run-perf-tests should timeout sooner
[WebKit-https.git] / Tools / ChangeLog
1 2012-01-25  Ryosuke Niwa  <rniwa@webkit.org>
2
3         run-perf-tests should timeout sooner
4         https://bugs.webkit.org/show_bug.cgi?id=77026
5
6         Reviewed by Adam Barth.
7
8         Bump off the timeout from 10 minutes to 4 minutes.
9
10         Also use float instead of str in test result values to compress the size of json files bots upload.
11
12         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
13         (PerfTestsRunner._parse_args):
14         (PerfTestsRunner._process_chromium_style_test_result):
15         (PerfTestsRunner._process_parser_test_result):
16         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
17         (test_run_test_set_for_parser_tests):
18         (test_run_test_set_with_json_output):
19         (test_run_test_set_with_json_source):
20
21 2012-01-25  Enrica Casucci  <enrica@apple.com>
22
23         WebView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
24         https://bugs.webkit.org/show_bug.cgi?id=76951
25         <rdar://problem/9325158>
26
27         Reviewed by Alexey Proskuryakov.
28
29         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Modified to reflect the changes to WebView
30         and WebHTMLView.
31         (TestWebKitAPI::TEST):
32
33 2012-01-23  MORITA Hajime  <morrita@google.com>
34
35         [Refactoring][Internals] Should have InternalSettings
36         https://bugs.webkit.org/show_bug.cgi?id=76424
37
38         Reviewed by Kent Tamura.
39
40         * GNUmakefile.am: Added so new files.
41
42 2012-01-25  Takashi Toyoshima  <toyoshim@chromium.org>
43
44         webkitpy: new-run-webkit-websocket is broken
45         https://bugs.webkit.org/show_bug.cgi?id=76999
46
47         Reviewed by Dirk Pranke.
48
49         r105674 broke new-run-webkit-websocket, since options.chromium
50         doesn't exist any more. new-run-webkit-httpd was also broken
51         and already fixed at r105835.
52         * Scripts/new-run-webkit-websocketserver:
53         (main):
54
55 2012-01-25  Nándor Huszka  <huszka.nandor@stud.u-szeged.hu>
56
57         [WK2] WebKitTestRunner needs layoutTestController.setPopupBlockingEnabled
58         https://bugs.webkit.org/show_bug.cgi?id=63458
59
60         Reviewed by Adam Roben.
61
62         Implement the setPopupBlockingEnabled method.
63
64         * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
65         * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
66         (WTR::LayoutTestController::setPopupBlockingEnabled):
67         * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
68
69 2012-01-24  Balazs Kelemen  <kbalazs@webkit.org>
70
71         [Qt] Cleanup testfonts handling
72         https://bugs.webkit.org/show_bug.cgi?id=76068
73
74         Reviewed by Simon Hausmann.
75
76         Factor test font initialization into it's own file  so test
77         font users can share the implementation by adding the file
78         to their build.
79         Use fonts.conf from the test font directory to not require the
80         current directory to be the WebKit root.
81         Fix the -use-test-fonts option of QtTestBrowser by initializing
82         the fonts earlier, before it is too late (namely before calling
83         show on the main window).
84
85         * DumpRenderTree/qt/DumpRenderTree.pro:
86         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
87         (WebCore::DumpRenderTree::open):
88         * DumpRenderTree/qt/DumpRenderTreeQt.h:
89         * DumpRenderTree/qt/QtInitializeTestFonts.cpp: Copied from Tools/WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp.
90         (WebKit::initializeTestFonts):
91         * DumpRenderTree/qt/QtInitializeTestFonts.h: Added.
92         * DumpRenderTree/qt/fonts.conf: Removed. No need for this since
93         we will use the one from the test fonts directory.
94         * DumpRenderTree/qt/main.cpp:
95         (main): Fix the ifdefs to make it clear that the code to
96         set up the Dpi has not been ported to Qt5 yet.
97         * QtTestBrowser/QtTestBrowser.pro:
98         * QtTestBrowser/launcherwindow.h:
99         (WindowOptions::WindowOptions):
100         * QtTestBrowser/main.cpp:
101         (launcherMain):
102         (LauncherApplication::handleUserOptions):
103         * WebKitTestRunner/InjectedBundle/Target.pri:
104         * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
105         (WTR::activateFonts): Remove the code to set up the Dpi
106         because it does not work with Qt5.
107
108 2012-01-25  Simon Hausmann  <simon.hausmann@nokia.com>
109
110         [Qt] build-webkit --qt requires a git-svn or SVN checkout.
111
112         Reviewed by Tor Arne Vestø.
113
114         Don't require it by making svnRevisionForDirectory return "unknown"
115         instead of bailing out with an error.
116
117         * Scripts/VCSUtils.pm:
118         (svnRevisionForDirectory):
119
120 2012-01-25  Adam Barth  <abarth@webkit.org>
121
122         Temporarily disable running the unit tests on the Chromium EWS.  These
123         tests don't work properly (see
124         https://bugs.webkit.org/show_bug.cgi?id=76979) and are currently
125         blocking the EWS.
126
127         * Scripts/webkitpy/common/config/ports.py:
128         (ChromiumPort):
129
130 2012-01-25  Adam Barth  <abarth@webkit.org>
131
132         Update results.html version of PixelZoomer
133         https://bugs.webkit.org/show_bug.cgi?id=76965
134
135         Reviewed by Ojan Vafai.
136
137         Improve readability as suggested by Ojan.
138
139         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
140
141 2012-01-25  Carlos Garcia Campos  <cgarcia@igalia.com>
142
143         [GTK] Page clients don't need to be GObjects anymore in WebKit2 GTK+ API
144         https://bugs.webkit.org/show_bug.cgi?id=76899
145
146         Reviewed by Martin Robinson.
147
148         * gtk/generate-gtkdoc:
149         (get_webkit2_options): WebKitWebLoaderClient has been renamed to
150         WebKitLoaderClient.
151
152 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
153
154         Build fix. Trigger Chromium Mac Release (Perf) on new commits.
155
156         * BuildSlaveSupport/build.webkit.org-config/config.json:
157
158 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
159
160         WebKit needs public performance tests running on public bots
161         https://bugs.webkit.org/show_bug.cgi?id=10266
162
163         Reviewed by Adam Barth.
164
165         Completes the initial implementation of WebKit Perf bots. Add a build step for run-perf-test
166         and make google-mac-4 run it, and upload the results to webkit-perf.webkit.org.
