.com>
1412
1413         Teach TestFailures that ORWT's results.html file might be missing due to all tests passing
1414
1415         There are three reasons why we might fail to fetch ORWT's results.html:
1416           - All tests passed, so no results.html was generated
1417           - Some error during the test run caused results.html not to be generated (e.g., ORWT
1418             timed out)
1419           - Some network error occurred when fetching results.html
1420
1421         We were failing to account for the first possibility in some cases. For test runs before
1422         r89610, we first check build.webkit.org/json to determine how many tests failed and whether
1423         ORWT exited early due to too many failures; if all tests passed then we don't fetch
1424         results.html at all. r89610 changed ORWT to put information in results.html about exiting
1425         early due to too many failures, so we no longer needed to check build.webkit.org/json to get
1426         that information, and in r89619 I changed TestFailures to do just that. But I forgot that we
1427         still needed to check build.webkit.org/json to find out if all tests passed!
1428
1429         Now, for test runs after r89610, we check results.html first, and then check
1430         build.webkit.org/json if we fail to fetch results.html. This lets us distinguish between all
1431         tests passing and the error cases.
1432
1433         Fixes <http://webkit.org/b/64280> TestFailures page can't pinpoint that r90699 caused 13
1434         tests to fail on Windows 7 Release (WebKit2 Tests)
1435
1436         Reviewed by David Kilzer.
1437
1438         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:
1439         (LayoutTestResultsLoader.prototype.start): Bumped the cache version so that old, buggy
1440         cached data will get evicted. We were marking builds where all tests passed as errors!
1441         (LayoutTestResultsLoader.prototype._fetchAndParseORWTResults): Added success/error callback
1442         parameters to the fetchAndParseResultsHTML helper function, and added a similar
1443         fetchNumberOfFailingTests function that fetches data from build.webkit.org/json (code came
1444         from later in the function). For test runs before r89610, we first check
1445         build.webkit.org/json then check results.html. For builds after r89610, we first check
1446         results.html and then check build.webkit.org/json if we couldn't fetch results.html.
1447
1448 2011-07-11  Csaba Osztrogonác  <ossy@webkit.org>
1449
1450         new-run-webkit-tests does not support qt-arm or qt-4.8 results
1451         https://bugs.webkit.org/show_bug.cgi?id=64071
1452
1453         [Qt] NRWT should pick up the right httpd config file
1454         https://bugs.webkit.org/show_bug.cgi?id=64086
1455
1456         Reviewed by Zoltán Herczeg.
1457
1458         * Scripts/run-webkit-tests: Disable NRWT for qt-mac, qt-arm and qt-4.8 platforms until fix.
1459         (useNewRunWebKitTests):
1460
1461 2011-07-11  Kenichi Ishibashi  <bashi@chromium.org>
1462
1463         Add TestNetscapePlugIn/Tests/FormValue.cpp to DRT build files
1464         https://bugs.webkit.org/show_bug.cgi?id=64248
1465
1466         Reviewed by Kent Tamura.
1467
1468         * DumpRenderTree/DumpRenderTree.gypi: Added FormValue.cpp.
1469         * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj: Ditto.
1470         * DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro: Ditto.
1471         * GNUmakefile.am: Ditto.
1472
1473 2011-07-10  Kenichi Ishibashi  <bashi@chromium.org>
1474
1475         Let plugins participate in form submission
1476         https://bugs.webkit.org/show_bug.cgi?id=13061
1477
1478         Adds NPPVformValue support to the testing plugin.
1479
1480         Reviewed by Anders Carlsson.
1481
1482         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: Added FormValue.cpp
1483         * DumpRenderTree/TestNetScapePlugIn/Tests/FormValue.cpp: Added.
1484         (FormValue::FormValue): Ditto.
1485         (FormValue::NPP_GetValue): Ditto.
1486
1487 2011-07-11  Roland Steiner  <rolandsteiner@chromium.org>
1488
1489         Reviewed by Tony Chang.
1490
1491         Complete functions in filesystem.py
1492         https://bugs.webkit.org/show_bug.cgi?id=63528
1493
1494         - missing text file functions added
1495         - functions sorted
1496         - removed 'append' optional parameters (were unused)
1497         - adapted filesystem_mock in the same way
1498
1499         * Scripts/webkitpy/common/system/filesystem.py:
1500         * Scripts/webkitpy/common/system/filesystem_mock.py:
1501         * Scripts/webkitpy/common/system/filesystem_unittest.py:
1502
1503 2011-07-10  Adam Barth  <abarth@webkit.org>
1504
1505         Clean up style in fallback path calculation
1506         https://bugs.webkit.org/show_bug.cgi?id=64239
1507
1508         Reviewed by Daniel Bates.
1509
1510         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1511         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
1512         * Scripts/webkitpy/layout_tests/port/mac.py:
1513         * Scripts/webkitpy/layout_tests/port/win.py:
1514
1515 2011-07-10  Balazs Kelemen  <kbalazs@webkit.org>
1516
1517         [Qt][WK2] Implement the rest of WTR::PlatformWebView
1518         https://bugs.webkit.org/show_bug.cgi?id=63630
1519
1520         Reviewed by Kenneth Rohde Christiansen.
1521
1522         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
1523         (WTR::PlatformWebView::windowFrame):
1524         (WTR::PlatformWebView::setWindowFrame):
1525
1526 2011-07-10  Mark Rowe  <mrowe@apple.com>
1527
1528         Teach build-webkit how to find the latest WebKitSystemInterface binary.
1529
1530         * Scripts/build-webkit:
1531
1532 2011-07-09  Adam Roben  <aroben@apple.com>
1533
1534         Teach TestFailures to abbreviate the examples of test flakiness
1535
1536         These lists can get quite long, and it's not really helpful in most cases to have soooooo
1537         many examples of flakiness.
1538
1539         Fixes <http://webkit.org/b/64203> Lists of flaky revisions on TestFailures page can get so
1540         long they're hard to navigate
1541
1542         Reviewed by Dan Bates.
1543
1544         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyLayoutTestDetector.js:
1545         (FlakyLayoutTestDetector.prototype.flakinessExamples): If we have more than a certain number
1546         of examples, replace the middle items with a separator.
1547
1548         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
1549         (.flakiness-example-separator): Added styles for the separator.
1550
1551         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1552         (ViewController.prototype._domForPossiblyFlakyTests): Use a vertical ellipsis to represent
1553         the separator.
1554
1555 2011-07-09  Dirk Pranke  <dpranke@chromium.org>
1556
1557         nrwt: stack traces from worker-side exceptions aren't very useful inside test-webkitpy
1558         https://bugs.webkit.org/show_bug.cgi?id=64218
1559
1560         Reviewed by Eric Seidel.
1561
1562         Exceptions aren't picklable and can't be sent across the
1563         manager/worker message queue without losing information. NRWT
1564         handles this by turning the stack trace into a set of strings,
1565         and logging the strings when we receive an exception from the
1566         worker. However, when you are running tests and something
1567         crashes on the worker side, test-webkitpy prints the
1568         manager-side stack trace, which is just confusing and useless.
1569
1570         This patch changes the logic so that exceptions are passed
1571         through as-is when the worker and manager are in the same
1572         process (the --worker-model=inline option). This increases the
1573         code paths slightly but makes crashes much more useful.
1574
1575         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1576         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
1577         * Scripts/webkitpy/layout_tests/controllers/message_broker.py:
1578         * Scripts/webkitpy/layout_tests/controllers/worker.py:
1579         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1580
1581 2011-07-09  Darin Fisher  <darin@chromium.org>
1582
1583         Eliminate bad dependency on gfx::Point.
1584         https://bugs.webkit.org/show_bug.cgi?id=64228
1585
1586         Reviewed by Kent Tamura.
1587
1588         * DumpRenderTree/chromium/EventSender.cpp:
1589         (initMouseEvent):
1590
1591 2011-07-08  Ojan Vafai  <ojan@chromium.org>
1592
1593         pull static dashboard files into the appengine server from the chromium repository
1594         https://bugs.webkit.org/show_bug.cgi?id=64208
1595
1596         Reviewed by Dirk Pranke.
1597
1598         These files belong in the WebKit repo since they are tied primarily to the webkit tests.
1599         They have some extra bits to support chromium's gtests, but that seems fine.
1600         Mainly, this will allow other WebKit hackers to hack on the dashboards.
1601
1602         As a nice side-effect, we can now push the dashboard files when we do appengine pushes
1603         instead of the weird thing we used to do of pulling them from the Chromium repository
1604         and storing them in the appengine datastore. This allows for cleaning up a lot of code
1605         and will likely make the dashboards load a bit faster.
1606
1607         The new JS files don't fully match WebKit style, but I'd like to clean that up in a
1608         followup patch if possible to maintain my sanity with this patch.
1609
1610         * TestResultServer/app.yaml:
1611         * TestResultServer/handlers/dashboardhandler.py: Removed.
1612         * TestResultServer/handlers/menu.py:
1613         * TestResultServer/main.py:
1614         * TestResultServer/model/dashboardfile.py: Removed.
1615         * TestResultServer/static-dashboards/LICENSE.dygraph.txt: Added.
1616         * TestResultServer/static-dashboards/README.dygraph.txt: Added.
1617         * TestResultServer/static-dashboards/README.webtreemap.txt: Added.
1618         * TestResultServer/static-dashboards/aggregate_results.html: Added.
1619         * TestResultServer/static-dashboards/builders.js: Added.
1620         * TestResultServer/static-dashboards/dashboard_base.js: Added.
1621         * TestResultServer/static-dashboards/dygraph-combined.js: Added.
1622         * TestResultServer/static-dashboards/flakiness_dashboard.html: Added.
1623         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js: Added.
1624         * TestResultServer/static-dashboards/timeline_explorer.html: Added.
1625         * TestResultServer/static-dashboards/treemap.html: Added.
1626         * TestResultServer/static-dashboards/webtreemap.css: Added.
1627         * TestResultServer/static-dashboards/webtreemap.js: Added.
1628         * TestResultServer/stylesheets/dashboardfile.css: Removed.
1629         * TestResultServer/stylesheets/menu.css:
1630         * TestResultServer/templates/dashboardfilelist.html: Removed.
1631         * TestResultServer/templates/menu.html:
1632
1633 2011-07-08  Adam Roben  <aroben@apple.com>
1634
1635         Make TestFailures's list of flaky tests look more like the list of non-flaky tests
1636
1637         Fixes <http://webkit.org/b/64204> TestFailures page's flaky tests list is ugly!
1638
1639         Reviewed by Daniel Bates.
1640
1641         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
1642         (#failure-history, #possibly-flaky-tests): Expanded this rule to apply to the list of flaky
1643         tests.
1644         (#failure-history > li, #possibly-flaky-tests > li): Ditto, but moved the 50px left padding
1645         from here...
1646         (#failure-history > li): ...to here.
1647
1648         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1649         (ViewController.prototype._domForPossiblyFlakyTests): Give the list an id attribute for
1650         styling purposes.
1651
1652 2011-07-08  Jeffrey Pfau  <jpfau@apple.com>
1653
1654         Unreviewed, add myself as committer.
1655
1656         * Scripts/webkitpy/common/config/committers.py:
1657
1658 2011-07-08  Adam Barth  <abarth@webkit.org>
1659
1660         TestResultsServer should keep old test results
1661         https://bugs.webkit.org/show_bug.cgi?id=64199
1662
1663         Reviewed by Ojan Vafai.
1664
1665         Having historical data will help us do failure archeology.
1666
1667         * TestResultServer/handlers/testfilehandler.py:
1668         * TestResultServer/model/testfile.py:
1669         * TestResultServer/templates/showfilelist.jsonp: Added.
