Build fix. Trigger Chromium Mac Release (Perf) on new commits.
[WebKit-https.git] / Tools / ChangeLog
1 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
2
3         Build fix. Trigger Chromium Mac Release (Perf) on new commits.
4
5         * BuildSlaveSupport/build.webkit.org-config/config.json:
6
7 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
8
9         WebKit needs public performance tests running on public bots
10         https://bugs.webkit.org/show_bug.cgi?id=10266
11
12         Reviewed by Adam Barth.
13
14         Completes the initial implementation of WebKit Perf bots. Add a build step for run-perf-test
15         and make google-mac-4 run it, and upload the results to webkit-perf.webkit.org.
16
17         * BuildSlaveSupport/build.webkit.org-config/config.json:
18         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
19         (RunAndUploadPerfTests):
20         (RunAndUploadPerfTests.start):
21         (UploadPerfTestResults):
22         (BuildAndPerfTestFactory):
23         (BuildAndPerfTestFactory.__init__):
24
25 2012-01-24  Ojan Vafai  <ojan@chromium.org>
26
27         Initializing the browser property of the Bugzilla class takes too long
28         https://bugs.webkit.org/show_bug.cgi?id=76960
29
30         Reviewed by Adam Barth.
31
32         Lazily initialize it. This saves ~150ms on a no-op run-webkit-tests call
33         on my Mac Pro.
34
35         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
36         (Bugzilla.__init__):
37         (Bugzilla._get_browser):
38         (Bugzilla):
39         (Bugzilla._set_browser):
40
41 2012-01-24  Mark Rowe  <mrowe@apple.com>
42
43         Death to JavaScriptGlue.
44
45         Rubber-stamped by Geoffrey Garen.
46
47         * BuildSlaveSupport/build-launcher-app:
48         (copyNightlyLauncher):
49         * Scripts/build-webkit:
50         * Scripts/do-file-rename:
51         * Scripts/do-webcore-rename:
52         * Scripts/webkitpy/common/config/build.py:
53         (_should_file_trigger_build):
54         * Scripts/webkitpy/common/config/build_unittest.py: Update the test to use another Mac-specific
55         path in conjunction with a cross-platform path.
56         (ShouldBuildTest):
57
58 2012-01-24  Kentaro Hara  <haraken@chromium.org>
59
60         webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl should test stdout and stderr
61         https://bugs.webkit.org/show_bug.cgi?id=75836
62
63         Reviewed by David Kilzer.
64
65         Previously parser_unittests.pl can just test the returned value of
66         get_function_line_ranges() in prepare-ChangeLog. This patch fixes
67         parser_unittests.pl so that it also tests the stdout and stderr of
68         get_function_line_ranges().
69
70         This patch adds css_unittests_warning.css, which contains test cases
71         for warning messages of the CSS parser of prepare-ChangeLog.
72
73         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
74         (captureOutput): Captures the stdout and stderr of a given method.
75         (convertAbsolutepathToWebKitPath): Converts absolute paths in a given
76         string to relative paths.
77
78         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning-expected.txt: Added.
79         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning.css: Added.
80         (element1):
81         (element2):
82         (element3):
83         (element4):
84         ():
85         (element5):
86         (element6):
87
88         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
89         Just updated the test results, i.e. added stdout and stderr results.
90         No essential changes in the test results.
91         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Ditto.
92         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/java_unittests-expected.txt: Ditto.
93         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Ditto.
94         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt: Ditto.
95         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/python_unittests-expected.txt: Ditto.
96
97 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
98
99         webkitpy: new-run-webkit-httpd is broken
100         https://bugs.webkit.org/show_bug.cgi?id=76963
101
102         Reviewed by Adam Barth.
103
104         r105674 broke new-run-webkit-httpd, since options.chromium
105         doesn't exist any more. We didn't catch this because we don't
106         run the integration tests by default anywhere :(.
107
108         * Scripts/new-run-webkit-httpd:
109         (run):
110         (main):
111
112 2012-01-24  Adam Barth  <abarth@webkit.org>
113
114         garden-o-matic pixel zoomer should let you look at every pixel of the image
115         https://bugs.webkit.org/show_bug.cgi?id=76962
116
117         Reviewed by Ojan Vafai.
118
119         Previously, you couldn't get up to the right and bottom pixels.
120
121         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
122
123 2012-01-24  Enrica Casucci  <enrica@apple.com>
124
125         WebHTMLView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
126         https://bugs.webkit.org/show_bug.cgi?id=76951
127         <rdar://problem/9325158>
128
129         Added test to verify that WebHTMLView supports the typingAttributes method and that it
130         works as expected.
131         
132         Reviewed by Alexey Proskuryakov.
133
134         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
135         (TestWebKitAPI::TEST):
136
137 2012-01-24  James Robinson  <jamesr@chromium.org>
138
139         Put a space between the trac link and ellipsis in sheriffbot rollout message so it linkifies better in some IRC clients
140         https://bugs.webkit.org/show_bug.cgi?id=76884
141
142         Reviewed by Adam Barth.
143
144         * Scripts/webkitpy/tool/bot/irc_command.py:
145         (Rollout.execute):
146         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
147         (SheriffIRCBotTest.test_rollout):
148         (SheriffIRCBotTest.test_revert):
149         (SheriffIRCBotTest.test_multi_rollout):
150         (SheriffIRCBotTest.test_rollout_with_r_in_svn_revision):
151         (SheriffIRCBotTest.test_multi_rollout_with_r_in_svn_revision):
152         (SheriffIRCBotTest.test_rollout_invalidate_reason):
153         (test_multi_rollout_invalidate_reason):
154
155 2012-01-24  Kentaro Hara  <haraken@chromium.org>
156
157         The cpp parser of prepare-ChangeLog cannot detect a change in classes and namespaces
158         https://bugs.webkit.org/show_bug.cgi?id=75531
159
160         Reviewed by David Kilzer.
161
162         Previously, the cpp parser of prepare-ChangeLog could not detect a change
163         outside methods. Consider the following cpp file.
164
165             namespace N {
166             int a;     // this change does not appear on ChangeLog.
167             class C {
168                 int b;     // this change does not appear on ChangeLog.
169                 void f()
170                 {
171                     int c;     // this change appears on ChangeLog.
172                 }
173                 int d;     // this change does not appear on ChangeLog.
174             };
175             int e;     // this change does not appear on ChangeLog.
176             };
177
178         The previous prepare-ChangeLog outputs just methods in which a change is found:
179             (N::C::f):
180
181         This patch fixes prepare-ChangeLog so that it outputs namespaces, classes
182         and methods in which a change is found:
183             (N):
184             (N::C):
185             (N::C::f):
186
187         * Scripts/prepare-ChangeLog:
188         (get_function_line_ranges_for_cpp): Modified as described above.
189
190         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests.cpp: Added test cases.
191         (Class104):
192         (Class105):
193         (Class106):
194         (Class106::func32):
195         (Class106::func33):
196         (NameSpace3):
197         (NameSpace4):
198         (NameSpace5):
199         (NameSpace6):
200         (Class107):
201         (NameSpace5::NameSpace6::Class107::func34):
202         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
203
204 2012-01-24  Kentaro Hara  <haraken@chromium.org>
205
206         prepare-ChangeLog outputs warnings for top-level { ... } blocks
207         https://bugs.webkit.org/show_bug.cgi?id=75943
208
209         Reviewed by David Kilzer.
210
211         Currently, prepare-ChangeLog outputs warnings for top-level { ... } blocks,
212         because it does not distinguish '}' at the end of a subroutine
213         with '}' at the end of a if or for statement.
214
215         foo.pl:
216             if (1) {
217             }
218             for (@v) {
219             }
220
221         output:
222             nested functions found at top-level at foo.pl:2
223             nested functions found at top-level at foo.pl:4
224
225         This patch fixes the bug.
226
227         * Scripts/prepare-ChangeLog:
228         (get_function_line_ranges_for_perl):
229         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
230         Added test cases. No change in the test results.
231
232 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
233
234         r105674 broke check-webkit-style for chromium's test_expectations.
235
236         Unreviewed, build fix.
237         
238         We now need to pass in a platform name of 'chromium' if we want
239         the chromium port; options.chromium is no longer recognized. I
240         will follow this patch up with a separate patch that has a test,
241         but this at least fixes the build.
242
243         * Scripts/webkitpy/style/checkers/test_expectations.py:
244         (TestExpectationsChecker._determine_port_from_exepectations_path):
245
246 2012-01-20  Ojan Vafai  <ojan@chromium.org>
247
248         check-webkit-style of the chromium test_expectations.txt file takes too long
249         https://bugs.webkit.org/show_bug.cgi?id=76745
250
251         Reviewed by Dimitri Glazkov.
252
253         When in lint mode, have TestExpectations test all configurations instead
254         of looping over each configuration. This also has the benefit of making
255         the error output considerably more concise.
256
257         Also, got rid of the double-printing of errors when linting through check-webkit-style.
258
259         * Scripts/webkitpy/layout_tests/controllers/manager.py:
260         (Manager.lint):
261         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
262         (TestExpectations._report_errors):
263         (TestExpectations._add_expectations):
264         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
265         (test_parse_error_nonfatal):
266         (test_error_on_different_platform):
267         * Scripts/webkitpy/style/checkers/test_expectations.py:
268         (TestExpectationsChecker.check_test_expectations):
269         (TestExpectationsChecker.check):
270         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
271         (TestExpectationsTestCase.test_determine_port_from_exepectations_path):
272         (TestExpectationsTestCase.assert_lines_lint):
273
274 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
275
276         Another build fix attempt after r105543.
277
278         * Scripts/webkitpy/common/net/file_uploader.py:
279         (FileUploader.upload_as_multipart_form_data):
280
281 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
282
283         Build fix attempt after r105543.
284
285         * Scripts/webkitpy/common/net/file_uploader.py:
286         (FileUploader.upload_as_multipart_form_data):
287
288 2012-01-24  Vincent Scheib  <scheib@chromium.org>
289
290         [Chromium] Add Pointer Lock test hooks and mock implementation to DumpRenderTree
291         https://bugs.webkit.org/show_bug.cgi?id=76411
292
293         Reviewed by Adam Barth.
294
295         Breaking up https://bugs.webkit.org/show_bug.cgi?id=75762
296         into a series of patches, this change adds a trivial
297         pointer lock implementation to the Chromium port's DumpRenderTree
298         and exposes LayoutTestController hooks to force failure conditions.
299
300         * DumpRenderTree/chromium/LayoutTestController.cpp:
301         (LayoutTestController::LayoutTestController):
302         (LayoutTestController::didLosePointerLock):
303         (LayoutTestController::setPointerLockWillFailAsynchronously):
304         (LayoutTestController::setPointerLockWillFailSynchronously):
305         * DumpRenderTree/chromium/LayoutTestController.h:
306         * DumpRenderTree/chromium/WebViewHost.cpp:
307         (WebViewHost::requestPointerLock):
308         (WebViewHost::requestPointerUnlock):
309         (WebViewHost::isPointerLocked):
310         (WebViewHost::didCompletePointerLock):
311         (WebViewHost::didNotCompletePointerLock):
312         (WebViewHost::didLosePointerLock):
313         (WebViewHost::reset):
314         * DumpRenderTree/chromium/WebViewHost.h:
315         (WebViewHost::setPointerLockWillFailAsynchronously):
316         (WebViewHost::setPointerLockWillFailSynchronously):
317
318 2012-01-24  Mario Sanchez Prada  <msanchez@igalia.com>
319
320         Unreviewed, skip WK2 API test timing out on some GTK bots.
321
322         WebKit2APITests/TestDownloads is consistently timing out on the
323         32bit release and 64bit debug bots.
324
325         * Scripts/run-gtk-tests:
326         (TestRunner): Skip WebKit2APITests/TestDownloads.
327
328 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
329
330         [Qt] Fix the build with the newes Qt5 hashes
331         https://bugs.webkit.org/show_bug.cgi?id=76657
332
333         Reviewed by Simon Hausmann.
334
335         * MiniBrowser/qt/MiniBrowserApplication.cpp:
336         (MiniBrowserApplication::MiniBrowserApplication): Disable the automatic touch->mouse event synthesis
337         for now to prevent an infinite loop where Qt always creates mouse events out of our touch mock events.
338
339 2012-01-24  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
340
341         [Qt] Distinguish between Cygwin and cmd.exe in the way we set env variables
342
343         We can't use win32-msvc* to decide whether or not to use (set FOO=bar), as
344         building with MinGW inside cmd.exe will require the same construct. Instead
345         we assume that when the first character in PWD is a '/' we're running under
346         Cygwin, and use the normal unix way of setting environment variables.
347
348         Reviewed by Simon Hausmann.
349
350         * qmake/mkspecs/features/default_pre.prf:
351
352 2012-01-24  Simon Hausmann  <simon.hausmann@nokia.com>
353
354         Some qmake build files must include Qt5's "quick" module.
355         https://bugs.webkit.org/show_bug.cgi?id=76671
356
357         Unreviewed build fix.
358
359         * MiniBrowser/qt/MiniBrowser.pro:
360         * WebKitTestRunner/Target.pri:
361
362 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
363
364         [Qt] Fix the event timestamps for touch mocking
365
366         Reviewed by Simon Hausmann.
367
368         * MiniBrowser/qt/MiniBrowserApplication.cpp:
369         (MiniBrowserApplication::sendTouchEvent): We should not set QEvent::None as the
370         timestamp argument in the QWindowSystemInterface::handleTouchEvent call.
371
372 2012-01-24  Carlos Garcia Campos  <cgarcia@igalia.com>
373
374         [GTK] Implement DownloadClient in WebKit2 GTK+ API
375         https://bugs.webkit.org/show_bug.cgi?id=72952
376
377         Reviewed by Martin Robinson.
378
379         * gtk/generate-gtkdoc:
380         (get_webkit2_options): Ignore WebKitDownloadClient.
381
382 2012-01-23  Zan Dobersek  <zandobersek@gmail.com>
383
384         [GTK] editing/deleting/5408255.html results are incorrect
385         https://bugs.webkit.org/show_bug.cgi?id=53644
386
387         Reviewed by Martin Robinson.
388
389         WEBKIT_TOP_LEVEL environment variable is now set directly in either
390         WebKitTestRunner or DumpRenderTree through usage of a compilation-time
391         macro. This way both tools can be run outside the test harness without
392         the need to manually set the environment variable.
393
394         * DumpRenderTree/gtk/DumpRenderTree.cpp:
395         (getTopLevelPath):
396         * GNUmakefile.am:
397         * Scripts/webkitpy/layout_tests/port/gtk.py:
398         (GtkPort.setup_environ_for_server):
399         * WebKitTestRunner/GNUmakefile.am:
400         * WebKitTestRunner/InjectedBundle/gtk/InjectedBundleGtk.cpp:
401         (WTR::InjectedBundle::platformInitialize):
402
403 2012-01-23  Dmitry Lomov  <dslomov@google.com>
404
405         [Chromium] Implement layoutTestController.workerThreadCount in DRT
406         https://bugs.webkit.org/show_bug.cgi?id=74653.
407
408         Reviewed by Darin Fisher.
409
410         * DumpRenderTree/chromium/LayoutTestController.cpp:
411         (LayoutTestController::LayoutTestController):
412         (LayoutTestController::workerThreadCount):
413         * DumpRenderTree/chromium/LayoutTestController.h:
414
415 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
416
417         nrwt: make --chromium work like --qt
418         https://bugs.webkit.org/show_bug.cgi?id=76875
419
420         Reviewed by Adam Barth.
421
422         --chromium used to have to be handled differently from --qt
423         due to the way the PortFactory was implemented; there's not
424         really a good reason for that any more so this patch makes
425         things slightly more consistent and eliminates the
426         options.chromium flag (--chromium is now truly a synonym for
427         --platform chromium).
428
429         * Scripts/webkitpy/layout_tests/port/factory.py:
430         (PortFactory._default_port):
431         (PortFactory.get):
432         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
433         (FactoryTest.setUp):
434         (FactoryTest.test_chromium_mac):
435         (FactoryTest.test_chromium_linux):
436         (FactoryTest.test_chromium_win):
437         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
438         (parse_args):
439
440 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
441
442         run-webkit-tests needs to propagate --chromium
443         https://bugs.webkit.org/show_bug.cgi?id=76870
444
445         Reviewed by Eric Seidel.
446
447         run-webkit-tests removes '--chromium' argument from @ARGV when
448         determining which port to run, which means that that doesn't
449         propagate to new-run-webkit-tests. That's bad (and is handled
450         for the other ports by re-adding the flag, but apparently we're
451         just now noticing for Chromium).
452
453         * Scripts/run-webkit-tests:
454
455 2012-01-23  Brian Weinstein  <bweinstein@apple.com>
456
457         More build fixing after r105646.
458
459         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
460         (TestWebKitAPI::TestNode::key):
461
462 2012-01-23  Gavin Barraclough  <barraclough@apple.com>
463
464         Unreviewed build fix - r105646 broke this.
465
466         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
467         (TestWebKitAPI::TestNode::TestNode):
468         (TestWebKitAPI::TestNode::key):
469         (TestWebKitAPI::RedBlackTreeTest::assertEqual):
470         (TestWebKitAPI::RedBlackTreeTest::assertSameValuesForKey):
471         (TestWebKitAPI::RedBlackTreeTest::testDriver):
472
473 2012-01-23  Sheriff Bot  <webkit.review.bot@gmail.com>
474
475         Unreviewed, rolling out r105658.
476         http://trac.webkit.org/changeset/105658
477         https://bugs.webkit.org/show_bug.cgi?id=76883
478
479         We want this eventually, but not right at this moment
480         (Requested by abarth on #webkit).