167
168         * BuildSlaveSupport/build.webkit.org-config/config.json:
169         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
170         (RunAndUploadPerfTests):
171         (RunAndUploadPerfTests.start):
172         (UploadPerfTestResults):
173         (BuildAndPerfTestFactory):
174         (BuildAndPerfTestFactory.__init__):
175
176 2012-01-24  Ojan Vafai  <ojan@chromium.org>
177
178         Initializing the browser property of the Bugzilla class takes too long
179         https://bugs.webkit.org/show_bug.cgi?id=76960
180
181         Reviewed by Adam Barth.
182
183         Lazily initialize it. This saves ~150ms on a no-op run-webkit-tests call
184         on my Mac Pro.
185
186         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
187         (Bugzilla.__init__):
188         (Bugzilla._get_browser):
189         (Bugzilla):
190         (Bugzilla._set_browser):
191
192 2012-01-24  Mark Rowe  <mrowe@apple.com>
193
194         Death to JavaScriptGlue.
195
196         Rubber-stamped by Geoffrey Garen.
197
198         * BuildSlaveSupport/build-launcher-app:
199         (copyNightlyLauncher):
200         * Scripts/build-webkit:
201         * Scripts/do-file-rename:
202         * Scripts/do-webcore-rename:
203         * Scripts/webkitpy/common/config/build.py:
204         (_should_file_trigger_build):
205         * Scripts/webkitpy/common/config/build_unittest.py: Update the test to use another Mac-specific
206         path in conjunction with a cross-platform path.
207         (ShouldBuildTest):
208
209 2012-01-24  Kentaro Hara  <haraken@chromium.org>
210
211         webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl should test stdout and stderr
212         https://bugs.webkit.org/show_bug.cgi?id=75836
213
214         Reviewed by David Kilzer.
215
216         Previously parser_unittests.pl can just test the returned value of
217         get_function_line_ranges() in prepare-ChangeLog. This patch fixes
218         parser_unittests.pl so that it also tests the stdout and stderr of
219         get_function_line_ranges().
220
221         This patch adds css_unittests_warning.css, which contains test cases
222         for warning messages of the CSS parser of prepare-ChangeLog.
223
224         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
225         (captureOutput): Captures the stdout and stderr of a given method.
226         (convertAbsolutepathToWebKitPath): Converts absolute paths in a given
227         string to relative paths.
228
229         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning-expected.txt: Added.
230         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests_warning.css: Added.
231         (element1):
232         (element2):
233         (element3):
234         (element4):
235         ():
236         (element5):
237         (element6):
238
239         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
240         Just updated the test results, i.e. added stdout and stderr results.
241         No essential changes in the test results.
242         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Ditto.
243         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/java_unittests-expected.txt: Ditto.
244         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Ditto.
245         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt: Ditto.
246         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/python_unittests-expected.txt: Ditto.
247
248 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
249
250         webkitpy: new-run-webkit-httpd is broken
251         https://bugs.webkit.org/show_bug.cgi?id=76963
252
253         Reviewed by Adam Barth.
254
255         r105674 broke new-run-webkit-httpd, since options.chromium
256         doesn't exist any more. We didn't catch this because we don't
257         run the integration tests by default anywhere :(.
258
259         * Scripts/new-run-webkit-httpd:
260         (run):
261         (main):
262
263 2012-01-24  Adam Barth  <abarth@webkit.org>
264
265         garden-o-matic pixel zoomer should let you look at every pixel of the image
266         https://bugs.webkit.org/show_bug.cgi?id=76962
267
268         Reviewed by Ojan Vafai.
269
270         Previously, you couldn't get up to the right and bottom pixels.
271
272         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/pixelzoomer.js:
273
274 2012-01-24  Enrica Casucci  <enrica@apple.com>
275
276         WebHTMLView should implement typingAttributes methods to work correctly with the Inspector bar in Mail.
277         https://bugs.webkit.org/show_bug.cgi?id=76951
278         <rdar://problem/9325158>
279
280         Added test to verify that WebHTMLView supports the typingAttributes method and that it
281         works as expected.
282         
283         Reviewed by Alexey Proskuryakov.
284
285         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
286         (TestWebKitAPI::TEST):
287
288 2012-01-24  James Robinson  <jamesr@chromium.org>
289
290         Put a space between the trac link and ellipsis in sheriffbot rollout message so it linkifies better in some IRC clients
291         https://bugs.webkit.org/show_bug.cgi?id=76884
292
293         Reviewed by Adam Barth.
294
295         * Scripts/webkitpy/tool/bot/irc_command.py:
296         (Rollout.execute):
297         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
298         (SheriffIRCBotTest.test_rollout):
299         (SheriffIRCBotTest.test_revert):
300         (SheriffIRCBotTest.test_multi_rollout):
301         (SheriffIRCBotTest.test_rollout_with_r_in_svn_revision):
302         (SheriffIRCBotTest.test_multi_rollout_with_r_in_svn_revision):
303         (SheriffIRCBotTest.test_rollout_invalidate_reason):
304         (test_multi_rollout_invalidate_reason):
305
306 2012-01-24  Kentaro Hara  <haraken@chromium.org>
307
308         The cpp parser of prepare-ChangeLog cannot detect a change in classes and namespaces
309         https://bugs.webkit.org/show_bug.cgi?id=75531
310
311         Reviewed by David Kilzer.
312
313         Previously, the cpp parser of prepare-ChangeLog could not detect a change
314         outside methods. Consider the following cpp file.
315
316             namespace N {
317             int a;     // this change does not appear on ChangeLog.
318             class C {
319                 int b;     // this change does not appear on ChangeLog.
320                 void f()
321                 {
322                     int c;     // this change appears on ChangeLog.
323                 }
324                 int d;     // this change does not appear on ChangeLog.
325             };
326             int e;     // this change does not appear on ChangeLog.
327             };
328
329         The previous prepare-ChangeLog outputs just methods in which a change is found:
330             (N::C::f):
331
332         This patch fixes prepare-ChangeLog so that it outputs namespaces, classes
333         and methods in which a change is found:
334             (N):
335             (N::C):
336             (N::C::f):
337
338         * Scripts/prepare-ChangeLog:
339         (get_function_line_ranges_for_cpp): Modified as described above.
340
341         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests.cpp: Added test cases.
342         (Class104):
343         (Class105):
344         (Class106):
345         (Class106::func32):
346         (Class106::func33):
347         (NameSpace3):
348         (NameSpace4):
349         (NameSpace5):
350         (NameSpace6):
351         (Class107):
352         (NameSpace5::NameSpace6::Class107::func34):
353         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/cpp_unittests-expected.txt:
354
355 2012-01-24  Kentaro Hara  <haraken@chromium.org>
356
357         prepare-ChangeLog outputs warnings for top-level { ... } blocks
358         https://bugs.webkit.org/show_bug.cgi?id=75943
359
360         Reviewed by David Kilzer.
361
362         Currently, prepare-ChangeLog outputs warnings for top-level { ... } blocks,
363         because it does not distinguish '}' at the end of a subroutine
364         with '}' at the end of a if or for statement.
