da6364b06b626efe9af188e9cc68cbf08f07b239
[WebKit-https.git] / Tools / ChangeLog
1 2011-08-12  Sadrul Habib Chowdhury  <sadrul@chromium.org>
2
3         DRT: Add support for gesture events.
4         https://bugs.webkit.org/show_bug.cgi?id=66105
5
6         Reviewed by Darin Fisher.
7
8         * DumpRenderTree/chromium/EventSender.cpp:
9         (EventSender::EventSender):
10         (EventSender::gestureScrollBegin):
11         (EventSender::gestureScrollEnd):
12         (EventSender::gestureTap):
13         (EventSender::gestureEvent):
14         * DumpRenderTree/chromium/EventSender.h:
15
16 2011-08-12  Andy Estes  <aestes@apple.com>
17
18         allowRoundingHacksCallback uses the wrong argument for 'thisObject'.
19         https://bugs.webkit.org/show_bug.cgi?id=66158
20
21         Reviewed by Dan Bernstein.
22
23         * DumpRenderTree/LayoutTestController.cpp:
24         (allowRoundingHacksCallback): The third argument is the 'this'
25         JSObjectRef, not the second.
26
27 2011-08-12  Andy Estes  <aestes@apple.com>
28
29         Cancel in onbeforeunload dialog sometime causes a button to stop working.
30         https://bugs.webkit.org/show_bug.cgi?id=26211
31
32         Reviewed by Alexey Proskuryakov.
33
34         Implement a new LayoutTestController method that a test can call to set
35         the return value of DumpRenderTree's beforeunload UI delegate.
36
37         * DumpRenderTree/LayoutTestController.cpp:
38         (LayoutTestController::LayoutTestController): Initialize
39         m_shouldStayOnPageAfterHandlingBeforeUnload.
40         (setShouldStayOnPageAfterHandlingBeforeUnloadCallback): Call
41         LayoutTestController::setShouldStayOnPageAfterHandlingBeforeUnload().
42         (LayoutTestController::staticFunctions): Register
43         'setShouldStayOnPageAfterHandlingBeforeUnload' as a static function.
44         * DumpRenderTree/LayoutTestController.h: 
45         (LayoutTestController::shouldStayOnPageAfterHandlingBeforeUnload):
46         (LayoutTestController::setShouldStayOnPageAfterHandlingBeforeUnload):
47         * DumpRenderTree/chromium/LayoutTestController.cpp:
48         (LayoutTestController::LayoutTestController): Initialize
49         m_shouldStayOnPageAfterHandlingBeforeUnload and bind
50         'setShouldStayOnPageAfterHandlingBeforeUnload' to its c++ setter.
51         (LayoutTestController::reset): Reset
52         m_shouldStayOnPageAfterHandlingBeforeUnload to false.
53         (LayoutTestController::setShouldStayOnPageAfterHandlingBeforeUnload):
54         Set m_shouldStayOnPageAfterHandlingBeforeUnload to the value passed in
55         from JavaScript.
56         * DumpRenderTree/chromium/LayoutTestController.h:
57         (LayoutTestController::shouldStayOnPageAfterHandlingBeforeUnload):
58         * DumpRenderTree/chromium/WebViewHost.cpp:
59         (WebViewHost::runModalBeforeUnloadDialog): Add the correct logging and
60         return the inverse of
61         LayoutTestController::shouldStayOnPageAfterHandlingBeforeUnload().
62         * DumpRenderTree/mac/UIDelegate.mm:
63         (-[UIDelegate webView:runBeforeUnloadConfirmPanelWithMessage:initiatedByFrame:]): Ditto.
64         * DumpRenderTree/win/UIDelegate.cpp:
65         (UIDelegate::runBeforeUnloadConfirmPanelWithMessage): Ditto.
66         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
67         Define setShouldStayOnPageAfterHandlingBeforeUnload().
68         * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
69         (WTR::InjectedBundle::postNewBeforeUnloadReturnValue): Post a message
70         to the Test Controller telling it what value it should return in its
71         onbeforeunload UI delegate.
72         * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
73         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
74         (WTR::LayoutTestController::setShouldStayOnPageAfterHandlingBeforeUnload):
75         Call InjectedBundle::postNewBeforeUnloadReturnValue().
76         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
77         * WebKitTestRunner/TestController.cpp:
78         (WTR::TestController::TestController): Initialize
79         m_beforeUnloadReturnValue to true.
80         (WTR::runBeforeUnloadConfirmPanel): Add the correct logging and return
81         the value of TestController::beforeUnloadReturnValue().
82         (WTR::TestController::resetStateToConsistentValues): Reset
83         m_beforeUnloadReturnValue to true.
84         * WebKitTestRunner/TestController.h:
85         (WTR::TestController::beforeUnloadReturnValue):
86         (WTR::TestController::setBeforeUnloadReturnValue):
87         * WebKitTestRunner/TestInvocation.cpp:
88         (WTR::TestInvocation::didReceiveMessageFromInjectedBundle): Handle the
89         message posted from the injected bundle by retrieving the message body
90         as a WKBoolean and calling TestController::setBeforeUnloadReturnValue().
91
92 2011-08-12  Sam Weinig  <sam@webkit.org>
93
94         Move compiler specific macros to their own header
95         https://bugs.webkit.org/show_bug.cgi?id=66119
96
97         Reviewed by Anders Carlsson.
98
99         * DumpRenderTree/ForwardingHeaders/wtf/Compiler.h: Added.
100
101 2011-08-11  Dimitri Glazkov  <dglazkov@chromium.org>
102
103         Add a way to extend DOM objects in garden-o-matic.
104         https://bugs.webkit.org/show_bug.cgi?id=66096
105
106         Reviewed by Adam Barth.
107
108         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js: Added base.extends.
109         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js: Added tests.
110
111 2011-08-11  MORITA Hajime  <morrita@google.com>
112
113         [Chromium][DRT] editing/spelling/spelling-backspace-between-lines.html fails
114         https://bugs.webkit.org/show_bug.cgi?id=45438
115
116         Reviewed by Kent Tamura.
117
118         MockSpellCheck was assuming the input of spellCheckWord() is a word.
119         But it can be a series of word. This change made spellCheckWord()
120         understand simple word boundary which appears in tests.
121         
122         * DumpRenderTree/chromium/MockSpellCheck.cpp:
123         (MockSpellCheck::spellCheckWord):
124
125 2011-08-11  Mark Rowe  <mrowe@apple.com>
126
127         Rework some Makefile logic to remove a double-negative.
128
129         Reviewed by Jon Honeycutt.
130
131         * DumpRenderTree/mac/PerlSupport/Makefile:
132
133 2011-08-11  Anders Carlsson  <andersca@apple.com>
134
135         Add a FIXME.
136
137         * TestWebKitAPI/TestsController.cpp:
138         (TestWebKitAPI::TestsController::TestsController):
139
140 2011-08-11  Anders Carlsson  <andersca@apple.com>
141
142         Initialize threading before running any tests
143         https://bugs.webkit.org/show_bug.cgi?id=66112
144
145         Reviewed by Adam Barth.
146
147         * TestWebKitAPI/TestsController.cpp:
148         (TestWebKitAPI::TestsController::TestsController):
149
150 2011-08-11  Adam Barth  <abarth@webkit.org>
151
152         Add unexpected-passes.html to TestFailures for marking tests as passing
153         https://bugs.webkit.org/show_bug.cgi?id=66102
154
155         Reviewed by Dimitri Glazkov.
156
157         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
158         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
159             - Move some code into the library so it can be shared.
160         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
161         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
162         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
163             - Add round-trip unit tests.
164         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/unexpected-passes.html: Added.
165         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/unexpected-passes.js: Added.
166             - I'm not entirely sure this feature should remain a separate HTML
167               file, but it seemed better than cluttering up the main HTML file
168               with too much extra stuff.
169         * Scripts/webkitpy/tool/servers/gardeningserver.py:
170         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
171             - The gardening server couldn't handle adding PASS expectations.  Now it can.
172
173 2011-08-11  Eric Seidel  <eric@webkit.org>
174
175         NRWT has wrong fallback order for Mac now that Lion exists
176         https://bugs.webkit.org/show_bug.cgi?id=66093
177
178         Reviewed by Adam Barth.
179
180         ORWT used a different system for fallback orders than NRWT did.
181         I moved win.py to match ORWT fallback orders in bug 64486.
182         This bug moves mac.py to match ORWT fallback order
183         and adds 'lion' as a supported OS version.
184
185         Because ChromiumMac shares OS version detection code with AppleMac
186         this also added support for 'lion' to chromium mac.  Hopefully that's a good thing.
187
188         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
189         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
190         * Scripts/webkitpy/layout_tests/port/mac.py:
191         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
192         * Scripts/webkitpy/layout_tests/port/win.py:
193
194 2011-08-11  Adam Barth  <abarth@webkit.org>
195
196         Update test results now that we use absolute URLs.
197
198         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js:
199
200 2011-08-11  Adam Barth  <abarth@webkit.org>
201
202         garden-o-matic should be able to determine when tests unexpectedly pass
203         https://bugs.webkit.org/show_bug.cgi?id=66092
204
205         Reviewed by Dimitri Glazkov.
206
207         This patch just contains the library methods for extracting this
208         information.  I'll add some UI in a subsequent patch.
209
210         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
211         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
212
213 2011-08-11  Dimitri Glazkov  <dglazkov@chromium.org>
214
215         New expectation entries should be placed next to existing ones.
216         https://bugs.webkit.org/show_bug.cgi?id=66066
217
218         Reviewed by Adam Barth.
219
220         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added the notion of insertion point,
221             which is computed as the last found reference to the same test.
222         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added and adjusted unit tests.
223
224 2011-08-11  Dimitri Glazkov  <dglazkov@chromium.org>
225
226         Unreviewed, rolling out r92847.
227         http://trac.webkit.org/changeset/92847
228         https://bugs.webkit.org/show_bug.cgi?id=65986
229
230         Broke check-webkit-style
231
232         * Scripts/webkitpy/style/checker.py:
233         * Scripts/webkitpy/style/checker_unittest.py:
234
235 2011-08-11  Dimitri Glazkov  <dglazkov@chromium.org>
236
237         Unreviewed, rolling out r92847.
238         http://trac.webkit.org/changeset/92847
239         https://bugs.webkit.org/show_bug.cgi?id=65986
240
241         Broke check-webkit-style
242
243         * Scripts/webkitpy/style/checker.py:
244         * Scripts/webkitpy/style/checker_unittest.py:
245
246 2011-08-11  Pierre Rossi  <pierre.rossi@gmail.com>
247
248         [Qt] autotests shouldn't require config.h in the style check
249         https://bugs.webkit.org/show_bug.cgi?id=65986
250
251         Remove the include rules from the style-check for Qt tests.
252         Since for the most part it's just a matter of using the
253         API provided by QtWebKit, enforcing the WebCore set of rules
254         only seems to make sense in the case of MIMESniffing.
255
256         Reviewed by Benjamin Poulain.
257
258         * Scripts/webkitpy/style/checker.py:
259         * Scripts/webkitpy/style/checker_unittest.py:
260
261 2011-08-11  Zsolt Fehér  <h490267@stud.u-szeged.hu>
262
263         [WK2] Implement waitForPolicyDelegate
264         https://bugs.webkit.org/show_bug.cgi?id=42330
265
266         Reviewed by Csaba Osztrogonác.
267
268         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
269         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
270         (WTR::navigationTypeToString):
271         (WTR::InjectedBundlePage::decidePolicyForNavigationAction):
272         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
273         (WTR::LayoutTestController::LayoutTestController):
274         (WTR::LayoutTestController::setCustomPolicyDelegate):
275         (WTR::LayoutTestController::waitForPolicyDelegate):
276         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
277         (WTR::LayoutTestController::isPolicyDelegateEnabled):
278         (WTR::LayoutTestController::isPolicyDelegatePermissive):
279
280 2011-08-11  Adam Barth  <abarth@webkit.org>
281
282         rebaseline-expectations should use the release bots, not the debug
283         bots.  The debug bots don't all run all the tests.  Also, delay
284         updating test_expectations a long as possible to avoid clearing out
285         lines that we'll need for other ports.
286
287         * Scripts/webkitpy/layout_tests/port/builders.py:
288         * Scripts/webkitpy/tool/commands/rebaseline.py:
289         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
290
291 2011-08-10  Ryosuke Niwa  <rniwa@webkit.org>
292
293         r92792 inadvertently removed PlatformSpecificScheduler for all Snow Leopard bots.
294         Revert it partially so that the remaining Snow Leopard bots can fulfill their duty.
295
296         * BuildSlaveSupport/build.webkit.org-config/config.json:
297
298 2011-08-10  Eric Seidel  <eric@webkit.org>
299
300         new-run-webkit-test's WinPort has no fallback logic
301         https://bugs.webkit.org/show_bug.cgi?id=64486
302
303         Unreviewed.
304
305         Turns out that we've long had a subtle bug whereby if you didn't
306         already have DumpRenderTree built, we might end up running unsupported tests
307         because the DumpRenderTree --print-supported-features check would silently fail.
308
309         When I moved the call from using popen to using Executive.run_command, this silent
310         failure became noisy.  (Breaking some bots.)
311
312         This re-orders our steps so that we always build DRT before collecting expectations,
313         as well as adds a warning message when DumpRenderTree can't be found.
314
315         * Scripts/webkitpy/layout_tests/port/webkit.py:
316         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
317
318 2011-08-10  Eric Seidel  <eric@webkit.org>
319
320         new-run-webkit-test's WinPort has no fallback logic
321         https://bugs.webkit.org/show_bug.cgi?id=64486
322
323         Unreviewed.  Fix test-webkitpy under Python 2.5.
324
325         * Scripts/webkitpy/layout_tests/port/win.py:
326
327 2011-08-10  Benjamin Poulain  <benjamin@webkit.org>
328
329         [Qt][WK2] MiniBrowser is firing twice the QDesktopWebView::mousePressEvent method
330         https://bugs.webkit.org/show_bug.cgi?id=65875
331
332         Reviewed by Kenneth Rohde Christiansen.
333
334         For some reason, QWidget send fake mouse event on some platform, under certains conditions.
335
336         This causes mouse events to be sent twice in some cases. We work around the problem
337         by defining the first touch point to be a primary touch point (case which hit some
338         weird condition in QWidget, skipping the generation of fake mouse events).
339
340         * MiniBrowser/qt/MiniBrowserApplication.cpp:
341         (MiniBrowserApplication::notify):
342
343 2011-08-10  David Hyatt  <hyatt@apple.com>
344
345         https://bugs.webkit.org/show_bug.cgi?id=66004
346
347         Remove the regions and exclusions bots and also the flags in build-webkit, since I'm just turning
348         the code on by default.
349
350         Reviewed by Adam Roben.
351
352         * BuildSlaveSupport/build.webkit.org-config/config.json:
353         * Scripts/build-webkit:
354
355 2011-08-10  Oliver Hunt  <oliver@apple.com>
356
357         JSEvaluteScript does not return the correct object when given JSONP data
358         https://bugs.webkit.org/show_bug.cgi?id=66003
359
360         Reviewed by Gavin Barraclough.
361
362         Add a test case.
363
364         * DumpRenderTree/mac/DumpRenderTree.mm:
365         (testStringByEvaluatingJavaScriptFromString):
366
367 2011-08-10  Eric Seidel  <eric@webkit.org>
368
369         WIN: NRWT runs compositing tests on configurations that don't support compositing
370         https://bugs.webkit.org/show_bug.cgi?id=64472
371
372         Reviewed by Adam Barth.
373
374         I think the code was just wrong.  It was never splitting the string
375         into list pieces before.
376         I've now tested the supported_features code and theoretically it
377         should now work with Windows DRT.
378
379         I also made the list-lookup functions always return lists, to make
380         it possible to clean up the list-transform code in the future
381         (I opted not to do that cleanup in this patch, but at least now
382         both types of feature lookup functions match return types.)
383
384         * Scripts/webkitpy/layout_tests/port/webkit.py:
385         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
386
387 2011-08-10  Eric Seidel  <eric@webkit.org>
388
389         Clean up ChromiumDriver a little
390         https://bugs.webkit.org/show_bug.cgi?id=65995
391
392         Unreviewed.
393
394         Fix NRWT for chromium.  Silly python.
395
396         * Scripts/webkitpy/layout_tests/port/chromium.py:
397
398 2011-08-10  Eric Seidel  <eric@webkit.org>
399
400         Clean up ChromiumDriver a little
401         https://bugs.webkit.org/show_bug.cgi?id=65995
402
403         Reviewed by Adam Barth.
404
405         - We no longer support threading, so remove threading specific code.
406         - Add FIXMEs about using ServerProcess instead.
407         - Condense option-mapping if-cascade into a for loop.
408         - Unindent long if blocks by using early return.
409         - Unwrap lines which are needlessly wrapped.
410
411         There should be no functional changes here, just code cleanup/dead-code removal.
412
413         * Scripts/webkitpy/layout_tests/port/chromium.py:
414
415 2011-08-10  Dimitri Glazkov  <dglazkov@chromium.org>
416
417         Unmuddle construction options for TestConfiguration.
418         https://bugs.webkit.org/show_bug.cgi?id=65988
419
420         The TestConfiguration instance is now constructed only using actual values.
421         Port-based construction is split into a "from_port" class method.
422
423         Reviewed by Eric Seidel.
424
425         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added TestConfiguration.from_port
426         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Adjusted code to accommodate the change.
427         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Ditto.
428         * Scripts/webkitpy/layout_tests/port/base.py: Ditto.
429         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: Ditto.