1670
1671 2011-07-08  Adam Roben  <aroben@apple.com>
1672
1673         Remove commit-log-editor's dependency on Module::Load::Conditional
1674
1675         This module isn't available in Perl 5.8.8 (the version used on Leopard).
1676
1677         Fixes <http://webkit.org/b/64198> REGRESSION (r90583):
1678         webkitpy.common.checkout.checkout_unittest failing on Leopard
1679
1680         Reviewed by Daniel Bates.
1681
1682         * Scripts/commit-log-editor: Use the new loadTermReadKey() function instead of
1683         Module::Load::Conditional::can_load.
1684         (loadTermReadKey): Added. Tries to load Term::ReadKey and returns true if it's successful.
1685
1686 2011-07-08  Adam Roben  <aroben@apple.com>
1687
1688         Make checkout_unittest more robust against files moving around
1689
1690         Fixes <http://webkit.org/b/64197> checkout_unittest contains ugly, fragile code to find the
1691         Scripts directory
1692
1693         Reviewed by Adam Barth.
1694
1695         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
1696         (CommitMessageForThisCommitTest.test_commit_message_for_this_commit): Instantiate a real SCM
1697         object and use it to get the path to the Scripts directory, rather than hard-coding the
1698         relative path from this file to Scripts.
1699
1700 2011-07-08  Adam Roben  <aroben@apple.com>
1701
1702         Make TestFailures remember that run-webkit-tests timed out, even across reloads
1703
1704         Fixes <http://webkit.org/b/64193> TestFailures page incorrectly thinks all tests passed in
1705         http://build.webkit.org/builders/Windows%207%20Release%20(Tests)/builds/14589 after a reload
1706
1707         Reviewed by Daniel Bates.
1708
1709         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:
1710         (LayoutTestResultsLoader.prototype.start): Store an "error" attribute in the cached data.
1711         When true, it indicates that there was an error retrieving the results for this build and
1712         that the errorCallback should be called.
1713
1714 2011-07-08  Adam Barth  <abarth@webkit.org>
1715
1716         Teach garden-o-matic how to display test results
1717         https://bugs.webkit.org/show_bug.cgi?id=64141
1718
1719         Reviewed by Ojan Vafai.
1720
1721         This patch includes basic infrastructure for probing build.chromium.org
1722         for test results.  We only handle text and image tests, not anything
1723         complicated like reftests.  Also, we're using the revision/build
1724         independent results store on the server, so we're avoiding that
1725         complication for now.
1726
1727         It's slightly hacky that we need to probe the server to see what kinds
1728         of results exist.  A better solution would be to add CORS support to
1729         the server or to use the local server to help.
1730
1731         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
1732         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
1733         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
1734         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
1735         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
1736         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
1737
1738 2011-07-08  Dirk Pranke  <dpranke@chromium.org>
1739
1740         REGRESSION(90419) NRWT's httpd locking is broken for --child-processes=1
1741         https://bugs.webkit.org/show_bug.cgi?id=64092
1742
1743         Reviewed by Tony Chang.
1744
1745         The code was incorrectly creating empty shards in the
1746         shard_in_two case.
1747
1748         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1749         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
1750
1751 2011-07-08  Adam Roben  <aroben@apple.com>
1752
1753         Teach buildbot to figure out how many webkitpy/webkitperl tests failed
1754
1755         Fixes <http://webkit.org/b/64192> It's hard to tell how many test-webkitpy/test-webkitperl
1756         tests failed when looking at build.webkit.org
1757
1758         Reviewed by Eric Seidel.
1759
1760         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1761         (TestWithFailureCount): New class that represents a test build step which has an associated
1762         failure count. Eventually we should move more of our test classes to deriving from this.
1763         (TestWithFailureCount.countFailures): Method for subclasses to override to say how many
1764         failures occurred.
1765
1766         (TestWithFailureCount.commandComplete):
1767         (TestWithFailureCount.evaluateCommand):
1768         (TestWithFailureCount.getText):
1769         (TestWithFailureCount.getText2):
1770         These were all based on RunGtkAPITests.
1771
1772         (RunPythonTests): Changed to inherit from TestWithFailureCount.
1773         (RunPythonTests.countFailures): Parses the test-webkitpy output looking for the count of
1774         failures.
1775         (RunPerlTests): Changed to inherit from TestWithFailureCount.
1776         (RunPerlTests.countFailures): Parses the test-webkitperl output looking for the count of
1777         failures.
1778
1779 2011-07-08  Adam Roben  <aroben@apple.com>
1780
1781         Ensure $CHANGE_LOG_EMAIL_ADDRESS is set when testing webkitpy's commit-log-editor integration
1782
1783         Fixes <http://webkit.org/b/64180> REGRESSION (r90564): test-webkitpy failing on multiple
1784         bots due to commit-log-editor errors
1785
1786         Reviewed by Adam Barth.
1787
1788         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
1789         (CommitMessageForThisCommitTest.test_commit_message_for_this_commit): Set
1790         $CHANGE_LOG_EMAIL_ADDRESS to the patch author's email address. This ensures that
1791         commit-log-editor can find a value for the committer's email, and that the committer and
1792         author email addresses match, which will prevent commit-log-editor from inserting a "Patch
1793         by" line in the commit message.
1794
1795 2011-07-08  Andreas Kling  <kling@webkit.org>
1796
1797         [Qt][WK2] Views should expose QActions for basic navigation.
1798         https://bugs.webkit.org/show_bug.cgi?id=64174
1799
1800         Reviewed by Benjamin Poulain.
1801
1802         Bring the toolbar in Qt's MiniBrowser back to life.
1803
1804         * MiniBrowser/qt/BrowserView.cpp:
1805         (BrowserView::navigationAction):
1806         * MiniBrowser/qt/BrowserView.h:
1807         * MiniBrowser/qt/BrowserWindow.cpp:
1808         (BrowserWindow::BrowserWindow):
1809
1810 2011-07-08  Adam Roben  <aroben@apple.com>
1811
1812         Teach TestFailures how to find test names in commit-log-editor-style commit messages
1813
1814         TestFailures was relying on Trac turning the list of modified files in our commit messages
1815         into an HTML list. But Trac only does this when the list of modified files is indented.
1816         commit-log-editor doesn't indent the file list, so the list wasn't being turned into an HTML
1817         list, which was confusing TestFailures.
1818
1819         TestFailures now does much simpler parsing of the commit message (i.e., just a substring
1820         search) without relying at all on its structure.
1821
1822         Fixes <http://webkit.org/b/64173> TestFailures page fails to blame r90608 for breaking
1823         fast/dom/HTMLProgressElement/progress-element-markup.html on Windows
1824
1825         Reviewed by David Kilzer.
1826
1827         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1828         (Trac.prototype.getCommitDataForRevisionRange): Instead of trying to parse the commit
1829         message, just return its text.
1830
1831         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1832         (ViewController.prototype._domForRegressionRange): Instead of searching for test names in
1833         each commit's list of modified files, just search for test names anywhere in the commit's
1834         message.
1835
1836 2011-07-08  Adam Barth  <abarth@webkit.org>
1837
1838         sheriffbot is too spammy in IRC
1839         https://bugs.webkit.org/show_bug.cgi?id=64153
1840
1841         Reviewed by Eric Seidel.
1842
1843         Reporting failures in IRC worked well when the tree was greener than it
1844         is today.  Nowadays, this feature mostly just results in spam about
1845         false positives.  If we reach a greener state, we can bring this code
1846         back.
1847
1848         (Another possibility is to restrict this feature to builder bots,
1849         rather than including testers, as we did before this patch.)
1850
1851         * Scripts/webkitpy/tool/commands/sheriffbot.py:
1852         * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
1853
1854 2011-07-08  Adam Roben  <aroben@apple.com>
1855
1856         Ensure commit-log-editor adds a "Patch by" line when the author and committer are different
1857
1858         Previously we were only adding a "Patch by" line when the ChangeLog contained a "Reviewed
1859         by" line. But some patches (like rollout patches) don't contain that line. Now we always add
1860         "Patch by" to the commit log regardless of the ChangeLog's contents.
1861
1862         Fixes <http://webkit.org/b/64127> Committer for r90588 is commit-queue@webkit.org, but
1863         should have been abarth@webkit.org
1864
1865         Reviewed by Anders Carlsson.
1866
1867         * Scripts/commit-log-editor:
1868         (createCommitMessage): Try to put the "Patch by" line just above the "Reviewed by" line, as
1869         before. If there is no "Reviewed by" line, try to put it just above the first modified file.
1870         If all else fails, put it at the end of the commit message.
1871         (patchAuthorshipString): Added. Code came from createCommitMessage.
1872
1873 2011-07-08  Herczeg Zoltan  <zherczeg@webkit.org>
1874
1875         Reviewed by Andreas Kling.
1876
1877         Adding myself as a reviewer.
1878
1879         * Scripts/webkitpy/common/config/committers.py:
1880
1881 2011-07-07  Kevin Ollivier  <kevino@theolliviers.com>
1882
1883         [wx] Unreviewed build fix, remove old files from the tree and update paths to new ones.
1884
1885         * wx/build: Removed.
1886         * wx/build/build_utils.py: Removed.
1887         * wx/build/settings.py: Removed.
1888         * wx/build/waf_extensions.py: Removed.
1889         * wx/build/wxpresets.py: Removed.
1890         * wx/packaging/build-mac-installer.py:
1891
1892 2011-07-07  Kevin Ollivier  <kevino@theolliviers.com>
1893
1894         [wx] Unreviewed build fix, more fixes for install name issues.
1895
1896         * wx/packaging/build-mac-installer.py:
1897
1898 2011-07-07  Dirk Pranke  <dpranke@chromium.org>
1899
1900         test-webkitpy fails on chromium win
1901         https://bugs.webkit.org/show_bug.cgi?id=64137
1902
1903         Reviewed by Eric Seidel.
1904
1905         test-webkitpy doesn't play nicely with the multiprocessing
1906         module on win32. An earlier change actually reenabled the tests
1907         on win32 by mistake.
1908
1909         This patch also fixes a few cases where path names will trip
1910         things up on win32 (e.g., by testing unix paths on windows).
1911         We do not lose any real coverage here with those fixes.
1912
1913         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
1914         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
1915         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
1916         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1917
1918 2011-07-07  Leandro Pereira  <leandro@profusion.mobi>
1919
1920         [ImageDiff] Calculate/print difference right after reading baseline image.
1921         https://bugs.webkit.org/show_bug.cgi?id=64117
1922
1923         Reviewed by Kent Tamura.
1924
1925         Otherwise, ImageDiff will block on fgets() until the universe ends or you ^C
1926         it.  Whichever happens first.
1927
1928         * DumpRenderTree/efl/ImageDiff.cpp:
1929         (main):
1930
1931 2011-07-07  Raphael Kubo da Costa  <kubo@profusion.mobi>
1932
1933         [EFL] DRT: Leak cairo_t to make the pixel tests stop crashing
1934         https://bugs.webkit.org/show_bug.cgi?id=64107
1935
1936         Reviewed by Kent Tamura.
1937
1938         The pointer is later managed by BitmapContext, but it was being
1939         de-refed and deleted earlier by our RefPtr, causing crashes in all
1940         pixel tests.
1941
1942         We now call leakRef() to make sure it is not removed when our
1943         createBitmapContextFromWebView goes out of context.
1944
1945         * DumpRenderTree/efl/PixelDumpSupportEfl.cpp:
1946         (createBitmapContextFromWebView):
1947
1948 2011-07-07  Adam Roben  <aroben@apple.com>
1949
1950         Update TestFailures's title and header immediately upon navigation
1951
1952         Fixes <http://webkit.org/b/64125> TestFailures page seems slow to react on link clicks
1953
1954         Reviewed by Daniel Bates.
1955
1956         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1957         (ViewController.prototype.loaded): Create and store an h1 element for displaying the page's
1958         title.