481
482         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
483
484 2012-01-23  Adam Barth  <abarth@webkit.org>
485
486         garden-o-matic should support Chromium Mac Lion
487         https://bugs.webkit.org/show_bug.cgi?id=76880
488
489         Reviewed by Eric Seidel.
490
491         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
492
493 2012-01-23  Emil A Eklund  <eae@chromium.org>
494
495         check-webkit-style whitespace/operators triggers on overloaded division operator
496         https://bugs.webkit.org/show_bug.cgi?id=76650
497
498         Reviewed by Darin Adler.
499
500         * Scripts/webkitpy/style/checkers/cpp.py:
501         (check_spacing):
502         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
503         (CppStyleTest.test_operator_methods):
504
505 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
506
507         run-perf-tests should report server-side errors
508         https://bugs.webkit.org/show_bug.cgi?id=76802
509
510         Reviewed by Tony Chang.
511
512         Report errors whenever server's response doesn't match "OK".
513
514         * Scripts/webkitpy/common/net/file_uploader.py:
515         (FileUploader.upload_single_text_file):
516         (FileUploader.upload_as_multipart_form_data):
517         (FileUploader._upload_data.callback):
518         (FileUploader):
519         (FileUploader._upload_data):
520         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
521         (PerfTestsRunner._upload_json):
522         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
523         (test_upload_json):
524         (test_upload_json.MockFileUploader.upload_single_text_file):
525
526 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
527
528         run-perf-tests ignore Skipped list on chromium
529         https://bugs.webkit.org/show_bug.cgi?id=76764
530
531         Reviewed by Dirk Pranke.
532
533         Move skipped_perf_tests from WebKit port to Base port so that Chromium port
534         can also find skipped list. Chromium port only uses test_expectations.txt for
535         layout tests but performacne tests don't use test_expectations.txt so Chromium port
536         also needs to use Skipped list.
537
538         * Scripts/webkitpy/layout_tests/port/base.py:
539         (Port._tests_from_skipped_file_contents):
540         (Port):
541         (Port._expectations_from_skipped_files):
542         (Port.skipped_perf_tests):
543         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
544         (PortTest.test_skipped_perf_tests):
545         (PortTest.test_skipped_perf_tests.add_text_file):
546         * Scripts/webkitpy/layout_tests/port/webkit.py:
547         (WebKitPort._skipped_tests_for_unsupported_features):
548         (WebKitPort._skipped_file_search_paths):
549         (WebKitPort.skipped_layout_tests):
550
551 2012-01-23  Andy Estes  <aestes@apple.com>
552
553         Fix the build after r105635.
554
555         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
556         (TestWebKitAPI::TEST):
557
558 2012-01-23  Xianzhu Wang  <wangxianzhu@chromium.org>
559
560         Basic enhancements to StringBuilder
561         https://bugs.webkit.org/show_bug.cgi?id=67081
562
563         Reviewed by Darin Adler.
564
565         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
566         (TestWebKitAPI::TEST):
567
568 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
569
570         [GTK] run-gtk-tests randomly fails while running the xprop comand
571         https://bugs.webkit.org/show_bug.cgi?id=76817
572
573         Reviewed by Gustavo Noronha Silva.
574
575         No need to use xprop to remove the AT_SPI_BUS property since
576         run-gtk-tests will always run new instances of Xvfb.
577
578         * Scripts/run-gtk-tests:
579         (TestRunner): Unskip TestWebKitAccessibility.
580         (TestRunner.run): Uncomment lines for launching the accessibility
581         bus and registry daemon, and remove lines for running xprop.
582
583 2012-01-23  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
584
585         [Qt] Don't warn about override and final being C++11 extensions
586
587         Clang will emit a warning when these extensions are used without passing
588         --std=c++11, but we use feature checking to decide if we have the right
589         extensions, so we can safely ignore these warnings. The XCode and Windows
590         project files have the same workaround.
591
592         The reason for adding the flag to QMAKE_OBJECTIVE_CFLAGS as well is that
593         we only have one extra compiler for Objective-C, which is also used for
594         Objective-C++ sources, so we have to pass the flag, even if it doesn't
595         make sense for Objective-C.
596
597         Reviewed by Simon Hausmann.
598
599         * qmake/mkspecs/features/unix/default_post.prf:
600
601 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
602
603         [Qt] Change how build-webkit decides when to do full incremental builds
604
605         Instead of relying on update-webkit (which isn't run on the bots) to
606         decide when to do a full incremental build (make qmake), we let the
607         build-webkit script itself check the current SVN revision against the
608         previous build (by storing it in .webkit.config).
609
610         If the two differ we assume a full incremental build is needed, since
611         the new revisions might have introduced problematic things like new
612         Q_OBJECT macros. If not, we assume the developer is doing changes
613         locally, and revert to doing a plain 'make'.
614
615         In addition, when the build fails in the latter case, we inform the
616         developer of possible pitfalls and how to manually run 'make qmake'.
617
618         Reviewed by Ossy.
619
620         * Scripts/update-webkit:
621         * Scripts/webkitdirs.pm:
622         (buildQMakeProjects):
623
624 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
625
626         [GTK] run-gtk-tests randomly fails while running the xprop comand
627         https://bugs.webkit.org/show_bug.cgi?id=76817
628
629         Reviewed by Philippe Normand.
630
631         Temporarily comment the lines related to launching the ATSPI bus
632         and registry daemon, which are making the bots to fail randomly.
633
634         * Scripts/run-gtk-tests:
635         (TestRunner): Skip TestWebKitAccessibility, as it won't run
636         properly if the ATSPI infrastructure is not properly initialized.
637         (TestRunner.run): Comment lines related to initialization of ATSPI.
638
639 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
640
641         Windows python test build fix.
642
643         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
644         (PerfTestsRunner.__init__):
645
646 2012-01-22  Sergio Villar Senin  <svillar@igalia.com>
647
648         [GTK] DumpRenderTree converts "file:///" to a path differently
649         https://bugs.webkit.org/show_bug.cgi?id=76631
650
651         Reviewed by Martin Robinson.
652
653         DumpRenderTree should print "/" as the last path component if the
654         path is a lone slash instead of empty output.
655
656         * DumpRenderTree/gtk/DumpRenderTree.cpp:
657         (webViewConsoleMessage):
658
659 2012-01-20  David Levin  <levin@chromium.org>
660
661         The leaky DC check should use adoptPtr as a signal instead of OwnPtr to get less false positives.
662         https://bugs.webkit.org/show_bug.cgi?id=76752
663
664         Reviewed by Eric Seidel.
665
666         * Scripts/webkitpy/style/checkers/cpp.py:
667         (check_for_leaky_patterns):
668         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
669         (LeakyPatternTest.test_create_dc):
670         (LeakyPatternTest.test_create_compatible_dc):
671
672 2012-01-20  Shawn Singh  <shawnsingh@chromium.org>
673
674         Changed my status in commiters.py
675
676         Unreviewed.
677
678         * Scripts/webkitpy/common/config/committers.py:
679
680 2012-01-20  Ojan Vafai  <ojan@chromium.org>
681
682         run-webkit-tests --lint-test-files crawls the whole LayoutTests subtree
683         https://bugs.webkit.org/show_bug.cgi?id=76748
684
685         Reviewed by Ryosuke Niwa.
686
687         It crawls the whole subtree and then doesn't use the data. Cutting this out
688         saves 4 seconds warm and 17 seconds cold on my Mac Pro.
689
690         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
691         (run):
692
693 2012-01-20  Adam Barth  <abarth@webkit.org>
694
695         Another tiny tweak to the garden-o-matic CSS.  This makes things line
696         up slightly nicer when there aren't any failures.
697
698         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
699         (#summary):
700
701 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
702
703         run-perf-tests should support --test-results-server option
704         https://bugs.webkit.org/show_bug.cgi?id=76680
705
706         Reviewed by Adam Barth.
707
708         Add --test-results-server, --builder-name, and --build-number options to run-perf-tests
709         to be used by perf bots. Also refactor file_uploader as needed.
710
711         * Scripts/webkitpy/common/net/file_uploader.py:
712         (FileUploader.__init__):
713         (FileUploader.upload_single_file):
714         (FileUploader.upload_as_multipart_form_data):
715         (FileUploader):
716         (FileUploader._upload_data):
717         (FileUploader._upload_data.callback):
718         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
719         (JSONResultsGeneratorBase.upload_json_files):
720         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
721         (PerfTestsRunner):
722         (PerfTestsRunner._parse_args):
723         (PerfTestsRunner.run):
724         (PerfTestsRunner._generate_json):
725         (PerfTestsRunner._upload_json):
726         (PerfTestsRunner._run_tests_set):
727         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
728         (create_runner):
729         (test_run_with_upload_json):
730         (test_run_with_upload_json.mock_upload_json):
731         (test_upload_json):
732         (test_upload_json.MockFileUploader):
733         (test_upload_json.MockFileUploader.__init__):
734         (test_upload_json.MockFileUploader.upload_single_file):
735         (test_parse_args):
736
737 2012-01-20  Adam Barth  <abarth@webkit.org>
738
739         The party time image overlaps real content!  This patch makes the
740         image centered vertically.
741
742         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
743         (#onebar.partytime #summary):
744
745 2012-01-19  Ojan Vafai  <ojan@chromium.org>
746
747         Refactor TestExpectationsParser in preparation for caching the results
748         https://bugs.webkit.org/show_bug.cgi?id=76669
749
750         Reviewed by Dimitri Glazkov.
751
752         Make everything private expect for the parse method. 
753         Eventually, we'll need the expectations lines to not be modified
754         outside of TestExpectationsParser so we can cache the results.
755         This makes check-webkit-style of the chromium test_expectations.txt file
756         go from ~17 seconds to ~12 seconds on my Mac Pro.
757
758         This patch is just a refactor in preparation, so no new tests.
759
760         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
761         (TestExpectationEditorTests.make_parsed_expectation_lines):
762         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
763         (TestExpectationParser.parse):
764         (TestExpectationParser):
765         (TestExpectationParser._parse_line):
766         (TestExpectationParser._tokenize):
767         (TestExpectationParser._tokenize_list):
768         (TestExpectationsModel._clear_expectations_for_test):
769         (TestExpectations.__init__):
770         (TestExpectations._add_expectations):
771         (TestExpectations._add_skipped_tests):
772         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
773         (TestExpectationParserTests.test_tokenize_blank):
774         (TestExpectationParserTests.test_tokenize_missing_colon):
775         (TestExpectationParserTests.test_tokenize_extra_colon):
776         (TestExpectationParserTests.test_tokenize_empty_comment):
777         (TestExpectationParserTests.test_tokenize_comment):
778         (TestExpectationParserTests.test_tokenize_missing_equal):
779         (TestExpectationParserTests.test_tokenize_extra_equal):
780         (TestExpectationParserTests.test_tokenize_valid):
781         (TestExpectationParserTests.test_tokenize_valid_with_comment):
782         (TestExpectationParserTests.test_tokenize_valid_with_multiple_modifiers):
783         (TestExpectationParserTests.test_parse_empty_string):
784         (TestExpectationSerializerTests.assert_round_trip):
785         (TestExpectationSerializerTests.assert_list_round_trip):
786         * Scripts/webkitpy/tool/commands/expectations.py:
787         (OptimizeExpectations.execute):
788         * Scripts/webkitpy/tool/servers/gardeningserver.py:
789         (GardeningExpectationsUpdater.update_expectations):
790
791 2012-01-20  Adam Barth  <abarth@webkit.org>
792
793         Follow-up to previous patch: don't produce NaN when the revision number
794         is missing.
795
796         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
797         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
798
799 2012-01-20  Adam Barth  <abarth@webkit.org>
800
801         Garden-o-matic should tell me which revisions have been checked by all the bots
802         https://bugs.webkit.org/show_bug.cgi?id=76722
803
804         Reviewed by Dimitri Glazkov.
805
806         This information helps me know when it's safe for me to roll a
807         candidate revision.
808
809         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
810         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
811         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
812
813 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
814
815         build-webkit: Don't spit out congratulations message on Ctrl+C
816
817         Reviewed by Andreas Kling.
818
819         * Scripts/build-webkit:
820
821 2012-01-20  Philippe Normand  <pnormand@igalia.com>
822
823         [GTK] Allow extra jhbuild modulesets
824         https://bugs.webkit.org/show_bug.cgi?id=76691
825
826         Reviewed by Gustavo Noronha Silva.
827
828         * gtk/jhbuildrc: Extra modulesets and modules can be added in
829         JHBuild using the WEBKIT_EXTRA_MODULESETS and WEBKIT_EXTRA_MODULES
830         env variables respectively. Use comma-separated values. Example:
831         WEBKIT_EXTRA_MODULES=file:///path/to/module.set,file:///other/path.
832
833 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
834
835         [GTK] Expose accessibility hierarchy in WebKit2 to ATK/AT-SPI based ATs
836         https://bugs.webkit.org/show_bug.cgi?id=72589
837
838         Reviewed by Martin Robinson.
839
840         Ensure the accessibility infrastructure is available when running
841         the unit tests, otherwise WK2's accessibility tests won't work.
842
843         * Scripts/run-gtk-tests:
844         (TestRunner):
845         (TestRunner._lookup_atspi2_binary): New. Looks for the directory
846         where at-spi2's stuff is located (like at-spi-bus-launcher and
847         at-spi2-registryd).
848         (TestRunner.run): Initialize the accessibility infrastructure.
849
850         * gtk/generate-gtkdoc:
851         (get_webkit2_options): Add 'WebKitWebViewBaseAccessible.*'.
852
853         * gtk/jhbuild.modules: Added at-spi2-core and at-spi2-atk.
854
855 2012-01-20  Hayato Ito  <hayato@chromium.org>
856
857         Add ShadowRoot.idl which is enabled by newly introduced SHADOW_DOM flag.
858         https://bugs.webkit.org/show_bug.cgi?id=76353
859
860         Reviewed by Hajime Morita.
861
862         Add ShadowRoot.idl, which is enabled only on chromium port since this is
863         under development feature.
864         ShadowRoot.idl contains minimum API so that we can test it.
865         Other APIs should be added on other changes so that we can isolate issues.
866
867         * DumpRenderTree/chromium/TestShell.cpp:
868         (TestShell::TestShell):
869
870 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
871
872         [GTK] GTK's DRT not to log events for already defunct objects
873         https://bugs.webkit.org/show_bug.cgi?id=76620
874
875         Reviewed by Martin Robinson.
876
877         Do not log 'state-change:defunct' events.
878
879         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
880         (printAccessibilityEvent): Early return if a
881         'state-change:defunct' signal is passed.
882         (axObjectEventListener): Pass the signal name and value in
883         separate parameters to printAccessibilityEvent.
884
885 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
886
887         Unreviewed build fix. Wrong variable being used as the cwd for git
888         remote update command.
889
890         * gtk/run-with-jhbuild:
891         (update_jhbuild):
892
893 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
894
895         [GTK] ensure the jhbuild used by webkit is as up-to-date as needed
896         https://bugs.webkit.org/show_bug.cgi?id=76585
897
898         Reviewed by Martin Robinson.
899
900         This allows us to specify a minimum version of jhbuild required,
901         while also providing stability against changes done to the master
902         branch, by letting us choose what is the exact version that will
903         be used.
904
905         * gtk/run-with-jhbuild: refactored for more hackability and easier
906         reading
907         (jhbuild_installed): checks whether jhbuild is installed
908         (jhbuild_cloned): checks whether jhbuild has been cloned
909         (jhbuild_at_expected_revision): checks whether jhbuild is at the
910         expected revision
911         (update_jhbuild): brings jhbuild to the expected revision
912         (clone_jhbuild): clones jhbuild
913         (install_jhbuild): build and installs jhbuild
914         (update_webkitgtk_libs): runs update-webkitgtk-libs
915         (ensure_jhbuild): high-level logic to decide whether and which of
916         the steps above are executed
917
918 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
919
920         [WK2] DumpRenderTree converts "file:///" to a path differently
921         https://bugs.webkit.org/show_bug.cgi?id=76653
922
923         Reviewed by John Sullivan.
924
925         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::lastFileURLPathComponent):
926         Do print a slash if there are no path components.
927
928 2012-01-19  Eric Seidel  <eric@webkit.org>
929
930         webkit-patch cannot rollout patches without changelog/bug number
931         https://bugs.webkit.org/show_bug.cgi?id=75962
932
933         Reviewed by Adam Barth.
934
935         * Scripts/webkitpy/tool/commands/download.py:
936         (AbstractRolloutPrepCommand._prepare_state):
937         * Scripts/webkitpy/tool/commands/download_unittest.py:
938         (AbstractRolloutPrepCommandTest.test_prepare_state):
939
940 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
941
942         [WK2] fast/url/degenerate-file-base.html fails
943         https://bugs.webkit.org/show_bug.cgi?id=76619
944
945         Reviewed by Adam Roben.
946
947         A URL generated form local path for loading had an extra slash, making document.URL
948         incorrect in all WebKit2 tests (file://localhost//path/to/test.html).
949
950         * WebKitTestRunner/TestInvocation.cpp: (WTR::createWKURL):
951
952 2012-01-18  Ojan Vafai  <ojan@chromium.org>
953
954         check-webkit-style of the chromium test_expectations.txt file doesn't test all chromium ports
955         https://bugs.webkit.org/show_bug.cgi?id=76510
956
957         Reviewed by Adam Barth.
958
959         Unlike other style rules, if you get the syntax of the test_expectations.txt
960         file wrong, the layout tests won't run. Also, this check is simple and only
961         slows down committing if you actually modify one of the test_expectations.txt files.