430
431 2011-08-10  Eric Seidel  <eric@webkit.org>
432
433         Leaks bot shows python logging prefixes as part of summary display
434         https://bugs.webkit.org/show_bug.cgi?id=65931
435
436         Reviewed by Adam Barth.
437
438         The big part of this change is adding a unittesting system for
439         master.cfg, so mere-mortals can make changes to build.webkit.org
440         with greatly-reduced risk of breaking the world.
441
442         Turns out that in typing up my trival logging fix I made 2 typos,
443         so hopefully the hour spent getting master.cfg to load was worth it. :)
444
445         I also tried to make mastercfg_unittest run as part of test-webkitpy
446         but right now test-webkitpy requires that unittest files be located in
447         a module-name-compatible directory.  'build.webkit.org-config' is not
448         a valid module name.
449
450         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
451          - Fixed _parseNewRunWebKitTestsOutput to strip NRWT's python formatter prefix
452            by using a new _strip_python_logging_prefix function.
453          - Also fixed this function to file to work with a more modern
454            simplejson version (which returns unicode objects instead of str objects)
455            and filed a related bug with buildbot.net due to their handling of unicode builder names.
456         * BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py: Added.
457          - BuildBotConfigLoader is a huge pile of code just to get master.cfg to load.
458          - The actual tiny unittest of the new _parseNewRunWebKitTestsOutput logic.
459         * Scripts/webkitpy/thirdparty/__init__.py:
460          - Add autoinstallation of buildbot (and jinja2) for use by mastercfg_unittest.py
461
462 2011-08-09  Dimitri Glazkov  <dglazkov@chromium.org>
463
464         garden-o-matic should extrapolate failures given a set of known results
465         https://bugs.webkit.org/show_bug.cgi?id=65882
466
467         Reviewed by Adam Barth.
468
469         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Spuriously covered my old sins.
470         * Scripts/webkitpy/layout_tests/port/builders.py: Augmented builders dictionary to contain build coverage specifiers,
471             adjusted all call sites.
472         * Scripts/webkitpy/tool/servers/gardeningserver.py: Added BuildCoverageExtrapolator, a thing that
473             makes wider net for specific failures.
474         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added unit tests.
475
476 2011-08-10  Adam Barth  <abarth@webkit.org>
477
478         The URL we were computing for retrieving baselines was missing a slash
479         character because our mocks had an extra slash character.  I've
480         corrected both the code and the mock.
481
482         * Scripts/webkitpy/tool/commands/rebaseline.py:
483         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
484         * Scripts/webkitpy/tool/mocktool.py:
485
486 2011-08-09  Adam Barth  <abarth@webkit.org>
487
488         Move the quote mark to the right place.
489
490         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
491
492 2011-08-09  Adam Barth  <abarth@webkit.org>
493
494         garden-o-matic should use huerstics to narrow regression range
495         https://bugs.webkit.org/show_bug.cgi?id=65948
496
497         Reviewed by Dimitri Glazkov.
498
499         When analyzing a regression range, if one of the commit messages
500         mentions the test in question, we now hueristically assume that
501         revision caused the failure.  (I'm shamelessly stealing this idea from
502         TestFailures.)
503
504         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
505
506 2011-07-13  Eric Seidel  <eric@webkit.org>
507
508         new-run-webkit-test's WinPort has no fallback logic
509         https://bugs.webkit.org/show_bug.cgi?id=64486
510
511         Reviewed by Adam Roben.
512
513         I've tried to write a patch for bug 64439 twice now, and both times
514         I've ended up re-writing half the port system.  So I'm breaking
515         things up into smaller pieces, this being the first.
516
517         WinPort still does not have any port_name parsing, so when instantiated
518         with the name "win-xp" (i.e. by the rebaseline server) it will just behave as the 'win' port.
519         I'll fix this in a second pass when I standardize port_name parsing for all webkit ports.
520
521         Otherwise this should "just work" for windows.  I've not been able to test the
522         version detection on my mac, but the unit tests show the code behaving as designed.
523
524         * Scripts/webkitpy/layout_tests/port/win.py:
525         * Scripts/webkitpy/layout_tests/port/win_unittest.py: Added.
526
527 2011-08-09  Adam Barth  <abarth@webkit.org>
528
529         Teach build.webkit.org's garden-o-matic how to talk to the local server
530         https://bugs.webkit.org/show_bug.cgi?id=65940
531
532         Reviewed by Dimitri Glazkov.
533
534         After this patch, the version of garden-o-matic on build.webkit.org is
535         fully functional.  It can interact with the local server via CORS.
536
537         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
538         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
539         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
540         * Scripts/webkitpy/tool/commands/gardenomatic.py:
541         * Scripts/webkitpy/tool/servers/gardeningserver.py:
542
543 2011-08-09  Adam Barth  <abarth@webkit.org>
544
545         Enable CORS for garden-o-matic
546         https://bugs.webkit.org/show_bug.cgi?id=65936
547
548         Reviewed by Dimitri Glazkov.
549
550         This patch will allow the hosted instance of garden-o-matic to
551         communicate with the local server.
552
553         * Scripts/webkitpy/tool/servers/gardeningserver.py:
554         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
555
556 2011-08-09  Adam Barth  <abarth@webkit.org>
557
558         Upgrade jquery in TestFailures
559         https://bugs.webkit.org/show_bug.cgi?id=65935
560
561         Reviewed by Dimitri Glazkov.
562
563         Upgrade to a newer version of jQuery with better support for CORS and CSP.
564
565         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
566         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
567         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
568
569 2011-08-09  Sheriff Bot  <webkit.review.bot@gmail.com>
570
571         Unreviewed, rolling out r92683.
572         http://trac.webkit.org/changeset/92683
573         https://bugs.webkit.org/show_bug.cgi?id=65921
574
575          "Breaks Qt --minimal build. Revert pending update of the Qt
576         minimal configuration" (Requested by simathur on #webkit).
577
578         * QtTestBrowser/cookiejar.cpp:
579         (TestBrowserCookieJar::TestBrowserCookieJar):
580         * QtTestBrowser/launcherwindow.cpp:
581         (LauncherWindow::createChrome):
582         (LauncherWindow::setDiskCache):
583         * QtTestBrowser/main.cpp:
584         (LauncherApplication::handleUserOptions):
585         * QtTestBrowser/webpage.cpp:
586         (WebPage::openUrlInDefaultBrowser):
587
588 2011-08-09  Siddharth Mathur  <siddharth.mathur@nokia.com>
589
590         [Qt] Simplify code by removing QT_NO_DESKTOPSERVICES and QT_NO_NETWORKDISKCACHE
591         https://bugs.webkit.org/show_bug.cgi?id=65880
592
593         Reviewed by Andreas Kling.
594
595         QDesktopServices and QNetworkDiskCache have been around for a while now. Kill flags which 
596         check for their feature flags. Assume that any reasonable Qt build provides them. 
597         * QtTestBrowser/cookiejar.cpp:
598         (TestBrowserCookieJar::TestBrowserCookieJar):
599         * QtTestBrowser/launcherwindow.cpp:
600         (LauncherWindow::createChrome):
601         (LauncherWindow::setDiskCache):
602         * QtTestBrowser/main.cpp:
603         (LauncherApplication::handleUserOptions):
604         * QtTestBrowser/webpage.cpp:
605         (WebPage::openUrlInDefaultBrowser):
606
607 2011-08-09  Adam Barth  <abarth@webkit.org>
608
609         BaselineOptimizer created the wrong baseline for fast/js/regexp-overflow.html
610         https://bugs.webkit.org/show_bug.cgi?id=65891
611
612         Reviewed by Eric Seidel.
613
614         The problem was that platform/chromium contained a bogus expectation
615         file that needed to be removed, but by the time we got around to
616         removing it, we'd already moved the correct baseline into its place.
617         After this patch, we copy the gold results into memory before
618         reshuffling things on disk (and we delete bad things before adding good
619         things).
620
621         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
622         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
623
624 2011-08-08  Adam Barth  <abarth@webkit.org>
625
626         Chromium Windows bots can't figure out what SVN revision they're running
627         https://bugs.webkit.org/show_bug.cgi?id=65893
628
629         Reviewed by Eric Seidel.
630
631         The comment in _engage_awesome_windows_hacks explains why we're making
632         this change.  It's ugly and rediculous, but this approach seems better
633         than using shell=True when calling popen.
634
635         * Scripts/webkitpy/common/checkout/scm/svn.py:
636         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
637
638 2011-08-08  Adam Barth  <abarth@webkit.org>
639
640         Remove deduplicate-tests
641         https://bugs.webkit.org/show_bug.cgi?id=65886
642
643         Reviewed by Tony Chang.
644
645         This script is redundant with "webkit-patch bulk-optimize-baselines",
646         which can find many more duplicates than deduplicate-tests because
647         bulk-optimize-baselines can move results around whereas
648         deduplicate-tests can only remove results.
649
650         * Scripts/deduplicate-tests: Removed.
651         * Scripts/webkitpy/to_be_moved/deduplicate_tests.py: Removed.
652         * Scripts/webkitpy/to_be_moved/deduplicate_tests_unittest.py: Removed.
653
654 2011-08-08  Adam Barth  <abarth@webkit.org>
655
656         scm.delete should delete empty parent directories as well
657         https://bugs.webkit.org/show_bug.cgi?id=65878
658
659         Reviewed by Eric Seidel.
660
661         This behavior makes SVN match GIT (which has no concept of empty
662         directories).  This bug comes up a lot when optimizing baselines, which
663         often create empty directories when we're able to optimize everything
664         out of a given folder.
665
666         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
667         * Scripts/webkitpy/common/checkout/scm/svn.py:
668
669 2011-08-08  Lucas Forschler  <lforschler@apple.com>
670
671         Add triggers for Lion WK2 test bots.
672
673         Reviewed by Stephanie Lewis.
674
675         * BuildSlaveSupport/build.webkit.org-config/config.json:
676
677 2011-08-08  Adam Barth  <abarth@webkit.org>
678
679         bulk-optimize-baselines should take directories on the command line
680         https://bugs.webkit.org/show_bug.cgi?id=65879
681
682         Reviewed by Dimitri Glazkov.
683
684         Passing lists of tests via stdin is a pain.  It's easier if the command
685         takes a list of tests on the command line, like run-webkit-tests.
686
687         * Scripts/webkitpy/tool/commands/rebaseline.py:
688
689 2011-08-08  Adam Barth  <abarth@webkit.org>
690
691         Deprecate rebaseline-chromium-webkit-tests
692         https://bugs.webkit.org/show_bug.cgi?id=65873
693
694         Reviewed by Dimitri Glazkov.
695
696         This script never worked correctly and now is redundant with more
697         modern infrastructure for rebaselining tests.  This patch deprecates
698         the script in order to flush out any issues with the new tools.  Once
699         folks are happy with the new tools, we'll delete this script.
700
701         * Scripts/rebaseline-chromium-webkit-tests:
702
703 2011-08-08  Dimitri Glazkov  <dglazkov@chromium.org>
704
705         garden-o-matic should only touch expectations that need updating.
706         https://bugs.webkit.org/show_bug.cgi?id=65876
707
708         Reviewed by Adam Barth.
709
710         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added reconstitute_only_these to TestExpectationSerializer.list_to_string,
711             so that a caller could influence which lines to serialize from values.
712         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests.
713         * Scripts/webkitpy/tool/servers/gardeningserver.py: Started using reconstitute_only_these.
714         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added tests.
715
716 2011-08-08  Dimitri Glazkov  <dglazkov@chromium.org>
717
718         TestExpectationsEditor should return a list of updated expectations.
719         https://bugs.webkit.org/show_bug.cgi?id=65872
720
721         Reviewed by Adam Barth.
722
723         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Added a list to store lines that are being updated.
724         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Changed tests to account for updated lines.
725
726 2011-08-08  Jochen Eisinger  <jochen@chromium.org>
727
728         Fix SCM webkitpy unit test failures
729         https://bugs.webkit.org/show_bug.cgi?id=65823
730
731         Reviewed by Eric Seidel.
732
733         * Scripts/webkitpy/common/checkout/scm/git.py:
734         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
735
736 2011-08-08  Eric Seidel  <eric@webkit.org>
737
738         Move the --leaks bot back over to NRWT now that I'm back
739         from vacation and able to debug it.
740
741         Unreviewed.
742
743         * Scripts/run-webkit-tests:
744         (usingWebKit2):
745         (useNewRunWebKitTests):
746
747 2011-08-05  Dimitri Glazkov  <dglazkov@chromium.org>
748
749         Wire up updating expectations in garden-o-matic.
750         https://bugs.webkit.org/show_bug.cgi?id=65794
751
752         Reviewed by Adam Barth.
753
754         * Scripts/webkitpy/layout_tests/port/factory.py: Moved _port_for_builder from rebaseline.py to here, rename to get_from_builder_name.
755         * Scripts/webkitpy/tool/commands/rebaseline.py: Adapted callsites to new location.
756         * Scripts/webkitpy/tool/servers/gardeningserver.py: Added GardeningExpectationsUpdater, which knows how to update expectations.
757         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added loads of tests.
758
759 2011-08-08  Jochen Eisinger  <jochen@chromium.org>
760
761         webkit-patch doesn't get along with renamed files
762         https://bugs.webkit.org/show_bug.cgi?id=48075
763
764         Possibly a bit heavy handed - I removed all instances of -C and
765         changed every instance of -M with '--no-renames' in git.py. This
766         forces git to not try to tell us about renames at all, which is
767         ultimately the behaviour we want. The old file is shown deleted,
768         then the new file is shown added, followed by any changes that
769         occurred. Also gets rid of the problem where deleting one file
770         and adding another file which has similar content would
771         unexpectedly show up as a rename, and fall out of a diff.
772
773         Based on a patch by Wyatt Carss.
774
775         Reviewed by Eric Seidel.
776
777         * Scripts/webkitpy/common/checkout/scm/git.py:
778
779 2011-08-07  Sam White  <samuel.white@rochester.edu>
780
781         Add the ability to search the AccessibilityObject cache
782         https://bugs.webkit.org/show_bug.cgi?id=64994
783         
784         Added testing support for AccessibilityObject cache searching. Currently,
785         only the mac platform is full supported and has had API exposed. Other
786         platforms have only have this new functionality stubbed. Full implementation
787         is a job suited only for an accessibility expert of each respective platform.
788
789         Reviewed by Chris Fleizach.
790
791         * DumpRenderTree/AccessibilityUIElement.cpp:
792         (uiElementForSearchPredicateCallback):
793         (AccessibilityUIElement::getJSClass):
794         * DumpRenderTree/AccessibilityUIElement.h:
795         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
796         (AccessibilityUIElement::uiElementForSearchPredicate):
797         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
798         (AccessibilityUIElement::uiElementForSearchPredicate):
799         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
800         (AccessibilityUIElement::uiElementForSearchPredicate):
801
802 2011-08-06  Adam Barth  <abarth@webkit.org>
803
804         Remove dead code in webkitpy
805         https://bugs.webkit.org/show_bug.cgi?id=65818
806
807         Reviewed by Dimitri Glazkov.
808
809         This code was added to support the rebaseline2 command six months ago,
810         but the rebaseline2 command was never finished and no one else has used
811         this code since it was landed.  This patch removes the dead code.  If
812         we revive rebaseline2, then we can revert this patch and recover the
813         code.  Until then, this code is just dead weight.
814
815         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
816         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
817         * Scripts/webkitpy/common/net/testoutput.py: Removed.
818         * Scripts/webkitpy/common/net/testoutput_unittest.py: Removed.
819         * Scripts/webkitpy/common/net/testoutputset.py: Removed.
820         * Scripts/webkitpy/common/net/testoutputset_unittest.py: Removed.
821         * Scripts/webkitpy/common/system/directoryfileset.py: Removed.
822         * Scripts/webkitpy/common/system/directoryfileset_unittest.py: Removed.
823         * Scripts/webkitpy/layout_tests/port/base.py:
824         * Scripts/webkitpy/layout_tests/port/chromium.py:
825         * Scripts/webkitpy/layout_tests/port/webkit.py:
826
827 2011-08-06  Adam Barth  <abarth@webkit.org>
828
829         webkit-patch optimize-baselines can't handle promotions to base results directory
830         https://bugs.webkit.org/show_bug.cgi?id=65819
831
832         Reviewed by Dimitri Glazkov.
833
834         This bug came up when testing the optimizer on some editing baselines.
835
836         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
837         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
838
839 2011-08-06  Adam Barth  <abarth@webkit.org>
840
841         Add a small command to optimize baselines in bulk.
842
843         * Scripts/webkitpy/tool/commands/rebaseline.py:
844
845 2011-08-05  Dimitri Glazkov  <dglazkov@chromium.org>
846
847         Fix unit test breakage by plumbing convert_404_to_None in MockWeb.
848
849         * Scripts/webkitpy/tool/mocktool.py: Added missing parameter to MockWeb.get_binary
850
851 2011-08-05  Dimitri Glazkov  <dglazkov@chromium.org>
852
853         Garden-o-matic updateexpectations needs a unit test.
854         https://bugs.webkit.org/show_bug.cgi?id=65780
855
856         Reviewed by Adam Barth.
857
858         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py: Added unit test.
859
860 2011-08-05  Anders Carlsson  <andersca@apple.com>
861
862         Future proof an Objective-C test
863         https://bugs.webkit.org/show_bug.cgi?id=65788
864
865         Reviewed by Dan Bernstein.
866
867         Add isObjectInstanceOf to ObjCController.
868
869         * DumpRenderTree/mac/ObjCController.m:
870         (+[ObjCController isSelectorExcludedFromWebScript:]):
871         (+[ObjCController webScriptNameForSelector:]):
872         (-[ObjCController isObject:instanceOf:]):
873
874 2011-08-05  Adam Barth  <abarth@webkit.org>
875
876         trac.js needs unittests
877         https://bugs.webkit.org/show_bug.cgi?id=65673
878
879         Reviewed by Dimitri Glazkov.
880
881         This tests also have somewhat large fixtures.  I've trimmed the example
882         XML down a bunch while still hitting interesting cases in the parsing.
883
884         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
885         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/trac_unittests.js: Added.