365
366         foo.pl:
367             if (1) {
368             }
369             for (@v) {
370             }
371
372         output:
373             nested functions found at top-level at foo.pl:2
374             nested functions found at top-level at foo.pl:4
375
376         This patch fixes the bug.
377
378         * Scripts/prepare-ChangeLog:
379         (get_function_line_ranges_for_perl):
380         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
381         Added test cases. No change in the test results.
382
383 2012-01-24  Dirk Pranke  <dpranke@chromium.org>
384
385         r105674 broke check-webkit-style for chromium's test_expectations.
386
387         Unreviewed, build fix.
388         
389         We now need to pass in a platform name of 'chromium' if we want
390         the chromium port; options.chromium is no longer recognized. I
391         will follow this patch up with a separate patch that has a test,
392         but this at least fixes the build.
393
394         * Scripts/webkitpy/style/checkers/test_expectations.py:
395         (TestExpectationsChecker._determine_port_from_exepectations_path):
396
397 2012-01-20  Ojan Vafai  <ojan@chromium.org>
398
399         check-webkit-style of the chromium test_expectations.txt file takes too long
400         https://bugs.webkit.org/show_bug.cgi?id=76745
401
402         Reviewed by Dimitri Glazkov.
403
404         When in lint mode, have TestExpectations test all configurations instead
405         of looping over each configuration. This also has the benefit of making
406         the error output considerably more concise.
407
408         Also, got rid of the double-printing of errors when linting through check-webkit-style.
409
410         * Scripts/webkitpy/layout_tests/controllers/manager.py:
411         (Manager.lint):
412         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
413         (TestExpectations._report_errors):
414         (TestExpectations._add_expectations):
415         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
416         (test_parse_error_nonfatal):
417         (test_error_on_different_platform):
418         * Scripts/webkitpy/style/checkers/test_expectations.py:
419         (TestExpectationsChecker.check_test_expectations):
420         (TestExpectationsChecker.check):
421         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
422         (TestExpectationsTestCase.test_determine_port_from_exepectations_path):
423         (TestExpectationsTestCase.assert_lines_lint):
424
425 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
426
427         Another build fix attempt after r105543.
428
429         * Scripts/webkitpy/common/net/file_uploader.py:
430         (FileUploader.upload_as_multipart_form_data):
431
432 2012-01-24  Ryosuke Niwa  <rniwa@webkit.org>
433
434         Build fix attempt after r105543.
435
436         * Scripts/webkitpy/common/net/file_uploader.py:
437         (FileUploader.upload_as_multipart_form_data):
438
439 2012-01-24  Vincent Scheib  <scheib@chromium.org>
440
441         [Chromium] Add Pointer Lock test hooks and mock implementation to DumpRenderTree
442         https://bugs.webkit.org/show_bug.cgi?id=76411
443
444         Reviewed by Adam Barth.
445
446         Breaking up https://bugs.webkit.org/show_bug.cgi?id=75762
447         into a series of patches, this change adds a trivial
448         pointer lock implementation to the Chromium port's DumpRenderTree
449         and exposes LayoutTestController hooks to force failure conditions.
450
451         * DumpRenderTree/chromium/LayoutTestController.cpp:
452         (LayoutTestController::LayoutTestController):
453         (LayoutTestController::didLosePointerLock):
454         (LayoutTestController::setPointerLockWillFailAsynchronously):
455         (LayoutTestController::setPointerLockWillFailSynchronously):
456         * DumpRenderTree/chromium/LayoutTestController.h:
457         * DumpRenderTree/chromium/WebViewHost.cpp:
458         (WebViewHost::requestPointerLock):
459         (WebViewHost::requestPointerUnlock):
460         (WebViewHost::isPointerLocked):
461         (WebViewHost::didCompletePointerLock):
462         (WebViewHost::didNotCompletePointerLock):
463         (WebViewHost::didLosePointerLock):
464         (WebViewHost::reset):
465         * DumpRenderTree/chromium/WebViewHost.h:
466         (WebViewHost::setPointerLockWillFailAsynchronously):
467         (WebViewHost::setPointerLockWillFailSynchronously):
468
469 2012-01-24  Mario Sanchez Prada  <msanchez@igalia.com>
470
471         Unreviewed, skip WK2 API test timing out on some GTK bots.
472
473         WebKit2APITests/TestDownloads is consistently timing out on the
474         32bit release and 64bit debug bots.
475
476         * Scripts/run-gtk-tests:
477         (TestRunner): Skip WebKit2APITests/TestDownloads.
478
479 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
480
481         [Qt] Fix the build with the newes Qt5 hashes
482         https://bugs.webkit.org/show_bug.cgi?id=76657
483
484         Reviewed by Simon Hausmann.
485
486         * MiniBrowser/qt/MiniBrowserApplication.cpp:
487         (MiniBrowserApplication::MiniBrowserApplication): Disable the automatic touch->mouse event synthesis
488         for now to prevent an infinite loop where Qt always creates mouse events out of our touch mock events.
489
490 2012-01-24  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
491
492         [Qt] Distinguish between Cygwin and cmd.exe in the way we set env variables
493
494         We can't use win32-msvc* to decide whether or not to use (set FOO=bar), as
495         building with MinGW inside cmd.exe will require the same construct. Instead
496         we assume that when the first character in PWD is a '/' we're running under
497         Cygwin, and use the normal unix way of setting environment variables.
498
499         Reviewed by Simon Hausmann.
500
501         * qmake/mkspecs/features/default_pre.prf:
502
503 2012-01-24  Simon Hausmann  <simon.hausmann@nokia.com>
504
505         Some qmake build files must include Qt5's "quick" module.
506         https://bugs.webkit.org/show_bug.cgi?id=76671
507
508         Unreviewed build fix.
509
510         * MiniBrowser/qt/MiniBrowser.pro:
511         * WebKitTestRunner/Target.pri:
512
513 2012-01-24  Andras Becsi  <andras.becsi@nokia.com>
514
515         [Qt] Fix the event timestamps for touch mocking
516
517         Reviewed by Simon Hausmann.
518
519         * MiniBrowser/qt/MiniBrowserApplication.cpp:
520         (MiniBrowserApplication::sendTouchEvent): We should not set QEvent::None as the
521         timestamp argument in the QWindowSystemInterface::handleTouchEvent call.