962
963         * Scripts/webkitpy/style/checkers/test_expectations.py:
964         (TestExpectationsChecker.check_test_expectations):
965         (TestExpectationsChecker.check):
966         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
967         (TestExpectationsTestCase.test_check_covers_all_configurations):
968         (TestExpectationsTestCase.test_check_covers_all_configurations.mock_check_test_expectations):
969         (TestExpectationsTestCase):
970         (assert_lines_lint):
971
972 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
973
974         Build fix. Revert an unintentional change to WebKitDriver in r105443.
975
976         * Scripts/webkitpy/layout_tests/port/driver.py:
977         (DriverProxy.__init__):
978
979 2012-01-19  Dirk Pranke  <dpranke@chromium.org>
980
981         remove the duplicated build rules from WebKit.gyp, Tools.gyp
982         https://bugs.webkit.org/show_bug.cgi?id=73384
983
984         Reviewed by Tony Chang.
985
986         This patch removes the no-longer-necessary conditional logic for
987         build_webkit_exes_from_webkit_gyp; we now always build the exes
988         from their own dedicated gyp files.
989
990         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
991
992 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
993
994         make-new-script-test should work with HTTP tests
995         https://bugs.webkit.org/show_bug.cgi?id=76603
996
997         Reviewed by Adam Roben.
998
999         * Scripts/make-new-script-test:
1000         (makePathToSharedSources): Renamed from makeTestRelativePathToSharedSources(), since it
1001         now also creates absolute paths for HTTP.
1002
1003 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
1004
1005         Some perf tests time out when ran by run-perf-tests
1006         https://bugs.webkit.org/show_bug.cgi?id=76612
1007
1008         Reviewed by Dirk Pranke and Eric Seidel.
1009
1010         Always pass --no-timeout to DumpRenderTree from run-perf-tests.
1011         Otherwise some tests such as Parser/xml-parser.html will timeout.
1012
1013         --no-timeout option is currently supported by Chromium and Mac ports.
1014
1015         * Scripts/webkitpy/layout_tests/port/base.py:
1016         (Port.to.create_driver):
1017         * Scripts/webkitpy/layout_tests/port/chromium.py:
1018         (ChromiumDriver.__init__):
1019         (ChromiumDriver._wrapper_options):
1020         * Scripts/webkitpy/layout_tests/port/driver.py:
1021         (Driver.__init__):
1022         (DriverProxy.__init__):
1023         * Scripts/webkitpy/layout_tests/port/webkit.py:
1024         (WebKitDriver.__init__):
1025         (WebKitDriver.cmd_line):
1026         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1027         (WebKitDriverTest.test_read_binary_block):
1028         (WebKitDriverTest):
1029         (WebKitDriverTest.test_no_timeout):
1030         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1031         (get_tests_run.RecordingTestDriver.__init__):
1032         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1033         (PerfTestsRunner._parse_args):
1034         (PerfTestsRunner._run_tests_set):
1035         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1036         (create_runner):
1037
1038 2012-01-19  Ben Wells  <benwells@chromium.org>
1039
1040         Add support for window.print to chromium DRT
1041         https://bugs.webkit.org/show_bug.cgi?id=76479
1042
1043         When used in a test window.print goes into print mode and then
1044         straight out again. This will allow the afterprint event to be
1045         tested by having the afterprint event happen before the dump
1046         happens.
1047
1048         Reviewed by Mihai Parparita.
1049
1050         * DumpRenderTree/chromium/WebViewHost.cpp:
1051         (WebViewHost::printPage):
1052         * DumpRenderTree/chromium/WebViewHost.h:
1053
1054 2012-01-18  Enrica Casucci  <enrica@apple.com>
1055
1056         editingAttributedStringFromRange in WebHTMLConverter does not handle NSUnderlineStyleAttributeName.
1057         https://bugs.webkit.org/show_bug.cgi?id=76588
1058         <rdar://problem/9325183>
1059
1060         Added test.
1061         
1062         Reviewed by Dan Bernstein.
1063
1064         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
1065         (TestWebKitAPI::TEST):
1066
1067 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1068
1069         [Qt] Do full incremental builds until bots use update-webkit
1070
1071         Rubber-stamped by Simon Hausmann.
1072
1073         * Scripts/webkitdirs.pm:
1074         (buildQMakeProjects):
1075
1076 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1077
1078         Fix the Qt build
1079
1080         Unreviewed build fix.
1081
1082         * Scripts/webkitdirs.pm:
1083         (buildQMakeProjects):
1084
1085 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1086
1087         [Qt] Make build-webkit always do safe incremental builds after update-webkit
1088
1089         When building against Qt5 with GCC we would just run 'make' in the build dir,
1090         and rely on the dependency files output by GCC (-MD) to handle dependency
1091         tracking, but that fails for special-cases like adding a Q_OBJECT macro
1092         to a header.
1093
1094         To guarantee that an incrmental build will work, we have to run 'make qmake',
1095         which we now do on every build-webkit that's followed by a successful run
1096         of update-webkit. The reasoning is that update-webkit can result in such
1097         potential corner-cases being applied, and since we can't know for sure
1098         unless we inspect the diff and account for all the corner cases we assume
1099         the worst and always run 'make qmake'.
1100
1101         After a succesful run of build-webkit we proceed to do just 'make' for any
1102         subsequent runs, since we assume that the developer knows what kind of
1103         changes he/she is doing, and when a 'make qmake' is needed.
1104
1105         Reviewed by Simon Hausmann.
1106
1107         * Scripts/update-webkit:
1108         * Scripts/webkitdirs.pm:
1109         (determineCurrentSVNRevision):
1110         (currentSVNRevision):
1111         (buildQMakeProjects):
1112         * qmake/mkspecs/features/default_post.prf:
1113
1114 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
1115
1116         run-perf-tests should support Skipped list
1117         https://bugs.webkit.org/show_bug.cgi?id=76594
1118
1119         Reviewed by Adam Barth.
1120
1121         Add a support for Skipped list in run-perf-tests; also skip files in resources directories.
1122
1123         * Scripts/webkitpy/layout_tests/port/base.py:
1124         (Port.perf_tests_dir):
1125         (Port.skipped_perf_tests):
1126         (Port.skips_perf_test):
1127         (Port.relative_perf_test_filename):
1128         * Scripts/webkitpy/layout_tests/port/test.py:
1129         (TestPort.perf_tests_dir):
1130         * Scripts/webkitpy/layout_tests/port/webkit.py:
1131         (WebKitPort._expectations_from_skipped_files):
1132         (WebKitPort):
1133         (WebKitPort.skipped_layout_tests):
1134         (WebKitPort.skipped_perf_tests):
1135         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1136         (PerfTestsRunner._collect_tests):
1137         (PerfTestsRunner.run):
1138         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1139         (test_run_test_set_with_json_output):
1140         (test_run_test_set_with_json_source):
1141         (test_collect_tests_with_skipped_list):
1142         (test_collect_tests_with_skipped_list.add_file):
1143
1144 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
1145
1146         Fix path to chromium_src_dir introduced in previous change.
1147
1148         Unreviewed, build fix.
1149
1150         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1151
1152 2012-01-18  Alexey Proskuryakov  <ap@apple.com>
1153
1154         Need infrastructure to test Content-Disposition filename encoding support
1155         https://bugs.webkit.org/show_bug.cgi?id=76572
1156         <rdar://problem/6421825>
1157
1158         Reviewed by Sam Weinig.
1159
1160         Made policy delegate log suggested file name when resource has Content-Disposition: attachment.
1161
1162         WebKitTestRunner does not have a policy delegate yet, bug 42546.
1163
1164         * DumpRenderTree/mac/PolicyDelegate.mm:
1165         (dispositionTypeFromContentDispositionHeader):
1166         (-[PolicyDelegate webView:decidePolicyForMIMEType:request:frame:decisionListener:]):
1167
1168 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1169
1170         Fix tests from http://trac.webkit.org/changeset/105332.
1171
1172         The code assumed that if you passed an options object with a configuration
1173         attribute that it would have a valid value. A bunch of the testing mocks
1174         do things like having a configuration value of None.
1175
1176         * Scripts/webkitpy/layout_tests/port/base.py:
1177         (Port.__init__):
1178
1179 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1180
1181         Shave 0.5 seconds off check-webkit-style runtime for test_expectations.txt
1182         https://bugs.webkit.org/show_bug.cgi?id=76576
1183
1184         Reviewed by Adam Barth.
1185
1186         Avoid calling default_configuration in base.py. We don't need it for
1187         checking test_expectations.txt style. It takes ~0.5 seconds to run on my
1188         Mac Pro. It's the call to "perl Tools/Scripts/webkit-build-directory --top-level"
1189         from common.executive.
1190
1191         At some point someone should probably look into why that call is so slow
1192         since it's on the critical path for run-webkit-tests and build-webkit startup.
1193
1194         * Scripts/webkitpy/layout_tests/port/base.py:
1195         (Port.__init__):
1196         * Scripts/webkitpy/style/checkers/test_expectations.py:
1197         (TestExpectationsChecker._determine_port_from_exepectations_path):
1198
1199 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
1200
1201         [chromium] move Tools.gyp, switch build-webkit --chromium to All.gyp
1202         https://bugs.webkit.org/show_bug.cgi?id=76505
1203
1204         Reviewed by Tony Chang.
1205
1206         Update chromium build script to use All.gyp instead of WebKit.gyp
1207
1208         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp: Renamed from Source/WebKit/chromium/Tools.gyp.
1209         * Scripts/webkitdirs.pm:
1210
1211 2012-01-18  Sam Weinig  <sam@webkit.org>
1212
1213         Don’t hardcode the path to ‘make‘.
1214
1215         Reviewed by Dan Bernstein.
1216
1217         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1218
1219 2012-01-18  Joshua Bell  <jsbell@chromium.org>
1220
1221         Adding myself to committers.py
1222         https://bugs.webkit.org/show_bug.cgi?id=76569
1223
1224         Unreviewed.
1225
1226         * Scripts/webkitpy/common/config/committers.py:
1227
1228 2012-01-18  Adam Barth  <abarth@webkit.org>
1229
1230         Up the cap on test failures tolerated by the EWS so the mac-ews can run to completion
1231         https://bugs.webkit.org/show_bug.cgi?id=76567
1232
1233         Reviewed by Eric Seidel.
1234
1235         Current the AppleMac port has slightly more failures than we allow on
1236         the EWS, causing the mac-ews to spin.  This patch ups the limit a bit,
1237         which should hopefully let the mac-ews run to completion.  Once
1238         AppleMac has fewer failures, we can lower the cap again.
1239
1240         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
1241         (LayoutTestResultsReaderTest.test_layout_test_results):
1242         * Scripts/webkitpy/tool/steps/runtests.py:
1243         (RunTests):
1244         * Scripts/webkitpy/tool/steps/steps_unittest.py:
1245
1246 2012-01-18  Eric Carlson  <eric.carlson@apple.com>
1247
1248         Provide access to user's list of preferred languages
1249         https://bugs.webkit.org/show_bug.cgi?id=76138
1250
1251         Reviewed by Alexey Proskuryakov.
1252
1253         * GNUmakefile.am: Add JSInternalsCustom.cpp.
1254
1255 2012-01-18  Adam Barth  <abarth@webkit.org>
1256
1257         Add Chromium Linux GPU 32 to the list of allowed Chromium Linux
1258         configurations.
1259
1260         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1261         (ChromiumLinuxPort.__init__):
1262
1263 2012-01-18  Dominic Mazzoni  <dmazzoni@google.com>
1264
1265         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
1266         https://bugs.webkit.org/show_bug.cgi?id=73460
1267
1268         Reviewed by Chris Fleizach.
1269
1270         * DumpRenderTree/AccessibilityUIElement.h:
1271         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1272         (AccessibilityUIElement::AccessibilityUIElement):
1273         (AccessibilityUIElement::scrollToMakeVisibleCallback):
1274         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
1275         (AccessibilityUIElement::scrollToGlobalPointCallback):
1276         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1277         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1278         (AccessibilityUIElement::scrollToMakeVisible):
1279         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1280         (AccessibilityUIElement::scrollToGlobalPoint):
1281         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1282         (AccessibilityUIElement::scrollToMakeVisible):
1283         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1284         (AccessibilityUIElement::scrollToGlobalPoint):
1285         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1286         (AccessibilityUIElement::scrollToMakeVisible):
1287         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1288         (AccessibilityUIElement::scrollToGlobalPoint):
1289
1290 2012-01-17  Enrica Casucci  <enrica@apple.com>
1291
1292         Missing NSForegroundColorAttributeName should be treated as black instead of transparent.
1293         https://bugs.webkit.org/show_bug.cgi?id=76490
1294         <rdar://problem/9460733>
1295
1296         Added TestWebKitAPI test.
1297         
1298         Reviewed by Dan Bernstein.
1299
1300         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added new test file.
1301         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Added test that calls
1302         changeAttributes to the WebView.
1303         
1304 2012-01-18  Balazs Kelemen  <kbalazs@webkit.org>
1305
1306         [Qt] Consolidate layout test crash logging
1307         https://bugs.webkit.org/show_bug.cgi?id=75088
1308
1309         Reviewed by Simon Hausmann.
1310
1311         Move backtrace generating logic into WTFReportBacktrace
1312         and add a way to deinstall signal handlers if we know
1313         that we have already printed the backtrace.
1314
1315         * DumpRenderTree/qt/main.cpp:
1316         (crashHandler):
1317         (setupSignalHandlers):
1318         (WTFCrashHook):
1319         (main):
1320         * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
1321         (WTR::crashHandler):
1322         (WTR::setupSignalHandlers):
1323         (WTR::crashHook):
1324         (WTR::InjectedBundle::platformInitialize):
1325
1326 2012-01-18  Ilya Tikhonovsky  <loislo@chromium.org>
1327
1328         Unreviewed build fix after r105256.
1329
1330         There was a cyclic dependency between self._port and self._host assigments.
1331
1332         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1333         (PerfTestsRunner.__init__):
1334
1335 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1336
1337         Add new watchlist rule for the Qt build system
1338
1339         Reviewed by Simon Hausmann.
1340
1341         * Scripts/webkitpy/common/config/watchlist:
1342
1343 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1344
1345         [Qt] Move OTHER_FILES from WebKit.pro to Tools.pro
1346
1347         Reviewed by Simon Hausmann.
1348
1349         * Tools.pro:
1350
1351 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1352
1353         [Qt] Allow building only a subset of the WebKit sub-projects
1354
1355         Fixes build-jsc, broken in r104825.
1356
1357         We no longer need the intermediate buildQMakeQtProject(), as we're the
1358         only port using qmake, and buildQMakeProject is now highly Qt specific.
1359
1360         <http://webkit.org/b/76179>
1361
1362         Reviewed by Simon Hausmann.
1363
1364         * Scripts/build-jsc:
1365         * Scripts/build-webkit:
1366         * Scripts/webkitdirs.pm:
1367         (buildQMakeProjects):
1368
1369 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
1370
1371         run-perf-tests should generate a json file that summaries the result
1372         https://bugs.webkit.org/show_bug.cgi?id=76504
1373
1374         Reviewed by Adam Barth.
1375
1376         Add an ability to generate a json file to run-perf-test in the preparation for perf bots.
1377         New option --outout-json-path specifies the json file's path, and --source-json-path specifies
1378         another json file to be merged into the generated json file.
1379
1380         Also fixed a bug that --build wasn't set by default.
1381
1382         * Scripts/webkitpy/layout_tests/port/test.py:
1383         (TestPort.webkit_base):
1384         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1385         (PerfTestsRunner.__init__):
1386         (PerfTestsRunner):
1387         (PerfTestsRunner._parse_args):
1388         (PerfTestsRunner.run):
1389         (PerfTestsRunner._generate_json_if_specified):
1390         (PerfTestsRunner._process_chromium_style_test_result):
1391         (PerfTestsRunner._process_parser_test_result):
1392         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1393         (MainTest):
1394         (create_runner):
1395         (test_run_test_set):
1396         (test_run_test_set_for_parser_tests):
1397         (test_run_test_set_with_summary_json):
1398         (test_collect_tests):
1399         (test_parse_args):
1400
1401 2012-01-17  Sheriff Bot  <webkit.review.bot@gmail.com>
1402
1403         Unreviewed, rolling out r105244.
1404         http://trac.webkit.org/changeset/105244
1405         https://bugs.webkit.org/show_bug.cgi?id=76518
1406
1407         broke Chromium Mac (Requested by rolandsteiner on #webkit).
1408
1409         * DumpRenderTree/AccessibilityUIElement.h:
1410         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1411         (AccessibilityUIElement::AccessibilityUIElement):
1412         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1413         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1414         (AccessibilityUIElement::removeSelection):
1415         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1416         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1417         (AccessibilityUIElement::removeSelection):
1418
1419 2012-01-17  Dominic Mazzoni  <dmazzoni@google.com>
1420
1421         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
1422         https://bugs.webkit.org/show_bug.cgi?id=73460
1423
1424         Reviewed by Chris Fleizach.
1425
1426         * DumpRenderTree/AccessibilityUIElement.h:
1427         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1428         (AccessibilityUIElement::AccessibilityUIElement):
1429         (AccessibilityUIElement::scrollToMakeVisibleCallback):
1430         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
1431         (AccessibilityUIElement::scrollToGlobalPointCallback):
1432         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1433         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1434         (AccessibilityUIElement::scrollToMakeVisible):
1435         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1436         (AccessibilityUIElement::scrollToGlobalPoint):
1437         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1438         (AccessibilityUIElement::scrollToMakeVisible):
1439         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1440         (AccessibilityUIElement::scrollToGlobalPoint):
1441         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1442         (AccessibilityUIElement::scrollToMakeVisible):
1443         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1444         (AccessibilityUIElement::scrollToGlobalPoint):
1445
1446 2012-01-17  Adam Barth  <abarth@webkit.org>
1447
1448         Enable tests on the mac-ews
1449         https://bugs.webkit.org/show_bug.cgi?id=76502
1450
1451         Reviewed by Simon Fraser.
1452
1453         Apple now has five machines running the mac-ews.  Lucas Forschler has
1454         asked that we try enabling tests to see if we have enough bandwidth.