886
887 2011-08-05  Adam Barth  <abarth@webkit.org>
888
889         Rebuild rebaseline-chromium-webkit-tests on top of modern infrastructure
890         https://bugs.webkit.org/show_bug.cgi?id=65759
891
892         Reviewed by Dimitri Glazkov.
893
894         rebaseline-chromium-webkit-tests doesn't really fit into webkitpy's
895         architecture, is poorly tested, and has a bunch of quirks.  This patch
896         rebuilds the core functionality of rebaseline-chromium-webkit-tests on
897         top of more modern infrastructure.  In the process, we get more code
898         re-use and better testing.
899
900         Once we're satisfied with this new implementation, we can delete the
901         old implementation.
902
903         * Scripts/webkitpy/common/checkout/baselineoptimizer.py:
904         * Scripts/webkitpy/common/net/web.py:
905         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
906         * Scripts/webkitpy/layout_tests/port/builders.py:
907         * Scripts/webkitpy/layout_tests/port/chromium.py:
908         * Scripts/webkitpy/layout_tests/port/factory.py:
909         * Scripts/webkitpy/layout_tests/port/webkit.py:
910         * Scripts/webkitpy/tool/commands/rebaseline.py:
911         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
912
913 2011-08-05  Anders Carlsson  <andersca@apple.com>
914
915         Remove PluginHalter
916         https://bugs.webkit.org/show_bug.cgi?id=65729
917
918         Reviewed by Darin Adler.
919
920         Remove call to set the allowed plug-in run time.
921
922         * DumpRenderTree/mac/DumpRenderTree.mm:
923         (resetDefaultsToConsistentValues):
924
925 2011-08-05  Ademar de Souza Reis Jr.  <ademar.reis@openbossa.org>
926
927         Unreviewed build fix for Qt-Minimal buildbot
928
929         Use ~/.QtTestBrowser for the cookiejar when QDesktopServices
930         is disabled.
931
932         Rubber-stamped by Holger Freyther.
933
934         * QtTestBrowser/cookiejar.cpp:
935         (TestBrowserCookieJar::TestBrowserCookieJar):
936
937 2011-08-05  Ademar de Souza Reis Jr  <ademar.reis@openbossa.org>
938
939         [Qt] QtTestBrowser: Add support for saving cookies on disk
940         https://bugs.webkit.org/show_bug.cgi?id=65328
941
942         Simple cookiejar that saves cookies on disk.
943
944         Added menu options to clear cookies (Edit->Clear cookies) and
945         disable disk cookies (Settings->[ ]Save Cookies on Disk).
946
947         Option -no-disk-cookies is also available.
948
949         Reviewed by Andreas Kling.
950
951         * QtTestBrowser/QtTestBrowser.pro:
952         * QtTestBrowser/cookiejar.cpp: Added.
953         * QtTestBrowser/cookiejar.h: Added.
954         * QtTestBrowser/launcherwindow.cpp:
955         * QtTestBrowser/launcherwindow.h:
956         * QtTestBrowser/main.cpp:
957
958 2011-08-04  Adam Barth  <abarth@webkit.org>
959
960         model.js needs some unit testing
961         https://bugs.webkit.org/show_bug.cgi?id=65730
962
963         Reviewed by Dimitri Glazkov.
964
965         These tests do not cover all the functions exported by the model
966         module.  We'll need to add the remaining ones in the future.
967
968         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model_unittests.js: Added.
969         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
970
971 2011-08-04  Adam Barth  <abarth@webkit.org>
972
973         Another speculative fix for the "svn revision on Windows" problem.
974
975         * Scripts/webkitpy/common/checkout/scm/git.py:
976
977 2011-08-04  Mark Rowe  <mrowe@apple.com>
978
979         Future-proof Xcode configuration settings.
980
981         * DumpRenderTree/mac/Configurations/Base.xcconfig:
982         * DumpRenderTree/mac/Configurations/CompilerVersion.xcconfig:
983         * DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
984         * MiniBrowser/Configurations/Base.xcconfig:
985         * MiniBrowser/Configurations/CompilerVersion.xcconfig:
986         * MiniBrowser/Configurations/DebugRelease.xcconfig:
987         * TestWebKitAPI/Configurations/Base.xcconfig:
988         * TestWebKitAPI/Configurations/CompilerVersion.xcconfig:
989         * TestWebKitAPI/Configurations/DebugRelease.xcconfig:
990         * WebKitTestRunner/Configurations/Base.xcconfig:
991         * WebKitTestRunner/Configurations/CompilerVersion.xcconfig:
992         * WebKitTestRunner/Configurations/DebugRelease.xcconfig:
993
994 2011-08-04  Adam Barth  <abarth@webkit.org>
995
996         Fancier debug logging.
997
998         * Scripts/webkitpy/layout_tests/controllers/manager.py:
999
1000 2011-08-04  David Levin  <levin@chromium.org>
1001
1002         [chromium] Fix assertion failure in ThreadIdentifierData::identifier in TestShell.
1003         https://bugs.webkit.org/show_bug.cgi?id=65685
1004
1005         Reviewed by Dmitry Titov.
1006
1007         * DumpRenderTree/chromium/WebViewHost.cpp:
1008         (WebViewHost::WebViewHost): Intialize WTF threading before
1009         anything in WTF is used. Usually this is done by WebViewImpl::WebViewImpl
1010         but in some cases code may use WTF without creating WebViewImpl and
1011         WTF code functionality may call WTF::currentThread, which requires
1012         this initialization now.
1013
1014 2011-08-04  Adam Barth  <abarth@webkit.org>
1015
1016         An educated guess at causing the Windows bots on build.webkit.org to
1017         find the SVN revision number.
1018
1019         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1020         * Scripts/webkitpy/layout_tests/port/base.py:
1021
1022 2011-08-04  Adam Barth  <abarth@webkit.org>
1023
1024         More logging to help debug the "no SVN rev on Chromium Windows bots"
1025         bug.  My current theory is that we're trying to detect the SCM system
1026         in the wrong directory.
1027
1028         * Scripts/webkitpy/layout_tests/port/base.py:
1029
1030 2011-08-04  Dimitri Glazkov  <dglazkov@chromium.org>
1031
1032         Speed up webkit-patch optimize-expectations significantly by not collecting test files.
1033         https://bugs.webkit.org/show_bug.cgi?id=65697
1034
1035         I neglected to realize that collecting test names is not necessary for this operation.
1036
1037         Reviewed by Adam Barth.
1038
1039         * Scripts/webkitpy/tool/commands/expectations.py: Changed to pass an empty list of test files.
1040
1041 2011-08-04  Adam Barth  <abarth@webkit.org>
1042
1043         checkout_unittest.js should have better coverage of checkout.js
1044         https://bugs.webkit.org/show_bug.cgi?id=65671
1045
1046         Reviewed by Dimitri Glazkov.
1047
1048         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
1049             - Remove unused functions.
1050         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js:
1051             - Test remaining public functions in the checkout module.
1052
1053 2011-08-04  Dmitry Lomov  <dslomov@google.com>
1054
1055         https://bugs.webkit.org/show_bug.cgi?id=65706
1056         Run run-unit-tests on release mode test bots.
1057         Reenabling after 61812 is fixed.
1058
1059         Reviewed by Adam Roben.
1060
1061         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1062
1063 2011-08-04  Adam Barth  <abarth@webkit.org>
1064
1065         builders.js needs unit tests
1066         https://bugs.webkit.org/show_bug.cgi?id=65670
1067
1068         Reviewed by Dimitri Glazkov.
1069
1070         This test requires somewhat large test fixtures, but using real
1071         examples seemed better than trimming them down too much.
1072
1073         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js:
1074         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders_unittests.js: Added.
1075         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
1076             - Note: trac_unittests.js doesn't exist (yet!).
1077
1078 2011-08-04  Alexis Menard  <alexis.menard@openbossa.org>
1079
1080         [Qt] Make navigation actions properly usable in QML.
1081         https://bugs.webkit.org/show_bug.cgi?id=65624
1082
1083         Fix compilation after the introduction of QWebNavigationController.
1084
1085         Reviewed by Benjamin Poulain.
1086
1087         * MiniBrowser/qt/BrowserView.cpp:
1088         (BrowserView::navigationAction):
1089
1090 2011-08-04  Adam Barth  <abarth@webkit.org>
1091
1092         Another subtle bug due to lack of testing in main.js.  :(
1093
1094         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1095
1096 2011-08-04  Adam Barth  <abarth@webkit.org>
1097
1098         Fix types.  Sadly, main.js has no test coverage.  (I need to think
1099         about how to test this part of the code.)
1100
1101         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1102
1103 2011-08-04  Adam Barth  <abarth@webkit.org>
1104
1105         bugzilla.js is missing unit tests
1106         https://bugs.webkit.org/show_bug.cgi?id=65656
1107
1108         Reviewed by Adam Roben.
1109
1110         Testing makes perfect.
1111
1112         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/bugzilla_unittests.js: Added.
1113         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net_unittests.js:
1114         (NetworkSimulator.prototype.runTest.get if):
1115         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
1116
1117 2011-08-04  Adam Barth  <abarth@webkit.org>
1118
1119         Add missing license blocks to garden-o-matic
1120         https://bugs.webkit.org/show_bug.cgi?id=65654
1121
1122         Reviewed by Dimitri Glazkov.
1123
1124         This patch is entirely boilerplate.
1125
1126         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
1127         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
1128         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js:
1129         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js:
1130         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
1131         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js:
1132         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
1133         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1134         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1135         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1136         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1137         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net.js:
1138         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net_unittests.js:
1139         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
1140         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
1141         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
1142         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1143         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
1144
1145 2011-08-04  Adam Barth  <abarth@webkit.org>
1146
1147         garden-o-matic needs a way to mock out the network
1148         https://bugs.webkit.org/show_bug.cgi?id=65653
1149
1150         Reviewed by Dimitri Glazkov.
1151
1152         This patch introduces the net module, which is a useful mock point for
1153         abstracting away the network.  Future patches will use this
1154         infrastructure to test some currently untested code.
1155
1156         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Bugzilla.js:
1157         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1158         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
1159         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js:
1160         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
1161         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1162         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
1163         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net.js: Added.
1164         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/net_unittests.js: Added.
1165         (NetworkSimulator):
1166         (NetworkSimulator.prototype.scheduleCallback):
1167         (NetworkSimulator.prototype.runTest):
1168         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
1169         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
1170         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
1171         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1172
1173 2011-08-03  Adam Barth  <abarth@webkit.org>
1174
1175         Use failureInfo more pervasively in garden-o-matic
1176         https://bugs.webkit.org/show_bug.cgi?id=65650
1177
1178         Reviewed by Dimitri Glazkov.
1179
1180         These three variable appears in concert together many places.  This
1181         patch unifies the remaining uses into being failureInfo objects, which
1182         saves some marshalling.
1183
1184         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1185         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
1186         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
1187         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1188         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
1189
1190 2011-08-03  Adam Barth  <abarth@webkit.org>
1191
1192         garden-o-matic should be able to update expecations from the details pane
1193         https://bugs.webkit.org/show_bug.cgi?id=65648
1194
1195         Reviewed by Dimitri Glazkov.
1196
1197         The new button queues up the updates, which get executed all at once.
1198
1199         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1200         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1201         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1202
1203 2011-08-03  Adam Barth  <abarth@webkit.org>
1204
1205         Add UI to garden-o-matic for updating expectations
1206         https://bugs.webkit.org/show_bug.cgi?id=65644
1207
1208         Reviewed by Dimitri Glazkov.
1209
1210         The patch plumbs the UI back to the gardening server, but the gardening
1211         server endpoint is just a stub at this point.
1212
1213         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
1214         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1215         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1216         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1217         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1218         * Scripts/webkitpy/tool/servers/reflectionhandler.py:
1219
1220 2011-08-03  Adam Barth  <abarth@webkit.org>
1221
1222         garden-o-matic details view should queue rebaselines for later execution
1223         https://bugs.webkit.org/show_bug.cgi?id=65636
1224
1225         Reviewed by Dimitri Glazkov.
1226
1227         This patch doesn't contain any confirmation UI, but we'll probably want
1228         that at some point.
1229
1230         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
1231         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1232         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1233         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1234
1235 2011-08-03  Kentaro Hara  <haraken@google.com>
1236
1237         Implement EventSender.scalePageBy()
1238         https://bugs.webkit.org/show_bug.cgi?id=58013
1239
1240         Reviewed by Darin Fisher.
1241
1242         Implemented EventSender.scalePageBy(f, x, y), which scales a page by a factor of f
1243         and then sets a scroll position to (x, y). Enabled the tests that had been waiting
1244         for the implementation of EventSender.scalePageBy(f, x, y).
1245
1246         Tests: compositing/scaling/tiled-layer-recursion.html
1247                fast/repaint/scale-page-shrink.html
1248                fast/dom/Element/scale-page-client-rects.html
1249                fast/dom/Range/scale-page-client-rects.html
1250                fast/events/scroll-in-scaled-page-with-overflow-hidden.html
1251                fast/dom/Element/scale-page-bounding-client-rect.html
1252                fast/dom/Range/scale-page-bounding-client-rect.html
1253
1254         * DumpRenderTree/chromium/EventSender.cpp:
1255         (EventSender::EventSender): Added bindings for scalePageBy().
1256         (EventSender::scalePageBy): A wrapper method for scalePage() in WebView.
1257         * DumpRenderTree/chromium/EventSender.h:
1258         * DumpRenderTree/chromium/TestShell.cpp:
1259         (TestShell::resetTestController): Resets the scale factor to 1.
1260
1261 2011-08-03  David Levin  <levin@chromium.org>
1262
1263         Rename WEBKIT_API to WEBKIT_EXPORT in check-webkit-style.
1264         https://bugs.webkit.org/show_bug.cgi?id=65652
1265
1266         Reviewed by Adam Barth.
1267
1268         * Scripts/webkitpy/style/checkers/cpp.py:
1269         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
1270
1271 2011-08-03  Filip Pizlo  <fpizlo@apple.com>
1272
1273         Adding Filip Pizlo to committer list.
1274
1275         Unreviewed.
1276
1277         * Scripts/webkitpy/common/config/committers.py:
1278
1279 2011-08-03  Dimitri Glazkov  <dglazkov@chromium.org>
1280
1281         Add "optimize-expectations" command to webkit-patch.
1282         https://bugs.webkit.org/show_bug.cgi?id=65633
1283
1284         For now, only operates on Chromium test_expectations.txt and does not report errors/warnings.
1285
1286         Reviewed by Adam Barth.
1287
1288         * Scripts/webkitpy/tool/commands/__init__.py: Added command.
1289         * Scripts/webkitpy/tool/commands/expectations.py: Added.
1290
1291 2011-08-03  Dimitri Glazkov  <dglazkov@chromium.org>
1292
1293         Allow multiple bug identifiers in test expectations.
1294         https://bugs.webkit.org/show_bug.cgi?id=65642
1295
1296         Reviewed by Adam Barth.
1297
1298         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Tweaked to allow multiple bug ids.
1299         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests.
1300
1301 2011-08-03  Dimitri Glazkov  <dglazkov@chromium.org>
1302
1303         Teach BuilderToPort to distinguish between Release and Debug builders.
1304         https://bugs.webkit.org/show_bug.cgi?id=65628
1305
1306         Reviewed by Adam Barth.
1307
1308         * Scripts/webkitpy/tool/commands/rebaseline.py: Added a smart options object and passed it to port factory.
1309         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py: Added tests.
1310
1311 2011-08-03  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
1312
1313         [Qt] Fix running qmltests for WK2 in the bot
1314         https://bugs.webkit.org/show_bug.cgi?id=65621
1315
1316         Reviewed by Csaba Osztrogonác.
1317
1318         The path for QML_IMPORT_PATH was assuming we were running the WK1 tests,
1319         but now there's the case for WK2 UIProcess tests as well.
1320
1321         * Scripts/run-qtwebkit-tests: add both possible paths to the QML_IMPORT_PATH.
1322
1323 2011-08-02  Dimitri Glazkov  <dglazkov@chromium.org>
1324
1325         Add a way to edit test expectations.
1326         https://bugs.webkit.org/show_bug.cgi?id=64922
1327
1328         Implements a TestExpectationsEditor, which provides a standard way to
1329         update and remove test expectations.
1330
1331         Reviewed by Adam Barth.
1332
1333         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed TestExpectationSerializer to skip lines that match no test configurations,
1334             added TestExpectationLine.is_flaky, added BugManager, a simple interface to be fleshed out later, and a TestExpectationsEditor, the big
1335             enchilada.
1336         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Loads of tests to capture desired behavior.
1337
1338 2011-08-03  Andreas Kling  <kling@webkit.org>
1339
1340         [Qt] MiniBrowser: Unbreak load progress indication.
1341         https://bugs.webkit.org/show_bug.cgi?id=65614
1342
1343         Reviewed by Kenneth Rohde Christiansen.
1344
1345         loadProgress() was renamed to loadProgressChanged().
1346
1347         * MiniBrowser/qt/BrowserWindow.cpp:
1348         (BrowserWindow::BrowserWindow):
1349         (BrowserWindow::onLoadProgressChanged):
1350         * MiniBrowser/qt/BrowserWindow.h:
1351
1352 2011-08-03  Martin Robinson  <mrobinson@igalia.com>
1353
1354         Build fix for WebKit2GTK+.
1355
1356         * WebKitTestRunner/GNUmakefile.am: Include libInternals as one of
1357         of the library dependencies of the injected bundle.