522
523 2012-01-24  Carlos Garcia Campos  <cgarcia@igalia.com>
524
525         [GTK] Implement DownloadClient in WebKit2 GTK+ API
526         https://bugs.webkit.org/show_bug.cgi?id=72952
527
528         Reviewed by Martin Robinson.
529
530         * gtk/generate-gtkdoc:
531         (get_webkit2_options): Ignore WebKitDownloadClient.
532
533 2012-01-23  Zan Dobersek  <zandobersek@gmail.com>
534
535         [GTK] editing/deleting/5408255.html results are incorrect
536         https://bugs.webkit.org/show_bug.cgi?id=53644
537
538         Reviewed by Martin Robinson.
539
540         WEBKIT_TOP_LEVEL environment variable is now set directly in either
541         WebKitTestRunner or DumpRenderTree through usage of a compilation-time
542         macro. This way both tools can be run outside the test harness without
543         the need to manually set the environment variable.
544
545         * DumpRenderTree/gtk/DumpRenderTree.cpp:
546         (getTopLevelPath):
547         * GNUmakefile.am:
548         * Scripts/webkitpy/layout_tests/port/gtk.py:
549         (GtkPort.setup_environ_for_server):
550         * WebKitTestRunner/GNUmakefile.am:
551         * WebKitTestRunner/InjectedBundle/gtk/InjectedBundleGtk.cpp:
552         (WTR::InjectedBundle::platformInitialize):
553
554 2012-01-23  Dmitry Lomov  <dslomov@google.com>
555
556         [Chromium] Implement layoutTestController.workerThreadCount in DRT
557         https://bugs.webkit.org/show_bug.cgi?id=74653.
558
559         Reviewed by Darin Fisher.
560
561         * DumpRenderTree/chromium/LayoutTestController.cpp:
562         (LayoutTestController::LayoutTestController):
563         (LayoutTestController::workerThreadCount):
564         * DumpRenderTree/chromium/LayoutTestController.h:
565
566 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
567
568         nrwt: make --chromium work like --qt
569         https://bugs.webkit.org/show_bug.cgi?id=76875
570
571         Reviewed by Adam Barth.
572
573         --chromium used to have to be handled differently from --qt
574         due to the way the PortFactory was implemented; there's not
575         really a good reason for that any more so this patch makes
576         things slightly more consistent and eliminates the
577         options.chromium flag (--chromium is now truly a synonym for
578         --platform chromium).
579
580         * Scripts/webkitpy/layout_tests/port/factory.py:
581         (PortFactory._default_port):
582         (PortFactory.get):
583         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
584         (FactoryTest.setUp):
585         (FactoryTest.test_chromium_mac):
586         (FactoryTest.test_chromium_linux):
587         (FactoryTest.test_chromium_win):
588         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
589         (parse_args):
590
591 2012-01-23  Dirk Pranke  <dpranke@chromium.org>
592
593         run-webkit-tests needs to propagate --chromium
594         https://bugs.webkit.org/show_bug.cgi?id=76870
595
596         Reviewed by Eric Seidel.
597
598         run-webkit-tests removes '--chromium' argument from @ARGV when
599         determining which port to run, which means that that doesn't
600         propagate to new-run-webkit-tests. That's bad (and is handled
601         for the other ports by re-adding the flag, but apparently we're
602         just now noticing for Chromium).
603
604         * Scripts/run-webkit-tests:
605
606 2012-01-23  Brian Weinstein  <bweinstein@apple.com>
607
608         More build fixing after r105646.
609
610         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
611         (TestWebKitAPI::TestNode::key):
612
613 2012-01-23  Gavin Barraclough  <barraclough@apple.com>
614
615         Unreviewed build fix - r105646 broke this.
616
617         * TestWebKitAPI/Tests/WTF/RedBlackTree.cpp:
618         (TestWebKitAPI::TestNode::TestNode):
619         (TestWebKitAPI::TestNode::key):
620         (TestWebKitAPI::RedBlackTreeTest::assertEqual):
621         (TestWebKitAPI::RedBlackTreeTest::assertSameValuesForKey):
622         (TestWebKitAPI::RedBlackTreeTest::testDriver):
623
624 2012-01-23  Sheriff Bot  <webkit.review.bot@gmail.com>
625
626         Unreviewed, rolling out r105658.
627         http://trac.webkit.org/changeset/105658
628         https://bugs.webkit.org/show_bug.cgi?id=76883
629
630         We want this eventually, but not right at this moment
631         (Requested by abarth on #webkit).
632
633         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
634
635 2012-01-23  Adam Barth  <abarth@webkit.org>
636
637         garden-o-matic should support Chromium Mac Lion
638         https://bugs.webkit.org/show_bug.cgi?id=76880
639
640         Reviewed by Eric Seidel.
641
642         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
643
644 2012-01-23  Emil A Eklund  <eae@chromium.org>
645
646         check-webkit-style whitespace/operators triggers on overloaded division operator
647         https://bugs.webkit.org/show_bug.cgi?id=76650
648
649         Reviewed by Darin Adler.
650
651         * Scripts/webkitpy/style/checkers/cpp.py:
652         (check_spacing):
653         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
654         (CppStyleTest.test_operator_methods):
655
656 2012-01-23  Ryosuke Niwa  <rniwa@webkit.org>
657
658         run-perf-tests should report server-side errors
659         https://bugs.webkit.org/show_bug.cgi?id=76802
660
661         Reviewed by Tony Chang.
662
663         Report errors whenever server's response doesn't match "OK".
664
665         * Scripts/webkitpy/common/net/file_uploader.py:
666         (FileUploader.upload_single_text_file):
667         (FileUploader.upload_as_multipart_form_data):
668         (FileUploader._upload_data.callback):
669         (FileUploader):
670         (FileUploader._upload_data):
671         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
672         (PerfTestsRunner._upload_json):
673         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
674         (test_upload_json):
675         (test_upload_json.MockFileUploader.upload_single_text_file):
676
677 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
678
679         run-perf-tests ignore Skipped list on chromium
680         https://bugs.webkit.org/show_bug.cgi?id=76764
681
682         Reviewed by Dirk Pranke.
683
684         Move skipped_perf_tests from WebKit port to Base port so that Chromium port
685         can also find skipped list. Chromium port only uses test_expectations.txt for
686         layout tests but performacne tests don't use test_expectations.txt so Chromium port
687         also needs to use Skipped list.