1455
1456         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1457         (MacEWS):
1458
1459 2012-01-17  Sanjoy Das  <sanjoy@playingwithpointers.com>
1460
1461         Add a check-webkit-style exception for GDBInterface.cpp
1462         https://bugs.webkit.org/show_bug.cgi?id=76187
1463
1464         Add an exception for Source/JavaScriptCore/jit/GDBInterface.cpp so
1465         that it does not fail the style check.
1466
1467         Reviewed by Martin Robinson.
1468
1469         * Scripts/webkitpy/style/checker.py: Tell the style checker to not run
1470         readability/naming tests on
1471         Source/JavaScriptCore/jit/GDBInterface.cpp.
1472         * Scripts/webkitpy/style/checker_unittest.py:
1473         (GlobalVariablesTest.test_path_rules_specifier): Add a test for
1474         the above exception.
1475
1476 2012-01-17  Hajime Morrita  <morrita@chromium.org>
1477
1478         https://bugs.webkit.org/show_bug.cgi?id=76245
1479         [Mac][DRT] should allow query string for the file URL.
1480
1481         Reviewed by Tony Chang.
1482
1483         Taught DRT about file:// as a url instead of a file path.
1484         We are now able to use a path file:///foo/bar.html?baz as a url
1485         with which DRT opens "/foo/bar.html" with "baz" as a query string.
1486
1487         * DumpRenderTree/mac/DumpRenderTree.mm:
1488         (runTest):
1489
1490 2012-01-17  Ojan Vafai  <ojan@chromium.org>
1491
1492         Check style on test_expectations.txt files before commit
1493         https://bugs.webkit.org/show_bug.cgi?id=76484
1494
1495         Reviewed by Adam Barth.
1496
1497         Unlike other style rules, if you get the syntax of the test_expectations.txt
1498         file wrong, the layout tests won't run. Also, this check is simple and only
1499         slows down committing if you actually modify one of the test_expectations.txt files.
1500
1501         * Scripts/webkitpy/tool/steps/commit.py:
1502         (Commit._check_test_expectations):
1503         (Commit.run):
1504         * Scripts/webkitpy/tool/steps/commit_unittest.py: Added.
1505         (CommitTest):
1506         (CommitTest.test_check_test_expectations):
1507
1508 2012-01-17  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
1509
1510         [Qt] Debug build fails with debug qt5
1511         https://bugs.webkit.org/show_bug.cgi?id=76463
1512
1513         Reviewed by Simon Hausmann.
1514
1515         Add QNetworkCookie include in order to satisfy sizeof in QTypeInfo in debug builds.
1516
1517         * QtTestBrowser/cookiejar.h:
1518
1519 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1520
1521         webkitpy: fix json import on linux
1522         https://bugs.webkit.org/show_bug.cgi?id=76481
1523
1524         Reviewed by Adam Barth.
1525
1526         For some reason, the version of simplejson autoinstalled on my linux
1527         box seems to be broken. We should really only be using
1528         simplejson if json isn't available anyway, so this change looks
1529         for json first.
1530
1531         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1532         (JSONGeneratorTest.test_test_timings_trie):
1533
1534 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1535
1536         Fix failures in test-webkitpy caused by r105177.
1537
1538         Unreviewed, build fix.
1539
1540         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1541         (ChromiumLinuxPort._determine_driver_path_statically):
1542         (ChromiumLinuxPort._static_build_path):
1543
1544 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1545
1546         webkitpy: make PortFactory.get() be fully data-driven
1547         https://bugs.webkit.org/show_bug.cgi?id=76360
1548
1549         Reviewed by Eric Seidel.
1550
1551         Refactor PortFactory.get() to be fully data-driven.
1552
1553         * Scripts/webkitpy/layout_tests/port/factory.py:
1554         (PortFactory):
1555         (PortFactory._default_port):
1556         (PortFactory.get):
1557         (PortFactory.get.in):
1558         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1559         (FactoryTest.assert_port):
1560
1561 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1562
1563         webkitpy: add determine_full_port_name(), clean up port.__init__()
1564         https://bugs.webkit.org/show_bug.cgi?id=76357
1565
1566         Reviewed by Adam Barth.
1567
1568         Each Port class is now required to implement a 'port_name' field
1569         that will match the start of all ports constructed by this class and
1570         a 'determine_full_port_name' field that will make the port
1571         fully-qualified.
1572
1573         * Scripts/webkitpy/layout_tests/port/apple.py:
1574         (ApplePort.__init__):
1575         * Scripts/webkitpy/layout_tests/port/base.py:
1576         (Port):
1577         (Port.determine_port_name):
1578         (Port.__init__):
1579         * Scripts/webkitpy/layout_tests/port/chromium.py:
1580         (ChromiumPort.for):
1581         (ChromiumPort.__init__):
1582         (ChromiumPort):
1583         (ChromiumPort._chromium_base_dir):
1584         (ChromiumPort.path_from_chromium_base):
1585         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
1586         (ChromiumGpuLinuxPort):
1587         (ChromiumGpuLinuxPort.__init__):
1588         (ChromiumGpuMacPort):
1589         (ChromiumGpuMacPort.__init__):
1590         (ChromiumGpuWinPort):
1591         (ChromiumGpuWinPort.__init__):
1592         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1593         (ChromiumLinuxPort):
1594         (ChromiumLinuxPort.determine_port_name):
1595         (ChromiumLinuxPort.__init__):
1596         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
1597         (ChromiumMacPort):
1598         (ChromiumMacPort.determine_port_name):
1599         (ChromiumMacPort.__init__):
1600         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
1601         (ChromiumWinPort):
1602         (ChromiumWinPort.determine_port_name):
1603         (ChromiumWinPort.__init__):
1604         * Scripts/webkitpy/layout_tests/port/dryrun.py:
1605         (DryRunPort):
1606         (DryRunPort.determine_port_name):
1607         (DryRunPort.__init__):
1608         * Scripts/webkitpy/layout_tests/port/efl.py:
1609         (EflPort):
1610         * Scripts/webkitpy/layout_tests/port/factory.py:
1611         (PortFactory):
1612         (PortFactory._default_port):
1613         (PortFactory.get):
1614         (PortFactory.get.in):
1615         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1616         (FactoryTest.assert_port):
1617         * Scripts/webkitpy/layout_tests/port/google_chrome.py:
1618         (GoogleChromeLinux32Port):
1619         (GoogleChromeLinux32Port.determine_port_name):
1620         (GoogleChromeLinux64Port):
1621         (GoogleChromeLinux64Port.determine_port_name):
1622         (GoogleChromeMacPort):
1623         (GoogleChromeMacPort.determine_port_name):
1624         (GoogleChromeWinPort):
1625         (GoogleChromeWinPort.determine_port_name):
1626         * Scripts/webkitpy/layout_tests/port/gtk.py:
1627         (GtkPort):
1628         * Scripts/webkitpy/layout_tests/port/mac.py:
1629         (MacPort):
1630         (MacPort.determine_port_name):
1631         (MacPort.__init__):
1632         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1633         (test_versions):
1634         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
1635         (MockDRTPort):
1636         (MockDRTPort.determine_port_name):
1637         (MockDRTPort.__init__):
1638         (MockDRT):
1639         (MockDRT.determine_port_name):
1640         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1641         (PortTestCase.make_port):
1642         * Scripts/webkitpy/layout_tests/port/qt.py:
1643         (QtPort):
1644         (QtPort.determine_port_name):
1645         (QtPort.__init__):
1646         * Scripts/webkitpy/layout_tests/port/test.py:
1647         (TestPort):
1648         (TestPort.determine_port_name):
1649         (TestPort.__init__):
1650         * Scripts/webkitpy/layout_tests/port/webkit.py:
1651         (WebKitPort.__init__):
1652         * Scripts/webkitpy/layout_tests/port/win.py:
1653         (WinPort):
1654         (WinPort.determine_port_name):
1655         * Scripts/webkitpy/tool/servers/rebaselineserver.py:
1656         (get_test_baselines.AllPlatformsPort.__init__):
1657         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
1658         (get_test_config.TestMacPort):
1659         (get_test_config):
1660
1661 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1662
1663         webkitpy: clean up port code in preparation for static port names
1664         https://bugs.webkit.org/show_bug.cgi?id=76356
1665
1666         Reviewed by Adam Barth.
1667
1668         To fix bug 76215 and be able to determine appropriate port names
1669         in webkitpy without actually constructing Port objects, we need
1670         to shuffle some logic in the chromium ports to be able to figure
1671         out whether we built DRT in 32 or 64-bit mode.
1672
1673         * Scripts/webkitpy/layout_tests/port/chromium.py:
1674         (ChromiumPort):
1675         (ChromiumPort._chromium_base_dir):
1676         (ChromiumPort.__init__):
1677         (ChromiumPort.path_from_chromium_base):
1678         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1679         (ChromiumLinuxPort):
1680         (ChromiumLinuxPort._determine_driver_path_statically):
1681         (ChromiumLinuxPort._static_build_path):
1682         (ChromiumLinuxPort._determine_architecture):
1683         (ChromiumLinuxPort.__init__):
1684         (ChromiumLinuxPort._build_path):
1685
1686 2012-01-17  Hao Zheng  <zhenghao@chromium.org>
1687
1688         Do not remove pixel failures for ref tests.
1689         https://bugs.webkit.org/show_bug.cgi?id=76243
1690
1691         Reviewed by Tony Chang.
1692
1693         'NRWT --no-pixel-tests' complains when reftests are expected to
1694         be image mismatch.
1695
1696         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1697         (Manager._update_summary_with_result):
1698         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1699         (is_reftest_failure):
1700         (determine_result_type):
1701
1702 2012-01-17  Balazs Ankes  <Ankes.Balazs@stud.u-szeged.hu>
1703
1704         REGRESSION(r81225): ORWT should ignore reftests
1705         https://bugs.webkit.org/show_bug.cgi?id=67936
1706
1707         Reviewed by Csaba Osztrogonác.
1708
1709         * Scripts/old-run-webkit-tests:
1710         (isUsedInReftest): $filename should be the first parameter
1711         (findTestsToRun): run test if it isn't reftest
1712
1713 2012-01-17  Gustavo Noronha Silva  <gns@gnome.org>
1714
1715         Unreviewed. Fix python unit test I broke with r105142
1716
1717         * Scripts/webkitpy/common/config/ports_unittest.py:
1718         (WebKitPortTest.test_gtk_port):
1719
1720 2012-01-17  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1721
1722         [Qt] Don't set the 'primary' flag when mocking touch-points in MiniBrowser
1723
1724         Qt 5 no longer has that flag.
1725
1726         Reviewed by Simon Hausmann.
1727
1728         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1729         (MiniBrowserApplication::notify):
1730
1731 2012-01-16  Gustavo Noronha Silva  <gustavo.noronha@collabora.com>
1732
1733         GTK+ EWS needs to run update-webkitgtk-libs after applying a patch
1734         https://bugs.webkit.org/show_bug.cgi?id=75857
1735
1736         Reviewed by Adam Barth.
1737
1738         * Scripts/webkitdirs.pm:
1739         (buildAutotoolsProject): move cleaning the build to happen early,
1740         and handle --update-gtk right after that
1741         * Scripts/webkitpy/common/config/ports.py:
1742         (GtkPort.build_webkit_command): pass --update-gtk as an option to
1743         build-webkit, so that it is used in the EWS
1744
1745 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1746
1747         [Qt] Don't add _debug postfix to targets when building debug only
1748
1749         Reviewed by Simon Hausmann.
1750
1751         * qmake/mkspecs/features/default_post.prf:
1752         * qmake/mkspecs/features/functions.prf:
1753
1754 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1755
1756         [Qt] Add more files to project-files for easier navigation in Qt Creator
1757
1758         Reviewed by Simon Hausmann.
1759
1760         * Tools.pro:
1761         * qmake/config.tests/README: Added.
1762
1763 2012-01-17  Philippe Normand  <pnormand@igalia.com>
1764
1765         [GTK] fast/dom/constructed-objects-prototypes.html fails
1766         https://bugs.webkit.org/show_bug.cgi?id=75432
1767
1768         Reviewed by Martin Robinson.
1769
1770         * Scripts/build-webkit: Enable video-track in GTK.
1771
1772 2012-01-16  Takashi Toyoshima  <toyoshim@chromium.org>
1773
1774         Add toyoshim as a committer.
1775         https://bugs.webkit.org/show_bug.cgi?id=76390
1776
1777         Reviewed by Hajime Morita.
1778
1779         * Scripts/webkitpy/common/config/committers.py:
1780
1781 2012-01-16  Brent Fulgham  <bfulgham@webkit.org>
1782
1783         [WinCairo] Build fix after r104155. Several debug targets were not
1784         properly including the '*Common' property sheets.
1785
1786         * DumpRenderTree/win/DumpRenderTreeLauncherDebugCairoCFLite.vsprops:
1787         * DumpRenderTree/win/ImageDiffLauncherDebugCairoCFLite.vsprops:
1788         * MiniBrowser/Configurations/MiniBrowserLauncherDebugCairoCFLite.vsprops:
1789         * WebKitTestRunner/win/WebKitTestRunnerLauncherDebugCairoCFLite.vsprops:
1790         * WinLauncher/WinLauncherLauncherDebugCairoCFLite.vsprops:
1791
1792 2012-01-16  Csaba Osztrogonác  <ossy@webkit.org>
1793
1794         REGRESSION(r105072): It broke committers_unittest.py
1795         https://bugs.webkit.org/show_bug.cgi?id=76396
1796
1797         Reviewed by Adam Barth.
1798
1799         * Scripts/webkitpy/common/config/committers_unittest.py:
1800         (CommittersTest._assert_fuzz_match): It should handle more names.
1801         (CommittersTest.test_contributors_by_fuzzy_match_with_legacy_names): Fix the testcase.
1802
1803 2012-01-16  Mario Sanchez Prada  <msanchez@igalia.com>
1804
1805         [GTK] Include gtk+3 in the jhbuild modules
1806         https://bugs.webkit.org/show_bug.cgi?id=76284
1807
1808         Reviewed by Philippe Normand.
1809
1810         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
1811
1812 2012-01-16  Dan Winship  <danw@gnome.org>
1813
1814         Add myself as a committer
1815
1816         Unreviewed
1817
1818         * Scripts/webkitpy/common/config/committers.py:
1819
1820 2012-01-16  Raphael Kubo da Costa  <kubo@profusion.mobi>
1821
1822         Unreviewed; fix the build when --as-needed and --no-copy-dt-needed-entries are passed to the linker.
1823
1824         Some recent versions of a few Linux distributions have started passing
1825         --as-needed and --no-copy-dt-needed-entries by default to the linker,
1826         which broke the build when SHARED_CORE was turned on, as libwtf_efl.so
1827         needed a few symbols from libjavascriptcore_efl.so.
1828
1829         * DumpRenderTree/efl/CMakeLists.txt: Explicitly pass WTF_LIBRARY_NAME
1830         before JavaScriptCore_LIBRARY_NAME in DumpRenderTree_LIBRARIES so the
1831         libraries are searched in the right order by the linker.
1832
1833 2012-01-16  Philippe Normand  <pnormand@igalia.com>
1834
1835         [GTK] set audio resources path in NRTW
1836         https://bugs.webkit.org/show_bug.cgi?id=76380
1837
1838         Reviewed by Martin Robinson.
1839
1840         * Scripts/webkitpy/layout_tests/port/gtk.py:
1841         (GtkPort.setup_environ_for_server): Set $AUDIO_RESOURCES_PATH so
1842         the uninstalled audio resources are used during webaudio tests.
1843
1844 2012-01-16  Carlos Garcia Campos  <cgarcia@igalia.com>
1845
1846         [GTK] ldflags and cflags should take precedence over the existing flags in gtkdoc.py
1847         https://bugs.webkit.org/show_bug.cgi?id=76369
1848
1849         Reviewed by Xan Lopez.
1850
1851         If there's an installed version of webkit and there are new
1852         symbols in the build tree, gtkdoc-scanobj fails due to undefined
1853         references, because the env ldflags are given before the local
1854         ones.
1855
1856         * gtk/gtkdoc.py:
1857         (GTKDoc._run_gtkdoc_scangobj):
1858
1859 2012-01-15  Pablo Flouret  <pablof@motorola.com>
1860
1861         Fix compilation errors on build-webkit --debug --no-video on mac.
1862         https://bugs.webkit.org/show_bug.cgi?id=75867
1863
1864         Reviewed by Philippe Normand.
1865
1866         Add --[no-]video-track to build-webkit, since video track feature
1867         depends on video.
1868
1869         * Scripts/build-webkit:
1870
1871 2012-01-14  James Robinson  <jamesr@chromium.org>
1872
1873         [chromium] Failing webkit_unit_tests does not make the cr-linux EWS bubble red
1874         https://bugs.webkit.org/show_bug.cgi?id=76313
1875
1876         Reviewed by Adam Barth.
1877
1878         The exit code of this script should be the exit code of webkit_unit_tests. Tested manually.
1879
1880         * Scripts/run-chromium-webkit-unit-tests:
1881
1882 2012-01-14  Sheriff Bot  <webkit.review.bot@gmail.com>
1883
1884         Unreviewed, rolling out r105017.
1885         http://trac.webkit.org/changeset/105017
1886         https://bugs.webkit.org/show_bug.cgi?id=76333
1887
1888         broke the build (Requested by philn-tp on #webkit).
1889
1890         * gtk/jhbuild.modules:
1891
1892 2012-01-14  Mario Sanchez Prada  <msanchez@igalia.com>
1893
1894         [GTK] Include gtk+3 in the jhbuild modules
1895         https://bugs.webkit.org/show_bug.cgi?id=76284
1896
1897         Reviewed by Philippe Normand.
1898
1899         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
1900
1901 2012-01-13  Adam Barth  <abarth@webkit.org>
1902
1903         Remove --dry-run support from webkit-patch
1904         https://bugs.webkit.org/show_bug.cgi?id=76300
1905
1906         Reviewed by Dirk Pranke.
1907
1908         This is the rest of the patch from Bug 76300.  I landed it separtely
1909         because main.py had some stray conflict markers that I needed to clean
1910         up first.