1358
1359 2011-08-03  Carlos Garcia Campos  <cgarcia@igalia.com>
1360
1361         [GTK] Install web and plugin processes in libexecdir instead of bindir
1362         https://bugs.webkit.org/show_bug.cgi?id=65600
1363
1364         Reviewed by Martin Robinson.
1365
1366         Set WEBKIT_EXEC_PATH environment variable to use web and plugin
1367         process from current directory instead of the intalled ones.
1368
1369         * MiniBrowser/gtk/GNUmakefile.am:
1370         * MiniBrowser/gtk/main.c:
1371         (main):
1372         * WebKitTestRunner/GNUmakefile.am:
1373         * WebKitTestRunner/gtk/main.cpp:
1374         (main):
1375
1376 2011-08-02  Anders Carlsson  <andersca@apple.com>
1377
1378         Add mac-lion platform to old-run-webkit-tests
1379         https://bugs.webkit.org/show_bug.cgi?id=65577
1380
1381         * Scripts/old-run-webkit-tests:
1382         * Scripts/webkitdirs.pm:
1383         (isLion):
1384
1385 2011-08-02  Alexandru Chiculita  <achicu@adobe.com>
1386
1387         Buildbot: The Properties object has no "hasProperty" method in buildbot 8.3
1388         https://bugs.webkit.org/show_bug.cgi?id=65554
1389
1390         Using "has_key" instead of "hasProperty" in order to support buildbot 8.3.
1391
1392         Reviewed by Adam Roben.
1393
1394         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1395
1396 2011-08-02  Jochen Eisinger  <jochen@chromium.org>
1397
1398         Implement WebPermissionClient::allowPlugins for layout tests
1399         https://bugs.webkit.org/show_bug.cgi?id=33991
1400
1401         Reviewed by Adam Barth.
1402
1403         * DumpRenderTree/chromium/LayoutTestController.cpp:
1404         (LayoutTestController::LayoutTestController):
1405         (LayoutTestController::setPluginsAllowed):
1406         * DumpRenderTree/chromium/LayoutTestController.h:
1407         * DumpRenderTree/chromium/WebPermissions.cpp:
1408         (WebPermissions::allowPlugins):
1409         (WebPermissions::setPluginsAllowed):
1410         (WebPermissions::reset):
1411         * DumpRenderTree/chromium/WebPermissions.h:
1412
1413 2011-08-02  Adam Barth  <abarth@webkit.org>
1414
1415         TestFailures unit tests leak a global variable
1416         https://bugs.webkit.org/show_bug.cgi?id=65500
1417
1418         Reviewed by Adam Roben.
1419
1420         qunit detects the leak if you check the "noglobals" box on run-unittests.html.
1421
1422         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Builder_unittests.js:
1423
1424 2011-08-02  Alexandru Chiculita  <achicu@adobe.com>
1425
1426         CSS Regions build bot should archive and upload output files
1427         https://bugs.webkit.org/show_bug.cgi?id=65200
1428
1429         Until now the buildbots would only upload the result archive if the 'trigger' parameters was specified. The
1430         upload was needed, because the testbots needed to download and use the archive. For CSS Regions we don't have
1431         a testbot yet, but we want to save the archive on the server for manual testing. Added new parameter
1432         in config.json, called "upload" that when set to "true" will force the upload to the buildmaster server.
1433
1434         The "features" argument is now taken into account when determining the name of the archive on the server. That
1435         was needed in order to avoid the CSS Regions Mac buildbot overwrite the SnowLeopard Mac build.
1436
1437         Reviewed by Adam Roben.
1438
1439         * BuildSlaveSupport/build.webkit.org-config/config.json:
1440         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
1441
1442 2011-08-02  Patrick Gansterer  <paroga@webkit.org>
1443
1444         Fix _win32_check_running_pid on 64bit windows python
1445         https://bugs.webkit.org/show_bug.cgi?id=64950
1446
1447         Reviewed by Adam Roben.
1448
1449         Use the correct types for the PROCESSENTRY32 struct.
1450         th32DefaultHeapID is declared as ULONG_PTR and not as DWORD.
1451         Using 32bit for pointers instead of 64bit results in an ERROR_BAD_LENGTH
1452         failure, when calling the Process32First function.
1453         This is covered by the executive unittest.
1454
1455         * Scripts/webkitpy/common/system/executive.py:
1456
1457 2011-08-01  Adam Barth  <abarth@webkit.org>
1458
1459         garden-o-matic should call optimize-baselines when rebaselining tests
1460         https://bugs.webkit.org/show_bug.cgi?id=65499
1461
1462         Reviewed by Dimitri Glazkov.
1463
1464         I took the opportunity to modernize this code to use some of our more
1465         powerful primitives from base.
1466
1467         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
1468         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js:
1469         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js:
1470         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1471         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
1472
1473 2011-08-01  Stephanie Lewis  <slewis@apple.com>
1474
1475         Finish reconfiguration started by Lucas Forschler
1476         Reconfigured slaves:
1477         Removed Leopard Intel Release (Build)
1478         Removed Leopard Intel Release (Tests)
1479         Remove apple-pixel-1
1480         Added Lion slaves from disabled Leopard slaves
1481
1482         Added builders:
1483         Lion Intel Release (Build)
1484         Lion Intel Debug (Build)
1485         Lion Intel Release (Tests)
1486         Lion Intel Release WK2 (Tests)
1487         Lion Intel Debug (Tests)
1488         Lion Intel Debug WK2 (Tests)
1489
1490         Added triggers:
1491         lion-intel-debug-tests
1492         lion-intel-debug-test-wk2
1493         lion-intel-release-tests
1494         lion-intel-release-tests-wk2
1495         
1496         Removed trigger:
1497         leopard-intel-release-tests
1498
1499         Added PlatformSpecificScheduler:
1500         mac-lion
1501
1502         Added "mac-lion" to build.py and build_unitttest.py
1503
1504         Reviewed by Mark Rowe.
1505
1506         * BuildSlaveSupport/build.webkit.org-config/config.json:
1507         * Scripts/webkitpy/common/config/build.py:
1508         * Scripts/webkitpy/common/config/build_unittest.py:
1509
1510  
1511 2011-08-01  Adam Barth  <abarth@webkit.org>
1512
1513         webkit-patch needs to be able to "optimize" the storage of baselines on disk
1514         https://bugs.webkit.org/show_bug.cgi?id=65418
1515
1516         Reviewed by Dimitri Glazkov.
1517
1518         If we're not careful when rebaselining tests, we can end up with lots
1519         of duplicate expected results files in the tree.  This patch adds a
1520         webkit-patch command that optimizes the storage of expected results on
1521         disk.
1522
1523         This command is similar to deduplicate-tests, except that it can move
1524         test results around rather than just remove duplicate results.
1525
1526         Unfortunately, this problem is very tricky because the baseline search
1527         structure is a hypergraph.  This patch include a huerstic optimizer
1528         that appears to work on a bunch of examples I've tried.  We'll likely
1529         need to refine it as gain experience.
1530
1531         * Scripts/webkitpy/common/system/filesystem.py:
1532         * Scripts/webkitpy/tool/commands/rebaseline.py:
1533
1534 2011-08-01  Dimitri Glazkov  <dglazkov@chromium.org>
1535
1536         Teach TestExpectationSerializer about parsed expectations.
1537         https://bugs.webkit.org/show_bug.cgi?id=65444
1538
1539         This is a bit largish in size, but the essentials are:
1540         * TestExpectationSerializer, given a parsed TestExpectationLine, can now spit out a pretty string.
1541         * We now store original content of the expectation line on TestExpectationLine.
1542         * Lots of tests.
1543
1544         Reviewed by Adam Barth.
1545
1546         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Fixed an issue where we didn't
1547             populate macros correctly, also fixed an issue where we incorrectly returned an empty list
1548             for all configurations.
1549         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added tests for those.
1550         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Removed vestiges of code long gone,
1551             refactored TestExpectationsSerializer to recognize and correctly serialize parsed TestExpectationLine instances,
1552             changed to store and use TestExpectationLine.original_string.
1553         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added tests.
1554
1555 2011-08-01  Adam Barth  <abarth@webkit.org>
1556
1557         Remove unused functionality from gardeningserver
1558         https://bugs.webkit.org/show_bug.cgi?id=65479
1559
1560         Reviewed by Dimitri Glazkov.
1561
1562         We now handle these operations directly from the front-end.  That means
1563         they work even without the gardening server running!
1564
1565         Also, add some missing files to the list of static files.
1566
1567         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1568         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
1569
1570 2011-08-01  Dimitri Glazkov  <dglazkov@chromium.org>
1571
1572         Introduce SpecifierSorter, a thing that knows how specifiers should be ordered.
1573         https://bugs.webkit.org/show_bug.cgi?id=65413
1574
1575         Reviewed by Adam Barth.
1576
1577         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added SpecifierSorter and converted
1578             existing code to use it.
1579         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added unit tests and cleaned up a bit.
1580
1581 2011-08-01  Adam Barth  <abarth@webkit.org>
1582
1583         Refactor bugzilla.js for use by garden-o-matic
1584         https://bugs.webkit.org/show_bug.cgi?id=65450
1585
1586         Reviewed by Dimitri Glazkov.
1587
1588         This patch refactors bugzilla.js to use the AsynchronousCache and
1589         updates the style to use a module instead of an object.  This patch
1590         then fixes all the existing code that uses this class to use the new
1591         API style.
1592
1593         This main benefit of this patch is we remove the tricky manual caching
1594         and this code is now available to use in garden-o-matic (since the
1595         dependency on Utilities.js is now gone).
1596
1597         I ran all the unit tests and poked around in TestFailures a bit to see
1598         that everything seemed to be working properly.
1599
1600         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Bugzilla.js:
1601         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js:
1602         (FailingTestsBugForm):
1603         (FailingTestsBugForm.prototype._createBugTitle):
1604         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js:
1605         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm.js:
1606         (FlakyTestBugForm):
1607         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm_unittests.js:
1608         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NewBugForm.js:
1609         (NewBugForm):
1610         (NewBugForm.prototype.domElement):
1611         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/NewBugForm_unittests.js:
1612         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm.js:
1613         (TestRelatedBugForm):
1614         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm_unittests.js:
1615         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
1616         (ViewController.prototype._displayBuilder.start):
1617         (ViewController.prototype._displayBuilder):
1618         (ViewController.prototype._domForAuxiliaryUIElements):
1619         (ViewController.prototype._domForNewAndExistingBugs.bugzilla.quickSearch):
1620         (ViewController.prototype._domForPossiblyFlakyTests.flakyList.appendChildren):
1621         (ViewController.prototype._domForPossiblyFlakyTests):
1622         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
1623         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1624         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
1625
1626 2011-08-01  Adam Barth  <abarth@webkit.org>
1627
1628         garden-o-matic shouldn't require local server to determine if compile failed
1629         https://bugs.webkit.org/show_bug.cgi?id=65446
1630
1631         Reviewed by Dimitri Glazkov.
1632
1633         This patch adds some basic bindings for the buildbot JSON API.  I
1634         wanted to use Buildbot.js an Builder.js from the original TestFailures,
1635         but they make a number of WebKit-specific assumptions.  We'll
1636         eventually need to reconcile these pieces of code.
1637
1638         I should really add tests for this code, but that will require me to
1639         build a better mock network.  I'll be happy to do that in a follow-up
1640         patch.
1641
1642         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/builders.js: Added.
1643             - It's very unclear to me whether didFail is correct.  Supposedly
1644               the first entry here is the exit code of the step, which is 0 on
1645               success.  I suspect we'll need to improve this function over time.
1646         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1647         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1648         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
1649
1650 2011-08-01  Adam Barth  <abarth@webkit.org>
1651
1652         garden-o-matic should show all failures, even ones that it can't find a regression range for
1653         https://bugs.webkit.org/show_bug.cgi?id=65447
1654
1655         Reviewed by Dimitri Glazkov.
1656
1657         Previously, we just wouldn't display test failures that we couldn't
1658         identify a regression range for (or whose regression range had fallen
1659         off the page).  This patch just displays them at the end of the list,
1660         which is slightly (but not much) better.
1661
1662         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1663
1664 2011-08-01  Jeff Miller  <jeffm@apple.com>
1665
1666         Windows build fix after r92096.
1667
1668         * TestWebKitAPI/win/InjectedBundleControllerWin.cpp: Use #include, not #import, and don't include Foundation.h.
1669         * TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj: Add InjectedBundleControllerWin.cpp.
1670
1671 2011-07-31  Anders Carlsson  <andersca@apple.com>
1672
1673         DownloadDecideDestinationCrash WebKit2 API tests crashes
1674         https://bugs.webkit.org/show_bug.cgi?id=65442
1675
1676         Reviewed by Sam Weinig.
1677
1678         Copy the 18-characters.html file to the destination directory so we won't
1679         crash trying to make an URL for it.
1680
1681         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1682
1683 2011-07-31  Anders Carlsson  <andersca@apple.com>
1684
1685         Fix SpacebarScrolling WebKit2 API test
1686         https://bugs.webkit.org/show_bug.cgi?id=65431
1687
1688         Reviewed by Sam Weinig.
1689
1690         Add a way to do platform-specific bundle initialization, and use it to disable smooth
1691         scrolling. Also make it possible for a test to load the injected bundle without actually running
1692         a test from it.
1693
1694         * TestWebKitAPI/InjectedBundleController.cpp:
1695         (TestWebKitAPI::InjectedBundleController::initialize):
1696         * TestWebKitAPI/InjectedBundleController.h:
1697         * TestWebKitAPI/PlatformUtilities.cpp:
1698         (TestWebKitAPI::Util::createContextWithInjectedBundle):
1699         * TestWebKitAPI/PlatformUtilities.h:
1700         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
1701         * TestWebKitAPI/Tests/WebKit2/SpacebarScrolling.cpp:
1702         (TestWebKitAPI::TEST):
1703         * TestWebKitAPI/mac/InjectedBundleControllerMac.mm: Added.
1704         (TestWebKitAPI::InjectedBundleController::platformInitialize):
1705         * TestWebKitAPI/win/InjectedBundleControllerWin.cpp: Added.
1706         (TestWebKitAPI::InjectedBundleController::platformInitialize):
1707         * TestWebKitAPI/win/TestWebKitAPI.vcproj:
1708
1709 2011-07-31  Dimitri Glazkov  <dglazkov@chromium.org>
1710
1711         Use set comparison to find duplicate or overlapping specifiers in test expectations.
1712         https://bugs.webkit.org/show_bug.cgi?id=65430
1713
1714         Reviewed by Adam Barth.
1715
1716         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed to use set comparisons, because length
1717             of a set is wrong.
1718         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py: Added some unit tests to actually catch
1719             regressions.
1720
1721 2011-07-31  Xan Lopez  <xlopez@igalia.com>
1722
1723         Group dependencies CFLAGS together.
1724
1725         * GNUmakefile.am: ditto.
1726
1727 2011-07-31  Daniel Bates  <dbates@webkit.org>
1728
1729         Attempt to fix the GTK build after <http://trac.webkit.org/changeset/92059>
1730         (https://bugs.webkit.org/show_bug.cgi?id=65419).
1731
1732         * GNUmakefile.am: Add $(CAIRO_CFLAGS) to libWebCoreInternals_la_CPPFLAGS.
1733
1734 2011-07-30  Patrick Gansterer  <paroga@webkit.org>
1735
1736         Fix ports_unittest.py on native windows
1737         https://bugs.webkit.org/show_bug.cgi?id=64959
1738
1739         Reviewed by Adam Barth.
1740
1741         Replace WebKitPort.script_path with WebKitPort.script_shell_command.
1742         The latter includes the interpreter executable required on native windows.
1743
1744         * Scripts/webkitpy/common/config/ports_unittest.py:
1745
1746 2011-07-30  Adam Barth  <abarth@webkit.org>
1747
1748         Properly escape regular expressions.  (Maybe these shouldn't be regular
1749         expressions?)
1750
1751         * Scripts/webkitpy/tool/commands/rebaseline.py:
1752
1753 2011-07-30  Adam Barth  <abarth@webkit.org>
1754
1755         Add support for Win7 to garden-o-matic.
1756
1757         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
1758         * Scripts/webkitpy/tool/commands/rebaseline.py:
1759
1760 2011-07-30  Adam Barth  <abarth@webkit.org>
1761
1762         garden-o-matic regression ranges expand after fast bots start passing
1763         https://bugs.webkit.org/show_bug.cgi?id=65406
1764
1765         Reviewed by Dimitri Glazkov.
1766
1767         Once the gardener has fixed a problem, the fast bots pick up the fix,
1768         which means we end up using the slow bots to compute the regression
1769         range.  Unfortunately, the slow bots give us large regression ranges,
1770         leading to this strange effect where the regression range quality
1771         degrades as fixes roll in.
1772
1773         This patch causes us to remember the previous regression ranges we've
1774         seen and to use that information in the future, as long as it's
1775         consistent with the new information we receive.
1776
1777         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1778
1779 2011-07-30  Adam Barth  <abarth@webkit.org>
1780
1781         garden-o-matic should have a Refresh button to update its display
1782         https://bugs.webkit.org/show_bug.cgi?id=65403
1783
1784         Reviewed by Dimitri Glazkov.
1785
1786         The refresh button should be faster and less disruptive than the browser's reload button.
1787
1788         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
1789         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
1790
1791 2011-07-30  Adam Barth  <abarth@webkit.org>
1792
1793         garden-o-matic should look for the "patch by" line in revisions ostensibly authored by the commit-queue
1794         https://bugs.webkit.org/show_bug.cgi?id=65231
1795
1796         Reviewed by Dimitri Glazkov.
1797
1798         The human readable name is easier to understand, especially when the patch is landed by the commit-queue.
1799
1800         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1801
1802 2011-07-30  Adam Barth  <abarth@webkit.org>
1803
1804         garden-o-matic should strikeout revisions that have been rolled out
1805         https://bugs.webkit.org/show_bug.cgi?id=65227
1806
1807         Reviewed by Dimitri Glazkov.
1808
1809         This helps gardeners keep track of which problems have been solved.
1810
1811         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
1812         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
1813         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
1814         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
1815
1816 2011-07-30  Adam Barth  <abarth@webkit.org>
1817
1818         Add missing files to the server.
1819
1820         * Scripts/webkitpy/tool/servers/gardeningserver.py:
1821
1822 2011-07-29  Adam Barth  <abarth@webkit.org>
1823
1824         Actually import the logging module!
1825
1826         * Scripts/webkitpy/common/checkout/scm/git.py:
1827
1828 2011-07-29  Adam Barth  <abarth@webkit.org>
1829
1830         More debug logging.  Maybe we're running the git version of this
1831         function somehow?  (That would explain why it doesn't work!)