688
689         * Scripts/webkitpy/layout_tests/port/base.py:
690         (Port._tests_from_skipped_file_contents):
691         (Port):
692         (Port._expectations_from_skipped_files):
693         (Port.skipped_perf_tests):
694         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
695         (PortTest.test_skipped_perf_tests):
696         (PortTest.test_skipped_perf_tests.add_text_file):
697         * Scripts/webkitpy/layout_tests/port/webkit.py:
698         (WebKitPort._skipped_tests_for_unsupported_features):
699         (WebKitPort._skipped_file_search_paths):
700         (WebKitPort.skipped_layout_tests):
701
702 2012-01-23  Andy Estes  <aestes@apple.com>
703
704         Fix the build after r105635.
705
706         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
707         (TestWebKitAPI::TEST):
708
709 2012-01-23  Xianzhu Wang  <wangxianzhu@chromium.org>
710
711         Basic enhancements to StringBuilder
712         https://bugs.webkit.org/show_bug.cgi?id=67081
713
714         Reviewed by Darin Adler.
715
716         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp:
717         (TestWebKitAPI::TEST):
718
719 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
720
721         [GTK] run-gtk-tests randomly fails while running the xprop comand
722         https://bugs.webkit.org/show_bug.cgi?id=76817
723
724         Reviewed by Gustavo Noronha Silva.
725
726         No need to use xprop to remove the AT_SPI_BUS property since
727         run-gtk-tests will always run new instances of Xvfb.
728
729         * Scripts/run-gtk-tests:
730         (TestRunner): Unskip TestWebKitAccessibility.
731         (TestRunner.run): Uncomment lines for launching the accessibility
732         bus and registry daemon, and remove lines for running xprop.
733
734 2012-01-23  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
735
736         [Qt] Don't warn about override and final being C++11 extensions
737
738         Clang will emit a warning when these extensions are used without passing
739         --std=c++11, but we use feature checking to decide if we have the right
740         extensions, so we can safely ignore these warnings. The XCode and Windows
741         project files have the same workaround.
742
743         The reason for adding the flag to QMAKE_OBJECTIVE_CFLAGS as well is that
744         we only have one extra compiler for Objective-C, which is also used for
745         Objective-C++ sources, so we have to pass the flag, even if it doesn't
746         make sense for Objective-C.
747
748         Reviewed by Simon Hausmann.
749
750         * qmake/mkspecs/features/unix/default_post.prf:
751
752 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
753
754         [Qt] Change how build-webkit decides when to do full incremental builds
755
756         Instead of relying on update-webkit (which isn't run on the bots) to
757         decide when to do a full incremental build (make qmake), we let the
758         build-webkit script itself check the current SVN revision against the
759         previous build (by storing it in .webkit.config).
760
761         If the two differ we assume a full incremental build is needed, since
762         the new revisions might have introduced problematic things like new
763         Q_OBJECT macros. If not, we assume the developer is doing changes
764         locally, and revert to doing a plain 'make'.
765
766         In addition, when the build fails in the latter case, we inform the
767         developer of possible pitfalls and how to manually run 'make qmake'.
768
769         Reviewed by Ossy.
770
771         * Scripts/update-webkit:
772         * Scripts/webkitdirs.pm:
773         (buildQMakeProjects):
774
775 2012-01-23  Mario Sanchez Prada  <msanchez@igalia.com>
776
777         [GTK] run-gtk-tests randomly fails while running the xprop comand
778         https://bugs.webkit.org/show_bug.cgi?id=76817
779
780         Reviewed by Philippe Normand.
781
782         Temporarily comment the lines related to launching the ATSPI bus
783         and registry daemon, which are making the bots to fail randomly.
784
785         * Scripts/run-gtk-tests:
786         (TestRunner): Skip TestWebKitAccessibility, as it won't run
787         properly if the ATSPI infrastructure is not properly initialized.
788         (TestRunner.run): Comment lines related to initialization of ATSPI.
789
790 2012-01-22  Ryosuke Niwa  <rniwa@webkit.org>
791
792         Windows python test build fix.
793
794         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
795         (PerfTestsRunner.__init__):
796
797 2012-01-22  Sergio Villar Senin  <svillar@igalia.com>
798
799         [GTK] DumpRenderTree converts "file:///" to a path differently
800         https://bugs.webkit.org/show_bug.cgi?id=76631
801
802         Reviewed by Martin Robinson.
803
804         DumpRenderTree should print "/" as the last path component if the
805         path is a lone slash instead of empty output.
806
807         * DumpRenderTree/gtk/DumpRenderTree.cpp:
808         (webViewConsoleMessage):
809
810 2012-01-20  David Levin  <levin@chromium.org>
811
812         The leaky DC check should use adoptPtr as a signal instead of OwnPtr to get less false positives.
813         https://bugs.webkit.org/show_bug.cgi?id=76752
814
815         Reviewed by Eric Seidel.
816
817         * Scripts/webkitpy/style/checkers/cpp.py:
818         (check_for_leaky_patterns):
819         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
820         (LeakyPatternTest.test_create_dc):
821         (LeakyPatternTest.test_create_compatible_dc):
822
823 2012-01-20  Shawn Singh  <shawnsingh@chromium.org>
824
825         Changed my status in commiters.py
826
827         Unreviewed.
828
829         * Scripts/webkitpy/common/config/committers.py:
830
831 2012-01-20  Ojan Vafai  <ojan@chromium.org>
832
833         run-webkit-tests --lint-test-files crawls the whole LayoutTests subtree
834         https://bugs.webkit.org/show_bug.cgi?id=76748
835
836         Reviewed by Ryosuke Niwa.
837
838         It crawls the whole subtree and then doesn't use the data. Cutting this out
839         saves 4 seconds warm and 17 seconds cold on my Mac Pro.
840
841         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
842         (run):
843
844 2012-01-20  Adam Barth  <abarth@webkit.org>
845
846         Another tiny tweak to the garden-o-matic CSS.  This makes things line
847         up slightly nicer when there aren't any failures.
848
849         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
850         (#summary):
851
852 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
853
854         run-perf-tests should support --test-results-server option
855         https://bugs.webkit.org/show_bug.cgi?id=76680
856
857         Reviewed by Adam Barth.
858
859         Add --test-results-server, --builder-name, and --build-number options to run-perf-tests
860         to be used by perf bots. Also refactor file_uploader as needed.