1911
1912         * Scripts/webkitpy/test/main.py:
1913         (Tester.parse_args):
1914         (Tester._run_tests):
1915
1916 2012-01-13  Adam Barth  <abarth@webkit.org>
1917
1918         test-webkitpy does not run
1919         https://bugs.webkit.org/show_bug.cgi?id=76318
1920
1921         Unreviewed.
1922
1923         Looks like some conflict markers got left in the file.
1924
1925         * Scripts/webkitpy/test/main.py:
1926         (Tester.run):
1927
1928 2012-01-13  Adam Barth  <abarth@webkit.org>
1929
1930         Remove --dry-run support from webkit-patch
1931         https://bugs.webkit.org/show_bug.cgi?id=76300
1932
1933         Reviewed by Dirk Pranke.
1934
1935         We don't use this functionality for anything, and it doesn't really
1936         work.  Someone in #webkit tried using it and and got sad/confused.  We
1937         should just remove it.
1938
1939         * Scripts/webkitpy/common/checkout/scm/git.py:
1940         (Git.push_local_commits_to_server):
1941         * Scripts/webkitpy/common/checkout/scm/scm.py:
1942         (SCM.__init__):
1943         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
1944         (_shared_test_commit_with_message):
1945         (GitSVNTest.test_commit_text_parsing):
1946         * Scripts/webkitpy/common/checkout/scm/svn.py:
1947         (SVN.commit_with_message):
1948         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
1949         (Bugzilla.__init__):
1950         (Bugzilla.quips):
1951         (Bugzilla.authenticate):
1952         (Bugzilla.add_attachment_to_bug):
1953         (Bugzilla.add_patch_to_bug):
1954         (Bugzilla.create_bug):
1955         (Bugzilla.clear_attachment_flags):
1956         (Bugzilla.set_flag_on_attachment):
1957         (Bugzilla.obsolete_attachment):
1958         (Bugzilla.add_cc_to_bug):
1959         (Bugzilla.post_comment_to_bug):
1960         (Bugzilla.close_bug_as_fixed):
1961         (Bugzilla.reassign_bug):
1962         (reopen_bug):
1963         * Scripts/webkitpy/test/main.py:
1964         (Tester.parse_args):
1965         (Tester.run_tests):
1966
1967 2012-01-13  David Levin  <levin@chromium.org>
1968
1969         check-webkit-style: should encourage the use of Own* classes for Windows DC.
1970         https://bugs.webkit.org/show_bug.cgi?id=76227
1971
1972         Reviewed by Dirk Pranke.
1973
1974         * Scripts/webkitpy/style/checkers/cpp.py:
1975         (check_for_leaky_patterns): The new check.
1976         (process_line): Added a call to the new check.
1977         (CppChecker): Added the new error type.
1978         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
1979         (CppStyleTestBase):
1980         (CppStyleTestBase.perform_leaky_pattern_check):
1981         The check for only leaky pattern errors.
1982         (LeakyPatternTest): Test cases.
1983
1984 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
1985
1986         test-webkitpy: clean up handling of test directories, QueueStatusServer
1987         https://bugs.webkit.org/show_bug.cgi?id=76238
1988
1989         Reviewed by Adam Barth.
1990
1991         This patch cleans up a bunch of hacks related to how we find all
1992         of the python unit tests and set up the sys.path variable in
1993         test-webkitpy.
1994
1995         This patch changes how test-webkitpy works slightly in that we
1996         will now look for test files under Tools/Scripts and Source/WebKit2/Scripts
1997         rather than just Tools/Scripts/webkitpy and Source/WebKit2/Scripts/webkit.
1998         This greatly simplifies the mapping of paths to module names and
1999         appears to have no significant performance impact or other drawbacks.
2000
2001         There are other ways that we could remove all of the hacks and
2002         still have a more flexible mapping of paths to module names, but
2003         the ones I've thought of so far add a decent amount of complexity to
2004         the implementation and don't seem to buy us much.
2005         
2006         * QueueStatusServer/__init__.py:
2007         * Scripts/test-webkitpy:
2008         * Scripts/webkitpy/test/main.py:
2009         (Tester._modules_from_paths):
2010         (Tester._is_module):
2011
2012 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2013
2014         test-webkitpy: should support classes and individual test names as well as modules
2015         https://bugs.webkit.org/show_bug.cgi?id=76233
2016
2017         Reviewed by Adam Barth.
2018
2019         The default unittest.main() supports a list of individual
2020         TestCase classes or individual tests as well as modules. This
2021         change adds support for the same to test-webkitpy, so now you
2022         can say
2023
2024           % test-wekbitpy webkitpy.test.cat_unittest.CatTest
2025
2026         and
2027
2028           % test-wekbitpy webkitpy.test.cat_unittest.CatTest.test_basic
2029
2030         in addition to
2031
2032           % test-wekbitpy webkitpy.test.cat_unittest
2033
2034         * Scripts/test-webkitpy:
2035         * Scripts/webkitpy/test/cat_unittest.py:
2036         * Scripts/webkitpy/test/main.py:
2037         (Tester.parse_args):
2038         (Tester.run):
2039         (Tester):
2040         (Tester._find_modules):
2041         (Tester._exclude):
2042         (Tester._run_tests):
2043         (Tester._is_module):
2044         (Tester._log_exception):
2045
2046 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2047
2048         test-webkitpy: add basic code coverage support
2049         https://bugs.webkit.org/show_bug.cgi?id=76201
2050
2051         Reviewed by Adam Barth.
2052
2053         This test adds basic code coverage support to test-webkitpy; if
2054         you pass the -c flag, we will generate a new coverage file. We
2055         can start simply with this and add more features as necessary.
2056
2057         This patch requires coverage to be already installed on the
2058         machine. I am not auto-installing this for now.
2059
2060         * Scripts/webkitpy/test/main.py:
2061         (Tester.parse_args):
2062         (Tester.run_tests):
2063
2064 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2065
2066         StatusBubbleTest.test_build_bubble is failing
2067         https://bugs.webkit.org/show_bug.cgi?id=76222
2068
2069         Reviewed by Adam Barth.
2070
2071         * QueueStatusServer/handlers/statusbubble_unittest.py:
2072         (StatusBubbleTest.test_build_bubble):
2073
2074 2012-01-13  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
2075
2076         nrwt should use multiple http shards
2077         https://bugs.webkit.org/show_bug.cgi?id=75958
2078
2079         Add --max-locked-shards option to the nrwt for manual set
2080         the maximum number of the locked shards
2081
2082         Reviewed by Dirk Pranke.
2083
2084         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2085         (Manager._max_locked_shards):
2086         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2087         (ShardingTests.get_shards):
2088         (ShardingTests.test_multiple_locked_shards):
2089         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2090         (parse_args):
2091
2092 2012-01-13  Raphael Kubo da Costa  <kubo@profusion.mobi>
2093
2094         prepare-ChangeLog: Look for a text editor in a few more variables.
2095         https://bugs.webkit.org/show_bug.cgi?id=76289
2096
2097         Reviewed by Ryosuke Niwa.
2098
2099         A text editor was looked for only in the CHANGE_LOG_EDITOR and
2100         CHANGE_LOG_EDIT_APPLICATION environment variables before falling back
2101         to the OS X-specific open(1) command.
2102
2103         As there does not seem to be a common way of looking for a text editor
2104         among the scripts in Tools/Scripts right now, just look at $VISUAL and
2105         $EDITOR (set much more often than the other variables in Unix systems)
2106         before falling back to open(1).
2107
2108         * Scripts/prepare-ChangeLog:
2109         (openChangeLogs):
2110
2111 2012-01-13  Gabor Rapcsanyi  <rgabor@webkit.org>
2112
2113         [GTK] gtkdoc.py ignoring LDFLAGS and CFLAGS environment variables
2114         https://bugs.webkit.org/show_bug.cgi?id=76269
2115
2116         Reviewed by Martin Robinson.
2117
2118         * gtk/gtkdoc.py:
2119         (GTKDoc._run_gtkdoc_scangobj):
2120
2121 2012-01-13  Csaba Osztrogonác  <ossy@webkit.org>
2122
2123         [Qt][WK2]REGRESSION(r104881):It broke hundreds of tests
2124         https://bugs.webkit.org/show_bug.cgi?id=76247
2125
2126         Reviewed by Kenneth Rohde Christiansen.
2127
2128         * WebKitTestRunner/TestController.cpp:
2129         (WTR::TestController::resetStateToConsistentValues): Disable mock scrollbars on Qt by default. (workaround)
2130
2131 2012-01-13  Hajime Morrita  <morrita@chromium.org>
2132
2133         [Mac][DRT] Should support --no-timeout
2134         https://bugs.webkit.org/show_bug.cgi?id=76242
2135
2136         Reviewed by Darin Adler.
2137
2138         * DumpRenderTree/mac/DumpRenderTree.mm:
2139         (initializeGlobalsFromCommandLineOptions): Added "--no-timeout" to the options list.
2140         (setWaitToDumpWatchdog): Added to encapsulate the waitToDumpWatchdog global variable.
2141         (shouldSetWaitToDumpWatchdog): Added to check the useTimeoutWatchdog flag set by the option.
2142         * DumpRenderTree/mac/DumpRenderTreeMac.h:
2143         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2144         (LayoutTestController::setWaitToDump): Now calls shouldSetWaitToDumpWatchdog() to check the flag.
2145
2146 2012-01-12  Simon Fraser  <simon.fraser@apple.com>
2147
2148         Mitigate scrollbar differences when running pixel tests
2149         https://bugs.webkit.org/show_bug.cgi?id=67217
2150
2151         Reviewed by Dan Bernstein.
2152         
2153         Enable mock scrollbars by default for tests.
2154
2155         * DumpRenderTree/mac/DumpRenderTree.mm:
2156         (resetDefaultsToConsistentValues):
2157         * WebKitTestRunner/TestController.cpp:
2158         (WTR::TestController::resetStateToConsistentValues):
2159
2160 2012-01-12  Ryosuke Niwa  <rniwa@webkit.org>
2161
2162         Need a script to run tests in PerformanceTests
2163         https://bugs.webkit.org/show_bug.cgi?id=76132
2164
2165         Reviewed by Adam Barth.
2166
2167         Add run-perf-tests to run performance tests using parser/resources/runner.js.
2168         Unfortunately, there isn't an easy way of telling which test uses which format
2169         so hard-code directory that uses Chromium perf-bot style (only inspector perf. tests for now).
2170
2171         All test outputs are re-formatted to match Chromium perf-bot style.
2172
2173         * Scripts/run-inspector-perf-tests.py:
2174         * Scripts/run-perf-tests: Added.
2175         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2176         (PerfTestsRunner):
2177         (PerfTestsRunner.__init__):
2178         (PerfTestsRunner._collect_tests):
2179         (PerfTestsRunner.run):
2180         (PerfTestsRunner._print_status):
2181         (PerfTestsRunner._run_tests_set):
2182         (PerfTestsRunner._process_chromium_style_test_result):
2183         (PerfTestsRunner._should_ignore_line_in_parser_test_result):
2184         (PerfTestsRunner._process_parser_test_result):
2185         (PerfTestsRunner._run_single_test):
2186         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2187         (MainTest.TestDriver.run_test):
2188         (create_runner):
2189         (run_test):
2190         (test_run_test_set):
2191         (test_run_test_set_for_parser_tests):
2192         (test_collect_tests):
2193
2194 2012-01-12  Dirk Pranke  <dpranke@chromium.org>
2195
2196         webkitpy: clean up port unit tests in preparation for making port_name mandatory
2197         https://bugs.webkit.org/show_bug.cgi?id=76123
2198
2199         Reviewed by Adam Barth.
2200
2201         This change updates the unit tests to be more consistent in how
2202         ports are created, in preparation for making port_name a
2203         mandatory parameter to the constructor and removing any version
2204         interpretation from it.
2205
2206         There should be no functional changes in this patch, but some
2207         tests that are no longer relevant have been removed or reworked.
2208
2209         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2210         (PortTest.test_get_option__default):
2211         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2212         (ChromiumLinuxPortTest):
2213         (ChromiumLinuxPortTest.assert_architecture):
2214         (ChromiumLinuxPortTest.test_determine_architecture_fails):
2215         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2216         (ChromiumMacPortTest):
2217         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2218         (ChromiumPortTest):
2219         (ChromiumPortTest.test_all_test_configurations):
2220         (ChromiumPortTest.TestMacPort.__init__):
2221         (ChromiumPortTest.TestLinuxPort.__init__):
2222         (ChromiumPortTest.TestWinPort.__init__):
2223         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2224         (ChromiumWinTest):
2225         * Scripts/webkitpy/layout_tests/port/efl_unittest.py:
2226         (EflPortTest):
2227         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
2228         (GtkPortTest):
2229         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2230         (MacTest):
2231         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2232         (MockDRTPortTest):
2233         (MockDRTPortTest.make_port):
2234         * Scripts/webkitpy/layout_tests/port/qt.py:
2235         (QtPort._port_flag_for_scripts):
2236         (QtPort.__init__):
2237         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2238         (QtPortTest):
2239         (QtPortTest._assert_search_path):
2240         (QtPortTest.test_baseline_search_path):
2241         (QtPortTest.test_operating_system):
2242         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2243         (WebKitPortTest):
2244         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2245         (WinPortTest):
2246         (WinPortTest._assert_search_path):
2247         (WinPortTest.test_baseline_search_path):
2248         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
2249         (get_test_config.TestMacPort.__init__):
2250
2251 2012-01-12  Dirk Pranke  <dpranke@chromium.org>
2252
2253         test-webkitpy: clean up logging and provide a real --help message
2254         https://bugs.webkit.org/show_bug.cgi?id=76142
2255
2256         Reviewed by Adam Barth.
2257
2258         This test is mostly more refactoring - it calls objects in
2259         unittest directly instead of routing through unittest.main() --
2260         in order to gain more control over the logging and to provide a
2261         correct --help message. This will also give us a better
2262         foundation for adding more feautures down the road.
2263
2264         This patch adds different levels of logging controlled by the
2265         '-s', '-q', and '-v' options (see --help for details), and
2266         removes the --verbose-logging option.
2267
2268         * Scripts/test-webkitpy:
2269         * Scripts/webkitpy/test/main.py:
2270         (Tester):
2271         (Tester.clean_packages):
2272         (Tester.__init__):
2273         (Tester.parse_args):
2274         (Tester.configure):
2275         (Tester._configure_logging):
2276         (Tester.find_modules):
2277         (Tester._exclude):
2278         (Tester._modules_from_paths):
2279         (Tester.run_tests):
2280
2281 2012-01-12  Simon Hausmann  <simon.hausmann@nokia.com>
2282
2283         Make the new WTF module build on Qt
2284         https://bugs.webkit.org/show_bug.cgi?id=76163
2285
2286         Reviewed by Tor Arne Vestbø.
2287
2288         * qmake/mkspecs/modules/wtf.prf: Pull in WTF from the new location.
2289
2290 2012-01-12  Vsevolod Vlasov  <vsevik@chromium.org>
2291
2292         Unreviewed, mac DumpRenderTree fix.
2293
2294         * DumpRenderTree/mac/UIDelegate.mm:
2295         (-[UIDelegate webView:addMessageToConsole:withSource:]):
2296
2297 2012-01-12  Vsevolod Vlasov  <vsevik@chromium.org>
2298
2299         WebKitTestRunner should not print console message line number when it is equal to zero.
2300         https://bugs.webkit.org/show_bug.cgi?id=76167
2301
2302         Reviewed by Csaba Osztrogonác.
2303
2304         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2305         (WTR::InjectedBundlePage::willAddMessageToConsole):
2306
2307 2012-01-11  Vsevolod Vlasov  <vsevik@chromium.org>
2308
2309         Make default console messages line numbers consistent.
2310         https://bugs.webkit.org/show_bug.cgi?id=74075
2311
2312         Reviewed by Pavel Feldman.
2313
2314         Unset line numbers are not printed to console now.
2315         Added Source/WebCore/inspector to efl DumpRenderTree include path.
2316
2317         * DumpRenderTree/chromium/WebViewHost.cpp:
2318         (WebViewHost::didAddMessageToConsole):
2319         * DumpRenderTree/efl/CMakeLists.txt:
2320         * DumpRenderTree/efl/DumpRenderTreeView.cpp:
2321         (onConsoleMessage):
2322         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2323         (webViewConsoleMessage):
2324         * DumpRenderTree/mac/UIDelegate.mm:
2325         (-[UIDelegate webView:addMessageToConsole:withSource:]):
2326         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2327         (WebCore::WebPage::javaScriptConsoleMessage):
2328         * DumpRenderTree/win/UIDelegate.cpp:
2329         (UIDelegate::webViewAddMessageToConsole):
2330         * DumpRenderTree/wx/DumpRenderTreeWx.cpp:
2331         (LayoutWebViewEventHandler::OnConsoleMessageEvent):
2332
2333 2012-01-10  Dirk Pranke  <dpranke@chromium.org>
2334
2335         test-webkitpy: push more logic into webkitpy.test.main, clean up code
2336         https://bugs.webkit.org/show_bug.cgi?id=76021
2337
2338         Reviewed by Eric Seidel.
2339
2340         I plan to add more functionality to test-webkitpy, but it is
2341         difficult to hack on now; this patch is the first of two that
2342         brings test-webkitpy more inline with current coding style by
2343         pushing more logic into the webkitpy.test.main.Tester class (so
2344         that it will be testable itself).