1959         (ViewController.prototype._displayBuilder): Set the page's title and clear out any old
1960         content immediately instead of waiting for the first set of results for the history
1961         analyzer. As we receive new results from the analyzer we'll just clear out the main content
1962         area and replace it with the new info.
1963         (ViewController.prototype._displayTesters): Set the page's title and clear out any old
1964         content immediately instead of waiting for the list of testers from the buildbot. Sprinkled
1965         in a little use of our removeAllChildren helper, too.
1966         (ViewController.prototype._setTitle): Added. Set's the page's title and the header text.
1967
1968 2011-07-07  Adam Roben  <aroben@apple.com>
1969
1970         Make Checkout use SCM's Executive instead of conjuring up its own
1971
1972         This will improve integration with the rest of webkitpy, particularly when invoked via
1973         webkit-patch.
1974
1975         Fixes <http://webkit.org/b/64115> REGRESSION (r90564): webkitpy's Checkout class uses
1976         Executive inappropriately
1977
1978         Reviewed by Adam Barth.
1979
1980         * Scripts/webkitpy/common/checkout/checkout.py:
1981         (Checkout.commit_message_for_this_commit): Use SCM.run instead of creating an Executive for
1982         our own use. SCM might have some super-special Executive that it uses under the covers, and
1983         we want to use it, too!
1984
1985         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
1986         (CommitMessageForThisCommitTest.test_commit_message_for_this_commit): Mock the SCM.run
1987         method to call through to Executive.run_command.
1988
1989 2011-07-07  Adam Roben  <aroben@apple.com>
1990
1991         Make Term/ReadKey.pm an optional dependency of commit-log-editor
1992
1993         When Term/ReadKey.pm isn't present, the --regenerate-log switch will be non-functional.
1994
1995         Fixes <http://webkit.org/b/64113> REGRESSION (r90564):
1996         webkitpy.common.checkout.checkout_unittest failing on GTK bots due to missing
1997         Term/ReadKey.pm module
1998
1999         Reviewed by Xan Lopez.
2000
2001         * Scripts/commit-log-editor: Use Module::Load::Conditional::can_load to only load
2002         Term/ReadKey.pm if it's present. If it isn't present, ignore the --regenerate-log switch so
2003         that existing commit messages will be preserved. (Users can manually remove existing commit
2004         messages using git-reset.)
2005
2006 2011-07-07  Adam Roben  <aroben@apple.com>
2007
2008         Keep testing that commit messages containing Unicode are handled properly by webkitpy
2009
2010         Fixes <http://webkit.org/b/64109> REGRESSION (r90571): test-webkitpy no longer tests that we
2011         correctly handle Unicode in commit messages
2012
2013         Reviewed by Eric Seidel.
2014
2015         * Scripts/webkitpy/common/checkout/checkout_unittest.py: Put some Unicode characters in
2016         places in the ChangeLog that will end up being included in the commit message.
2017
2018 2011-07-07  Andreas Kling  <kling@webkit.org>
2019
2020         [Qt][WK2] Remove Symbian code.
2021         https://bugs.webkit.org/show_bug.cgi?id=64101
2022
2023         Reviewed by Benjamin Poulain.
2024
2025         * MiniBrowser/qt/BrowserWindow.cpp:
2026         (BrowserWindow::BrowserWindow):
2027         (BrowserWindow::screenshot):
2028         * MiniBrowser/qt/MiniBrowser.pro:
2029         * MiniBrowser/qt/MiniBrowserApplication.h:
2030         (WindowOptions::WindowOptions):
2031
2032 2011-07-07  Andreas Kling  <kling@webkit.org>
2033
2034         [Qt][WK2] Don't support ridiculous matrix of QT_NO_FEATURE combinations.
2035         https://bugs.webkit.org/show_bug.cgi?id=64099
2036
2037         Reviewed by Benjamin Poulain.
2038
2039         * MiniBrowser/qt/BrowserWindow.cpp:
2040         (BrowserWindow::openFile):
2041         (BrowserWindow::screenshot):
2042         (BrowserWindow::loadURLListFromFile):
2043
2044 2011-07-07  Adam Roben  <aroben@apple.com>
2045
2046         Completely ignore too-many-failures builds in TestFailures in most circumstances
2047
2048         Because a semi-arbitrary subset of tests are run in a too-many-failures build, we can't
2049         really use them to perform regression analysis. The only time we want to pay attention to
2050         too-many-failures builds is when we're trying to explain when the current bout of
2051         too-many-failures started.
2052
2053         Fixes <http://webkit.org/b/64106> TestFailures page sometimes claims a test started failing
2054         in a build that didn't even run it (because it exited early due to too many failues)
2055
2056         Reviewed by David Kilzer.
2057
2058         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestHistoryAnalyzer.js:
2059         (LayoutTestHistoryAnalyzer.prototype._incorporateBuildHistory): Removed old, broken
2060         too-many-failures handling that would cause us to blame builds that didn't even run a given
2061         test for breaking it. Instead, skip over all too-many-failures builds unless the most recent
2062         build was itself a too-many-failures build.
2063
2064 2011-07-07  Adam Roben  <aroben@apple.com>
2065
2066         Teach webkitpy's Checkout class to use commit-log-editor to create commit messages
2067
2068         Fixes <http://webkit.org/b/26755> webkit-patch's commit messages are less readable than
2069         commit-log-editor's
2070
2071         Reviewed by David Kilzer.
2072
2073         * Scripts/webkitpy/common/checkout/checkout.py:
2074         (Checkout.commit_message_for_this_commit): Run commit-log-editor, passing it the paths of
2075         the modified ChangeLogs, to generate the commit message, rather than trying to generate one
2076         ourselves.
2077
2078         * Scripts/webkitpy/common/checkout/checkout_unittest.py: Updated the expected commit message
2079         to match commit-log-editor's format.
2080         (CommitMessageForThisCommitTest.setUp): Write the ChangeLogs into Tools and LayoutTests
2081         directories so we can see how the various entries get labeled in the commit message.
2082         (CommitMessageForThisCommitTest.test_commit_message_for_this_commit): Create a mock SCM
2083         instance that knows how to find commit-log-editor and pass it to our Checkout instance.
2084         Don't bother capturing output, since there shouldn't be any.
2085
2086 2011-07-07  Adam Roben  <aroben@apple.com>
2087
2088         Add a --print-log option to commit-log-editor
2089
2090         When specified, commit-log-editor takes a set of ChangeLog paths on the command line,
2091         generates a commit message from those ChangeLogs, prints it to stdout, and exits.
2092
2093         Prep work for fixing <http://webkit.org/b/26755> webkit-patch's commit messages are less
2094         readable than commit-log-editor's
2095
2096         Reviewed by David Kilzer.
2097
2098         * Scripts/commit-log-editor: Changed to use Getopt::Long to parse options. Added --print-log
2099         option, which calls through to createCommitMessage, prints the result, and exits.
2100         (printUsageAndExit): Renamed from "usage" for clarity. Beefed up the usage statement to
2101         explain commit-log-editor's 2.5 (normal, --print-log, and --help) modes of operation.
2102
2103 2011-07-07  Adam Roben  <aroben@apple.com>
2104
2105         Extract commit-log-editor's commit-message-generation code into a separate function
2106
2107         As a bonus, we now skip a bunch of work in the case where we are generating a commit message
2108         for staged git changes.
2109
2110         Prep work for fixing <http://webkit.org/b/26755> webkit-patch's commit messages are less
2111         readable than commit-log-editor's
2112
2113         Reviewed by David Kilzer.
2114
2115         * Scripts/commit-log-editor: Moved the default definition of $endl up toward the top of the
2116         file so that it will be defined even if we aren't operating on an existing log message.
2117         Moved a bunch of top-level code to generate the commit message from here...
2118         (createCommitMessage): ...to here.
2119
2120 2011-07-07  Carlos Garcia Campos  <cgarcia@igalia.com>
2121
2122         [GTK] Open links in a new window when clicking with the middle button in MiniBrowser
2123         https://bugs.webkit.org/show_bug.cgi?id=63988
2124
2125         Reviewed by Martin Robinson.
2126
2127         * MiniBrowser/gtk/BrowserWindow.c:
2128         (browserWindowConstructed):
2129         (decidePolicyForNavigationAction): Ignore the action if a link is
2130         clicked with the middle buttonm and open the link in a new window.
2131         (browserWindowPolicyClientInit):
2132
2133 2011-07-07  Adam Barth  <abarth@webkit.org>
2134
2135         Remove style_references.py
2136         https://bugs.webkit.org/show_bug.cgi?id=64070
2137
2138         Reviewed by Eric Seidel.
2139
2140         We never really adopted the _references design, and this file is a bit
2141         of an orphan at this point.
2142
2143         * Scripts/check-webkit-style:
2144         * Scripts/webkitpy/style/checker.py:
2145         * Scripts/webkitpy/style/checker_unittest.py:
2146         * Scripts/webkitpy/style/checkers/python.py:
2147         * Scripts/webkitpy/style/checkers/test_expectations.py:
2148         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
2149         * Scripts/webkitpy/style/main_unittest.py:
2150         * Scripts/webkitpy/style_references.py: Removed.
2151
2152 2011-07-07  Eric Seidel  <eric@webkit.org>
2153
2154         Fix WebKit2 expected results search paths for Mac and Qt under new-run-webkit-tests
2155         https://bugs.webkit.org/show_bug.cgi?id=64056
2156
2157         Unreviewed.
2158
2159         I had written this unittest prior to landing my change, but
2160         had named it qt_unitest.py, so it wasn't running (nor was
2161         it added to my git repo).
2162
2163         Once I fixe the name of the unittest file, it was very easy
2164         to fix the error in webkit.py (which already had a FIXME).
2165
2166         * Scripts/webkitpy/layout_tests/port/qt_unittest.py: Added.
2167         * Scripts/webkitpy/layout_tests/port/webkit.py:
2168
2169 2011-07-07  Eric Seidel  <eric@webkit.org>
2170
2171         Unreviewed.  Add more debug logging about which test expectations we're using.
2172
2173         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2174         * Scripts/webkitpy/layout_tests/port/webkit.py:
2175
2176 2011-07-06  Adam Barth  <abarth@webkit.org>
2177
2178         Remove unexpected_results.json
2179         https://bugs.webkit.org/show_bug.cgi?id=64065
2180
2181         Reviewed by Eric Seidel.
2182
2183         No one should be using this file any more.  This patch removes it from
2184         the face of the earth.  This patch also removes two NRWT-specific
2185         command line options that were implemented using
2186         unexpected_results.json.
2187
2188         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
2189         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2190         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2191         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2192         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2193         * Scripts/webkitpy/tool/commands/rebaselineserver.py:
2194         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
2195
2196 2011-07-07  Adam Barth  <abarth@webkit.org>
2197
2198         REGRESSION(90520): chromium GPU tests produce output in the LayoutTests directory
2199         https://bugs.webkit.org/show_bug.cgi?id=64067
2200
2201         Reviewed by Eric Seidel.
2202
2203         The author of r90520 forgot to change the chromium_gpu.py implemenation
2204         of this method as well.
2205
2206         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
2207
2208 2011-07-06  Eric Seidel  <eric@webkit.org>
2209
2210         Fix WebKit2 expected results search paths for Mac and Qt under new-run-webkit-tests
2211         https://bugs.webkit.org/show_bug.cgi?id=64056
2212
2213         Reviewed by Adam Barth.
2214
2215         This fixes the MacPort's version of baseline_search_path to include
2216         wk2 fallback like WebKitPort's does.  This also re-works the QtPort
2217         to set self._name (as other ports do) so that WebKitPort's default
2218         implementations will understand Qt operating system flavors correctly
2219         (including when running wk2 tests).