861
862         * Scripts/webkitpy/common/net/file_uploader.py:
863         (FileUploader.__init__):
864         (FileUploader.upload_single_file):
865         (FileUploader.upload_as_multipart_form_data):
866         (FileUploader):
867         (FileUploader._upload_data):
868         (FileUploader._upload_data.callback):
869         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:
870         (JSONResultsGeneratorBase.upload_json_files):
871         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
872         (PerfTestsRunner):
873         (PerfTestsRunner._parse_args):
874         (PerfTestsRunner.run):
875         (PerfTestsRunner._generate_json):
876         (PerfTestsRunner._upload_json):
877         (PerfTestsRunner._run_tests_set):
878         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
879         (create_runner):
880         (test_run_with_upload_json):
881         (test_run_with_upload_json.mock_upload_json):
882         (test_upload_json):
883         (test_upload_json.MockFileUploader):
884         (test_upload_json.MockFileUploader.__init__):
885         (test_upload_json.MockFileUploader.upload_single_file):
886         (test_parse_args):
887
888 2012-01-20  Adam Barth  <abarth@webkit.org>
889
890         The party time image overlaps real content!  This patch makes the
891         image centered vertically.
892
893         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/onebar.css:
894         (#onebar.partytime #summary):
895
896 2012-01-19  Ojan Vafai  <ojan@chromium.org>
897
898         Refactor TestExpectationsParser in preparation for caching the results
899         https://bugs.webkit.org/show_bug.cgi?id=76669
900
901         Reviewed by Dimitri Glazkov.
902
903         Make everything private expect for the parse method. 
904         Eventually, we'll need the expectations lines to not be modified
905         outside of TestExpectationsParser so we can cache the results.
906         This makes check-webkit-style of the chromium test_expectations.txt file
907         go from ~17 seconds to ~12 seconds on my Mac Pro.
908
909         This patch is just a refactor in preparation, so no new tests.
910
911         * Scripts/webkitpy/layout_tests/controllers/test_expectations_editor_unittest.py:
912         (TestExpectationEditorTests.make_parsed_expectation_lines):
913         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
914         (TestExpectationParser.parse):
915         (TestExpectationParser):
916         (TestExpectationParser._parse_line):
917         (TestExpectationParser._tokenize):
918         (TestExpectationParser._tokenize_list):
919         (TestExpectationsModel._clear_expectations_for_test):
920         (TestExpectations.__init__):
921         (TestExpectations._add_expectations):
922         (TestExpectations._add_skipped_tests):
923         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
924         (TestExpectationParserTests.test_tokenize_blank):
925         (TestExpectationParserTests.test_tokenize_missing_colon):
926         (TestExpectationParserTests.test_tokenize_extra_colon):
927         (TestExpectationParserTests.test_tokenize_empty_comment):
928         (TestExpectationParserTests.test_tokenize_comment):
929         (TestExpectationParserTests.test_tokenize_missing_equal):
930         (TestExpectationParserTests.test_tokenize_extra_equal):
931         (TestExpectationParserTests.test_tokenize_valid):
932         (TestExpectationParserTests.test_tokenize_valid_with_comment):
933         (TestExpectationParserTests.test_tokenize_valid_with_multiple_modifiers):
934         (TestExpectationParserTests.test_parse_empty_string):
935         (TestExpectationSerializerTests.assert_round_trip):
936         (TestExpectationSerializerTests.assert_list_round_trip):
937         * Scripts/webkitpy/tool/commands/expectations.py:
938         (OptimizeExpectations.execute):
939         * Scripts/webkitpy/tool/servers/gardeningserver.py:
940         (GardeningExpectationsUpdater.update_expectations):
941
942 2012-01-20  Adam Barth  <abarth@webkit.org>
943
944         Follow-up to previous patch: don't produce NaN when the revision number
945         is missing.
946
947         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
948         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
949
950 2012-01-20  Adam Barth  <abarth@webkit.org>
951
952         Garden-o-matic should tell me which revisions have been checked by all the bots
953         https://bugs.webkit.org/show_bug.cgi?id=76722
954
955         Reviewed by Dimitri Glazkov.
956
957         This information helps me know when it's safe for me to roll a
958         candidate revision.
959
960         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/garden-o-matic.js:
961         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
962         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model_unittests.js:
963
964 2012-01-20  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
965
966         build-webkit: Don't spit out congratulations message on Ctrl+C
967
968         Reviewed by Andreas Kling.
969
970         * Scripts/build-webkit:
971
972 2012-01-20  Philippe Normand  <pnormand@igalia.com>
973
974         [GTK] Allow extra jhbuild modulesets
975         https://bugs.webkit.org/show_bug.cgi?id=76691
976
977         Reviewed by Gustavo Noronha Silva.
978
979         * gtk/jhbuildrc: Extra modulesets and modules can be added in
980         JHBuild using the WEBKIT_EXTRA_MODULESETS and WEBKIT_EXTRA_MODULES
981         env variables respectively. Use comma-separated values. Example:
982         WEBKIT_EXTRA_MODULES=file:///path/to/module.set,file:///other/path.
983
984 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
985
986         [GTK] Expose accessibility hierarchy in WebKit2 to ATK/AT-SPI based ATs
987         https://bugs.webkit.org/show_bug.cgi?id=72589
988
989         Reviewed by Martin Robinson.
990
991         Ensure the accessibility infrastructure is available when running
992         the unit tests, otherwise WK2's accessibility tests won't work.
993
994         * Scripts/run-gtk-tests:
995         (TestRunner):
996         (TestRunner._lookup_atspi2_binary): New. Looks for the directory
997         where at-spi2's stuff is located (like at-spi-bus-launcher and
998         at-spi2-registryd).
999         (TestRunner.run): Initialize the accessibility infrastructure.
1000
1001         * gtk/generate-gtkdoc:
1002         (get_webkit2_options): Add 'WebKitWebViewBaseAccessible.*'.
1003
1004         * gtk/jhbuild.modules: Added at-spi2-core and at-spi2-atk.
1005
1006 2012-01-20  Hayato Ito  <hayato@chromium.org>
1007
1008         Add ShadowRoot.idl which is enabled by newly introduced SHADOW_DOM flag.
1009         https://bugs.webkit.org/show_bug.cgi?id=76353
1010
1011         Reviewed by Hajime Morita.
1012
1013         Add ShadowRoot.idl, which is enabled only on chromium port since this is
1014         under development feature.
1015         ShadowRoot.idl contains minimum API so that we can test it.
1016         Other APIs should be added on other changes so that we can isolate issues.
1017
1018         * DumpRenderTree/chromium/TestShell.cpp:
1019         (TestShell::TestShell):
1020
1021 2012-01-20  Mario Sanchez Prada  <msanchez@igalia.com>
1022
1023         [GTK] GTK's DRT not to log events for already defunct objects
1024         https://bugs.webkit.org/show_bug.cgi?id=76620
1025
1026         Reviewed by Martin Robinson.
1027
1028         Do not log 'state-change:defunct' events.
1029
1030         * DumpRenderTree/gtk/AccessibilityCallbacks.cpp:
1031         (printAccessibilityEvent): Early return if a
1032         'state-change:defunct' signal is passed.