2345         
2346         There should be no functional changes in this patch.
2347
2348         * Scripts/test-webkitpy:
2349         (_path_from_webkit_root):
2350         * Scripts/webkitpy/test/main.py:
2351         (Tester):
2352         (Tester.init):
2353         (Tester.configure_logging):
2354         (Tester.configure_logging.filter):
2355         (Tester.clean_packages):
2356         (Tester.run_tests):
2357
2358 2012-01-11  Wei Jia  <wjia@chromium.org>
2359
2360         implement layout tests for <video> with media stream
2361         https://bugs.webkit.org/show_bug.cgi?id=74882
2362
2363         Reviewed by Kent Tamura.
2364
2365         * DumpRenderTree/DumpRenderTree.gypi:
2366         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp: Added.
2367         (WebKit::WebUserMediaClientMock::create):
2368         (WebKit::WebUserMediaClientMock::IsMockStream):
2369         (WebKit::WebUserMediaClientMock::requestUserMedia):
2370         (WebKit::WebUserMediaClientMock::cancelUserMediaRequest):
2371         * DumpRenderTree/chromium/WebUserMediaClientMock.h: Added.
2372         (WebKit::WebUserMediaClientMock::~WebUserMediaClientMock):
2373         (WebKit::WebUserMediaClientMock::WebUserMediaClientMock):
2374         * DumpRenderTree/chromium/WebViewHost.cpp:
2375         (WebViewHost::userMediaClient):
2376         (WebViewHost::userMediaClientMock):
2377         (WebViewHost::createMediaPlayer):
2378         (WebViewHost::mediaStreamUtil):
2379         (WebViewHost::testMediaStreamClient):
2380         * DumpRenderTree/chromium/WebViewHost.h:
2381
2382 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2383
2384         Unreviewed, build fix.
2385
2386         Fix chromium-win's version remapping to work correctly on Win 7 SP1.
2387
2388         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2389         (ChromiumWinPort.__init__):
2390
2391 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2392
2393         Sheriffbot build fix. Don't emit empty lines on IRC.
2394
2395         * Scripts/webkitpy/tool/bot/irc_command.py:
2396         (LastGreenRevision.execute):
2397
2398 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2399
2400         last-green-revision should report the revision that succeeded on all bots
2401         https://bugs.webkit.org/show_bug.cgi?id=76109
2402
2403         Reviewed by Adam Barth.
2404
2405         Add an algorithm to find the last known good revision among the specified bots. For a revision to be
2406         considered green by this algorithm, all matching builders must have a successful run at the revision
2407         or two consecutive successful runs before and after the revision.
2408
2409         Also fixed a bug in irc_command that the result wasn't posted on IRC properly and a bug in queries.py
2410         that resulted in an exception when the user didn't pass BUILDER_NAME argument.
2411
2412         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2413         (BuildBot._revisions_for_builder):
2414         (BuildBot):
2415         (BuildBot._find_green_revision):
2416         (BuildBot.last_green_revision):
2417         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2418         (MockBuildBot.last_green_revision):
2419         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2420         (test_revisions_for_builder):
2421         (test_find_green_revision):
2422         (test_last_green_revision):
2423         * Scripts/webkitpy/tool/bot/irc_command.py:
2424         (LastGreenRevision.execute):
2425         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2426         (SheriffIRCBotTest.test_lgr):
2427         * Scripts/webkitpy/tool/commands/queries.py:
2428         (LastGreenRevision):
2429
2430 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2431
2432         Unreviewed, fix build bustage on win32 introduced by r104725.
2433
2434         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2435         (MockDRTPortTest.make_port):
2436         (MockChromiumDRTTest.test_pixeltest__fails):
2437
2438 2012-01-11  Jochen Eisinger  <jochen@chromium.org>
2439
2440         [Chromium] mimic the (old) behavior of the mac port for (re)setting the color profile
2441         https://bugs.webkit.org/show_bug.cgi?id=75618
2442
2443         Eventually, we should get away from setting the display's color profile, as the mac port is doing now.
2444
2445         Reviewed by Tony Chang.
2446
2447         * DumpRenderTree/chromium/LayoutTestHelper.mm:
2448         (installLayoutTestColorProfile):
2449         (restoreUserColorProfile):
2450         (main):
2451
2452 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2453
2454         test-webkitpy: fix -v and eliminate some spurious warnings
2455         https://bugs.webkit.org/show_bug.cgi?id=76099
2456
2457         Reviewed by Adam Barth.
2458
2459         when we added support for --xml output, we broke the -v flag.
2460
2461         Also, fix the code to swallow a harmless warning from
2462         VCSUtils.pm and fix a deprecated warning in jsonchecker.py.
2463
2464         test-webkitpy is now silent again except for the one known
2465         logging issue.
2466
2467         * Scripts/webkitpy/common/checkout/checkout.py:
2468         (Checkout.commit_message_for_this_commit):
2469         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
2470         (test_commit_message_for_this_commit):
2471         * Scripts/webkitpy/style/checkers/jsonchecker.py:
2472         (JSONChecker.check):
2473         (JSONChecker.line_number_from_json_exception):
2474         * Scripts/webkitpy/test/main.py:
2475         (Tester.run_tests):
2476
2477 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2478
2479         webkitpy: webkitpy.layout_tests.port tests should only use MockSystemHost, not MockHost
2480         https://bugs.webkit.org/show_bug.cgi?id=76084
2481
2482         Reviewed by Adam Barth.
2483
2484         This will help make it easier to catch layering violations. This
2485         change just refactors test code; there are no changes to
2486         production code.
2487
2488         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2489         (PortTest.make_port):
2490         (PortTest.test_find_no_paths_specified):
2491         (PortTest.test_find_one_test):
2492         (PortTest.test_find_glob):
2493         (PortTest.test_find_with_skipped_directories):
2494         (PortTest.test_find_with_skipped_directories_2):
2495         (PortTest.test_parse_reftest_list):
2496         (PortTest.test_httpd_returns_error_code):
2497         (PortTest.test_virtual_methods):
2498         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
2499         (ChromiumGpuTest.assert_port_works):
2500         (ChromiumGpuTest._assert_baseline_path):
2501         (ChromiumGpuTest.test_graphics_type):
2502         (ChromiumGpuTest.test_default_tests_paths.test_paths):
2503         (ChromiumGpuTest.test_test_files):
2504         (ChromiumGpuTest.test_test_files.test_paths):
2505         (ChromiumGpuTest):
2506         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2507         (ChromiumLinuxPortTest.assert_architecture):
2508         (ChromiumLinuxPortTest.test_check_illegal_port_names):
2509         (ChromiumLinuxPortTest.test_determine_architecture_fails):
2510         * Scripts/webkitpy/layout_tests/port/driver_unittest.py:
2511         (DriverTest.make_port):
2512         (DriverTest):
2513         (DriverTest.assertVirtual):
2514         (DriverTest._assert_wrapper):
2515         (DriverTest.test_virtual_driver_methods):
2516         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2517         (MockDRTPortTest.make_port):
2518         (MockDRTPortTest.test_port_name_in_constructor):
2519         (MockDRTTest.assertTest):
2520         (MockDRTTest.test_main):
2521         (MockChromiumDRTTest.test_pixeltest__fails):
2522         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2523         (QtPortTest._assert_search_path):
2524         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2525         (TestWebKitPort.__init__):
2526         (WebKitPortUnitTests.test_default_options):
2527         (test_runtime_feature_list):
2528         (test_test_expectations):
2529
2530 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2531
2532         Style skipped tests in the flakiness dashboard
2533         https://bugs.webkit.org/show_bug.cgi?id=76091
2534
2535         Reviewed by Tony Chang.
2536
2537         * TestResultServer/static-dashboards/dashboard_base.js:
2538         Removed "FLAKY" from this list since we don't ever mark individual runs
2539         as flaky. Also deleted an outdated line.
2540         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2541
2542 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2543
2544         Remove gtest normalization code from the test results server
2545         https://bugs.webkit.org/show_bug.cgi?id=76089
2546
2547         Reviewed by Tony Chang.
2548
2549         The normalization now happens on the gtest side.
2550
2551         * TestResultServer/model/jsonresults.py:
2552         (JsonResults.merge):
2553         * TestResultServer/model/jsonresults_unittest.py:
2554         (JsonResultsTest.test_gtest):
2555
2556 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2557
2558         Be more aggressive about deleting unnecessary results from the test-results-server json files
2559         https://bugs.webkit.org/show_bug.cgi?id=76086
2560
2561         Reviewed by Tony Chang.
2562
2563         Remove any tests that only contain skip/pass/nodata values whose runtime
2564         is less than 5 seconds.
2565
2566         * TestResultServer/model/jsonresults.py:
2567         (JsonResults._normalize_results_json):
2568         * TestResultServer/model/jsonresults_unittest.py:
2569         (JsonResultsTest.test_merge_remove_test):
2570         (JsonResultsTest.test_merge_keep_test_with_all_pass_but_slow_time):
2571
2572 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2573
2574         last-green-revision should give us per-bot information
2575         https://bugs.webkit.org/show_bug.cgi?id=76011
2576
2577         Reviewed by Adam Barth.
2578
2579         Rewrote last-green-revision command. Instead of finding a revision for which all bots succeeded,
2580         we report the latest green run on each bot from the last 100 runs.
2581
2582         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2583         (BuildBot._fetch_builder_page):
2584         (BuildBot):
2585         (BuildBot._green_revision_for_builder):
2586         (BuildBot.last_green_revision):
2587         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2588         (test_green_revision_for_builder):
2589         (test_last_green_revision):
2590         * Scripts/webkitpy/tool/bot/irc_command.py:
2591         (LastGreenRevision.execute):
2592         * Scripts/webkitpy/tool/commands/queries.py:
2593         (LastGreenRevision.execute):
2594
2595 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2596
2597         webkitpy: clean up version detection in webkitpy.layout_tests.port
2598         https://bugs.webkit.org/show_bug.cgi?id=76016
2599
2600         Reviewed by Adam Barth.
2601
2602         This change updates the code in webkitpy.layout_tests.port to
2603         use the new os_name and os_version fields on host.platform. This
2604         gets layout_tests.port out of the version parsing business. Most
2605         of the diffs are just cleaning up test code to accomodate the
2606         changes.
2607
2608         * Scripts/webkitpy/common/system/executive_mock.py:
2609         (MockExecutive):
2610         (MockExecutive.ignore_error):
2611         (MockExecutive2):
2612         (MockExecutive2.ignore_error):
2613         * Scripts/webkitpy/common/system/platforminfo_mock.py:
2614         (MockPlatformInfo.is_mac):
2615         (MockPlatformInfo.is_linux):
2616         (MockPlatformInfo.is_win):
2617         * Scripts/webkitpy/common/system/systemhost_mock.py:
2618         (MockSystemHost.__init__):
2619         * Scripts/webkitpy/layout_tests/port/apple.py:
2620         (ApplePort.__init__):
2621         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
2622         (ChromiumMacPort.__init__):
2623         (ChromiumMacPort.check_wdiff):
2624         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2625         (ChromiumMacPortTest):
2626         (ChromiumMacPortTest.assert_name):
2627         (ChromiumMacPortTest.test_versions):
2628         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2629         (ChromiumPortTest.test_all_test_configurations):
2630         (ChromiumPortTest.TestMacPort.__init__):
2631         (ChromiumPortTest.TestLinuxPort.__init__):
2632         (ChromiumPortTest.TestWinPort.__init__):
2633         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2634         (ChromiumWinPort.__init__):
2635         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2636         (ChromiumWinTest):
2637         (ChromiumWinTest.assert_name):
2638         (ChromiumWinTest.test_versions):
2639         (ChromiumWinTest.test_baseline_path):
2640         (ChromiumWinTest.test_build_path):
2641         * Scripts/webkitpy/layout_tests/port/factory.py:
2642         (PortFactory._port_name_from_arguments_and_options):
2643         (PortFactory.get):
2644         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
2645         (FactoryTest):
2646         (FactoryTest.setUp):
2647         (FactoryTest.assert_port):
2648         (FactoryTest.test_mac):
2649         (FactoryTest.test_win):
2650         (FactoryTest.test_google_chrome):
2651         (FactoryTest.test_gtk):
2652         (FactoryTest.test_qt):
2653         (FactoryTest.test_chromium_gpu):
2654         (FactoryTest.test_chromium_gpu_linux):
2655         (FactoryTest.test_chromium_gpu_mac):
2656         (FactoryTest.test_chromium_gpu_win):
2657         (FactoryTest.test_chromium_mac):
2658         (FactoryTest.test_chromium_linux):
2659         (FactoryTest.test_chromium_win):
2660         (FactoryTest.test_unknown_specified):
2661         (FactoryTest.test_unknown_default):
2662         * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
2663         (TestGoogleChromePort):
2664         (TestGoogleChromePort._verify_baseline_path):
2665         (TestGoogleChromePort._verify_expectations_overrides):
2666         (TestGoogleChromePort.test_get_google_chrome_port):
2667         * Scripts/webkitpy/layout_tests/port/mac.py:
2668         (MacPort):
2669         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2670         (MacTest):
2671         (assert_name):
2672         (test_tests_for_other_platforms):
2673         (test_version):
2674         (test_versions):
2675         (test_is_version_methods):
2676         (test_setup_environ_for_server):
2677         (_assert_search_path):
2678         (test_show_results_html_file):
2679         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2680         (PortTestCase):
2681         (PortTestCase.make_port):
2682         * Scripts/webkitpy/layout_tests/port/win.py:
2683         (WinPort):
2684         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2685         (WinPortTest):
2686         (WinPortTest.test_show_results_html_file):
2687         (WinPortTest._assert_search_path):
2688         (WinPortTest._assert_version):
2689
2690 2012-01-11  Hugo Parente Lima  <hugo.lima@openbossa.org>
2691
2692         [Qt][WK2] Remove workaround introduced into bug 75470
2693         https://bugs.webkit.org/show_bug.cgi?id=75889
2694
2695         Reviewed by Tor Arne Vestbø.
2696
2697         * MiniBrowser/qt/qml/BrowserWindow.qml:
2698
2699 2012-01-10  Dmitry Lomov  <dslomov@google.com>
2700
2701         [Chromium] Remove obsolete references to WebWorker class.
2702         https://bugs.webkit.org/show_bug.cgi?id=76020
2703
2704         Reviewed by David Levin.
2705
2706         * DumpRenderTree/chromium/TestWebWorker.h: Removed.
2707         * DumpRenderTree/chromium/WebViewHost.cpp:
2708         * DumpRenderTree/chromium/WebViewHost.h:
2709
2710 2012-01-11  Csaba Osztrogonác  <ossy@webkit.org>
2711
2712         [Qt] Some css3 filter tests are failing after r104698
2713         https://bugs.webkit.org/show_bug.cgi?id=76055
2714
2715         Reviewed by Simon Hausmann.
2716
2717         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2718         (LayoutTestController::overridePreference): Add WebKitAcceleratedCompositingEnabled preference.
2719
2720 2012-01-11  Tony Chang  <tony@chromium.org>
2721
2722         Don't remove 0 byte files when running build-webkit --chromium
2723         https://bugs.webkit.org/show_bug.cgi?id=75997
2724
2725         Reviewed by Csaba Osztrogonác.
2726
2727         If QTDIR is set, isQt() is true even when running build-webkit --chromium.
2728
2729         * Scripts/webkitdirs.pm:
2730         (determineIsQt):
2731
2732 2012-01-10  David Kilzer  <ddkilzer@apple.com>
2733
2734         Add TestWebKitAPI tests for KURL
2735         <http://webkit.org/b/75774>
2736
2737         Reviewed by Adam Barth.
2738
2739         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add a
2740         WebCore folder and KURL.cpp source file to the project.
2741         * TestWebKitAPI/Tests/WebCore/KURL.cpp: Added.
2742         (TestWebKitAPI::WebCore::SetUp): Initialize the main thread
2743         during setup since this test doesn't go through WebKit or
2744         WebKit2.
2745         (TestWebKitAPI::TEST_F): Add two basic tests.
2746
2747 2012-01-10  David Kilzer  <ddkilzer@apple.com>
2748
2749         Extract common code for comparing WTF::String objects into WTFStringUtilities.h
2750
2751         Reviewed by Adam Barth.
2752
2753         Needed for: <http://webkit.org/b/75774> Add TestWebKitAPI tests for KURL
2754
2755         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add
2756         WTFStringUtilities.h to the project.
2757         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Update to use
2758         WTFStringUtilities.h.
2759         * TestWebKitAPI/WTFStringUtilities.h: Added.
2760         (WTF::operator<<):
2761
2762 2012-01-10  Vincent Scheib  <scheib@chromium.org>
2763
2764         Clarify run-webkit-tests options reset-results and new-baseline
2765         https://bugs.webkit.org/show_bug.cgi?id=75993
2766
2767         Reviewed by Dirk Pranke.
2768
2769         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2770         (parse_args):
2771
2772 2012-01-10  Simon Fraser  <simon.fraser@apple.com>
2773
2774         Disabled mock scrollbars should draw differently
2775         https://bugs.webkit.org/show_bug.cgi?id=75995
2776
2777         Reviewed by James Robinson.
2778
2779         When the scrollbar is disabled, paint the entire track of
2780         mock scrollbars with a lighter gray, and hide the thumb.