2220
2221         In trying to test this, I realized that nearly all of our port-tests
2222         use the default FileSystem(), User() and Executive() objects, which
2223         is really bad, because they expose details of your system in the unittest results!
2224
2225         Once I fixed many of the Port() calls to pass MockFileSystem(), then
2226         MacPort.test_tests_for_other_platforms started failing, and I had to
2227         teach MockFileSystem.glob how to handle directories to make it work again.
2228
2229         I removed the useless PortTestCase.test_baseline_search_path and replaced
2230         it with more useful test_baseline_search_path tests in MacPortTest and QtPortTest.
2231
2232         * Scripts/webkitpy/common/config/ports.py:
2233         * Scripts/webkitpy/common/system/filesystem_mock.py:
2234         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2235         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2236         * Scripts/webkitpy/layout_tests/port/mac.py:
2237         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2238         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2239         * Scripts/webkitpy/layout_tests/port/qt.py:
2240         * Scripts/webkitpy/layout_tests/port/test_files.py:
2241         * Scripts/webkitpy/layout_tests/port/win.py:
2242         * Scripts/webkitpy/to_be_moved/deduplicate_tests.py:
2243         * Scripts/webkitpy/tool/mocktool.py:
2244
2245 2011-07-06  Adam Barth  <abarth@webkit.org>
2246
2247         Wean rebaseline-server off unexpected_results.json
2248         https://bugs.webkit.org/show_bug.cgi?id=64063
2249
2250         Reviewed by Eric Seidel.
2251
2252         In the process of changing this code to use full_results.json, I
2253         noticed that the code was broken (because it wasn't tested).  This
2254         patch also adds test coverage for the broken code.
2255
2256         * Scripts/webkitpy/tool/commands/rebaselineserver.py:
2257         * Scripts/webkitpy/tool/servers/rebaselineserver.py:
2258         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
2259
2260 2011-07-06  Adam Barth  <abarth@webkit.org>
2261
2262         Wean resultsjsonparser off unexpected_results.json
2263         https://bugs.webkit.org/show_bug.cgi?id=64061
2264
2265         Reviewed by Eric Seidel.
2266
2267         We're about to delete unexpected_results.json, so we need to remove all
2268         the users.
2269
2270         * Scripts/webkitpy/common/config/ports.py:
2271         * Scripts/webkitpy/common/net/resultsjsonparser.py:
2272         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
2273         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2274
2275 2011-07-06  Adam Barth  <abarth@webkit.org>
2276
2277         Cleanup result_summary.py
2278         https://bugs.webkit.org/show_bug.cgi?id=64057
2279
2280         Reviewed by Eric Seidel.
2281
2282         This class had a bunch of out-of-date docstrings that no longer make
2283         any sense.
2284
2285         * Scripts/webkitpy/layout_tests/models/result_summary.py:
2286
2287 2011-07-06  Adam Barth  <abarth@webkit.org>
2288
2289         Tweak the ChangeLog for DEPS rolls to say "Unreviewed" so that the
2290         patches can be landed by the commit-queue.
2291
2292         * Scripts/webkitpy/tool/steps/preparechangelogfordepsroll.py:
2293
2294 2011-07-06  Adam Barth  <abarth@webkit.org>
2295
2296         Normalize callers of logging.getLogger to use __name__
2297         https://bugs.webkit.org/show_bug.cgi?id=64054
2298
2299         Reviewed by Eric Seidel.
2300
2301         Most callers were already using __name__.  This patch makes everyone
2302         consistent.
2303
2304         * Scripts/webkitpy/common/checkout/diff_parser.py:
2305         * Scripts/webkitpy/common/checkout/scm/svn.py:
2306         * Scripts/webkitpy/common/net/statusserver.py:
2307         * Scripts/webkitpy/common/system/executive.py:
2308         * Scripts/webkitpy/common/system/file_lock.py:
2309         * Scripts/webkitpy/common/system/user.py:
2310         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2311         * Scripts/webkitpy/layout_tests/models/result_summary.py:
2312         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2313         * Scripts/webkitpy/layout_tests/port/chromium.py:
2314         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2315         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
2316         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2317         * Scripts/webkitpy/layout_tests/port/gtk.py:
2318         * Scripts/webkitpy/layout_tests/port/http_lock.py:
2319         * Scripts/webkitpy/layout_tests/port/mac.py:
2320         * Scripts/webkitpy/layout_tests/port/qt.py:
2321         * Scripts/webkitpy/layout_tests/port/server_process.py:
2322         * Scripts/webkitpy/layout_tests/port/webkit.py:
2323         * Scripts/webkitpy/layout_tests/port/win.py:
2324         * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
2325         * Scripts/webkitpy/layout_tests/servers/http_server.py:
2326         * Scripts/webkitpy/layout_tests/servers/http_server_base.py:
2327         * Scripts/webkitpy/layout_tests/servers/websocket_server.py:
2328         * Scripts/webkitpy/layout_tests/views/printing.py:
2329         * Scripts/webkitpy/python24/versioning.py:
2330         * Scripts/webkitpy/style/checker.py:
2331         * Scripts/webkitpy/style/checkers/test_expectations.py:
2332         * Scripts/webkitpy/style/patchreader.py:
2333         * Scripts/webkitpy/to_be_moved/update_webgl_conformance_tests.py:
2334
2335 2011-07-06  Adam Barth  <abarth@webkit.org>
2336
2337         Move model classes out of layout_package and into models
2338         https://bugs.webkit.org/show_bug.cgi?id=64053
2339
2340         Reviewed by Eric Seidel.
2341
2342         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2343         * Scripts/webkitpy/common/net/layouttestresults.py:
2344         * Scripts/webkitpy/common/net/layouttestresults_unittest.py:
2345         * Scripts/webkitpy/common/net/resultsjsonparser.py:
2346         * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
2347         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2348         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2349         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py:
2350         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2351         * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
2352         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2353         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
2354         * Scripts/webkitpy/layout_tests/layout_package/result_summary.py: Removed.
2355         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py: Removed.
2356         * Scripts/webkitpy/layout_tests/layout_package/test_expectations_unittest.py: Removed.
2357         * Scripts/webkitpy/layout_tests/layout_package/test_failures.py: Removed.
2358         * Scripts/webkitpy/layout_tests/layout_package/test_failures_unittest.py: Removed.
2359         * Scripts/webkitpy/layout_tests/layout_package/test_input.py: Removed.
2360         * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
2361         * Scripts/webkitpy/layout_tests/layout_package/test_results.py: Removed.
2362         * Scripts/webkitpy/layout_tests/layout_package/test_results_unittest.py: Removed.
2363         * Scripts/webkitpy/layout_tests/models: Added.
2364         * Scripts/webkitpy/layout_tests/models/__init__.py: Added.
2365         * Scripts/webkitpy/layout_tests/models/result_summary.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/result_summary.py.
2366         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_expectations.py.
2367         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_expectations_unittest.py.
2368         * Scripts/webkitpy/layout_tests/models/test_failures.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_failures.py.
2369         * Scripts/webkitpy/layout_tests/models/test_failures_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_failures_unittest.py.
2370         * Scripts/webkitpy/layout_tests/models/test_input.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_input.py.
2371         * Scripts/webkitpy/layout_tests/models/test_results.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_results.py.
2372         * Scripts/webkitpy/layout_tests/models/test_results_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/test_results_unittest.py.
2373         * Scripts/webkitpy/layout_tests/port/chromium.py:
2374         * Scripts/webkitpy/layout_tests/views/printing.py:
2375         * Scripts/webkitpy/layout_tests/views/printing_unittest.py:
2376         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
2377         * Scripts/webkitpy/style_references.py:
2378         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
2379         * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py:
2380         * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py:
2381         * Scripts/webkitpy/tool/commands/queues_unittest.py:
2382         * Scripts/webkitpy/tool/commands/rebaseline.py:
2383
2384 2011-07-06  Dirk Pranke  <dpranke@chromium.org>
2385
2386         nrwt: fix typo preventing http startup on windows
2387         https://bugs.webkit.org/show_bug.cgi?id=64050
2388
2389         Reviewed by Eric Siedel.
2390
2391         NRWT checks whether it needs to start the servers by looking
2392         for "/http/" in the test names. After r90520, the leading slash
2393         is not present, so the check isn't quite right.
2394
2395         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2396         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2397
2398 2011-07-06  Dirk Pranke  <dpranke@chromium.org>
2399
2400         nrwt: remove --use-apache from the command line
2401         https://bugs.webkit.org/show_bug.cgi?id=63358
2402
2403         Reviewed by Adam Barth.
2404
2405         Re-land the fix in bug 63358 - the prior fix was flawed in that
2406         we were trying to start the http server during check_sys_deps()
2407         but not actually setting up the environment properly. This
2408         broken prior to this change, but exposed by it, since the patch
2409         changes the chromium port to call the base class's checks.
2410
2411         * Scripts/webkitpy/common/system/executive.py:
2412         * Scripts/webkitpy/layout_tests/port/base.py:
2413         * Scripts/webkitpy/layout_tests/port/chromium.py:
2414         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2415         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2416         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2417         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2418         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2419         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2420
2421 2011-07-06  Adam Barth  <abarth@webkit.org>
2422
2423         Move view-related code out of layout_package and into views
2424         https://bugs.webkit.org/show_bug.cgi?id=64051
2425
2426         Reviewed by Eric Seidel.
2427
2428         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2429         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2430         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py:
2431         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py:
2432         * Scripts/webkitpy/layout_tests/layout_package/metered_stream.py: Removed.
2433         * Scripts/webkitpy/layout_tests/layout_package/metered_stream_unittest.py: Removed.
2434         * Scripts/webkitpy/layout_tests/layout_package/printing.py: Removed.
2435         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py: Removed.
2436         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2437         * Scripts/webkitpy/layout_tests/views: Added.
2438         * Scripts/webkitpy/layout_tests/views/__init__.py: Added.
2439         * Scripts/webkitpy/layout_tests/views/metered_stream.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/metered_stream.py.
2440         * Scripts/webkitpy/layout_tests/views/metered_stream_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/metered_stream_unittest.py.
2441         * Scripts/webkitpy/layout_tests/views/printing.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/printing.py.
2442         * Scripts/webkitpy/layout_tests/views/printing_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py.
2443
2444 2011-07-06  Adam Barth  <abarth@webkit.org>
2445
2446         Add goofy comments to explain why we have these files.
2447
2448         * Scripts/webkitpy/layout_tests/controllers/__init__.py:
2449         * Scripts/webkitpy/layout_tests/layout_package/__init__.py:
2450
2451 2011-07-06  Adam Barth  <abarth@webkit.org>
2452
2453         Move a bunch of classes out of layout_tests.layout_package and into layout_tests.controllers
2454         https://bugs.webkit.org/show_bug.cgi?id=64048
2455
2456         Reviewed by Eric Seidel.
2457
2458         * Scripts/webkitpy/layout_tests/controllers: Added.
2459         * Scripts/webkitpy/layout_tests/controllers/__init__.py: Added.
2460         * Scripts/webkitpy/layout_tests/controllers/manager.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/manager.py.
2461         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/manager_unittest.py.
2462         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker.py.
2463         * Scripts/webkitpy/layout_tests/controllers/manager_worker_broker_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker_unittest.py.
2464         * Scripts/webkitpy/layout_tests/controllers/message_broker.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/message_broker.py.
2465         * Scripts/webkitpy/layout_tests/controllers/message_broker_unittest.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/message_broker_unittest.py.
2466         * Scripts/webkitpy/layout_tests/controllers/single_test_runner.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py.