1832
1833         * Scripts/webkitpy/common/checkout/scm/git.py:
1834
1835 2011-07-29  Adam Barth  <abarth@webkit.org>
1836
1837         Add some temporary debug logging messages to try to understand why the
1838         windows bots can't figure out their current SVN revision.
1839
1840         * Scripts/webkitpy/common/checkout/scm/svn.py:
1841
1842 2011-07-29  Adam Barth  <abarth@webkit.org>
1843
1844         Chromium try servers should use downstream test_expectations
1845         https://bugs.webkit.org/show_bug.cgi?id=65390
1846
1847         Reviewed by Dirk Pranke.
1848
1849         Otherwise the try servers will report bogus failures.
1850
1851         * Scripts/webkitpy/layout_tests/port/chromium.py:
1852
1853 2011-07-29  Dimitri Glazkov  <dglazkov@chromium.org>
1854
1855         Fix an error in macro collapsing, refactor code to be pretty.
1856         https://bugs.webkit.org/show_bug.cgi?id=65387
1857
1858         My set math was wrong.
1859
1860         Reviewed by Adam Barth.
1861
1862         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Moved collapsing into a classmethod, fixed the math to be correct.
1863         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added unit tests.
1864
1865 2011-07-27  Ryuan Choi  <ryuan.choi@samsung.com>
1866
1867         Style bot complains about EFL-specific forward declaration naming.
1868         https://bugs.webkit.org/show_bug.cgi?id=64680
1869
1870         Exclude EFL-specific forward declarations from indentifiers with underscores checks.
1871
1872         Reviewed by Adam Barth.
1873
1874         * Scripts/webkitpy/style/checkers/cpp.py:
1875         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
1876
1877 2011-07-27  Adam Roben  <aroben@apple.com>
1878
1879         Use wacky !svn/bc URLs consistently in Leaks Viewer
1880
1881         This ensures that any resources loaded via relative URLs will still get pulled from the same
1882         revision of Subversion.
1883
1884         Fixes <http://webkit.org/b/65267> Focus/exclude button glyphs don't load on Leaks Viewer
1885
1886         Reviewed by David Kilzer.
1887
1888         * BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/index.html: Use the
1889         !svn/bc-style URL for inspector.css, too, just like we do for all other resources we pull
1890         from Subversion.
1891
1892 2011-07-27  Raphael Kubo da Costa  <kubo@profusion.mobi>
1893
1894         [EFL] Drop WorkQueueItemEfl.h in favour of the general WorkQueueItem.h
1895         https://bugs.webkit.org/show_bug.cgi?id=65269
1896
1897         Reviewed by Kent Tamura.
1898
1899         WorkQueueItemEfl.h was being included by efl's
1900         LayoutTestControllerEfl.cpp, however the top-level
1901         LayoutTestController still included WorkQueueItem.h, causing a lot of
1902         type mismatches when running the WorkQueueItem code.
1903
1904         We can stick to the original WorkQueueItem header by making a few
1905         changes to EFL's LayoutTestController.
1906
1907         This should make tests such as
1908         fast/loader/non-deferred-substitute-load.html stop crashing.
1909
1910         * DumpRenderTree/efl/LayoutTestControllerEfl.cpp:
1911         (LayoutTestController::queueLoad):
1912         * DumpRenderTree/efl/WorkQueueItemEfl.cpp:
1913         (LoadItem::invoke):
1914         (LoadHTMLStringItem::invoke):
1915         (ScriptItem::invoke):
1916         * DumpRenderTree/efl/WorkQueueItemEfl.h: Removed.
1917
1918 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1919
1920         Re-add unit tests I accidentally zapped in r91878.
1921
1922         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Re-added unit tests.
1923
1924 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1925
1926         Replace SpecificityCalculator with TestConfiguration-driven logic.
1927         https://bugs.webkit.org/show_bug.cgi?id=65206
1928
1929         Use our newly-acquired ability to expand modifiers into a set of
1930         matching TestConfiguration instances to calculate specificity and
1931         determine whether expectation applies to a given test configuration.
1932
1933         Also, store bug modifier on TestExpectationsLine.
1934
1935         Reviewed by Adam Barth.
1936
1937         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Replaced the logic of calculating specificity with
1938             test configuration matching, removed a bunch of code.
1939         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Removed a bunch of tests that aren't useful anymore.
1940
1941
1942 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1943
1944         Allow ports to specify their own test expectation specifier macros.
1945         https://bugs.webkit.org/show_bug.cgi?id=65291
1946
1947         Reviewed by Adam Barth.
1948
1949         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed SpecificityCalculator to use port macros.
1950         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Ditto.
1951         * Scripts/webkitpy/layout_tests/port/base.py: Added configuration_specifier_macros to retrieve the macros.
1952         * Scripts/webkitpy/layout_tests/port/chromium.py: Overrode it to return Chromium's macros.
1953         * Scripts/webkitpy/layout_tests/port/test.py: Added test macros.
1954
1955 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1956
1957         Allow TestConfigurationConverter report conversion errors.
1958         https://bugs.webkit.org/show_bug.cgi?id=65287
1959
1960         Also, fix a few miscellaneous testing issues.
1961
1962         Reviewed by Adam Barth.
1963
1964         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added a way to report unknown modifier errors.
1965         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added a test for reporting errors.
1966         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: Fixed an error where an invalid TestConfiguration instance
1967             was created. 'default' is not a valid build type.
1968         * Scripts/webkitpy/layout_tests/port/test.py: Ditto. '' is not a valid version.
1969
1970 2011-07-27  Dimitri Glazkov  <dglazkov@chromium.org>
1971
1972         Teach all ports about their test configurations.
1973         https://bugs.webkit.org/show_bug.cgi?id=65278
1974
1975         Each Port object should return a valid set of all TestConfiguration instances,
1976         available for this port.
1977
1978         Reviewed by Philippe Normand.
1979
1980         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Removed TestConfiguration.all_test_configurations,
1981             since it is different for each port.
1982         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Removed related test.
1983         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Changed SpecificityCalculator to take a list of
1984             all test configurations as a param.
1985         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Adjusted arguments accordingly.
1986         * Scripts/webkitpy/layout_tests/port/base.py: Forced all ports to override generation of all test configurations.
1987         * Scripts/webkitpy/layout_tests/port/chromium.py: Implemented _generate_all_test_configurations for the port.
1988         * Scripts/webkitpy/layout_tests/port/gtk.py: Ditto.
1989         * Scripts/webkitpy/layout_tests/port/mac.py: Ditto.
1990         * Scripts/webkitpy/layout_tests/port/qt.py: Ditto.
1991         * Scripts/webkitpy/layout_tests/port/win.py:
1992         * Scripts/webkitpy/layout_tests/port/port_testcase.py: Added an extra test to ensure we're generating a good set.
1993         * Scripts/webkitpy/layout_tests/port/test.py: Implemented a fixed-in-time all_test_configurations for testing.
1994         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py: Added a mock all_test_configurations implementation.
1995         * Scripts/webkitpy/layout_tests/port/win_unittest.py: Added a unit test for the port.
1996
1997 2011-07-27  Benjamin Poulain  <benjamin@webkit.org>
1998
1999         [WK2][Qt] Move from QGraphicsView to Qt Scene Graph
2000         https://bugs.webkit.org/show_bug.cgi?id=62969
2001
2002         Reviewed by Andreas Kling.
2003
2004         Co-authored with Simon Hausmann  <simon.hausmann@nokia.com>
2005
2006         With the scene graph, the interface to have a window on screen is
2007         QSGCanvas. QGraphicsView is not used, and openGL is an enforced
2008         dependency.
2009
2010         Both the BrowserView and the WebKitTest runner are changed from
2011         QGraphicsView to QSGCanvas.
2012
2013         * MiniBrowser/qt/BrowserView.cpp:
2014         (BrowserView::BrowserView):
2015         (BrowserView::resizeEvent):
2016         (BrowserView::view):
2017         * MiniBrowser/qt/BrowserView.h:
2018         * MiniBrowser/qt/BrowserWindow.cpp:
2019         (BrowserWindow::BrowserWindow):
2020         (BrowserWindow::urlChanged):
2021         (BrowserWindow::~BrowserWindow):
2022         * MiniBrowser/qt/BrowserWindow.h:
2023         * MiniBrowser/qt/MiniBrowser.pro:
2024         * MiniBrowser/qt/main.cpp:
2025         (main):
2026         * WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro:
2027         * WebKitTestRunner/qt/PlatformWebViewQt.cpp:
2028         (WTR::WebView::WebView):
2029         (WTR::PlatformWebView::PlatformWebView):
2030         * WebKitTestRunner/qt/WebKitTestRunner.pro:
2031
2032 2011-07-27  Sheriff Bot  <webkit.review.bot@gmail.com>
2033
2034         Unreviewed, rolling out r91792.
2035         http://trac.webkit.org/changeset/91792
2036         https://bugs.webkit.org/show_bug.cgi?id=65242
2037
2038         It broke NRWT on Qt and GTK (Requested by Ossy_night on
2039         #webkit).
2040
2041         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2042         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py:
2043         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2044         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2045         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2046         * Scripts/webkitpy/layout_tests/port/test.py:
2047
2048 2011-07-26  Adam Barth  <abarth@webkit.org>
2049
2050         Switch to an older idiom that works on Python 2.5.
2051
2052         * Scripts/webkitpy/layout_tests/models/test_configuration.py:
2053
2054 2011-07-26  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2055
2056         Add an exception to coding style check for EFL port.
2057         https://bugs.webkit.org/show_bug.cgi?id=65162
2058
2059         Add an exception of whitespace/declaration for EFL port. Because,
2060         '*' operator is next to variable in EFL coding style.
2061
2062         Reviewed by Kent Tamura.
2063
2064         * Scripts/webkitpy/style/checker.py:
2065
2066 2011-07-26  Keith Kyzivat  <keith.kyzivat@nokia.com>
2067
2068         [Qt] Add option to turn on disk caching in QtTestBrowser
2069         https://bugs.webkit.org/show_bug.cgi?id=65007
2070
2071         Reviewed by Noam Rosenthal.
2072
2073         Add menu item and command line option to turn on disk caching in
2074         QtTestBrowser. Simple QNetworkDiskCache added to the page's
2075         QNetworkAccssManager.
2076
2077         * QtTestBrowser/launcherwindow.cpp:
2078         (LauncherWindow::initializeView):
2079         (LauncherWindow::createChrome):
2080         (LauncherWindow::setDiskCache):
2081         * QtTestBrowser/launcherwindow.h:
2082         (WindowOptions::WindowOptions):
2083         * QtTestBrowser/main.cpp:
2084         (LauncherApplication::handleUserOptions):
2085
2086 2011-07-26  Sadrul Habib Chowdhury  <sadrul@chromium.org>
2087
2088         Add support for download='filename' attribute in anchors.
2089         https://bugs.webkit.org/show_bug.cgi?id=64580
2090
2091         Reviewed by Adam Barth.
2092
2093         * DumpRenderTree/chromium/WebViewHost.cpp:
2094         (WebViewHost::loadURLExternally):
2095         * DumpRenderTree/chromium/WebViewHost.h:
2096
2097 2011-07-26  Dan Bernstein  <mitz@apple.com>
2098
2099         Rubber-stamped by Anders Carlsson.
2100
2101         Update DumpRenderTree and API tests for the pictograph font family preference.
2102
2103         * DumpRenderTree/mac/DumpRenderTree.mm:
2104         (resetDefaultsToConsistentValues):
2105         * DumpRenderTree/win/DumpRenderTree.cpp:
2106         (resetDefaultsToConsistentValues):
2107         * TestWebKitAPI/Tests/WebKit2/WKPreferences.cpp:
2108         (TestWebKitAPI::TEST):
2109         * WebKitTestRunner/TestController.cpp:
2110         (WTR::TestController::resetStateToConsistentValues):
2111
2112 2011-07-26  Dimitri Glazkov  <dglazkov@chromium.org>
2113
2114         Replace SpecificityCalculator with TestConfiguration-driven logic.
2115         https://bugs.webkit.org/show_bug.cgi?id=65206
2116
2117         Use our newly-acquired ability to expand modifiers into a set of
2118         matching TestConfiguration instances to calculate specificity and
2119         determine whether expectation applies to a given test configuration.
2120
2121         Also, store bug modifier on TestExpectationsLine.
2122
2123         Reviewed by Adam Barth.
2124
2125         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added a way to report unknown modifier errors.
2126         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added a test for reporting errors.
2127         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Replaced the logic of calculating specificity with
2128             test configuration matching, removed a bunch of code.
2129         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Removed a bunch of tests that aren't useful anymore.
2130         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py: Fixed an error where an invalid TestConfiguration instance
2131             was created. 'default' is not a valid build type.
2132         * Scripts/webkitpy/layout_tests/port/test.py: Ditto. '' is not a valid version.
2133
2134 2011-07-26  Adam Barth  <abarth@webkit.org>
2135
2136         Increase the information density in garden-o-matic
2137         https://bugs.webkit.org/show_bug.cgi?id=65141
2138
2139         Reviewed by Dimitri Glazkov.
2140
2141         This patch changes lets us show each revision on a single line,
2142         allowing use to show an entire day's worth of commits at once.
2143
2144         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
2145         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
2146         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
2147
2148 2011-07-26  Dimitri Glazkov  <dglazkov@chromium.org>
2149
2150         TestConfigurationConverter should not barf on invalid specifiers.
2151         https://bugs.webkit.org/show_bug.cgi?id=65194
2152
2153         Reviewed by Adam Barth.
2154
2155         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Refactored TestConfigurationConverter.to_config_set a bit,
2156             made it return empty set whenever an invalid specifier is encountered.
2157         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added corresponding tests.
2158
2159 2011-07-26  Dimitri Glazkov  <dglazkov@chromium.org>
2160
2161         Introduce a way to convert between specifier lists and TestConfiguration sets.
2162         https://bugs.webkit.org/show_bug.cgi?id=65094
2163
2164         Add TestConfigurationConverter, a facility to convert:
2165         1) a set of specifiers into a set of TestConfiguration instances;
2166         2) a set of TestConfiguration instances into a list of sets of specifiers.
2167
2168         Reviewed by Adam Barth.
2169
2170         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added TestConfigurationConverter.
2171         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added tests for it.
2172
2173 2011-07-26  Alexandru Chiculita  <achicu@adobe.com>
2174
2175         CSS Regions build bot should build with "--css-regions" and "--css-exclusion" flags
2176         https://bugs.webkit.org/show_bug.cgi?id=65171
2177
2178         Added "features" parameter in config.json for "builder" objects. The parameter is optional and must be an array. All the items
2179         of the array will be passed to the build-webkit script. The items will automatically be prefixed with "--".
2180
2181         Reviewed by Adam Roben.
2182
2183         * BuildSlaveSupport/build.webkit.org-config/config.json:
2184         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2185
2186 2011-07-26  Chris Rogers  <crogers@google.com>
2187
2188         Avoid outputting #EOF twice for web audio tests in chromium DRT
2189         https://bugs.webkit.org/show_bug.cgi?id=65201
2190
2191         Reviewed by Dirk Pranke.
2192
2193         * DumpRenderTree/chromium/TestShell.cpp:
2194         (TestShell::dump):
2195
2196 2011-07-26  Adam Barth  <abarth@webkit.org>
2197
2198         Windows bots on build.chromium.org can't figure out the head_svn_revision
2199         https://bugs.webkit.org/show_bug.cgi?id=65156
2200
2201         Reviewed by Tony Chang.
2202
2203         This patch is a bit of a guess as to the issue based on the
2204         investigation in https://bugs.webkit.org/show_bug.cgi?id=65148.  I'm
2205         not sure whether it will fix the issue, but it seems worth a try.
2206
2207         * Scripts/webkitpy/common/checkout/scm/svn.py:
2208
2209 2011-07-26  Carlos Garcia Campos  <cgarcia@igalia.com>
2210
2211         Unreviewed. Fix WebKitGtk build with GTK+ 3.x after r91707.
2212
2213         * DumpRenderTree/gtk/SelfScrollingWebKitWebView.cpp:
2214         (getPreferredSizeMethod):
2215
2216 2011-07-26  MORITA Hajime  <morrita@google.com>
2217
2218         [Chromium][DRT] MockSpellCheck.cpp should be clang clean
2219         https://bugs.webkit.org/show_bug.cgi?id=65166
2220
2221         Reviewed by Kent Tamura.
2222
2223         * DumpRenderTree/chromium/MockSpellCheck.cpp:
2224         (MockSpellCheck::spellCheckWord): -Wsign-compare warning.