1033         (axObjectEventListener): Pass the signal name and value in
1034         separate parameters to printAccessibilityEvent.
1035
1036 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
1037
1038         Unreviewed build fix. Wrong variable being used as the cwd for git
1039         remote update command.
1040
1041         * gtk/run-with-jhbuild:
1042         (update_jhbuild):
1043
1044 2012-01-19  Gustavo Noronha Silva  <gns@gnome.org>
1045
1046         [GTK] ensure the jhbuild used by webkit is as up-to-date as needed
1047         https://bugs.webkit.org/show_bug.cgi?id=76585
1048
1049         Reviewed by Martin Robinson.
1050
1051         This allows us to specify a minimum version of jhbuild required,
1052         while also providing stability against changes done to the master
1053         branch, by letting us choose what is the exact version that will
1054         be used.
1055
1056         * gtk/run-with-jhbuild: refactored for more hackability and easier
1057         reading
1058         (jhbuild_installed): checks whether jhbuild is installed
1059         (jhbuild_cloned): checks whether jhbuild has been cloned
1060         (jhbuild_at_expected_revision): checks whether jhbuild is at the
1061         expected revision
1062         (update_jhbuild): brings jhbuild to the expected revision
1063         (clone_jhbuild): clones jhbuild
1064         (install_jhbuild): build and installs jhbuild
1065         (update_webkitgtk_libs): runs update-webkitgtk-libs
1066         (ensure_jhbuild): high-level logic to decide whether and which of
1067         the steps above are executed
1068
1069 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
1070
1071         [WK2] DumpRenderTree converts "file:///" to a path differently
1072         https://bugs.webkit.org/show_bug.cgi?id=76653
1073
1074         Reviewed by John Sullivan.
1075
1076         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: (WTR::lastFileURLPathComponent):
1077         Do print a slash if there are no path components.
1078
1079 2012-01-19  Eric Seidel  <eric@webkit.org>
1080
1081         webkit-patch cannot rollout patches without changelog/bug number
1082         https://bugs.webkit.org/show_bug.cgi?id=75962
1083
1084         Reviewed by Adam Barth.
1085
1086         * Scripts/webkitpy/tool/commands/download.py:
1087         (AbstractRolloutPrepCommand._prepare_state):
1088         * Scripts/webkitpy/tool/commands/download_unittest.py:
1089         (AbstractRolloutPrepCommandTest.test_prepare_state):
1090
1091 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
1092
1093         [WK2] fast/url/degenerate-file-base.html fails
1094         https://bugs.webkit.org/show_bug.cgi?id=76619
1095
1096         Reviewed by Adam Roben.
1097
1098         A URL generated form local path for loading had an extra slash, making document.URL
1099         incorrect in all WebKit2 tests (file://localhost//path/to/test.html).
1100
1101         * WebKitTestRunner/TestInvocation.cpp: (WTR::createWKURL):
1102
1103 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1104
1105         check-webkit-style of the chromium test_expectations.txt file doesn't test all chromium ports
1106         https://bugs.webkit.org/show_bug.cgi?id=76510
1107
1108         Reviewed by Adam Barth.
1109
1110         Unlike other style rules, if you get the syntax of the test_expectations.txt
1111         file wrong, the layout tests won't run. Also, this check is simple and only
1112         slows down committing if you actually modify one of the test_expectations.txt files.
1113
1114         * Scripts/webkitpy/style/checkers/test_expectations.py:
1115         (TestExpectationsChecker.check_test_expectations):
1116         (TestExpectationsChecker.check):
1117         * Scripts/webkitpy/style/checkers/test_expectations_unittest.py:
1118         (TestExpectationsTestCase.test_check_covers_all_configurations):
1119         (TestExpectationsTestCase.test_check_covers_all_configurations.mock_check_test_expectations):
1120         (TestExpectationsTestCase):
1121         (assert_lines_lint):
1122
1123 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
1124
1125         Build fix. Revert an unintentional change to WebKitDriver in r105443.
1126
1127         * Scripts/webkitpy/layout_tests/port/driver.py:
1128         (DriverProxy.__init__):
1129
1130 2012-01-19  Dirk Pranke  <dpranke@chromium.org>
1131
1132         remove the duplicated build rules from WebKit.gyp, Tools.gyp
1133         https://bugs.webkit.org/show_bug.cgi?id=73384
1134
1135         Reviewed by Tony Chang.
1136
1137         This patch removes the no-longer-necessary conditional logic for
1138         build_webkit_exes_from_webkit_gyp; we now always build the exes
1139         from their own dedicated gyp files.
1140
1141         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1142
1143 2012-01-19  Alexey Proskuryakov  <ap@apple.com>
1144
1145         make-new-script-test should work with HTTP tests
1146         https://bugs.webkit.org/show_bug.cgi?id=76603
1147
1148         Reviewed by Adam Roben.
1149
1150         * Scripts/make-new-script-test:
1151         (makePathToSharedSources): Renamed from makeTestRelativePathToSharedSources(), since it
1152         now also creates absolute paths for HTTP.
1153
1154 2012-01-19  Ryosuke Niwa  <rniwa@webkit.org>
1155
1156         Some perf tests time out when ran by run-perf-tests
1157         https://bugs.webkit.org/show_bug.cgi?id=76612
1158
1159         Reviewed by Dirk Pranke and Eric Seidel.
1160
1161         Always pass --no-timeout to DumpRenderTree from run-perf-tests.
1162         Otherwise some tests such as Parser/xml-parser.html will timeout.
1163
1164         --no-timeout option is currently supported by Chromium and Mac ports.
1165
1166         * Scripts/webkitpy/layout_tests/port/base.py:
1167         (Port.to.create_driver):
1168         * Scripts/webkitpy/layout_tests/port/chromium.py:
1169         (ChromiumDriver.__init__):
1170         (ChromiumDriver._wrapper_options):
1171         * Scripts/webkitpy/layout_tests/port/driver.py:
1172         (Driver.__init__):
1173         (DriverProxy.__init__):
1174         * Scripts/webkitpy/layout_tests/port/webkit.py:
1175         (WebKitDriver.__init__):
1176         (WebKitDriver.cmd_line):
1177         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
1178         (WebKitDriverTest.test_read_binary_block):
1179         (WebKitDriverTest):
1180         (WebKitDriverTest.test_no_timeout):
1181         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
1182         (get_tests_run.RecordingTestDriver.__init__):
1183         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1184         (PerfTestsRunner._parse_args):
1185         (PerfTestsRunner._run_tests_set):
1186         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1187         (create_runner):
1188
1189 2012-01-19  Ben Wells  <benwells@chromium.org>
1190
1191         Add support for window.print to chromium DRT
1192         https://bugs.webkit.org/show_bug.cgi?id=76479
1193
1194         When used in a test window.print goes into print mode and then
1195         straight out again. This will allow the afterprint event to be
1196         tested by having the afterprint event happen before the dump
1197         happens.