2467         * Scripts/webkitpy/layout_tests/controllers/worker.py: Copied from Tools/Scripts/webkitpy/layout_tests/layout_package/worker.py.
2468         * Scripts/webkitpy/layout_tests/layout_package/manager.py: Removed.
2469         * Scripts/webkitpy/layout_tests/layout_package/manager_unittest.py: Removed.
2470         * Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker.py: Removed.
2471         * Scripts/webkitpy/layout_tests/layout_package/manager_worker_broker_unittest.py: Removed.
2472         * Scripts/webkitpy/layout_tests/layout_package/message_broker.py: Removed.
2473         * Scripts/webkitpy/layout_tests/layout_package/message_broker_unittest.py: Removed.
2474         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
2475         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py: Removed.
2476         * Scripts/webkitpy/layout_tests/layout_package/worker.py: Removed.
2477         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2478
2479 2011-07-06  Adam Barth  <abarth@webkit.org>
2480
2481         Attempt to fix unittest on Leopard.  Let's enumerate the keys in sorted
2482         order so the test results are predictable.
2483
2484         * Scripts/webkitpy/tool/bot/irc_command.py:
2485         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2486
2487 2011-07-06  Dirk Pranke  <dpranke@chromium.org>
2488
2489         Reviewed by Eric Siedel.
2490
2491         nrwt: make the code be consistent about using test names instead of filenames or paths
2492
2493         https://bugs.webkit.org/show_bug.cgi?id=63597
2494
2495         It can be hard to figure out where in the code we're assuming
2496         test names are listed as unix-style relative filenames and where
2497         they are either absolute paths or relative paths following the
2498         host filesystem convention.
2499
2500         This patch changes things so that everything outside of the
2501         Port object uses (and must assume) unix-style relative
2502         filenames (with one exception, which is specifying host-local
2503         filenames as a list of test arguments on the command line).
2504
2505         This should make things clearer and more consistent, and also
2506         removes a lot of path conversion calls.
2507
2508         The changes in this patch outside of port/base.py are largely
2509         cosmetic changes of "filename" to "test". We add a few routines
2510         to the base Port implementation to manipulate test names to
2511         split them, figure out if we're referring to a directory of
2512         tests, etc.
2513
2514         * Scripts/webkitpy/common/net/layouttestresults.py:
2515         * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
2516         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
2517         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
2518         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
2519         * Scripts/webkitpy/layout_tests/layout_package/manager_unittest.py:
2520         * Scripts/webkitpy/layout_tests/layout_package/printing.py:
2521         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
2522         * Scripts/webkitpy/layout_tests/layout_package/result_summary.py:
2523         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
2524         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:
2525         * Scripts/webkitpy/layout_tests/layout_package/test_expectations_unittest.py:
2526         * Scripts/webkitpy/layout_tests/layout_package/test_input.py:
2527         * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
2528         * Scripts/webkitpy/layout_tests/layout_package/test_results.py:
2529         * Scripts/webkitpy/layout_tests/layout_package/test_results_unittest.py:
2530         * Scripts/webkitpy/layout_tests/layout_package/worker.py:
2531         * Scripts/webkitpy/layout_tests/port/base.py:
2532         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2533         * Scripts/webkitpy/layout_tests/port/chromium.py:
2534         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2535         * Scripts/webkitpy/layout_tests/port/dryrun.py:
2536         * Scripts/webkitpy/layout_tests/port/test.py:
2537         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2538         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
2539         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
2540         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
2541         * Scripts/webkitpy/tool/bot/flakytestreporter.py:
2542         * Scripts/webkitpy/tool/commands/rebaseline.py:
2543         * Scripts/webkitpy/tool/commands/rebaselineserver.py:
2544
2545 2011-07-06  Chris Guillory  <ctguil@chromium.org>
2546
2547         No review necessary.
2548
2549         Adding myself to committers list in committers.py.
2550         https://bugs.webkit.org/show_bug.cgi?id=63478
2551
2552         * Scripts/webkitpy/common/config/committers.py:
2553
2554 2011-07-06  Johnny Ding  <jnd@chromium.org>
2555
2556         https://bugs.webkit.org/show_bug.cgi?id=63857
2557         Implement createPopupMenu for Chromium DRT.
2558
2559         Reviewed by Tony Chang.
2560
2561         * DumpRenderTree/chromium/WebViewHost.cpp:
2562         (WebViewHost::createPopupMenu):
2563         (WebViewHost::~WebViewHost):
2564         * DumpRenderTree/chromium/WebViewHost.h:
2565
2566 2011-07-06  Mihai Parparita  <mihaip@chromium.org>
2567
2568         Unreviewed, rolling out r90503.
2569         http://trac.webkit.org/changeset/90503
2570         https://bugs.webkit.org/show_bug.cgi?id=63358
2571
2572         Makes Chromium Windows canary bots hang.
2573
2574         * Scripts/webkitpy/layout_tests/port/base.py:
2575         * Scripts/webkitpy/layout_tests/port/chromium.py:
2576         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2577         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2578         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2579         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2580         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2581         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2582
2583 2011-07-06  Adam Barth  <abarth@webkit.org>
2584
2585         Move buildbot URL in webkitpy to config.urls
2586         https://bugs.webkit.org/show_bug.cgi?id=64041
2587
2588         Reviewed by Eric Seidel.
2589
2590         This patch prepares for using the buildbot class to talk to the
2591         Chromium buildbots.
2592
2593         * Scripts/webkitpy/common/config/urls.py:
2594         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2595         * Scripts/webkitpy/tool/commands/download_unittest.py:
2596         * Scripts/webkitpy/tool/commands/queries.py:
2597         * Scripts/webkitpy/tool/mocktool.py:
2598         * Scripts/webkitpy/tool/steps/ensurebuildersaregreen.py:
2599
2600 2011-07-06  Kevin Ollivier  <kevino@theolliviers.com>
2601
2602         [wx] Unreviewed build fix, add the correct path to the waf build settings.
2603         
2604         * wx/packaging/build-win-installer.py:
2605
2606 2011-07-06  Adam Barth  <abarth@webkit.org>
2607
2608         Add missing include.
2609
2610         * Scripts/webkitpy/tool/bot/irc_command.py:
2611
2612 2011-07-06  Adam Barth  <abarth@webkit.org>
2613
2614         Add roll-chromium-deps command to sheriff-bot
2615         https://bugs.webkit.org/show_bug.cgi?id=64037
2616
2617         Reviewed by Eric Seidel.
2618
2619         Lo, and the DEPS were rolled on command.
2620
2621         * Scripts/webkitpy/tool/bot/irc_command.py:
2622         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
2623         * Scripts/webkitpy/tool/bot/sheriff.py:
2624         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2625         * Scripts/webkitpy/tool/commands/roll.py:
2626         * Scripts/webkitpy/tool/commands/roll_unittest.py:
2627         * Scripts/webkitpy/tool/steps/updatechromiumdeps.py:
2628
2629 2011-07-06  Eric Seidel  <eric@webkit.org>
2630
2631         Split Driver, DriverOutput and DriverInput out into a new driver.py file
2632         https://bugs.webkit.org/show_bug.cgi?id=64031
2633
2634         Reviewed by Adam Barth.
2635
2636         This ended up kinda large and I'm happy to break it up if desired.
2637
2638         As part of cleaning up the Port module, I split Driver, DriverInput and DriverOutput
2639         out into its own driver.py file.
2640
2641         I also cleaned up how we import Port objects, by making port/__init__.py
2642         expose Port, Driver, DriverInput, and DriverOutput.  I removed exposure of
2643         some various test objects as they were causing circular dependencies.
2644         base.py is now an implementation detail of port.  No longer is "base" referenced
2645         anywhere in the code.
2646
2647         I also made Driver.__init__ non-virtual so that sub-classers could all share
2648         the data storage on Driver.  This more closely matches how other subclasses
2649         work throughout our python code.
2650
2651         Standard line-unwrapping and doc-string removal as I went through the code.
2652
2653         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2654         * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
2655         * Scripts/webkitpy/layout_tests/port/__init__.py:
2656         * Scripts/webkitpy/layout_tests/port/base.py:
2657         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2658         * Scripts/webkitpy/layout_tests/port/chromium.py:
2659         * Scripts/webkitpy/layout_tests/port/driver.py: Added.
2660         * Scripts/webkitpy/layout_tests/port/dryrun.py:
2661         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
2662         * Scripts/webkitpy/layout_tests/port/test.py:
2663         * Scripts/webkitpy/layout_tests/port/webkit.py:
2664         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
2665         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
2666         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
2667
2668 2011-07-06  Adam Roben  <aroben@apple.com>
2669
2670         Clear TestFailures's getNumberOfFailingTests cache
2671
2672         Now that we support NRWT, we need to clear out old zero values for the NRWT bots that might
2673         have been cached by old versions of TestFailures.
2674
2675         Fixes <http://webkit.org/b/64029> TestFailures page still doesn't show NRWT bots if you've
2676         used it recently
2677
2678         Reviewed by Adam Barth.
2679
2680         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
2681         (Builder.prototype.getNumberOfFailingTests): Added a .version property to the cache data
2682         that must match the current version. Otherwise we'll ignore the cached data.
2683
2684 2011-06-24  Dirk Pranke  <dpranke@chromium.org>
2685
2686         Reviewed by Adam Barth.
2687
2688         nrwt: remove --use-apache from the command line
2689         https://bugs.webkit.org/show_bug.cgi?id=63358
2690
2691         This change removes the --use-apache command line argument. It
2692         was initially put in when we were trying to get the cygwin
2693         apache instance to work with Chromium win, but that code has
2694         bitrotted and doesn't work at all now.
2695
2696         Arguably we should remove all of the code to allow for a choice
2697         of web servers, but since we may still want to switch off of
2698         LigHTTPd at some point on Windows, I'll leave the rest in for
2699         now (or at least yank it in a different change.)
2700
2701         * Scripts/webkitpy/layout_tests/port/base.py:
2702         * Scripts/webkitpy/layout_tests/port/chromium.py:
2703         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
2704         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2705         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2706         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2707
2708 2011-07-06  Dimitri Glazkov  <dglazkov@chromium.org>
2709
2710         garden-o-matic should use a favicon to indicate current state.
2711         https://bugs.webkit.org/show_bug.cgi?id=64027
2712
2713         Reviewed by Adam Barth.
2714
2715         * Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-green.png: Added.
2716         * Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-red.png: Added.
2717         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html: Added blurb about icons.
2718         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js: Added setting icon state.
2719         * Scripts/webkitpy/tool/servers/gardeningserver.py: Allowed the use of favicons.
2720
2721 2011-07-06  Daniel Bates  <dbates@rim.com>
2722
2723         Fix webkitpy error after changeset 90483 <http://trac.webkit.org/changeset/90483>; Add missing
2724         parameter to _fetch_file_from_results lamdba function.
2725
2726         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2727
2728 2011-07-06  Kevin Ollivier  <kevino@theolliviers.com>
2729
2730         [wx] Unreviewed build fix. Generate the proper install_name for dylibs on Mac.
2731
2732         * wx/packaging/build-mac-installer.py:
2733
2734 2011-07-06  Xan Lopez  <xlopez@igalia.com>
2735
2736         [GTK] Only set env. variables in start when strictly needed
2737         https://bugs.webkit.org/show_bug.cgi?id=64026
2738
2739         Reviewed by Eric Seidel.
2740
2741         * Scripts/webkitpy/layout_tests/port/gtk.py: add a
2742         setup_environ_for_server method in GtkPort and set as many
2743         env. variables as possible there.