2781
2782         * DumpRenderTree/mac/DumpRenderTree.mm:
2783         (-[DRTMockScroller drawKnob]):
2784         (-[DRTMockScroller drawRect:]):
2785
2786 2012-01-10  James Robinson  <jamesr@chromium.org>
2787
2788         [chromium] Run webkit_unit_tests as part of webkit-patch build-and-test
2789         https://bugs.webkit.org/show_bug.cgi?id=75988
2790
2791         Reviewed by Adam Barth.
2792
2793         * Scripts/webkitpy/common/config/ports.py:
2794         (WebKitPort.run_webkit_unit_tests_command):
2795         (WebKitPort):
2796         (ChromiumPort.run_webkit_unit_tests_command):
2797         (ChromiumPort):
2798         * Scripts/webkitpy/common/config/ports_mock.py:
2799         (MockPort.run_webkit_unit_tests_command):
2800         * Scripts/webkitpy/tool/commands/download_unittest.py:
2801         (DownloadCommandsTest.test_build_and_test):
2802         (test_land):
2803         (test_land_red_builders):
2804         * Scripts/webkitpy/tool/steps/runtests.py:
2805         (RunTests.run):
2806         * Scripts/webkitpy/tool/steps/runtests_unittest.py:
2807
2808 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
2809
2810         webkitpy: add os_name, os_version to platforminfo
2811         https://bugs.webkit.org/show_bug.cgi?id=75931
2812
2813         Reviewed by Eric Siedel.
2814
2815         As a first step in cleaning up the version parsing logic in
2816         webkitpy.layout_tests.port, this adds common routines for
2817         getting the os_name and os_version fields to the PlatformInfo
2818         class. Nothing uses them yet but I've added FIXME's to some of
2819         the code that needs to be deleted in the port files.
2820
2821         * Scripts/webkitpy/common/system/platforminfo.py:
2822         (PlatformInfo.__init__):
2823         (PlatformInfo):
2824         (PlatformInfo._determine_os_name):
2825         (PlatformInfo._determine_mac_version):
2826         (PlatformInfo._determine_linux_version):
2827         (PlatformInfo._determine_win_version):
2828         (PlatformInfo.display_name):
2829         (PlatformInfo.total_bytes_memory):
2830         (PlatformInfo._compute_bytes_from_vm_stat_output):
2831         (PlatformInfo.free_bytes_memory):
2832         * Scripts/webkitpy/common/system/platforminfo_unittest.py: Added.
2833         (TestPlatformInfo):
2834         (TestPlatformInfo.make_info):
2835         (TestPlatformInfo.make_info.mock_run_command):
2836         (TestPlatformInfo.setUp):
2837         (TestPlatformInfo.tearDown):
2838         (TestPlatformInfo.test_basic):
2839         (TestPlatformInfo.integration_test_basic):
2840         (TestPlatformInfo.test_display_name_mac):
2841         (TestPlatformInfo.test_display_name_win32):
2842         (TestPlatformInfo.test_memory_mac):
2843         (TestPlatformInfo.test_memory_win32):
2844         (TestPlatformInfo.test_determine_os_name):
2845         (TestPlatformInfo.test_determine_mac_version):
2846         (TestPlatformInfo.test_determine_linux_version):
2847         (TestPlatformInfo.test_determine_win_version_from_tuple):
2848         (TestPlatformInfo.test_determine_win_version_from_cmd):
2849         * Scripts/webkitpy/layout_tests/port/apple.py:
2850         (ApplePort.__init__):
2851         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
2852         (ChromiumMacPort.__init__):
2853         * Scripts/webkitpy/layout_tests/port/mac.py:
2854         (MacPort):
2855         * Scripts/webkitpy/layout_tests/port/win.py:
2856         (WinPort):
2857
2858 2012-01-10  Adam Roben  <aroben@apple.com>
2859
2860         Make it possible to type data: URLs into MiniBrowser on Windows
2861
2862         Fixes <http://webkit.org/b/75084> Crash when trying to navigate to a data: URL in
2863         MiniBrowser on Windows
2864
2865         Reviewed by Darin Adler.
2866
2867         * MiniBrowser/win/BrowserWindow.cpp:
2868         (BrowserWindow::handleMessage): Instead of checking for an "http://" prefix when deciding
2869         whether to prepend "http://", just look for whether the typed URL contains a colon. If it
2870         does, we assume it already has a scheme and don't modify it.
2871
2872 2012-01-10  Adam Roben  <aroben@apple.com>
2873
2874         Make MiniBrowser automatically escape invalid URL characters typed in its URL field
2875
2876         Fixes <http://webkit.org/b/75086> Crash when typing a data: URL containing double-quotes in
2877         MiniBrowser on Windows
2878
2879         Reviewed by Darin Adler.
2880
2881         * MiniBrowser/win/BrowserView.cpp:
2882         (BrowserView::goToURL): Pass the typed string through
2883         CFURLCreateStringByAddingPercentEscapes before trying to create a URL from it.
2884
2885 2012-01-10  Ben Wells  <benwells@chromium.org>
2886
2887         Adding myself (benwells) to committers.py
2888         https://bugs.webkit.org/show_bug.cgi?id=75971
2889
2890         Reviewed by Ojan Vafai.
2891
2892         * Scripts/webkitpy/common/config/committers.py:
2893
2894 2012-01-10  Mario Sanchez Prada  <msanchez@igalia.com>
2895
2896         AX: support helpText() in DumpRenderTree
2897         https://bugs.webkit.org/show_bug.cgi?id=40193
2898
2899         Reviewed by Chris Fleizach.
2900
2901         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
2902         (AccessibilityUIElement::helpText): Implemented.
2903
2904 2012-01-10  Sheriff Bot  <webkit.review.bot@gmail.com>
2905
2906         Unreviewed, rolling out r104572.
2907         http://trac.webkit.org/changeset/104572
2908         https://bugs.webkit.org/show_bug.cgi?id=75967
2909
2910         It broke zillion tests (Requested by Ossy on #webkit).
2911
2912         * DumpRenderTree/qt/EventSenderQt.cpp:
2913         (EventSender::EventSender):
2914         (EventSender::sendOrQueueEvent):
2915         (EventSender::replaySavedEvents):
2916         (EventSender::eventFilter):
2917         * DumpRenderTree/qt/EventSenderQt.h:
2918
2919 2012-01-10  Hugo Parente Lima  <hugo.lima@openbossa.org>
2920
2921         [Qt] fast/events/dont-loose-last-event test fails on WK1
2922         https://bugs.webkit.org/show_bug.cgi?id=73894
2923
2924         Reviewed by Simon Hausmann.
2925
2926         Empty the entire event queue when requested besides using
2927         sendEvent instead of postEvent to make sure all events were
2928         delivered.
2929         This commit also removes not working DRT code to handle drag and drop,
2930         for more refs see: https://bugs.webkit.org/show_bug.cgi?id=31332
2931
2932         * DumpRenderTree/qt/EventSenderQt.cpp:
2933         (EventSender::EventSender):
2934         (EventSender::sendOrQueueEvent):
2935         (EventSender::replaySavedEvents):
2936         * DumpRenderTree/qt/EventSenderQt.h:
2937
2938 2012-01-10  No'am Rosenthal  <noam.rosenthal@nokia.com>
2939
2940         [Qt] Enable CSS_FILTERS in Qt build
2941         https://bugs.webkit.org/show_bug.cgi?id=75777
2942
2943         Add the CSS_FILTERS feature flag to Qt.
2944
2945         Reviewed by Kenneth Rohde Christiansen.
2946
2947         * qmake/mkspecs/features/features.prf:
2948
2949 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
2950
2951         [Qt] Unreviewed -Werror build fix.
2952
2953         * MiniBrowser/qt/MiniBrowserApplication.cpp:
2954         (MiniBrowserApplication::MiniBrowserApplication): Re-order initialization to match
2955         declaration.
2956
2957 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
2958
2959         [Qt] Remove dependency to uitools
2960         https://bugs.webkit.org/show_bug.cgi?id=75952
2961
2962         Reviewed by Tor Arne Vestbø.
2963
2964         QUiLoader is used for creating QLabel/QProgressBar widgets for
2965         a layout test, to verify some widget embedding feature. Unfortunately
2966         in Qt 5, QUiLoader is part of qttools, which depends on webkit, which
2967         depends on qttools. Since we don't really _need_ QUiLoader for our tests,
2968         let's break the circular dependency.
2969
2970         * DumpRenderTree/qt/DumpRenderTree.pro:
2971         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2972         (WebCore::WebPage::createPlugin):
2973         * QtTestBrowser/QtTestBrowser.pro:
2974         * QtTestBrowser/launcherwindow.h:
2975         * QtTestBrowser/webpage.cpp:
2976         (WebPage::createPlugin):
2977         * qmake/mkspecs/features/minimal_qt.prf:
2978         * qmake/mkspecs/features/uitools.prf: Removed.
2979
2980 2012-01-10  Alexander Færøy  <alexander.faeroy@nokia.com>
2981
2982         [Qt] Don't expose m_windowOptions as part of the public interface in MiniBrowserApplication
2983         https://bugs.webkit.org/show_bug.cgi?id=75894
2984
2985         Reviewed by Tor Arne Vestbø.
2986
2987         * MiniBrowser/qt/MiniBrowserApplication.h:
2988         (MiniBrowserApplication::windowOptions):
2989         * MiniBrowser/qt/main.cpp:
2990         (main):
2991
2992 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
2993
2994         [Qt] DRT crashes with Qt 5
2995         https://bugs.webkit.org/show_bug.cgi?id=75951
2996
2997         Reviewed by Ossy.
2998
2999         * DumpRenderTree/qt/fonts.conf: Add missing last-resort font fallback for Qt 5 where
3000         Qt relies on font-config to define the default font family.
3001
3002 2012-01-10  Csaba Osztrogonác  <ossy@webkit.org>
3003
3004         [Qt] REGRESSION(r100130): Clean build feature is broken.
3005         https://bugs.webkit.org/show_bug.cgi?id=75863
3006
3007         Reviewed by Tor Arne Vestbø.
3008
3009         * Scripts/webkitdirs.pm:
3010         (buildQMakeProject):
3011
3012 2012-01-10  Zeno Albisser  <zeno@webkit.org>
3013
3014         [Qt][WK2] Fix keyboard shortcuts in MiniBrowser.
3015         https://bugs.webkit.org/show_bug.cgi?id=75885
3016
3017         In QML events are propagated through parents. But since the
3018         WebView may consume key events, a shortcut might never reach
3019         the top QtQuickItem.
3020         Therefore an application wide event handling function in C++
3021         needs to take care of shortcuts for MiniBrowser.
3022
3023         Reviewed by Kenneth Rohde Christiansen.
3024
3025         * MiniBrowser/qt/BrowserWindow.cpp:
3026         (BrowserWindow::reload):
3027         (BrowserWindow::focusAddressBar):
3028         * MiniBrowser/qt/BrowserWindow.h:
3029         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3030         (MiniBrowserApplication::notify):
3031         * MiniBrowser/qt/qml/BrowserWindow.qml:
3032
3033 2012-01-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
3034
3035         Add --efl alias for --platform=efl to run-webkit-test
3036         https://bugs.webkit.org/show_bug.cgi?id=75937
3037
3038         Reviewed by Andreas Kling.
3039         
3040         * Scripts/run-webkit-tests: Add --efl alias.
3041         * Scripts/webkitpy/layout_tests/run_webkit_tests.py: Add --efl alias.
3042         (parse_args):
3043
3044 2012-01-09  Dan Bernstein  <mitz@apple.com>
3045
3046         -[DOMRange renderedImageForcingBlackText:] fails with non-user-selectable text
3047         https://bugs.webkit.org/show_bug.cgi?id=75920
3048
3049         Reviewed by Darin Adler.
3050
3051         * TestWebKitAPI/Tests/mac/RenderedImageFromDOMRange.mm:
3052         (TestWebKitAPI::TEST): Added a test for a range in a
3053         -webkit-user-select: none; block.
3054
3055 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
3056
3057         webkitpy.layout_tests.port.mock_drt_unittest.MockChromiumDRTTest has been failing on cr-win
3058         https://bugs.webkit.org/show_bug.cgi?id=75884
3059
3060         Reviewed by Ryosuke Niwa.
3061
3062         Fix regression introduced (revealed?) in r104340. Chromium file         
3063         url syntax is different on win32 and the test results weren't           
3064         expecting that.                                                         
3065
3066         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3067         (TestExpectations._report_errors):
3068         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3069         (test_parse_error_nonfatal):
3070         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
3071         (MockDRTTest.input_line):
3072         (MockChromiumDRTTest.test_pixeltest__fails):
3073         * Scripts/webkitpy/layout_tests/port/test.py:
3074         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
3075         (TestRebaseliner.make_rebaseliner):
3076         (TestRebaseliner.test_one_platform):
3077         (TestRebaseliner.test_all_platforms):
3078         (TestRebaseliner.test_png_file_with_comment):
3079         (TestRebaseliner.test_png_file_with_comment_remove_old_checksum):
3080         (TestRebaseliner.test_png_file_with_comment_as_duplicate):
3081         (TestRealMain.test_all_platforms):
3082
3083 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
3084
3085         Fix regression in test-webkitpy introduced by removing the webkitpy/python24 directory.
3086
3087         Unreviewed, build fix.
3088
3089         test-webkitpy had a built-in self-test routine to verify that
3090         deleting the .pyc files worked correctly. When I removed the
3091         python24 directory in r104482, I broke the self-test.
3092
3093         I have deleted the self-test for now. Since I am reworking
3094         test-webkitpy I will make sure the new version is adequately
3095         tested.
3096
3097         * Scripts/test-webkitpy:
3098         (configure_logging):
3099         (_clean_pyc_files):
3100         (_clean_packages):
3101         (init):
3102
3103 2012-01-09  Justin Novosad  <junov@chromium.org>
3104
3105         [Chromium] remove all references to the legacy accelerated 2d Canvas
3106         implementation
3107         https://bugs.webkit.org/show_bug.cgi?id=75108
3108
3109         Purging an old settings flag that is no longer referenced
3110
3111         Reviewed by Darin Fisher.
3112
3113         * DumpRenderTree/chromium/DumpRenderTree.cpp:
3114         (main):
3115         * DumpRenderTree/chromium/TestShell.cpp:
3116         (TestShell::TestShell):
3117         (TestShell::resetWebSettings):
3118         * DumpRenderTree/chromium/TestShell.h:
3119         (TestShell::setAccelerated2dCanvasEnabled):
3120         * DumpRenderTree/chromium/WebPreferences.cpp:
3121         (WebPreferences::reset):
3122         (WebPreferences::applyTo):
3123         * DumpRenderTree/chromium/WebPreferences.h:
3124
3125 2012-01-09  Adam Roben  <aroben@apple.com>
3126
3127         Convert Cygwin paths to Windows paths before passing them to DRT
3128
3129         Fixes <http://webkit.org/b/64468> <rdar://problem/10663409> WIN: DumpRenderTree hangs under
3130         NRWT
3131
3132         I couldn't find a way to test this because I couldn't figure out how to override the
3133         cygpath function as it is used by WebKitDriver.
3134
3135         Reviewed by Dirk Pranke.
3136
3137         * Scripts/webkitpy/layout_tests/port/webkit.py:
3138         (WebKitDriver._command_from_driver_input): On Cygwin, convert the test's path to a
3139         Windows-style path before passing it to DRT.
3140
3141 2012-01-06  Dirk Pranke  <dpranke@chromium.org>
3142
3143         webkitpy: revamp version checking
3144         https://bugs.webkit.org/show_bug.cgi?id=75765
3145
3146         Reviewed by Adam Barth.
3147
3148         The version-checking code in webkitpy/python24/versioning seems really
3149         awkward and overdesigned, especially since we don't support python 2.4
3150         any more.
3151
3152         In addition, I am tired of getting warnings about Python 2.6 being "a
3153         newer release than what is supported" :).
3154
3155         I'm taking a page from the depot_tools python code and just creating a
3156         new webkitpy.common.version_check module that will print a message on
3157         stderr and call sys.exit() if we're running an unsupported version.
3158
3159         * Scripts/check-webkit-style:
3160         * Scripts/new-run-webkit-httpd:
3161         * Scripts/new-run-webkit-tests:
3162         * Scripts/rebaseline-chromium-webkit-tests:
3163         * Scripts/test-webkitpy:
3164         (configure_logging):
3165         (init):
3166         * Scripts/webkit-patch:
3167         (main):
3168         * Scripts/webkitpy/common/system/executive.py:
3169         (Executive._child_process_encoding):
3170         (Executive._should_encode_child_process_arguments):
3171         * Scripts/webkitpy/common/version_check.py: Copied from Tools/Scripts/rebaseline-chromium-webkit-tests.
3172         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3173         * Scripts/webkitpy/python24/__init__.py: Removed.
3174         * Scripts/webkitpy/python24/versioning.py: Removed.
3175         * Scripts/webkitpy/python24/versioning_unittest.py: Removed.
3176
3177 2012-01-09  Adam Barth  <abarth@webkit.org>
3178
3179         Remove Chromium Mac CG from the flakiness dashboard
3180         https://bugs.webkit.org/show_bug.cgi?id=75873
3181
3182         Reviewed by Ojan Vafai.
3183
3184         This configuration no longer exists.
3185
3186         * TestResultServer/static-dashboards/builders.js:
3187         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3188         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3189         (testOverrideJustBuildType):
3190         (testPlatformAndBuildType):
3191
3192 2012-01-06  Adam Roben  <aroben@apple.com>
3193
3194         Print an error message when NRWT can't run Apache
3195
3196         Fixes <http://webkit.org/b/75712> NRWT mysteriously exits when Apache returns an error code
3197
3198         Reviewed by Dirk Pranke.
3199
3200         * Scripts/webkitpy/layout_tests/port/base.py:
3201         (Port.check_httpd): Print an error message when Apache returns an error code instead of
3202         failing silently.
3203
3204         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3205         (PortTest.test_check_httpd_success): Test that we don't print anything when we successfully
3206         run Apache.
3207         (PortTest.test_httpd_returns_error_code): Test that we print an error message when Apache
3208         fails.
3209
3210 2012-01-09  Gabor Rapcsanyi  <rgabor@webkit.org>
3211
3212         [Gtk][ARM] Cross compiler buildbot is failing after r103040
3213         https://bugs.webkit.org/show_bug.cgi?id=75846
3214
3215         Reviewed by Philippe Normand.
3216
3217         * Scripts/webkitdirs.pm:
3218         (runAutogenForAutotoolsProjectIfNecessary):
3219
3220 2012-01-09  Alexander Færøy  <ahf@0x90.dk>
3221
3222         [Qt] Add support for dynamically enabling and disabling touch mocking in the Qt MiniBrowser
3223         https://bugs.webkit.org/show_bug.cgi?id=75807
3224
3225         This patch adds a new icon to navigationbar of the Qt MiniBrowser
3226         which allows you to dynamically toggle whether touch mocking is
3227         enabled or disabled.