2225
2226 2011-07-25  Patrick Gansterer  <paroga@webkit.org>
2227
2228         Ignore line endings in ChangeLogTests
2229         https://bugs.webkit.org/show_bug.cgi?id=64956
2230
2231         Reviewed by Adam Barth.
2232
2233         Split the lines of the ChangeLog text before comparing them.
2234         This fixes 3 tests on windows, since only the content importent.
2235
2236         * Scripts/webkitpy/common/checkout/changelog_unittest.py:
2237         * Scripts/webkitpy/tool/steps/preparechangelog_unittest.py:
2238
2239 2011-07-25  Adam Barth  <abarth@webkit.org>
2240
2241         Add some debug logging to NRWT to diagnose why the windows bots can't determine their SVN revision
2242         https://bugs.webkit.org/show_bug.cgi?id=65148
2243
2244         Reviewed by Tony Chang.
2245
2246         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2247
2248 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2249
2250         Tweak whitespace wrapping to improve readability.
2251         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2252
2253 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2254
2255         Minor tweak to wrap label elements as atomic units.
2256
2257         * TestResultServer/static-dashboards/dashboard_base.js:
2258         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2259
2260 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2261
2262         always show flakiness details on the builder page
2263         https://bugs.webkit.org/show_bug.cgi?id=65139
2264
2265         Reviewed by Tony Chang.
2266
2267         In order to simplify the UI, we hide the flakiness details by default,
2268         but there's no actual use-case for this page without the flakiness details.
2269         So get rid of that option and always show the flakiness details.
2270
2271         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2272
2273 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2274
2275         simplify the UI of the header above the test list table
2276         https://bugs.webkit.org/show_bug.cgi?id=65136
2277
2278         Reviewed by Adam Barth.
2279
2280         Use checkboxes instead of links since this is really about which tests to include.
2281         Ends up using less text and put the checkboxes to the right of the header instead of below it.
2282
2283         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2284         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2285
2286 2011-07-25  Chris Rogers  <crogers@google.com>
2287
2288         Update new-run-webkit-tests to accomodate binary audio data
2289         https://bugs.webkit.org/show_bug.cgi?id=65135
2290
2291         Reviewed by Dirk Pranke.
2292
2293         * Scripts/webkitpy/layout_tests/port/chromium.py:
2294
2295 2011-07-25  Chris Rogers  <crogers@google.com>
2296
2297         Update Chromium DRT to output binary (instead of base64-encoded) data for web audio testing
2298         https://bugs.webkit.org/show_bug.cgi?id=65039
2299
2300         Reviewed by Tony Chang.
2301
2302         * DumpRenderTree/chromium/LayoutTestController.cpp:
2303         (LayoutTestController::LayoutTestController):
2304         (LayoutTestController::setAudioData):
2305         * DumpRenderTree/chromium/LayoutTestController.h:
2306         (LayoutTestController::audioData):
2307         * DumpRenderTree/chromium/TestEventPrinter.cpp:
2308         (DRTPrinter::handleAudioHeader):
2309         (TestShellPrinter::handleAudioHeader):
2310         * DumpRenderTree/chromium/TestShell.cpp:
2311         (TestShell::dump):
2312
2313 2011-06-07  Martin Robinson  <mrobinson@igalia.com>
2314
2315         Reviewed by Gustavo Noronha Silva.
2316
2317         [GTK] Use WebCore scrollbars for the DRT instead of GtkScrolledWindow
2318         https://bugs.webkit.org/show_bug.cgi?id=62252
2319
2320         Add a new subclass of WebKitWebView that overrides the size request to
2321         return 1 pixel by 1 pixel. This ensure that it operates properly when
2322         not packed into a GtkScrolledWindow. Also construct the WebKitWebView
2323         with self-scrolling set to true.
2324
2325         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2326         (webInspectorInspectWebView): Use the new SelfScrollingWebKitWebview class here.
2327         (createWebView): Ditto.
2328         (main): Pack the widget into a GtkBox instead of a GtkScrolledWindow.
2329         * DumpRenderTree/gtk/SelfScrollingWebKitWebView.cpp: Added.
2330         (self_scrolling_webkit_web_view_class_init):
2331         (self_scrolling_webkit_web_view_init):
2332         (self_scrolling_webkit_web_view_new):
2333         (sizeRequestMethod):
2334         (getPreferredSizeMethod):
2335         * DumpRenderTree/gtk/SelfScrollingWebKitWebView.h: Added.
2336         * GNUmakefile.am: Added SelfScrollingWebKitWebView files to the source list.
2337
2338 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2339
2340         by default hide tests that have passes for all recorded runs
2341         https://bugs.webkit.org/show_bug.cgi?id=65127
2342
2343         Reviewed by Adam Barth.
2344
2345         This is part of simplifying the default views of the dashboard to
2346         make it show less information for the common use cases.
2347
2348         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2349         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2350
2351 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2352
2353         stop generating expectations.json now that it's unused
2354         https://bugs.webkit.org/show_bug.cgi?id=65130
2355
2356         Reviewed by Adam Barth.
2357
2358         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2359         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
2360         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
2361         * TestResultServer/handlers/menu.py:
2362         * TestResultServer/static-dashboards/dashboard_base.js:
2363         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2364
2365 2011-07-25  Ojan Vafai  <ojan@chromium.org>
2366
2367         don't show builders in the flakiness dashboard that have stopped running a given test suite
2368         https://bugs.webkit.org/show_bug.cgi?id=65126
2369
2370         Reviewed by Mihai Parparita.
2371
2372         * TestResultServer/static-dashboards/dashboard_base.js:
2373
2374 2011-07-25  Adam Barth  <abarth@webkit.org>
2375
2376         Overhaul garden-o-matic frontend to integrate revision history and failures
2377         https://bugs.webkit.org/show_bug.cgi?id=65089
2378
2379         Reviewed by Dimitri Glazkov.
2380
2381         This is a major overhaul of the garden-o-matic frontend.  The new UI
2382         integrates the revision history information with failures.  Failures
2383         appear in context adjacent to the revisions that caused them (which are
2384         highlighted).  Each buildbot receives a collumn to the left, which
2385         indicates which revisions that bot has tested and whether that bot
2386         showed the indicated failures.
2387
2388         Failures are selectable with checkboxes, and can then either be
2389         examined or rebaselined.
2390
2391         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
2392         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
2393         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2394         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
2395         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
2396         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
2397         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:
2398         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
2399         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
2400
2401 2011-07-25  Holger Hans Peter Freyther  <zecke@selfish.org>
2402
2403         [buildbot] Add Qt Linux MIPSEL Configuration
2404         https://bugs.webkit.org/show_bug.cgi?id=65116
2405
2406         Reviewed by Adam Roben.
2407
2408         Introduce a Qt Linux MIPS EL build slave. The configuration
2409         is a build only configuration.
2410
2411         * BuildSlaveSupport/build.webkit.org-config/config.json: Add MIPS buildbot.
2412
2413 2011-07-25  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
2414
2415         [Qt][WK2] Add multi-point touch mocking to MiniBrowser
2416         https://bugs.webkit.org/show_bug.cgi?id=64374
2417
2418         Reviewed by Benjamin Poulain.
2419
2420         Each mouse button generate a touch point. The touch point stay
2421         virtually on screen if the mouse release happened while the Ctrl
2422         modifier was pressed.
2423
2424         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2425         (MiniBrowserApplication::notify):
2426         * MiniBrowser/qt/MiniBrowserApplication.h:
2427
2428 2011-07-25  Balazs Kelemen  <kbalazs@webkit.org>
2429
2430         [Qt][WK2] Use NRWT for Qt-WK2
2431         https://bugs.webkit.org/show_bug.cgi?id=65097
2432
2433         Reviewed by Andreas Kling.
2434
2435         * Scripts/run-webkit-tests:
2436         (useNewRunWebKitTests):
2437
2438 2011-07-24  Sheriff Bot  <webkit.review.bot@gmail.com>
2439
2440         Unreviewed, rolling out r91635.
2441         http://trac.webkit.org/changeset/91635
2442         https://bugs.webkit.org/show_bug.cgi?id=65083
2443
2444         It broke the minimal build (Requested by Ossy___OFFLINE on
2445         #webkit).
2446
2447         * QtTestBrowser/launcherwindow.cpp:
2448         (LauncherWindow::initializeView):
2449         (LauncherWindow::createChrome):
2450         * QtTestBrowser/launcherwindow.h:
2451         (WindowOptions::WindowOptions):
2452         * QtTestBrowser/main.cpp:
2453         (LauncherApplication::handleUserOptions):
2454
2455 2011-07-24  Adam Barth  <abarth@webkit.org>
2456
2457         Rename garden-o-matic "alert" to "infobar" and increase the width to 100%.
2458
2459         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
2460         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2461         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
2462         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
2463         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
2464         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
2465
2466 2011-07-23  Dimitri Glazkov  <dglazkov@chromium.org>
2467
2468         TestConfiguration should be hashable and comparable for equality.
2469         https://bugs.webkit.org/show_bug.cgi?id=65070
2470
2471         Reviewed by Adam Barth.
2472
2473         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added __hash__ and __eq__ methods.
2474         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added unit tests.
2475
2476 2011-07-23  Dimitri Glazkov  <dglazkov@chromium.org>
2477
2478         Extract TestConfiguration from base and write a few unit tests to capture current behavior.
2479         https://bugs.webkit.org/show_bug.cgi?id=65066
2480
2481         Reviewed by Adam Barth.
2482
2483         * Scripts/webkitpy/layout_tests/models/test_configuration.py: Added.
2484         * Scripts/webkitpy/layout_tests/models/test_configuration_unittest.py: Added.
2485         * Scripts/webkitpy/layout_tests/port/base.py: Moved code to test_configuration.py
2486
2487 2011-07-23  Alok Priyadarshi  <alokp@chromium.org>
2488
2489         Switching off acceleration for small canvas broke gpu tests
2490         https://bugs.webkit.org/show_bug.cgi?id=65053
2491
2492         Reviewed by Stephen White.
2493
2494         * DumpRenderTree/chromium/WebPreferences.cpp:
2495         (WebPreferences::reset):
2496         (WebPreferences::applyTo):
2497         * DumpRenderTree/chromium/WebPreferences.h:
2498
2499 2011-07-23  Keith Kyzivat  <keith.kyzivat@nokia.com>
2500
2501         [Qt] Add option to turn on disk caching in QtTestBrowser
2502         https://bugs.webkit.org/show_bug.cgi?id=65007
2503
2504         Reviewed by Noam Rosenthal.
2505
2506         Add menu item and command line option to turn on disk caching in
2507         QtTestBrowser. Simple QNetworkDiskCache added to the page's
2508         QNetworkAccssManager.
2509
2510         * QtTestBrowser/launcherwindow.cpp:
2511         (LauncherWindow::initializeView):
2512         (LauncherWindow::createChrome):
2513         (LauncherWindow::setDiskCache):
2514         * QtTestBrowser/launcherwindow.h:
2515         (WindowOptions::WindowOptions):
2516         * QtTestBrowser/main.cpp:
2517         (LauncherApplication::handleUserOptions):
2518
2519 2011-07-22  Ojan Vafai  <ojan@chromium.org>
2520
2521         Don't match 'DEBUG' when trying to filter BUG modifiers.
2522         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2523         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2524
2525 2011-07-22  Ojan Vafai  <ojan@chromium.org>
2526
2527         Fix issue loading gtests since there's no g_expectations.
2528         Also, make filing a bug file a webkit.org bug, not a crbug since
2529         the link is only shown for layout tests.
2530         * TestResultServer/static-dashboards/dashboard_base.js:
2531         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2532
2533 2011-07-22  Ojan Vafai  <ojan@chromium.org>
2534
2535         pull test expectations into the dashboard from wvn.webkit.org
2536         https://bugs.webkit.org/show_bug.cgi?id=65054
2537
2538         Reviewed by Adam Barth.
2539
2540         Now that svn.webkit.org supports cors headers, we can just grab the txt
2541         file from there instead of generating a JSON file that we upload
2542         to appengine. We'll stop generating the JSON file in a followup patch.
2543
2544         * TestResultServer/static-dashboards/dashboard_base.js:
2545         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2546         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2547
2548 2011-07-22  Adam Barth  <abarth@webkit.org>
2549
2550         Fix typo so the tool works on a case-sensitive file system.
2551
2552         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2553
2554 2011-07-22  Adam Barth  <abarth@webkit.org>
2555
2556         garden-o-matic should know what patches are landing
2557         https://bugs.webkit.org/show_bug.cgi?id=64978
2558
2559         Reviewed by Dimitri Glazkov.
2560
2561         This patch adds a display of the recently committed patches to
2562         garden-o-matic.  The UI needs iteration, but this patch establishes the
2563         plumbing.
2564
2565         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
2566             - Teach trac how to extract more information from the commit log.
2567         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
2568             - Add a data attribute for storing the revision.
2569         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
2570             - Add a container for the recent commit history.
2571         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css:
2572             - CSS for displaying non-ugly commit history.
2573         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
2574             - Wire up trac to the UI container.
2575         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
2576             - A basic, compact display for a list of commits.
2577
2578 2011-07-22  Ojan Vafai  <ojan@chromium.org>
2579
2580         Fix bug when switching from layout-tests to another test-type if
2581         ToT webkit.org is the selected group.
2582         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2583
2584 2011-07-22  Ojan Vafai  <ojan@chromium.org>
2585
2586         Add back in line accidentally deleted in http://trac.webkit.org/changeset/91612.
2587         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2588
2589 2011-07-22  Ojan Vafai  <ojan@chromium.org>
2590
2591         simplify the flakiness dashboard to only show modifiers/expectations for the current builder
2592         https://bugs.webkit.org/show_bug.cgi?id=65049
2593
2594         Reviewed by Tony Chang.
2595
2596         This communicates less information, but probably makes the dashboard more usable.
2597         Also make it show that bug lines that don't point to bug numbers still
2598         show up in the bugs column (e.g. BUG_TONY) and stop showing the extra/missing
2599         columns in any view. The data is still used, but only in the view that updates
2600         the test_expectations.txt file.
2601
2602         * TestResultServer/static-dashboards/dashboard_base.js:
2603         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2604         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2605
2606 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2607
2608         hook in the rest of the bots to the dashboard now they run NRWT
2609         https://bugs.webkit.org/show_bug.cgi?id=64985
2610
2611         Reviewed by Adam Barth.
2612
2613         * TestResultServer/static-dashboards/builders.js:
2614
2615 2011-07-22  Alexey Proskuryakov  <ap@apple.com>
2616
2617         Adding Tim Horton to contributor list for Bugzilla auto-completion. He is CC'ed quite frequently.
2618
2619         Unreviewed.
2620
2621         * Scripts/webkitpy/common/config/committers.py:
2622
2623 2011-07-22  Michal Pakula vel Rutka  <m.pakula@samsung.com>
2624
2625         [EFL] ewk_frame_hit_test_new enchancement
2626         https://bugs.webkit.org/show_bug.cgi?id=64260
2627
2628         Apply changes done in ewk_frame_hit_test_new in EWebLauncher:
2629         Replacing Ewk_Hit_Test_Result_Context structure 'flags'
2630         in hit test output to by enum 'context'.
2631
2632         Reviewed by Antonio Gomes.
2633
2634         * EWebLauncher/main.c:
2635         (on_key_down):
2636
2637 2011-07-21  Adam Roben  <aroben@apple.com>
2638
2639         Fix typo in TestFailures
2640
2641         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
2642
2643 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2644
2645         simplify gtest display now that we strip modifiers from the JSON
2646         https://bugs.webkit.org/show_bug.cgi?id=64990
2647
2648         Reviewed by Adam Barth.
2649
2650         Now that names are normalized, we can remove all the code that
2651         handles gtest name changes (e.g. for adding modifiers like DISABLED_).
2652         Instead, if you try to list a test with a modifier in it, we need
2653         to strip the modifier so we get the normalized value.
2654
2655         We also get rid of the concept of extra/missing expectations for gtests.
2656         In a patch soon, we'll stop showing extra/missing expectations from the UI
2657         entirely and only leave it for the special updating test_expectations.txt
2658         view of the dashboard, which doesn't apply to gtests.
2659
2660         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2661         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
2662
2663 2011-07-21  Adam Barth  <abarth@webkit.org>
2664
2665         Refactor Trac.js for use in garden-o-matic
2666         https://bugs.webkit.org/show_bug.cgi?id=64998
2667
2668         Reviewed by Adam Roben.
2669
2670         This patch refactors Trac.js to match the architecture and style of
2671         garden-o-matic.  I've also factored AsynchronousCache out of Trac (and
2672         moved it to base) because it was redundant with some other code we
2673         already had.
2674
2675         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js:
2676         (FailingTestsBugForm):
2677         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js:
2678         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/Trac.js:
2679         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
2680         (ViewController.prototype._displayBuilder.start):
2681         (ViewController.prototype._displayBuilder):
2682         (ViewController.prototype._domForRegressionRange.trac.getCommitDataForRevisionRange):
2683         (ViewController.prototype._domForRegressionRange):
2684         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js:
2685         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js:
2686         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html:
2687         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
2688         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2689
2690 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2691
2692         Fix issue throwing a 500 to the bots. We were modifying
2693         the dictionary during an iteration over it's keys and would get
2694         RuntimeError: dictionary changed size during iteration
2695
2696         * TestResultServer/model/jsonresults.py:
2697
2698 2011-07-21  Adam Barth  <abarth@webkit.org>
2699
2700         ORWT fails to consume some NRWT arguments
2701         https://bugs.webkit.org/show_bug.cgi?id=64988
2702
2703         Reviewed by Adam Roben.
2704
2705         * Scripts/old-run-webkit-tests:
2706
2707 2011-07-21  Adam Barth  <abarth@webkit.org>
2708
2709         Merge TestFailures and garden-o-matic unit tests harness
2710         https://bugs.webkit.org/show_bug.cgi?id=64984
2711
2712         Reviewed by Adam Roben.
2713
2714         One unit testing framework is better than two.
2715
2716         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-more-unittests.html: Removed.