2744
2745 2011-07-06  Adam Roben  <aroben@apple.com>
2746
2747         Teach TestFailures how to load, parse, and interpret NRWT test results
2748
2749         Fixes <http://webkit.org/b/61877> TestFailures page doesn't show testers that use
2750         new-run-webkit-tests
2751
2752         Reviewed by Adam Barth.
2753
2754         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder.js:
2755         (Builder.prototype.failureDiagnosisTextAndURL): Added support for the new 'flaky' failure
2756         type. For now we don't account for image-only flakes (but TestFailures doesn't deal with
2757         pixel tests at all currently).
2758         (Builder.prototype.getNumberOfFailingTests): Relaxed the regex used for parsing the number
2759         of failing tests from the buildbot output. Make sure not to count "new passes" (a new
2760         category introduced by NRWT) as failures.
2761
2762         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:
2763         (LayoutTestResultsLoader.prototype.start): Moved a bunch of code from here to
2764         _fetchAndParseORWTResults. This function now attempts to load NRWT results, then falls back
2765         to loading ORWT results.
2766         (LayoutTestResultsLoader.prototype._fetchAndParseNRWTResults): Added. Tries to load and
2767         parse the full_results.json file for the given build.
2768         (LayoutTestResultsLoader.prototype._fetchAndParseORWTResults): Added. Code came from start.
2769         Fixed a bug along the way where we were sometimes calling the error callback instead of the
2770         success callback when all tests passed.
2771
2772         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NRWTResultsParser.js: Added.
2773         (NRWTResultsParser): Do-nothing constructor.
2774         (NRWTResultsParser.prototype.parse): Uses eval() (eww!) to get the test results out of the
2775         JS string, then iterates over all the tests in the results data and builds up a data
2776         structure matching what ORWTResultsParser returns.
2777
2778         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Utilities.js:
2779         (Array.prototype.contains):
2780         (String.prototype.contains):
2781         Added these simple helper functions.
2782
2783         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
2784         NRWTResultsParser.js.
2785
2786 2011-07-06  Adam Roben  <aroben@apple.com>
2787
2788         Extract code to parse ORWT's results.html file into its own class
2789
2790         Prep work for fixing <http://webkit.org/b/61877> TestFailures page doesn't show testers that
2791         use new-run-webkit-tests
2792
2793         Reviewed by Adam Barth.
2794
2795         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/LayoutTestResultsLoader.js:
2796         (LayoutTestResultsLoader.prototype.start): Moved a bunch of code from here...
2797
2798         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ORWTResultsParser.js: Added.
2799         (ORWTResultsParser):
2800         (ORWTResultsParser.prototype.parse):
2801         ...to here.
2802
2803         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
2804         ORWTResultsParser.js.
2805
2806 2011-07-06  Eric Seidel  <eric@webkit.org>
2807
2808         Unreviewed.  Fix a unittest from hitting the network. :)
2809
2810         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2811
2812 2011-07-06  Eric Seidel  <eric@webkit.org>
2813
2814         Unreviewed.  Fixing an exception in test-webkitpy, caused by bug 64006.
2815
2816         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2817
2818 2011-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.com>
2819
2820         Run an Xvfb instance to run the API tests on, like we do for the
2821         layout tests.
2822
2823         Reviewed by Xan Lopez.
2824
2825         * Scripts/run-gtk-tests:
2826
2827 2011-07-06  Daniel Bates  <dbates@webkit.org>
2828
2829         Make SCM unit tests faster
2830         https://bugs.webkit.org/show_bug.cgi?id=63883
2831
2832         Speeds up the SCM unit tests by a factor of 4. Currently, we create a
2833         mock SVN repo for each test_ method in SVNTest and GitTest and creating
2834         this repo is expensive.
2835
2836         Instead, it is significantly faster to create the mock SVN repo once
2837         and then perform a filesystem copy of it for each test_ method.
2838
2839         Note, Python 2.7's unittest module implements support for per class and
2840         per module setup and tear down methods which could be used to implement
2841         similar functionality. At the time of writing, test-webkitpy is designed
2842         to support Python 2.5. So, we can't take advantage of these Python 2.7
2843         features :(
2844
2845         Reviewed by Eric Seidel.
2846
2847         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
2848
2849 2011-07-06  Adam Barth  <abarth@webkit.org>
2850
2851         Remove unused parameter from _add_test in test_expectations.py
2852         https://bugs.webkit.org/show_bug.cgi?id=64008
2853
2854         Reviewed by Eric Seidel.
2855
2856         * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:
2857
2858 2011-07-06  Adam Barth  <abarth@webkit.org>
2859
2860         Use old-run-webkit-tests when --leaks is present
2861         https://bugs.webkit.org/show_bug.cgi?id=64012
2862
2863         Reviewed by Adam Roben.
2864
2865         We'll remove this branch once support of --leaks is added to
2866         new-run-webkit-tests.
2867
2868         * Scripts/run-webkit-tests:
2869
2870 2011-07-06  Benjamin Poulain  <benjamin@webkit.org>
2871
2872         [Qt] [WK2] Add touch mocking to Qt's MiniBrowser
2873         https://bugs.webkit.org/show_bug.cgi?id=63995
2874
2875         Reviewed by Andreas Kling.
2876
2877         The mouse events are intercepted through QCoreApplication::notify(),
2878         and fake touch events are generated for mouse events when necessary.
2879
2880         If touch events are received from the platform, we assume the current
2881         hardware has a touch screen and we stop generating fake events.
2882
2883         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2884         (isTouchEvent):
2885         (isMouseEvent):
2886         (MiniBrowserApplication::MiniBrowserApplication):
2887         (MiniBrowserApplication::notify):
2888         * MiniBrowser/qt/MiniBrowserApplication.h:
2889
2890 2011-07-06  Xan Lopez  <xlopez@igalia.com>
2891
2892         Reviewed by Gustavo Noronha.
2893
2894         * Scripts/webkitpy/layout_tests/port/gtk.py: use Popen correctly.
2895
2896 2011-07-06  Eric Seidel  <eric@webkit.org>
2897
2898         webkit-patch failure-reason does not understand NRWT results
2899         https://bugs.webkit.org/show_bug.cgi?id=64006
2900
2901         Reviewed by Adam Barth.
2902
2903         Adam Barth tells me failure-reason should be deleted soon,
2904         but aroben's fancy new TestFailures/ page doesn't work for
2905         NRWT yet, and I needed to know when
2906         third-party-cookie-relaxing started failing.
2907
2908         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2909         * Scripts/webkitpy/tool/commands/queries.py:
2910
2911 2011-07-06  Xan Lopez  <xlopez@igalia.com>
2912
2913         [GTK] Add missing environment variables in NWRT
2914         https://bugs.webkit.org/show_bug.cgi?id=64004
2915
2916         Reviewed by Gustavo Noronha Silva.
2917
2918         Add missing environment variables needed by the GTK+ port.
2919
2920         * Scripts/webkitpy/layout_tests/port/gtk.py: ditto.
2921
2922 2011-07-06  Adam Roben  <aroben@apple.com>
2923
2924         Make run-api-tests output intelligible on the bots
2925
2926         run-api-tests's output on the bots was very hard to read for two reasons:
2927           1) It was mixing its own output with gtest's
2928           2) It was using ANSI escape sequences to print colored text, but the bots don't support
2929              that
2930
2931         Now, the --verbose flag turns off almost all of run-api-tests's own output so that gtest's
2932         output will not be obscured. We still print "Timeout" messages even in verbose mode, since
2933         gtest doesn't have any native support for timeouts. Also, when our output is being
2934         redirected to a file, we don't print ANSI escape sequences.
2935
2936         Fixes <http://webkit.org/b/63996> It's very hard to read run-api-tests output on the bots
2937
2938         Reviewed by Sam Weinig.
2939
2940         * Scripts/run-api-tests: Don't use the :constants interface of Term::ANSIColor. We want to
2941         use the colored function instead.
2942         (runAllTestsInSuite): Don't print out suite names in verbose mode; gtest will give us enough
2943         context that they aren't needed.
2944         (runTest): Don't print out test names or pass/fail messages in verbose mode; gtest will do
2945         that for us. Also, use the new possiblyColored function instead of always coloring output.
2946         (possiblyColored): Added. When printing to a tty, returns the string with the appropriate
2947         ANSI color escape sequences added. Otherwise just returns the string unmodified.
2948
2949 2011-07-06  Xan Lopez  <xlopez@igalia.com>
2950
2951         Reviewed by Gustavo Noronha.
2952
2953         Launch Xvfb (mostly) the same way we were launching it in the
2954         bots, for maximum compatibility.
2955
2956         * Scripts/webkitpy/layout_tests/port/gtk.py:
2957
2958 2011-07-06  Andreas Kling  <kling@webkit.org>
2959
2960         [Qt][WK2] Split Qt API into two different web views (touch and desktop)
2961         https://bugs.webkit.org/show_bug.cgi?id=63950
2962
2963         Co-authored with Benjamin Poulain <benjamin@webkit.org>.
2964
2965         Reviewed by Kenneth Rohde Christiansen.
2966
2967         Port WTR/Qt to use QDesktopWebView and force a 800x600 viewport.
2968
2969         The Qt MiniBrowser now defaults to QDesktopWebView but can be started
2970         in QTouchWebView mode by passing -touch on the command line.
2971
2972         A lot of functionality has been removed and/or disabled from
2973         MiniBrowser until we reintroduce the necessary API's.
2974
2975         * MiniBrowser/qt/BrowserView.cpp:
2976         (BrowserView::BrowserView):
2977         (BrowserView::~BrowserView):
2978         (BrowserView::load):
2979         (BrowserView::view):
2980         (BrowserView::touchWebView):
2981         (BrowserView::desktopWebView):
2982         * MiniBrowser/qt/BrowserView.h:
2983         * MiniBrowser/qt/BrowserWindow.cpp:
2984         (newPageFunction):
2985         (BrowserWindow::BrowserWindow):
2986         (BrowserWindow::webView):
2987         (BrowserWindow::newWindow):
2988         (BrowserWindow::showUserAgentDialog):
2989         (BrowserWindow::updateUserAgentList):
2990         * MiniBrowser/qt/BrowserWindow.h:
2991         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2992         (MiniBrowserApplication::handleUserOptions):
2993         * MiniBrowser/qt/MiniBrowserApplication.h:
2994         (WindowOptions::WindowOptions):
2995         * MiniBrowser/qt/UrlLoader.cpp:
2996         (UrlLoader::UrlLoader):
2997         * MiniBrowser/qt/main.cpp:
2998         (main):
2999         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
3000         (WTR::WebView::wkView):
3001         (WTR::WebView::pageRef):
3002         (WTR::WebView::WebView):
3003         (WTR::PlatformWebView::page):
3004         (WTR::PlatformWebView::windowFrame):
3005         (WTR::PlatformWebView::setWindowFrame):
3006
3007 2011-07-06  Eric Seidel  <eric@webkit.org>
3008
3009         Unreviewed..
3010
3011         * Scripts/run-webkit-tests:
3012         (useNewRunWebKitTests):
3013          - Fix the pass-thru of --gtk and --qt for old-run-webkit-tests,
3014            it was previously broken by the existence of isGtk/isQt checks which
3015            removed the --gtk/--qt args from @ARGV.
3016
3017 2011-07-06  Eric Seidel  <eric@webkit.org>
3018
3019         new-run-webkit-tests uses a 35s timeout for Mac and 6s for all other ports
3020         https://bugs.webkit.org/show_bug.cgi?id=63983
3021
3022         Reviewed by Xan Lopez.
3023
3024         All WebKit ports want to use a 35s timeout.  Only chromium currently
3025         uses a 6s timeout.  Eventually we'll move all WebKit ports to 6s too
3026         to further speed up new-run-webkit-tests runs, but that's going to be a while.