3228
3229         Reviewed by Tor Arne Vestbø.
3230
3231         * MiniBrowser/qt/MiniBrowser.qrc:
3232         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3233         (MiniBrowserApplication::notify):
3234         * MiniBrowser/qt/MiniBrowserApplication.h:
3235         (WindowOptions::WindowOptions):
3236         (WindowOptions::touchMockingEnabled):
3237         (WindowOptions::setTouchMockingEnabled):
3238         * MiniBrowser/qt/icons/touch.png: Added.
3239         * MiniBrowser/qt/qml/BrowserWindow.qml:
3240
3241 2012-01-09  Carlos Garcia Campos  <cgarcia@igalia.com>
3242
3243         Unreviewed. Fix make distcheck issues.
3244
3245         * MiniBrowser/gtk/GNUmakefile.am:
3246
3247 2012-01-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3248
3249         Unreviewed, rolling out r104421.
3250         http://trac.webkit.org/changeset/104421
3251         https://bugs.webkit.org/show_bug.cgi?id=75816
3252
3253         Need to rebaseline some tests on Linux (Requested by noamr on
3254         #webkit).
3255
3256         * qmake/mkspecs/features/features.prf:
3257
3258 2012-01-08  Adam Barth  <abarth@webkit.org>
3259
3260         Remove support for Chromium Mac CG from garden-o-matic
3261         https://bugs.webkit.org/show_bug.cgi?id=75814
3262
3263         Reviewed by Eric Seidel.
3264
3265         This configuration no longer exists.  This patch is mostly a matter of
3266         updating config.js and removing the examples from the unit tests.
3267
3268         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
3269         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
3270         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js:
3271         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
3272         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
3273
3274 2012-01-08  No'am Rosenthal  <noam.rosenthal@nokia.com>
3275
3276         [Qt] Enable CSS_FILTERS in Qt build
3277         https://bugs.webkit.org/show_bug.cgi?id=75777
3278
3279         Add the CSS_FILTERS feature flag to Qt.
3280
3281         Reviewed by Kenneth Rohde Christiansen.
3282
3283         * qmake/mkspecs/features/features.prf:
3284
3285 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3286
3287         Categorize bots by ports instead of core/non-core separation
3288         https://bugs.webkit.org/show_bug.cgi?id=75766
3289
3290         Reviewed by Eric Seidel.
3291
3292         Rename categorizes so as to preserve the original ordering of bots per Eric's suggestion
3293         (Chromium still comes before GTK and Qt).
3294
3295         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3296         (loadBuilderConfig):
3297         * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
3298
3299 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3300
3301         Remove the concept of core builders from webkitpy
3302         https://bugs.webkit.org/show_bug.cgi?id=75809
3303
3304         Reviewed by Ryosuke Niwa.
3305
3306         Treat all bots equally since core/non-core sepration has been removed from build.webkit.org,
3307         and remove the relevant code.
3308
3309         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3310         (BuildBot.__init__):
3311         (BuildBot._matches_regexps):
3312         (BuildBot._builder_statuses_with_names_matching_regexps):
3313         (BuildBot.builder_statuses):
3314         (BuildBot.failure_map):
3315         (BuildBot._latest_builds_from_builders):
3316         (BuildBot.last_green_revision):
3317         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3318         (MockBuildBot.builder_statuses):
3319         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3320         (test_status_parsing):
3321         (test_last_green_revision):
3322         (test_last_green_revision.mock_builds_from_builders):
3323
3324 2012-01-08  Adam Barth  <abarth@webkit.org>
3325
3326         Remove Chromium Mac CG concept from webkitpy
3327         https://bugs.webkit.org/show_bug.cgi?id=75810
3328
3329         Reviewed by Ryosuke Niwa.
3330
3331         This configuration no longer exists.  We can remove all the supporting
3332         code from webkitpy.
3333
3334         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
3335         (BaselineOptimizerTest.test_move_baselines):
3336         (BaselineOptimizerTest.test_chromium_covers_mac_win_linux):
3337         (BaselineOptimizerTest.test_chromium_mac_redundant_with_apple_mac):
3338         (BaselineOptimizerTest.test_common_directory_includes_root):
3339         (BaselineOptimizerTest.test_complex_shadowing):
3340         * Scripts/webkitpy/common/config/build.py:
3341         (_should_file_trigger_build):
3342         * Scripts/webkitpy/common/config/build_unittest.py:
3343         (ShouldBuildTest):
3344         * Scripts/webkitpy/layout_tests/port/builders.py:
3345         * Scripts/webkitpy/layout_tests/port/chromium.py:
3346         (ChromiumPort):
3347         (ChromiumPort._generate_all_test_configurations):
3348         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
3349         (_default_tests_paths):
3350         (ChromiumGpuLinuxPort.tests):
3351         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
3352         (ChromiumGpuTest.integration_test_chromium_gpu_mac):
3353         (ChromiumGpuTest.assert_port_works):
3354         (ChromiumGpuTest.test_baseline_paths):
3355         (ChromiumGpuTest.test_graphics_type):
3356         (ChromiumGpuTest.test_default_tests_paths):
3357         (ChromiumGpuTest.test_test_files):
3358         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
3359         (ChromiumMacPort):
3360         (ChromiumMacPort.__init__):
3361         (ChromiumMacPort.baseline_search_path):
3362         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3363         (ChromiumMacPortTest.test_graphics_type):
3364         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3365         (ChromiumPortTest.test_all_test_configurations):
3366         * Scripts/webkitpy/layout_tests/port/dryrun.py:
3367         * Scripts/webkitpy/layout_tests/port/factory.py:
3368         (PortFactory._port_name_from_arguments_and_options):
3369         (PortFactory.get):
3370         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3371         (FactoryTest.test_chromium_gpu_mac):
3372         (FactoryTest.test_chromium_mac):
3373         * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
3374         (GetGoogleChromePortTest._verify_expectations_overrides):
3375         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
3376         (MockChromiumDRTTest.test_chromium_parse_options):
3377         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
3378         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
3379         (test_url_fetcher):
3380         (test_zip_factory):
3381         (test_archive):
3382         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3383         (TestRebaseline.test_rebaseline_expectations):
3384
3385 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3386
3387         Categorize bots by ports instead of core/non-core separation
3388         https://bugs.webkit.org/show_bug.cgi?id=75766
3389
3390         Reviewed by Adam Barth.
3391
3392         Get rid of the concept of core/non-core builders, and categorize build bots by ports instead.
3393
3394         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3395         (loadBuilderConfig):
3396         * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
3397
3398 2012-01-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3399
3400         Unreviewed, rolling out r104403.
3401         http://trac.webkit.org/changeset/104403
3402         https://bugs.webkit.org/show_bug.cgi?id=75803
3403
3404         It broke all tests on Qt5 (Requested by Ossy_weekend on
3405         #webkit).
3406
3407         * qmake/mkspecs/features/features.prf:
3408
3409 2012-01-08  No'am Rosenthal  <noam.rosenthal@nokia.com>
3410
3411         [Qt] Enable CSS_FILTERS in Qt build
3412         https://bugs.webkit.org/show_bug.cgi?id=75777
3413
3414         Add the CSS_FILTERS feature flag to Qt.
3415
3416         Reviewed by Kenneth Rohde Christiansen.
3417
3418         * qmake/mkspecs/features/features.prf:
3419
3420 2012-01-04  Kentaro Hara  <haraken@chromium.org>
3421
3422         Rewrite the CSS parser of prepare-ChangeLog with unittests.
3423         https://bugs.webkit.org/show_bug.cgi?id=75202
3424
3425         Reviewed by David Kilzer.
3426
3427         The current CSS parser can just parse simple CSSes like
3428
3429             foo bar baz {
3430                 property1: value;
3431                 property2: value;
3432             }
3433
3434         , and cannot parse comments nor a CSS in which a selector and {
3435         appears in different lines. This patch rewrites the CSS parser
3436         (i.e. get_selector_line_ranges_for_css()) so that it can parse more CSSes
3437         shown in css_unittests.css.
3438
3439         Test: Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests.css
3440
3441         * Scripts/prepare-ChangeLog:
3442         (get_selector_line_ranges_for_css):
3443         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
3444         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Added.
3445         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests.css: Added.
3446         (element1):
3447         (element2):
3448         (element3):
3449         (element4.p):
3450         (element5.p.q.r.s):
3451         (element6#p):
3452         (element7 element8):
3453         (element9.p element10.q):
3454         (element11#p element12#q):
3455         (element13, element14):
3456         (.p):
3457         (#p):
3458         (.p element15 #q element16.r element17#s):
3459         (element18:target):
3460         (element19):
3461         (element20):
3462         (element21):
3463         (element22):
3464
3465 2012-01-04  Kentaro Hara  <haraken@chromium.org>
3466
3467         The Perl parser of prepare-ChangeLog can parse here-documents
3468         https://bugs.webkit.org/show_bug.cgi?id=73208
3469
3470         Reviewed by David Kilzer.
3471
3472         Currently prepare-ChangeLog cannot recognize here-documents in Perl,
3473         which results in wrong subroutine names in ChangeLogs.
3474         With this patch, prepare-ChangeLog can judge whether the line of "}"
3475         is the end of a subroutine or a line inside a here-document.
3476
3477         Test: Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl
3478
3479         * Scripts/prepare-ChangeLog:
3480         (get_function_line_ranges_for_perl):
3481         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt:
3482         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
3483         Added test cases for here-documents.
3484         (func7):
3485         (func8):
3486         (func9):
3487         (func10):
3488         (func11):
3489         (func12):
3490         (func13):
3491         (func14):
3492         (func15):
3493         (func16):
3494
3495 2012-01-08  Kentaro Hara  <haraken@chromium.org>
3496
3497         Add unittests for the JavaScript parser of prepare-ChangeLog
3498         https://bugs.webkit.org/show_bug.cgi?id=75201
3499
3500         Reviewed by David Kilzer.
3501
3502         javascript_unittests.js is a unittest for get_function_line_ranges_for_javascript()
3503         of prepare-ChangeLog.
3504
3505         * Scripts/prepare-ChangeLog:
3506         (get_function_line_ranges_for_javascript): Before this patch, a string found was just
3507         omitted from parsing. Thus,
3508
3509             str = "foo"
3510             function func() {
3511             }
3512
3513         was recognized as equivalent to
3514
3515             str = function func() {
3516             }
3517
3518         This patch replaces a string with a dummy identifier 'string_appeared_here'
3519         to tell the parser that a string appeared there.
3520
3521         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
3522         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Added.
3523         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests.js: Added.
3524         (func1):
3525         (func2):
3526         (func3):
3527         (func4):
3528         (func5):
3529         (func6):
3530         (func7):
3531         (func8):
3532         (func9):
3533         (func10):
3534         (func11):
3535         (func12):
3536         (funcOverloaded):
3537         (Func1.prototype.get x1):
3538         (Func1.prototype.get x2):
3539         (Func1.prototype.set x1):
3540         (Func1.prototype.set x3):
3541         (Func2.prototype.func13):
3542         (Func2.prototype.func14):
3543         (Func2.prototype.func15):
3544         (func16.func17):
3545         (func16.func18):
3546         (func16.func19):
3547         (func16):
3548
3549 2012-01-08  David Levin  <levin@chromium.org>
3550
3551         [chromium] Add layout test support for autosize code in FrameView.
3552         https://bugs.webkit.org/show_bug.cgi?id=73631
3553
3554         Reviewed by Dmitry Titov.
3555
3556         * DumpRenderTree/chromium/LayoutTestController.cpp:
3557         (LayoutTestController::LayoutTestController): Expose the new autoresize method.
3558         (LayoutTestController::reset): Set the autoresize mode back to false.
3559         (LayoutTestController::enableAutoResizeMode): A way to turn on autoresizing from javascript code in DumpRenderTree.
3560         * DumpRenderTree/chromium/LayoutTestController.h:
3561         * DumpRenderTree/chromium/WebViewHost.cpp:
3562         (WebViewHost::didAutoResize): Handle the autoresize callback so that the outer bounds change appropriately.
3563         * DumpRenderTree/chromium/WebViewHost.h:
3564
3565 2012-01-07  Zan Dobersek  <zandobersek@gmail.com>
3566
3567         [WK2][GTK] Each test takes roughly half a second to load
3568         https://bugs.webkit.org/show_bug.cgi?id=70699
3569
3570         Reviewed by Martin Robinson.
3571
3572         Set the XDG_CACHE_HOME environment variable when testing with
3573         WebKitTestRunner. This way the testing is started with a non-existing
3574         application cache. If the cache exists and is not empty, the emptying
3575         and vacuuming that is performed before every test can consume a lot of
3576         time.
3577
3578         This is just a temporary workaround until it becomes possible to set
3579         a desired directory to be used as the application cache location through
3580         WebKitWebContext.
3581
3582         * Scripts/webkitpy/layout_tests/port/gtk.py:
3583         (GtkPort.setup_environ_for_server):
3584
3585 2012-01-07  Zan Dobersek  <zandobersek@gmail.com>
3586
3587         [GTK] Enable requestAnimationFrame in build-webkit
3588         https://bugs.webkit.org/show_bug.cgi?id=75773
3589
3590         Reviewed by Martin Robinson.
3591
3592         Enable requestAnimationFrame for the Gtk port.
3593
3594         * Scripts/build-webkit:
3595
3596 2012-01-06  Jarred Nicholls  <jarred@sencha.com>
3597
3598         Unreviewed build-webkit fix for Chromium to properly use make if gyp generates Makefiles.
3599
3600         * Scripts/webkitdirs.pm:
3601         (buildChromium):
3602         Missing parentheses.
3603
3604 2012-01-06  Simon Fraser  <simon.fraser@apple.com>
3605
3606         Fix DRT build on case-sensitive file systems.
3607
3608         Fix case of include of WebDynamicScrollBarsView.h
3609         
3610         * DumpRenderTree/mac/DumpRenderTree.mm:
3611
3612 2012-01-06  Simon Fraser  <simon.fraser@apple.com>
3613
3614         Pixel results from DumpRenderTree and WebKitTestRunner don't match because of colorspace issues
3615         https://bugs.webkit.org/show_bug.cgi?id=75662
3616
3617         Reviewed by Dan Bernstein.
3618         
3619         The pixel results generated by DumpRenderTree and WebKitTestRunner differed because
3620         of color profile issues. Fix by keeping everything in device RGB and ensuring that the
3621         test window uses the main display's color space, so that the pixel values in the bitmap
3622         that gets checksummed are identical to the CSS colors.
3623         
3624         Removed the code that switches the display profile in DRT, since that is no longer required.
3625
3626         * DumpRenderTree/PixelDumpSupport.h: Remove unused function declarations.
3627         * DumpRenderTree/mac/DumpRenderTree.mm:
3628         (crashHandler): Remove code that switches display profiles.
3629         (prepareConsistentTestingEnvironment): Ditto.
3630         (dumpRenderTree): Ditto.
3631         * DumpRenderTree/mac/PixelDumpSupportMac.mm: Ditto.
3632         (createBitmapContext): Use DeviceRGB for the bitmap colorspace.
3633         (createBitmapContextFromWebView): Add comment about the colorspace handling.
3634         * WebKitTestRunner/cg/TestInvocationCG.cpp:
3635         (WTR::createCGContextFromImage): Use a RetainPtr, add comment.
3636         (WTR::computeMD5HashStringForContext): Simplify the #ifdefs around the braces.
3637         * WebKitTestRunner/mac/PlatformWebViewMac.mm:
3638         (WTR::PlatformWebView::PlatformWebView): Set the window's colorspace to that of the main display.
3639         (WTR::PlatformWebView::windowSnapshotImage): Add comment about colorspaces.
3640
3641 2012-01-06  David Kilzer  <ddkilzer@apple.com>
3642
3643         run-api-tests: specify individual suites and tests on the command-line
3644
3645         Reviewed by Adam Roben.
3646
3647         Fixes: <http://webkit.org/b/75065> run-api-tests should be able to run individual suites and tests
3648
3649         The run-api-tests script will now accept a list of arguments on
3650         the command-line that are used as prefix filters.  To run all
3651         tests in a suite that starts with "WTF" (WTF and WTF_Vector)
3652         use:
3653
3654             $ run-api-tests WTF
3655
3656         To run only the tests in the "WTF" suite, not the "WTF_Vector"
3657         suite, use:
3658
3659             $ run-api-tests WTF.
3660
3661         * Scripts/run-api-tests: Filter tests if any prefix arguments
3662         are passed on the command-line.
3663
3664 2012-01-06  David Kilzer  <ddkilzer@apple.com>
3665
3666         run-api-tests: change internal representation of tests to array of "SuiteName.TestName" strings
3667
3668         Reviewed by Adam Roben.
3669
3670         Part of: <http://webkit.org/b/75065> run-api-tests should be able to run individual suites and tests
3671
3672         * Scripts/run-api-tests:
3673         (dumpTestsBySuite): Update to accept array of tests instead of
3674         hash data structure.
3675         (runTestsBySuite): Ditto.
3676         (listAllTests): Rename from populateTests().  Update to return
3677         an array of tests instad of the hash data structure.
3678
3679 2012-01-05  Dirk Pranke  <dpranke@chromium.org>
3680
3681         webkitpy: clean up test/uri conversion routines
3682         https://bugs.webkit.org/show_bug.cgi?id=75648
3683
3684         Reviewed by Eric Seidel.
3685
3686         This change moves test_to_uri and uri_to_test from the Port
3687         class to the Driver class (the routines were only being
3688         called by the drivers), and&n