2717         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
2718
2719 2011-07-21  Adam Barth  <abarth@webkit.org>
2720
2721         Rubber-stamped by Adam Roben.
2722
2723         Move garden-o-matic files into the same directory as TestFailures to
2724         make it easier to integrate the codebases.
2725
2726         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/base.js.
2727         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/base_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js.
2728         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js.
2729         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/checkout_unittests.js.
2730         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/config.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/config.js.
2731         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/favicon-green.png: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-green.png.
2732         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/favicon-red.png: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-red.png.
2733         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/index.html.
2734         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.css: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/main.css.
2735         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/main.js.
2736         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/partytime.gif: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/partytime.gif.
2737         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/results.js.
2738         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js.
2739         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-more-unittests.html: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html.
2740         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js.
2741         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js: Copied from Tools/Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js.
2742         * Scripts/webkitpy/tool/commands/abstractlocalservercommand.py:
2743         * Scripts/webkitpy/tool/commands/gardenomatic.py:
2744         * Scripts/webkitpy/tool/servers/data/gardeningserver: Removed.
2745         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js: Removed.
2746         * Scripts/webkitpy/tool/servers/data/gardeningserver/base_unittests.js: Removed.
2747         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js: Removed.
2748         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout_unittests.js: Removed.
2749         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js: Removed.
2750         * Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-green.png: Removed.
2751         * Scripts/webkitpy/tool/servers/data/gardeningserver/favicon-red.png: Removed.
2752         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html: Removed.
2753         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css: Removed.
2754         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js: Removed.
2755         * Scripts/webkitpy/tool/servers/data/gardeningserver/partytime.gif: Removed.
2756         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js: Removed.
2757         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js: Removed.
2758         * Scripts/webkitpy/tool/servers/data/gardeningserver/run-unittests.html: Removed.
2759         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js: Removed.
2760         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js: Removed.
2761         * Scripts/webkitpy/tool/servers/gardeningserver.py:
2762
2763 2011-07-19  Ojan Vafai  <ojan@chromium.org>
2764
2765         fix links on the flakiness dashboard
2766         https://bugs.webkit.org/show_bug.cgi?id=64832
2767
2768         Reviewed by Tony Chang.
2769
2770         Make them actual anchor tags and linkify the test name instead of
2771         adding an extra link. Also simplify the update view since we now
2772         show the linkified test name.
2773
2774         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2775
2776 2011-07-21  Adam Barth  <abarth@webkit.org>
2777
2778         webkit.org builders should upload results to test-results AppEngine instance
2779         https://bugs.webkit.org/show_bug.cgi?id=64979
2780
2781         Reviewed by Adam Roben.
2782
2783         The master just needs to pass some flags to NRWT.  This patch also lets
2784         ORWT consume the flags (ignoring them) so we can pass the flags
2785         unconditionally.
2786
2787         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2788         * Scripts/old-run-webkit-tests:
2789
2790 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2791
2792         normalize gtest names in the result json
2793         https://bugs.webkit.org/show_bug.cgi?id=64971
2794
2795         Reviewed by Eric Seidel.
2796
2797         This will make the resulting JSON smaller and make the
2798         flakiness dashboard much more sane (e.g. we won't need to
2799         show you the FLAKY_ version and the normal version).
2800
2801         * TestResultServer/model/jsonresults.py:
2802         * TestResultServer/model/jsonresults_unittest.py:
2803
2804 2011-07-21  Eric Seidel  <eric@webkit.org>
2805
2806         Unreviewed.  Reenable ORWT for the --leaks bot.
2807         Something is still wrong with the leaks summarizing
2808         but I'm about to head out on vacation for 2 weeks,
2809         so it's best to just revert back to ORWT until I'm back.
2810
2811         * Scripts/run-webkit-tests:
2812         (usingLeaks):
2813         (useNewRunWebKitTests):
2814
2815 2011-07-21  Ojan Vafai  <ojan@chromium.org>
2816
2817         cleanup jsonresults.py style in preparation for making some changes
2818         https://bugs.webkit.org/show_bug.cgi?id=64968
2819
2820         Reviewed by Adam Barth.
2821
2822         No code changes. Unittest still passes. Mostly just removing
2823         useless comments and 80 character wrapping.
2824
2825         * TestResultServer/model/jsonresults.py:
2826         * TestResultServer/model/jsonresults_unittest.py:
2827
2828 2011-07-21  Adam Barth  <abarth@webkit.org>
2829
2830         Add missing column header.
2831
2832         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2833
2834 2011-07-21  Eric Seidel  <eric@webkit.org>
2835
2836         Unreviewed.  Just fixing (and unittesting) a previous typo.
2837
2838         Fix typo in print_leaks_summary regexp which was causing
2839         leaks summarizing to fail on the --leaks bot.
2840
2841         * Scripts/webkitpy/layout_tests/port/mac.py:
2842         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2843
2844 2011-07-21  Adam Barth  <abarth@webkit.org>
2845
2846         garden-o-matic should be able to rebaseline many tests at once
2847         https://bugs.webkit.org/show_bug.cgi?id=64963
2848
2849         Reviewed by Dimitri Glazkov.
2850
2851         This patch contains some UI for rebaselining tests in bulk.  I don't
2852         expect this UI to be the final UI, but the underlying infrastructure
2853         will be useful.
2854
2855         * Scripts/webkitpy/tool/servers/data/gardeningserver/checkout.js:
2856         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
2857         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
2858         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
2859         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2860
2861 2011-07-21  Adam Barth  <abarth@webkit.org>
2862
2863         garden-o-matic has trouble when the same test fails in different ways on diffrent bots
2864         https://bugs.webkit.org/show_bug.cgi?id=64830
2865
2866         Unreviewed.  (Missing the patch is blocking me from fixing the tree.)
2867
2868         This patch is just some minor bug fixes.  One of the bugs results from
2869         the sadness of having two representations (a list and a space-separated
2870         string) for the set of failure types.
2871
2872         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
2873         * Scripts/webkitpy/tool/servers/data/gardeningserver/results_unittests.js:
2874         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
2875         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
2876
2877 2011-07-20  Ryosuke Niwa  <rniwa@webkit.org>
2878
2879         Buildbot marks a nrwt bot red when tests are missing results
2880         https://bugs.webkit.org/show_bug.cgi?id=64812
2881
2882         Reviewed by Adam Barth.
2883
2884         The bug was caused by multiple expressions matching on the single output.
2885         Fixed it by exiting the loop as soon as one expression matches.
2886
2887         Because the regular expression for 'failures' is most general,
2888         moved it to the end of the list to avoid it catching other cases.
2889
2890         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2891
2892 2011-07-20  Chang Shu  <cshu@webkit.org>
2893
2894         Adding myself to the reviewers list; No review needed.
2895
2896         * Scripts/webkitpy/common/config/committers.py:
2897
2898 2011-07-20  Eric Seidel  <eric@webkit.org>
2899
2900         new-run-webkit-tests --leaks does not restart DumpRenderTree often enough
2901         https://bugs.webkit.org/show_bug.cgi?id=64912
2902
2903         Reviewed by Adam Barth.
2904
2905         The bug was that set_default_option only works if the option's value
2906         is "None".  So I fixed --batch-size to default to None instead of 0
2907         (which kinda makes sense).
2908
2909         * Scripts/webkitpy/layout_tests/controllers/worker.py:
2910         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2911
2912 2011-07-20  Eric Seidel  <eric@webkit.org>
2913
2914         Make kill-old-processes kill httpd on mac
2915         https://bugs.webkit.org/show_bug.cgi?id=64898
2916
2917         Reviewed by Adam Barth.
2918
2919         * BuildSlaveSupport/kill-old-processes:
2920          - The line endings seem to have flipped again,
2921            Thus making this diff gigantic.  The only line
2922            changed is the "httpd" line.  I suspect we'll need
2923            to set the line ending style from an svn checkout.
2924
2925 2011-07-20  Ojan Vafai  <ojan@chromium.org>
2926
2927         show a list of average test times in the treemap
2928         https://bugs.webkit.org/show_bug.cgi?id=64899
2929
2930         Reviewed by Adam Roben.
2931
2932         For now, you click a link and it replaces the treemap
2933         with a reverse sorted list of directories based on
2934         average test runtime within that directory.
2935
2936         To cut some of the noise, we don't show directories
2937         that only have one test in them or where the average
2938         runtime is less than 100ms.
2939
2940         * TestResultServer/static-dashboards/treemap.html:
2941
2942 2011-07-20  Eric Seidel  <eric@webkit.org>
2943
2944         Teach build.webkit.org how to identify leaks in NRWT output
2945         https://bugs.webkit.org/show_bug.cgi?id=64891
2946
2947         Reviewed by Adam Barth.
2948
2949         This was the last remaining piece blocking turning on
2950         NRWT for the leaks bot, so I did that in this change too.
2951         If for some reason we see trouble on the bots, I'll
2952         revert that part of this change.
2953
2954         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
2955         * Scripts/run-webkit-tests:
2956
2957 2011-07-20  Ojan Vafai  <ojan@chromium.org>
2958
2959         make the treemap display more human friendly
2960         https://bugs.webkit.org/show_bug.cgi?id=64896
2961
2962         Reviewed by Tony Chang.
2963
2964         * TestResultServer/static-dashboards/treemap.html:
2965
2966 2011-07-20  Eric Seidel  <eric@webkit.org>
2967
2968         Move HttpLock to using a FileSystem object
2969         https://bugs.webkit.org/show_bug.cgi?id=64885
2970
2971         Reviewed by Adam Barth.
2972
2973         I made a typo in my previous change, taking the value
2974         of read_text_file(pid_file) and setting it to lock_pid_file
2975         instead of current_pid.  Fixed now. :)
2976
2977         In order to test my new change I had to overhaul the unittests
2978         for this class to create a separate set of tests which work off
2979         of Mock objects instead of the real filesystem.
2980
2981         Since Executive doesn't yet wrap os.getpid() I added a FIXME
2982         in several places where we're currently calling os.getpid().  I
2983         felt adding Executive.getpid was outside of the scope of this change
2984         but once it exists some of this code will be much simpler to mock.
2985
2986         * Scripts/webkitpy/layout_tests/port/http_lock.py:
2987         * Scripts/webkitpy/layout_tests/port/http_lock_unittest.py:
2988         * Scripts/webkitpy/tool/mocktool.py:
2989
2990 2011-07-20  Adam Roben  <aroben@apple.com>
2991
2992         Fix typo in TestFailures's Bugzilla constants
2993
2994         This was preventing the "Version" field from being pre-filled when filing new bugs.
2995
2996         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/WebKitBugzilla.js:
2997
2998 2011-07-20  Eric Seidel  <eric@webkit.org>
2999
3000         Move HttpLock to using a FileSystem object
3001         https://bugs.webkit.org/show_bug.cgi?id=64885
3002
3003         Reviewed by Adam Barth.
3004
3005         There is no functional change here.  Just moving off
3006         of direct os.foo calls to their filesystem equivalent.
3007
3008         This makes it possible to now unittest HttpLock using a
3009         MockFileSystem instead of needing to talk to the filesystem directly.
3010
3011         This also will have changed any port test cases which were using a
3012         mock file system to now have an HttpLock based on a mock filesystem
3013         instead.  I saw no change in the unittests so it's possible
3014         we're not testing that configuration (or more likely that it "just works").
3015
3016         * Scripts/webkitpy/layout_tests/port/base.py:
3017         * Scripts/webkitpy/layout_tests/port/http_lock.py:
3018         * Scripts/webkitpy/layout_tests/port/http_lock_unittest.py:
3019
3020 2011-07-20  Adam Barth  <abarth@webkit.org>
3021
3022         NRWT results uses too much space per run
3023         https://bugs.webkit.org/show_bug.cgi?id=64845
3024
3025         Reviewed by Ojan Vafai.
3026
3027         Removing these files saves about 750K per build.
3028
3029         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3030
3031 2011-07-20  Zan Dobersek  <zandobersek@gmail.com>
3032
3033         [Gtk] [NRWT] Xvfb produces a lot of stderr output
3034         https://bugs.webkit.org/show_bug.cgi?id=64669
3035
3036         Reviewed by Philippe Normand.
3037
3038         Redirect the stderr output of the driver's Xvfb process
3039         to /dev/null to suppress unneeded output.
3040
3041         * Scripts/webkitpy/layout_tests/port/gtk.py:
3042
3043 2011-07-19  Eric Seidel  <eric@webkit.org>
3044
3045         parse-malloc-history always exits 1, causing NRWT to fail
3046         https://bugs.webkit.org/show_bug.cgi?id=64835
3047
3048         Reviewed by Adam Barth.
3049
3050         ORWT just never checked the return code. :)
3051
3052         * Scripts/parse-malloc-history:
3053         (main):
3054         * Scripts/webkitpy/layout_tests/port/mac.py:
3055
3056 2011-07-19  Kevin Ollivier  <kevino@theolliviers.com>
3057
3058         [wx] Unreviewed build fix, don't add the debug prefix as 2.9 no longer uses one.
3059
3060         * waf/build/wxpresets.py:
3061
3062 2011-07-19  Eric Seidel  <eric@webkit.org>
3063
3064         new-run-webkit-tests should only enable MallocStackLogging for DRT
3065         https://bugs.webkit.org/show_bug.cgi?id=64792
3066
3067         Reviewed by Adam Barth.
3068
3069         The previous code would enable it for all servers launched
3070         by the port, which included Apache, the python websocket server
3071         as well as ImageDiff.  Now only DumpRenderTree will have
3072         MallocStackLogging enabled or the GuardMalloc library injected.
3073
3074         I also cleaned up the websocket_server code to use filesystem
3075         while I was in it.
3076
3077         I also made DRT restart every 1000 tests when running with
3078         --leaks enabled.  I believe this made the --leaks run slightly
3079         faster, but it still takes over an hour on my machine. :(
3080
3081         * Scripts/webkitpy/layout_tests/controllers/worker.py:
3082         * Scripts/webkitpy/layout_tests/port/base.py:
3083         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
3084         * Scripts/webkitpy/layout_tests/port/gtk.py:
3085         * Scripts/webkitpy/layout_tests/port/mac.py:
3086         * Scripts/webkitpy/layout_tests/port/qt.py:
3087         * Scripts/webkitpy/layout_tests/port/server_process.py:
3088         * Scripts/webkitpy/layout_tests/port/webkit.py:
3089         * Scripts/webkitpy/layout_tests/servers/http_server.py:
3090         * Scripts/webkitpy/layout_tests/servers/websocket_server.py:
3091
3092 2011-07-19  Ojan Vafai  <ojan@chromium.org>
3093
3094         remove the concept of platform fallbacks
3095         https://bugs.webkit.org/show_bug.cgi?id=64829
3096
3097         Reviewed by Adam Barth.
3098         
3099         In the process, wrote tests for this code and fixed a pretty major
3100         bug (now covered by the test).
3101
3102         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3103         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3104
3105 2011-07-19  Ojan Vafai  <ojan@chromium.org>
3106
3107         cleanup some of the PLATFORM logic in the flakiness dashboard
3108         https://bugs.webkit.org/show_bug.cgi?id=64821
3109
3110         Reviewed by Adam Barth.
3111
3112         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3113         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3114
3115 2011-07-19  Ojan Vafai  <ojan@chromium.org>
3116
3117         cleanup some of the PLATFORM logic in the flakiness dashboard
3118         https://bugs.webkit.org/show_bug.cgi?id=64821
3119
3120         Reviewed by Adam Barth.
3121
3122         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3123         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3124
3125 2011-07-19  Dimitri Glazkov  <dglazkov@chromium.org>
3126
3127         Store line number on TestExpectationLine.
3128         https://bugs.webkit.org/show_bug.cgi?id=64800
3129
3130         Reviewed by Adam Barth.
3131
3132         * Scripts/webkitpy/layout_tests/models/test_expectations.py: Refactored to store line number on TestExpectationLine instances.
3133
3134 2011-07-19  Ojan Vafai  <ojan@chromium.org>
3135
3136         fix flakiness dashboard for XP bots with XP in the name instead of WIN
3137         https://bugs.webkit.org/show_bug.cgi?id=64826
3138
3139         Reviewed by Mihai Parparita.
3140
3141         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3142         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3143
3144 2011-07-19  Cary Clark  <caryclark@google.com>
3145
3146         add myself to committers.py
3147         https://bugs.webkit.org/show_bug.cgi?id=64816
3148
3149         No review needed.
3150
3151         * Scripts/webkitpy/common/config/committers.py:
3152
3153 2011-07-19  Daniel Bates  <dbates@webkit.org>
3154
3155         Rename variable isEmptyTextPatch in svn-{apply, unapply}
3156         https://bugs.webkit.org/show_bug.cgi?id=64648
3157
3158         Reviewed by Adam Roben.
3159
3160         Rename the variable isEmptyTextPatch to hasTextChunks, which is
3161         more descriptive and makes its usage read well in control statements.
3162
3163         * Scripts/svn-apply:
3164         (patch):
3165         * Scripts/svn-unapply:
3166         (patch):
3167
3168 2011-07-19  Adam Roben  <aroben@apple.com>
3169
3170         Make TestFailures show existing bugs and a new bug link for flaky tests
3171
3172         Fixes <http://webkit.org/b/63728> TestFailures page should make it easy to file bugs about
3173         flaky tests
3174         and
3175         <http://webkit.org/b/63830> TestFailures page doesn't show related bugs for possibly-flaky
3176         tests, but should
3177
3178         Reviewed by Sam Weinig.
3179
3180         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm.js:
3181         Added. This is what's used to file new bugs about flaky tests.
3182         (FlakyTestBugForm): Calls up to the base class, stores the arguments, and sets our
3183         title, description, and URL.