3027
3028         * Scripts/webkitpy/layout_tests/port/mac.py:
3029         * Scripts/webkitpy/layout_tests/port/webkit.py:
3030         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3031
3032 2011-07-06  Xan Lopez  <xlopez@igalia.com>
3033
3034         Unreviewed.
3035
3036         NWRT actually seems to work locally, so it should work in the bots
3037         as long as we take care of the TIMEOUT issue (which is being fixed
3038         as I write this). Let's give this another shot and fix whatever
3039         breaks.
3040
3041         * Scripts/run-webkit-tests:
3042         (useNewRunWebKitTests):
3043
3044 2011-07-06  Xan Lopez  <xlopez@igalia.com>
3045
3046         Unreviewed.
3047
3048         Disable NWRT on GTK. At the very least it needs an updated exected
3049         results file since NWRT is more sensitive to timeouts, not sure
3050         what the exact problem is.
3051
3052         * Scripts/run-webkit-tests:
3053         (useNewRunWebKitTests):
3054
3055 2011-07-06  Adam Barth  <abarth@webkit.org>
3056
3057         Fix global variable leak detected by noglobals.
3058
3059         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3060
3061 2011-07-06  Adam Barth  <abarth@webkit.org>
3062
3063         garden-o-matic shouldn't show flaky tests by default
3064         https://bugs.webkit.org/show_bug.cgi?id=63979
3065
3066         Reviewed by Ojan Vafai.
3067
3068         If any of the actual test results are PASS, the test is flaky and we
3069         can ignore it (at least for now).  If we get awesome at dealing with
3070         failing tests, we might want to tackle flaky tests by refering the user
3071         to the flakiness dashboard, which is way more awesome at dealing with
3072         flaky tests that this tool ever will be.
3073
3074         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3075         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
3076
3077 2011-07-05  Adam Barth  <abarth@webkit.org>
3078
3079         garden-o-matic should list the failing tests
3080         https://bugs.webkit.org/show_bug.cgi?id=63978
3081
3082         Reviewed by Eric Seidel.
3083
3084         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js: Added.
3085         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3086         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3087         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3088         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
3089         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html:
3090         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js: Added.
3091         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js: Added.
3092         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3093
3094
3095 2011-07-05  Eric Seidel  <eric@webkit.org>
3096
3097         Re-enable NRWT on Leopard now that bug 63973 is fixed.
3098
3099         Reviewed by Adam Barth.
3100
3101         I also made the should-enable expression a black-list instead
3102         of a white-list since every bot on build.webkit.org is
3103         now using NRWT except windows and WebKit2 bots.
3104
3105         * Scripts/run-webkit-tests:
3106         (useNewRunWebKitTests):
3107
3108 2011-07-05  Eric Seidel  <eric@webkit.org>
3109
3110         Perl string concat failure on Leopard (causes NRWT to fail)
3111         https://bugs.webkit.org/show_bug.cgi?id=63973
3112
3113         Reviewed by Daniel Bates.
3114
3115         Leopard's "xcodebuild -version" has a different format from Snow Leopard/Lion
3116         thus the version parsing logic in webkitdirs.pm seems to fail.
3117
3118         Default to "3.0" instead of undef when we can't parse the version number.
3119         This should send us down the correct path for Leopard.
3120
3121         * Scripts/webkitdirs.pm:
3122         (determineBaseProductDir):
3123
3124 2011-07-05  Kevin Ollivier  <kevino@theolliviers.com>
3125
3126         [wx] Unreviewed build fix after addition of Source/WebCore/xml/parser dir.
3127
3128         * waf/build/settings.py:
3129
3130 2011-07-05  Eric Seidel  <eric@webkit.org>
3131
3132         Disable NRWT for Leopard until https://bugs.webkit.org/show_bug.cgi?id=63973 can be fixed.
3133
3134         Reviewed by Adam Barth.
3135
3136         * Scripts/run-webkit-tests:
3137         (useNewRunWebKitTests):
3138
3139 2011-07-05  Eric Seidel  <eric@webkit.org>
3140
3141         Make run-webkit-tests pass --verbose to new-run-webkit-tests on qt/gtk bots.
3142
3143         Reviewed by Adam Barth.
3144
3145         * Scripts/run-webkit-tests:
3146         (runningOnBuildBot):
3147
3148 2011-07-05  Adam Barth  <abarth@webkit.org>
3149
3150         Add results_unittests.js to garden-o-matic
3151         https://bugs.webkit.org/show_bug.cgi?id=63971
3152
3153         Reviewed by Eric Seidel.
3154
3155         Adding these unit tests forced me to commit to an API for the results
3156         module.  I'm not sure if this is the API we'll end up with, but it's
3157         something not entirely unreasonable.
3158
3159         Also, I've changed results.js over to WebKit style, as discussed previously.
3160
3161         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3162         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3163         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js: Added.
3164         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html:
3165
3166 2011-07-05  Eric Seidel  <eric@webkit.org>
3167
3168         Teach run-webkit-tests how to translate --qt and --gtk for new-run-webkit-tests.
3169
3170         Reviewed by Adam Barth.
3171
3172         * Scripts/run-webkit-tests:
3173
3174 2011-07-05  Adam Barth  <abarth@webkit.org>
3175
3176         Add first unit tests for garden-o-matic
3177         https://bugs.webkit.org/show_bug.cgi?id=63969
3178
3179         Reviewed by Eric Seidel.
3180
3181         I've also reformated base.js to match WebKit style, as discussed in the
3182         previous bug.
3183
3184         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
3185         * Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js: Added.
3186         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html: Added.
3187
3188 2011-07-05  Eric Seidel  <eric@webkit.org>
3189
3190         Move Qt, Gtk and Leopard to NRWT.
3191
3192         We'll roll out any of these if we see problems.
3193
3194         Reviewed by Adam Barth.
3195
3196         * Scripts/run-webkit-tests:
3197         (useNewRunWebKitTests):
3198
3199 2011-07-05  Adam Barth  <abarth@webkit.org>
3200
3201         garden-o-matic should understand which tests have unexpected failures
3202         https://bugs.webkit.org/show_bug.cgi?id=63965
3203
3204         Reviewed by Eric Seidel.
3205
3206         Currently, this code just logs the list of unexpected failures to the
3207         console.  A future patch will do something useful with this
3208         information.  A testing harness will also come in the next patch.
3209
3210         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js: Added.
3211         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3212         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3213         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3214
3215 2011-07-05  Adam Barth  <abarth@webkit.org>
3216
3217         garden-o-matic should know how to fetch test results from the (NRWT) bots
3218         https://bugs.webkit.org/show_bug.cgi?id=63959
3219
3220         Reviewed by Eric Seidel.
3221
3222         Currently, this code just fetches the results and expectations JSON
3223         blob for Linux from the server and logs it to the console.
3224
3225         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3226         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js: Added.
3227         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3228
3229 2011-07-05  Dirk Pranke  <dpranke@chromium.org>
3230
3231         Reviewed by Eric Seidel.
3232
3233         Python tests are failing on leopard
3234         https://bugs.webkit.org/show_bug.cgi?id=63842
3235
3236         Fix the missing flush() call that appears to be needed by the
3237         logging package in python 2.5. Also fix the regressions
3238         introduced in the run_webkit_test tests when we switched from
3239         thread to processes (since processes aren't available on 2.5).
3240
3241         Also fix a missing "from __future__ import with_statement' in
3242         testfilehandler (unreviewed).
3243
3244         * Scripts/webkitpy/layout_tests/layout_package/metered_stream.py:
3245         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3246         * TestResultServer/handlers/testfilehandler.py:
3247
3248 2011-07-05  Eric Seidel  <eric@webkit.org>
3249
3250         Reviewed by Adam Barth.
3251
3252         re-enabling new-run-webkit-tests on the Snow Leopard WebKit1 bots
3253         now that we believe the stray http-server issue to be solved.
3254
3255         * Scripts/run-webkit-tests:
3256         (useNewRunWebKitTests):
3257
3258 2011-07-05  Adam Barth  <abarth@webkit.org>
3259
3260         Turns out we need to support numerals too because that's what jQuery
3261         uses by default.
3262
3263         * TestResultServer/handlers/testfilehandler.py:
3264
3265 2011-07-05  Dirk Pranke  <dpranke@chromium.org>
3266
3267         nrwt: allow for multiple http shards
3268         https://bugs.webkit.org/show_bug.cgi?id=63116
3269
3270         Reviewed by Tony Chang.
3271
3272         This modifies the sharding logic to support multiple http
3273         shards, but for now we clamp to one shard until we can test
3274         perf impact and flakiness impact.
3275
3276         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3277         * Scripts/webkitpy/layout_tests/layout_package/manager_unittest.py:
3278
3279 2011-07-05  Dirk Pranke  <dpranke@chromium.org>
3280
3281         Re-land nrwt: make sharding tests needing locks less hard-coded
3282         https://bugs.webkit.org/show_bug.cgi?id=63112
3283
3284         Reviewed by Ojan Vafai.
3285
3286         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3287         * Scripts/webkitpy/layout_tests/layout_package/manager_unittest.py:
3288
3289 2011-07-05  Adam Barth  <abarth@webkit.org>
3290
3291         TestResultServer should support callback parameter for JSON
3292         https://bugs.webkit.org/show_bug.cgi?id=63961
3293
3294         Reviewed by Ojan Vafai.
3295
3296         This makes using jQuery.ajax so much more pleasant.
3297
3298         * TestResultServer/handlers/testfilehandler.py:
3299
3300 2011-07-05  Eric Seidel  <eric@webkit.org>
3301
3302         buildbot needs to understand whether NRWT exited early after having too many failures
3303         https://bugs.webkit.org/show_bug.cgi?id=63839
3304
3305         Reviewed by Adam Barth.
3306
3307         Fix-up the exited early messages printed by NRWT so that
3308         the buildbot can parse them as expected.
3309         It looks for lines using "if line.find('Exiting early') >= 0:"
3310
3311         I also plumbed the "early exit" status through to results.json
3312         in the form of an "interrupted" bool.  It was unclear to me
3313         if results.json already had enough information to compute this bool
3314         itself.  It's possible Ojan could come up with a better fix.
3315
3316         * Scripts/webkitpy/layout_tests/layout_package/manager.py:
3317         * Scripts/webkitpy/layout_tests/layout_package/manager_unittest.py:
3318         * Scripts/webkitpy/layout_tests/layout_package/printing_unittest.py:
3319         * Scripts/webkitpy/layout_tests/layout_package/result_summary.py:
3320
3321 2011-07-05  Eric Seidel  <eric@webkit.org>
3322
3323         new-run-webkit-tests fails to start http server if one is already running
3324         https://bugs.webkit.org/show_bug.cgi?id=63956
3325
3326         Reviewed by Adam Barth.
3327
3328         * BuildSlaveSupport/kill-old-processes:
3329          - Removed ^M line endings and sorted the process names.
3330          - There are no functional changes to this file.
3331            (except that now that its using unix line endings it can be executed directly!)
3332         * Scripts/webkitpy/layout_tests/servers/apache_http_server.py:
3333          - Change the name to 'httpd' to match old-run-webkit-test paths.
3334
3335 2011-07-05  Adam Barth  <abarth@webkit.org>
3336
3337         Add basic ajax support to garden-o-matic
3338         https://bugs.webkit.org/show_bug.cgi?id=63874
3339
3340         Reviewed by Eric Seidel.
3341
3342         This patch adds some basic infrastructure to garden-o-matic.
3343         Currently, the infrastructure is just used to make the "quit" command
3344         use Ajax, but in the future, this infrastructure will be used to do
3345         more sophistocated remote proceedure calls.
3346
3347         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3348         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js: Added.
3349         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3350         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
3351
3352 2011-07-05  Adam Barth  <abarth@webkit.org>
3353
3354         Add trivial garden-o-matic command (with server)
3355         https://bugs.webkit.org/show_bug.cgi?id=63872
3356
3357         Reviewed by Eric Seidel.
3358
3359         This patch adds a trivial "Hello, world" garden-o-matic command,
3360         complete with an HTTP server.  This command re-uses a bunch of code
3361         from the existing rebaseline-server command.  Over time, this command
3362         will grow to be a tool useful for gardening the WebKit tree.
3363
3364         * Scripts/webkitpy/tool/commands/__init__.py:
3365         * Scripts/webkitpy/tool/commands/abstractservercommand.py: Added.
3366         * Scripts/webkitpy/tool/commands/gardenomatic.py: Added.
3367         * Scripts/webkitpy/tool/commands/rebaselineserver.py:
3368         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html: Added.
3369         * Scripts/webkitpy/tool/servers/gardeningserver.py: Added.
3370         * Scripts/webkitpy/tool/servers/rebaselineserver.py:
3371
3372 2011-07-05  Adam Roben  <aroben@apple.com>
3373
3374         Make prepare-ChangeLog include modified Perl functions in its ChangeLog template
3375
3376         This is a very simple first cut. Functions must start with a line that starts with "sub "
3377         and end with a line that starts with a closing brace. No leading whitespace is allowed.
3378         Package names aren't parsed at all.
3379
3380         Fixes <http://webkit.org/b/21591> prepare-ChangeLog should know how to find functions in
3381         Perl files
3382
3383         Reviewed by David Kilzer.
3384
3385         * Scripts/prepare-ChangeLog: Removed redundant code that ignored certain files when
3386         searching for function line ranges. This is already done inside the get_function_line_ranges
3387         function.
3388         (get_function_line_ranges): Cleaned up coding style a little bit. Call
3389         get_function_line_ranges_for_perl for files with .pl and .pm extensions. For files with an
3390         unknown extension or no extension, read the shebang line to try to determine the script
3391         interpreter. Call get_function_line_ranges_for_perl if the interpreter seems to be Perl.
3392         (get_function_line_ranges_for_perl): Added. Does extremely basic parsing of the file to find
3393         lines starting with "sub " or "}".
3394
3395 2011-07-05  Adam Roben  <aroben@apple.com>
3396
3397         Clean up run-api-tests output on Windows
3398
3399         We were mixing run-api-tests output with gtest output, and the result was a mess.
3400
3401         Fixes <http://webkit.org/b/63954> run-api-tests output is very confusing on Windows
3402
3403         Reviewed by Dan Bates.
3404
3405         * Scripts/run-api-tests:
3406         (runTest):
3407         (populateTests):
3408         Made Windows use the formerly-Mac-only codepaths which correctly suppress gtest output
3409         except in verbose mode. Most changes are due to indentation. Use of the arch utility is now
3410         the only Mac-specific thing in these functions. Some other platform-specific code was moved
3411         from here...
3412
3413         (prepareEnvironmentForRunningTestTool):
3414         (testToolPath):
3415         ...to here.
3416
3417 2011-07-04  Gavin Barraclough  <barraclough@apple.com>
3418
3419         Reviewed by Sam Weinig.
3420
3421         https://bugs.webkit.org/show_bug.cgi?id=16652
3422         Firefox and JavaScriptCore differ in Number.toString(integer)
3423
3424         Added forwarding header.
3425
3426         * DumpRenderTree/ForwardingHeaders/wtf/StdLibExtras.h: Added.
3427
3428 2011-07-04  Anders Carlsson  <andersca@apple.com>
3429
3430         Try to fix Qt build.
3431
3432         * DumpRenderTree/TestNetscapePlugIn/Tests/NPRuntimeRemoveProperty.cpp:
3433
3434 2011-07-04  Anders Carlsson  <andersca@apple.com>
3435
3436         NP_RemoveProperty is not called back by Safari when delete npObject.prop is encountered in JavaScript
3437         https://bugs.webkit.org/show_bug.cgi?id=63915
3438         <rdar://problem/7124300>
3439
3440         Reviewed by Sam Weinig.
3441
3442         Extend the NPRuntimeRemoveProperty to handle delete object.property from JavaScript.
3443
3444         * DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp:
3445         (PluginTest::NPN_IdentifierIsString):
3446         (PluginTest::NPN_UTF8FromIdentifier):
3447         (PluginTest::NPN_IntFromIdentifier):
3448         (PluginTest::NPN_RetainObject):
3449         (PluginTest::NPN_ReleaseObject):
3450         * DumpRenderTree/TestNetscapePlugIn/PluginTest.h:
3451         (PluginTest::Object::removeProperty):
3452         (PluginTest::Object::identifierIs):
3453         (PluginTest::Object::NP_RemoveProperty):
3454         (PluginTest::Object::npClass):
3455         * DumpRenderTree/TestNetscapePlugIn/Tests/NPRuntimeRemoveProperty.cpp:
3456         (NPRuntimeRemoveProperty::TestObject::TestObject):
3457         (NPRuntimeRemoveProperty::TestObject::hasProperty):
3458         (NPRuntimeRemoveProperty::TestObject::getProperty):
3459         (NPRuntimeRemoveProperty::TestObject::removeProperty):
3460         (NPRuntimeRemoveProperty::PluginObject::PluginObject):
3461         (NPRuntimeRemoveProperty::PluginObject::~PluginObject):
3462         (NPRuntimeRemoveProperty::PluginObject::hasMethod):
3463         (NPRuntimeRemoveProperty::PluginObject::invoke):
3464         (NPRuntimeRemoveProperty::PluginObject::hasProperty):
3465         (NPRuntimeRemoveProperty::PluginObject::getProperty):
3466         (NPRuntimeRemoveProperty::NPP_GetValue):
3467
3468 2011-07-04  Leandro Pereira  <leandro@profusion.mobi>
3469
3470         Reviewed by Kent Tamura.
3471
3472         [EFL] Add LayoutTestControllerEfl.
3473         https://bugs.webkit.org/show_bug.cgi?id=61974
3474
3475         Only the core features are implemented. Most functions are actually
3476         stubs and will be implemented as soon as its features are needed.
3477         Build system changes will be made as soon as DRT is completely
3478         upstreamed.
3479
3480         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp: Added.
3481         (LayoutTestController::~LayoutTestController):
3482         (LayoutTestController::addDisallowedURL):
3483         (LayoutTestController::clearBackForwardList):
3484         (LayoutTestController::copyDecodedHostName):
3485         (LayoutTestController::copyEncodedHostName):
3486         (LayoutTestController::dispatchPendingLoadRequests):
3487         (LayoutTestController::display):
3488         (LayoutTestController::counterValueForElementById):
3489         (LayoutTestController::keepWebHistory):
3490         (LayoutTestController::computedStyleIncludingVisitedInfo):
3491         (LayoutTestController::nodesFromRect):
3492         (LayoutTestController::layerTreeAsText):
3493         (LayoutTestController::pageNumberForElementById):
3494         (LayoutTestController::numberOfPages):
3495         (LayoutTestController::pageProperty):
3496         (LayoutTestController::isPageBoxVisible):
3497         (LayoutTestController::pageSizeAndMarginsInPixels):
3498         (LayoutTestController::webHistoryItemCount):
3499         (LayoutTestController::workerThreadCount):
3500         (LayoutTestController::notifyDone):
3501         (LayoutTestController::pathToLocalResource):
3502         (LayoutTestController::queueLoad):
3503         (LayoutTestController::setAcceptsEditing):
3504         (LayoutTestController::setAlwaysAcceptCookies):
3505         (LayoutTestController::setCustomPolicyDelegate):
3506         (LayoutTestController::waitForPolicyDelegate):
3507         (LayoutTestController::setScrollbarPolicy):
3508         (LayoutTestController::addOriginAccessWhitelistEntry):
3509         (LayoutTestController::removeOriginAccessWhitelistEntry):
3510         (LayoutTestController::setMainFrameIsFirstResponder):
3511         (LayoutTestController::setTabKeyCyclesThroughElements):
3512         (LayoutTestController::setUseDashboardCompatibilityMode):
3513         (LayoutTestController::setUserStyleSheetEnabled):
3514         (LayoutTestController::setUserStyleSheetLocation):
3515         (LayoutTestController::setValueForUser):
3516         (LayoutTestController::setViewModeMediaFeature):
3517         (LayoutTestController::setWindowIsKey):
3518         (LayoutTestController::setSmartInsertDeleteEnabled):
3519         (waitToDumpWatchdogFired):
3520         (LayoutTestController::setWaitToDump):
3521         (LayoutTestController::windowCount):
3522         (LayoutTestController::setPrivateBrowsingEnabled):
3523         (LayoutTestController::setJavaScriptCanAccessClipboard):
3524         (LayoutTestController::setXSSAuditorEnabled):
3525         (LayoutTestController::setFrameFlatteningEnabled):
3526         (LayoutTestController::setSpatialNavigationEnabled):
3527         (LayoutTestController::setAllowUniversalAccessFromFileURLs):
3528         (LayoutTestController::setAllowFileAccessFromFileURLs):
3529         (LayoutTestController::setAuthorAndUserStylesEnabled):
3530         (LayoutTestController::setAutofilled):
3531         (LayoutTestController::disableImageLoading):
3532         (LayoutTestController::setMockDeviceOrientation):
3533         (LayoutTestController::setMockGeolocationPosition):
3534         (LayoutTestController::setMockGeolocationError):
3535         (LayoutTestController::setGeolocationPermission):
3536         (LayoutTestController::numberOfPendingGeolocationPermissionRequests):
3537         (LayoutTestController::addMockSpeechInputResult):
3538         (LayoutTestController::setIconDatabaseEnabled):
3539         (LayoutTestController::setJavaScriptProfilingEnabled):
3540         (LayoutTestController::setSelectTrailingWhitespaceEnabled):
3541         (LayoutTestController::setPopupBlockingEnabled):
3542         (LayoutTestController::setPluginsEnabled):
3543         (LayoutTestController::elementDoesAutoCompleteForElementWithId):
3544         (LayoutTestController::execCommand):
3545         (LayoutTestController::findString):
3546         (LayoutTestController::isCommandEnabled):
3547         (LayoutTestController::setCacheModel):
3548         (LayoutTestController::setPersistentUserStyleSheetLocation):
3549         (LayoutTestController::clearPersistentUserStyleSheet):
3550         (LayoutTestController::clearAllApplicationCaches):
3551         (LayoutTestController::setApplicationCacheOriginQuota):
3552         (LayoutTestController::clearApplicationCacheForOrigin):
3553         (LayoutTestController::localStorageDiskUsageForOrigin):
3554         (LayoutTestController::originsWithApplicationCache):
3555         (LayoutTestController::applicationCacheDiskUsageForOrigin):
3556         (LayoutTestController::clearAllDatabases):
3557         (LayoutTestController::setDatabaseQuota):
3558         (LayoutTestController::originsWithLocalStorage):
3559         (LayoutTestController::deleteAllLocalStorage):
3560         (LayoutTestController::deleteLocalStorageForOrigin):
3561         (LayoutTestController::observeStorageTrackerNotifications):
3562         (LayoutTestController::syncLocalStorage):
3563         (LayoutTestController::setDomainRelaxationForbiddenForURLScheme):
3564         (LayoutTestController::goBack):
3565         (LayoutTestController::setDefersLoading):
3566         (LayoutTestController::setAppCacheMaximumSize):
3567         (LayoutTestController::pauseAnimationAtTimeOnElementWithId):
3568         (LayoutTestController::pauseTransitionAtTimeOnElementWithId):
3569         (LayoutTestController::sampleSVGAnimationForElementAtTime):