3184
3185         (FlakyTestBugForm.prototype._createBugDescription):
3186         (FlakyTestBugForm.prototype._createBugTitle):
3187         These do what they say.
3188
3189         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FlakyTestBugForm_unittests.js:
3190         Added. Tests of the above.
3191
3192         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestFailures.css:
3193         (#failure-history > li):
3194         (#possibly-flaky-tests > li):
3195         (#failure-history > li, #possibly-flaky-tests > li):
3196         (#possibly-flaky-tests > li > :first-child):
3197         (.flakiness-examples-list):
3198         Styling to account for the new elements.
3199
3200         (.expandable):
3201         (.expanded > .expandable):
3202         Generalized the .flakiness-example-list code to a generic .expandable class. Removed the
3203         transition properties because we can't transition to/from 'auto' (but really would like
3204         to!).
3205
3206         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ViewController.js:
3207         (ViewController.prototype._displayBuilder): Moved code to create the FailingTestsBugForm
3208         here from _domForNewAndExistingBugs. Updated for changes to _domForPossiblyFlakyTests.
3209         (ViewController.prototype._domForNewAndExistingBugs): Changed to take a NewBugForm as an
3210         argument instead of creating one ourselves.
3211         (ViewController.prototype._domForPossiblyFlakyTests): Changed to take all analyzed builds as
3212         an argument instead of just the number of analyzed builds. Put the examples list inside a
3213         container <div>, which is also used to hold the new/existing bugs UI. Removed code to deal
3214         with animating the height of the examples list; we don't do this anymore because it's hard
3215         to make it work correctly with the asynchronous loading of existing bugs. When we populate
3216         the examples list, also set up the new/existing bug UI.
3217
3218         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
3219         FlakyTestBugForm.
3220
3221         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
3222         Pull in FlakyTestBugForm and its tests.
3223
3224 2011-07-19  Ryosuke Niwa  <rniwa@webkit.org>
3225
3226         Buildbot marks a nrwt bot red when tests are missing results
3227         https://bugs.webkit.org/show_bug.cgi?id=64812
3228
3229         Reviewed by Tony Chang.
3230
3231         Added "missing results" label for tests that are missing results.
3232
3233         buildbot Now reports shows a label "X missing results" on a run when X tests were missing
3234         expected results on that run.
3235
3236         Also rephrased "X failed" and "X were flaky" to "X failures" and "X flakes".
3237
3238         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3239
3240 2011-07-19  Adam Roben  <aroben@apple.com>
3241
3242         Extract some of FailingTestsBugForm's code into a base class
3243
3244         Prep work for fixing <http://webkit.org/b/63728> TestFailures page should make it easy to
3245         file bugs about flaky tests
3246
3247         Reviewed by Sam Weinig.
3248
3249         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm.js:
3250         Moved BugzillaConstants to new WebKitBugzilla file. Moved a bunch of other code from here to
3251         TestRelatedBugForm.js.
3252
3253         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/FailingTestsBugForm_unittests.js:
3254         Moved some tests to TestRelatedBugForm_unittests.js.
3255
3256         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm.js: Added.
3257         (TestRelatedBugForm):
3258         (TestRelatedBugForm.prototype.domElement):
3259         (TestRelatedBugForm.prototype._computeOperatingSystem):
3260         (TestRelatedBugForm.prototype._computePlatform):
3261         Code came from FailingTestsBugForm.
3262
3263         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/TestRelatedBugForm_unittests.js:
3264         Added. Tests came from FailingTestsBugForm_unittests.js
3265
3266         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/WebKitBugzilla.js:
3267         Added. Code came from FailingTestsBugForm.js.
3268
3269         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/index.html: Pull in
3270         TestRelatedBugForm.js and WebKitBugzilla.js. Moved Bugzilla.js out of the list of files that
3271         need to be pulled in early for parsing reasons.
3272
3273         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/run-unittests.html:
3274         Pull in TestRelatedBugForm and tests and WebKitBugzilla.
3275
3276 2011-07-19  Sam Weinig  <sam@webkit.org>
3277
3278         Remove obsolete Mac NRWT test bots
3279         https://bugs.webkit.org/show_bug.cgi?id=64806
3280
3281         Reviewed by Adam Roben.
3282
3283         Remove "Leopard Intel Release (NRWT)" and "SnowLeopard Intel Release (NRWT)"
3284         which have been disconnected for some time and have been obsoleted by
3285         the main testers using NRWT.
3286
3287         * BuildSlaveSupport/build.webkit.org-config/config.json:
3288
3289 2011-07-19  Sam Weinig  <sam@webkit.org>
3290
3291         webkit-patch rebaseline crashes on use
3292         https://bugs.webkit.org/show_bug.cgi?id=64775
3293
3294         Reviewed by Adam Roben.
3295
3296         Fix incorrectly spelled attribute name.
3297
3298         * Scripts/webkitpy/common/net/layouttestresults.py:
3299         The attribute is test_name, not filename.
3300
3301         * Scripts/webkitpy/common/net/layouttestresults_unittest.py:
3302         Add test for tests_matching_failure_types.
3303
3304 2011-07-18  Eric Seidel  <eric@webkit.org>
3305
3306         new-run-webkit-tests should support --leaks
3307         https://bugs.webkit.org/show_bug.cgi?id=63832
3308
3309         Reviewed by Dirk Pranke.
3310
3311         This may not be sufficient to actually transition over the leaks bot,
3312         but this is a huge step in the right direction.
3313
3314         I had to make parse-malloc-history understand being passed more than
3315         one file (to avoid the silly cat | nonsense in old-run-webkit-tests).
3316
3317         I removed some dead code relating to previous iterations of our crash detection.
3318
3319         I created a new class "LeakDetector" to encapsulate all this logic.
3320         Eventually we should consider pulling that class out of mac.py and
3321         sharing with other ports.  However given that ORWT has had
3322         --leaks support on Mac for almost 7 years and no other port has added
3323         it, leaves me to believe we're in no rush to move LeakDetector.
3324
3325         I've tested --leaks locally.  I suspect there are more bugs to shake out
3326         but it seems to work well enough to start.
3327
3328         I also added support for --guard-malloc, but have not tested it much.  It
3329         should be viewed as experimental at this time.
3330
3331         I also fixed various os.path uses to self._filesystem as I was reading
3332         through the various files to understand how best to fix this bug.
3333
3334         * Scripts/old-run-webkit-tests:
3335         (parseLeaksandPrintUniqueLeaks):
3336         * Scripts/parse-malloc-history:
3337         (main):
3338         * Scripts/webkitpy/common/system/crashlogs.py:
3339         * Scripts/webkitpy/layout_tests/controllers/manager.py:
3340         * Scripts/webkitpy/layout_tests/controllers/worker.py:
3341         * Scripts/webkitpy/layout_tests/port/base.py:
3342         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
3343         * Scripts/webkitpy/layout_tests/port/gtk.py:
3344         * Scripts/webkitpy/layout_tests/port/mac.py:
3345         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3346         * Scripts/webkitpy/layout_tests/port/server_process.py:
3347         * Scripts/webkitpy/layout_tests/port/webkit.py:
3348         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
3349
3350 2011-07-19  Adam Barth  <abarth@webkit.org>
3351
3352         garden-o-matic should have a "Triage Failures" button for iterating over failures
3353         https://bugs.webkit.org/show_bug.cgi?id=64769
3354
3355         Reviewed by Eric Seidel.
3356
3357         This patch adds some global static state and refactors things a bit so
3358         that we can iterate over all the failures in the details pane.
3359
3360         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
3361             - Add a generic callback iterator to iterate through a series of
3362               callbacks. We use this to iterate through the failures we want to
3363               display the details of.
3364         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3365             - Add a toolbar for the results summary and change the buttons to
3366               be real buttons.
3367         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3368             - Make the detail pane 75% of the window.
3369             - Make the buttons pretty.
3370             - Allow the failure type badges to apply to the details pane as well.
3371         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3372             - Restructure how we bring up the details pane so we can iterate
3373               through a bunch of failures.
3374             - Remove the transition between results details because it's
3375               annoying when you want to click through a bunch of failures.
3376         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
3377             - Add some more structure to the title bar for the results details
3378               so it's clearer which results we're currently displaying.
3379
3380 2011-07-19  Adam Barth  <abarth@webkit.org>
3381
3382         garden-o-matic should live update as conditions change on the buildbot
3383         https://bugs.webkit.org/show_bug.cgi?id=64646
3384
3385         Reviewed by Eric Seidel.
3386
3387         This patch causes the garden-o-matic display to update every 10 minutes
3388         to show the latest failures.  This patch required a moderately large
3389         refactoring of main.js to make the display incremental instead of
3390         all-at-once.
3391
3392         * Scripts/webkitpy/tool/servers/data/gardeningserver/base.js:
3393         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3394         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3395         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
3396
3397 2011-07-18  Tab Atkins  <jackalmage@gmail.com>
3398
3399         Tab Atkins is not in committers.py (as a non-committer)
3400         https://bugs.webkit.org/show_bug.cgi?id=64770
3401
3402         Reviewed by Eric Seidel.
3403
3404         * Scripts/webkitpy/common/config/committers.py:
3405
3406 2011-07-18  Ojan Vafai  <ojan@chromium.org>
3407
3408         Load ToT chromium bots by default in the flakiness dashboard
3409         https://bugs.webkit.org/show_bug.cgi?id=64756
3410
3411         Reviewed by Adam Barth.
3412
3413         For now, these are the most useful bots to look at. Eventually,
3414         we may want to load the build.webkit.org bots by default once
3415         most of the bots there upload results.
3416
3417         * TestResultServer/static-dashboards/dashboard_base.js:
3418
3419 2011-07-13  Jon Honeycutt  <jhoneycutt@apple.com>
3420
3421         Focus and selection events are not fired when a <select>'s selection
3422         changes
3423         https://bugs.webkit.org/show_bug.cgi?id=64504
3424         <rdar://problem/9319881>
3425
3426         Reviewed by Alice Liu.
3427
3428         * DumpRenderTree/AccessibilityController.h:
3429         Added m_notificationsEventHook for addNotificationListener().
3430         m_allEventsHook will now be used for setLogAccessibilityEvents().
3431
3432         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
3433         (AccessibilityController::AccessibilityController):
3434         Initialize m_notificationsEventHook.
3435         (AccessibilityController::~AccessibilityController):
3436         Turn off logging of all accessibility events. If
3437         m_notificationsEventHook is non-null, unhook it.
3438         (logEventProc):
3439         Add handling of EVENT_OBJECT_SELECTION.
3440         (AccessibilityController::setLogAccessibilityEvents):
3441         If the state of logging is not changing, return early. If we're turning
3442         off logging, unhook m_allEventsHook, and zero it out. Otherwise, add a
3443         hook for all events.
3444         (AccessibilityController::addNotificationListener):
3445         Use m_notificationsEventHook rather than m_allEventsHook.
3446
3447 2011-07-18  Eric Seidel  <eric@webkit.org>
3448
3449         webkit-patch apply-attachment does not work when not called from the root of the checkout
3450         https://bugs.webkit.org/show_bug.cgi?id=64751
3451
3452         Unreviewed.  Just updating unit test results after bug 64751.
3453
3454         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
3455         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
3456
3457 2011-07-18  Adam Barth  <abarth@webkit.org>
3458
3459         simplejson has trouble on chromium-linux
3460         https://bugs.webkit.org/show_bug.cgi?id=64757
3461
3462         Reviewed by Eric Seidel.
3463
3464         Use the native JSON, if available.
3465
3466         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
3467
3468 2011-07-18  Ojan Vafai  <ojan@chromium.org>
3469
3470         update the flakiness dashboard to understand the new platforms/formats in test_expectations
3471         https://bugs.webkit.org/show_bug.cgi?id=64743
3472
3473         Reviewed by Adam Barth.
3474
3475         The test expectations format changed a long time ago and the flakiness dashboard
3476         was never updated to match.
3477
3478         * TestResultServer/static-dashboards/dashboard_base.js:
3479         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3480         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3481         * TestResultServer/static-dashboards/timeline_explorer.html:
3482
3483 2011-07-18  Eric Seidel  <eric@webkit.org>
3484
3485         webkit-patch apply-attachment does not work when not called from the root of the checkout
3486         https://bugs.webkit.org/show_bug.cgi?id=64751
3487
3488         Reviewed by Adam Barth.
3489
3490         Last week I removed a os.chdir in EnsureWorkingDirectoryClean, which
3491         is a step that we run for most commands (as one of the earliest steps).
3492         EnsureWorkingDirectoryClean was incorrectly calling os.chdir to change
3493         the CWD to the checkout root.  This magically made a bunch of otherwise
3494         wrong code work.
3495
3496         When I realized that apply-attachment no longer worked as expected today,
3497         I went and fixed the bug, then realizing that we were not testing
3498         what the cwd was when running various commands.  I fixed our MockExecutive
3499         to always log what the cwd is and fixed a whole bunch of places
3500         where we needed to be setting the cwd.
3501
3502         Hopefully this will solve our cwd problems once and for-all, and webkit-patch
3503         will again correctly work when called from any directory (including outside
3504         of a webkit checkout).
3505
3506         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
3507         * Scripts/webkitpy/common/system/executive.py:
3508         * Scripts/webkitpy/common/system/workspace_unittest.py:
3509         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
3510         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
3511         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
3512         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
3513         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
3514         * Scripts/webkitpy/tool/bot/irc_command.py:
3515         * Scripts/webkitpy/tool/bot/irc_command_unittest.py:
3516         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
3517         * Scripts/webkitpy/tool/commands/download_unittest.py:
3518         * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
3519         * Scripts/webkitpy/tool/commands/queues.py:
3520         * Scripts/webkitpy/tool/commands/queues_unittest.py:
3521         * Scripts/webkitpy/tool/commands/queuestest.py:
3522         * Scripts/webkitpy/tool/mocktool.py:
3523         * Scripts/webkitpy/tool/servers/gardeningserver.py:
3524         * Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
3525         * Scripts/webkitpy/tool/steps/build.py:
3526         * Scripts/webkitpy/tool/steps/checkstyle.py:
3527         * Scripts/webkitpy/tool/steps/editchangelog.py:
3528         * Scripts/webkitpy/tool/steps/preparechangelog.py:
3529         * Scripts/webkitpy/tool/steps/preparechangelogforrevert.py:
3530         * Scripts/webkitpy/tool/steps/runtests.py:
3531         * Scripts/webkitpy/tool/steps/steps_unittest.py:
3532         * Scripts/webkitpy/tool/steps/update.py:
3533
3534 2011-07-18  Adam Barth  <abarth@webkit.org>
3535
3536         Increase information garden-o-matic information density by switching to a table
3537         https://bugs.webkit.org/show_bug.cgi?id=64642
3538
3539         Reviewed by Dimitri Glazkov.
3540
3541         The old layout couldn't handle the large number of failures we had
3542         today.  This patch switches garden-o-matic to a table-based layout,
3543         which can handle many more failures gracefully.  (I expect we'll need
3544         more UI iterations.)
3545
3546         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3547         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3548         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
3549         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
3550
3551 2011-07-18  Adam Barth  <abarth@webkit.org>
3552
3553         garden-o-matic should be able to determine when compile breaks
3554         https://bugs.webkit.org/show_bug.cgi?id=64190
3555
3556         Reviewed by Dimitri Glazkov.
3557
3558         This patch adds a red-ish box to the top of the page whenever there is
3559         a compile error on the bots.  The box automatically opens and closes as
3560         appropriate and links to the waterfall display.  In the future, we
3561         might want to compute a regression range.
3562
3563         * Scripts/webkitpy/tool/servers/data/gardeningserver/config.js:
3564             - Add the build-only bots to the config.  We use these to check
3565               whether the build failed, which is faster than waiting for the
3566               tester bots to cycle.
3567         * Scripts/webkitpy/tool/servers/data/gardeningserver/index.html:
3568             - Add DOM for the alert bar.
3569         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.css:
3570             - CSS to support the alert bar.
3571         * Scripts/webkitpy/tool/servers/data/gardeningserver/main.js:
3572             - Wiring up events to poll the buildbot to see whether compile has
3573               failed.
3574         * Scripts/webkitpy/tool/servers/data/gardeningserver/results.js:
3575             - Infrastructure for fetching and parsing the buildbot status JSON
3576               blob.  This code could be better factored for testability.  :(
3577         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui.js:
3578             - UI for displaying compile errors.
3579         * Scripts/webkitpy/tool/servers/data/gardeningserver/ui_unittests.js:
3580
3581 2011-07-18  Sheriff Bot  <webkit.review.bot@gmail.com>
3582
3583         Unreviewed, rolling out r91132 and r91135.
3584         http://trac.webkit.org/changeset/91132
3585         http://trac.webkit.org/changeset/91135
3586         https://bugs.webkit.org/show_bug.cgi?id=64681
3587
3588         Broke GTK and Chromium debug bots (Requested by rniwa on
3589         #webkit).
3590
3591         * DumpRenderTree/AccessibilityController.h:
3592         * DumpRenderTree/win/AccessibilityControllerWin.cpp:
3593         (AccessibilityController::AccessibilityController):
3594         (AccessibilityController::~AccessibilityController):
3595         (logEventProc):
3596         (AccessibilityController::setLogAccessibilityEvents):
3597         (AccessibilityController::addNotificationListener):
3598
3599 2011-07-17  Philippe Normand  <pnormand@igalia.com>
3600
3601         test-webkitpy failing with Python 2.5
3602         https://bugs.webkit.org/show_bug.cgi?id=64594
3603
3604         Reviewed by Eric Seidel.
3605
3606         Updated simplejson to version 2.1.6 that fixes an issue when
3607         dumping slashes. Also use cgi.parse_qs instead of the urlparse
3608         version to be compatible with Python2.5. Verified the fix with
3609         test-webkitpy on python 2.5 and python 2.6.
3610
3611         * Scripts/webkitpy/thirdparty/simplejson/README.txt:
3612         * Scripts/webkitpy/thirdparty/simplejson/__init__.py:
3613        &