1198
1199         Reviewed by Mihai Parparita.
1200
1201         * DumpRenderTree/chromium/WebViewHost.cpp:
1202         (WebViewHost::printPage):
1203         * DumpRenderTree/chromium/WebViewHost.h:
1204
1205 2012-01-18  Enrica Casucci  <enrica@apple.com>
1206
1207         editingAttributedStringFromRange in WebHTMLConverter does not handle NSUnderlineStyleAttributeName.
1208         https://bugs.webkit.org/show_bug.cgi?id=76588
1209         <rdar://problem/9325183>
1210
1211         Added test.
1212         
1213         Reviewed by Dan Bernstein.
1214
1215         * TestWebKitAPI/Tests/mac/InspectorBar.mm:
1216         (TestWebKitAPI::TEST):
1217
1218 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1219
1220         [Qt] Do full incremental builds until bots use update-webkit
1221
1222         Rubber-stamped by Simon Hausmann.
1223
1224         * Scripts/webkitdirs.pm:
1225         (buildQMakeProjects):
1226
1227 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1228
1229         Fix the Qt build
1230
1231         Unreviewed build fix.
1232
1233         * Scripts/webkitdirs.pm:
1234         (buildQMakeProjects):
1235
1236 2012-01-19  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1237
1238         [Qt] Make build-webkit always do safe incremental builds after update-webkit
1239
1240         When building against Qt5 with GCC we would just run 'make' in the build dir,
1241         and rely on the dependency files output by GCC (-MD) to handle dependency
1242         tracking, but that fails for special-cases like adding a Q_OBJECT macro
1243         to a header.
1244
1245         To guarantee that an incrmental build will work, we have to run 'make qmake',
1246         which we now do on every build-webkit that's followed by a successful run
1247         of update-webkit. The reasoning is that update-webkit can result in such
1248         potential corner-cases being applied, and since we can't know for sure
1249         unless we inspect the diff and account for all the corner cases we assume
1250         the worst and always run 'make qmake'.
1251
1252         After a succesful run of build-webkit we proceed to do just 'make' for any
1253         subsequent runs, since we assume that the developer knows what kind of
1254         changes he/she is doing, and when a 'make qmake' is needed.
1255
1256         Reviewed by Simon Hausmann.
1257
1258         * Scripts/update-webkit:
1259         * Scripts/webkitdirs.pm:
1260         (determineCurrentSVNRevision):
1261         (currentSVNRevision):
1262         (buildQMakeProjects):
1263         * qmake/mkspecs/features/default_post.prf:
1264
1265 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
1266
1267         run-perf-tests should support Skipped list
1268         https://bugs.webkit.org/show_bug.cgi?id=76594
1269
1270         Reviewed by Adam Barth.
1271
1272         Add a support for Skipped list in run-perf-tests; also skip files in resources directories.
1273
1274         * Scripts/webkitpy/layout_tests/port/base.py:
1275         (Port.perf_tests_dir):
1276         (Port.skipped_perf_tests):
1277         (Port.skips_perf_test):
1278         (Port.relative_perf_test_filename):
1279         * Scripts/webkitpy/layout_tests/port/test.py:
1280         (TestPort.perf_tests_dir):
1281         * Scripts/webkitpy/layout_tests/port/webkit.py:
1282         (WebKitPort._expectations_from_skipped_files):
1283         (WebKitPort):
1284         (WebKitPort.skipped_layout_tests):
1285         (WebKitPort.skipped_perf_tests):
1286         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1287         (PerfTestsRunner._collect_tests):
1288         (PerfTestsRunner.run):
1289         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1290         (test_run_test_set_with_json_output):
1291         (test_run_test_set_with_json_source):
1292         (test_collect_tests_with_skipped_list):
1293         (test_collect_tests_with_skipped_list.add_file):
1294
1295 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
1296
1297         Fix path to chromium_src_dir introduced in previous change.
1298
1299         Unreviewed, build fix.
1300
1301         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp:
1302
1303 2012-01-18  Alexey Proskuryakov  <ap@apple.com>
1304
1305         Need infrastructure to test Content-Disposition filename encoding support
1306         https://bugs.webkit.org/show_bug.cgi?id=76572
1307         <rdar://problem/6421825>
1308
1309         Reviewed by Sam Weinig.
1310
1311         Made policy delegate log suggested file name when resource has Content-Disposition: attachment.
1312
1313         WebKitTestRunner does not have a policy delegate yet, bug 42546.
1314
1315         * DumpRenderTree/mac/PolicyDelegate.mm:
1316         (dispositionTypeFromContentDispositionHeader):
1317         (-[PolicyDelegate webView:decidePolicyForMIMEType:request:frame:decisionListener:]):
1318
1319 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1320
1321         Fix tests from http://trac.webkit.org/changeset/105332.
1322
1323         The code assumed that if you passed an options object with a configuration
1324         attribute that it would have a valid value. A bunch of the testing mocks
1325         do things like having a configuration value of None.
1326
1327         * Scripts/webkitpy/layout_tests/port/base.py:
1328         (Port.__init__):
1329
1330 2012-01-18  Ojan Vafai  <ojan@chromium.org>
1331
1332         Shave 0.5 seconds off check-webkit-style runtime for test_expectations.txt
1333         https://bugs.webkit.org/show_bug.cgi?id=76576
1334
1335         Reviewed by Adam Barth.
1336
1337         Avoid calling default_configuration in base.py. We don't need it for
1338         checking test_expectations.txt style. It takes ~0.5 seconds to run on my
1339         Mac Pro. It's the call to "perl Tools/Scripts/webkit-build-directory --top-level"
1340         from common.executive.
1341
1342         At some point someone should probably look into why that call is so slow
1343         since it's on the critical path for run-webkit-tests and build-webkit startup.
1344
1345         * Scripts/webkitpy/layout_tests/port/base.py:
1346         (Port.__init__):
1347         * Scripts/webkitpy/style/checkers/test_expectations.py:
1348         (TestExpectationsChecker._determine_port_from_exepectations_path):
1349
1350 2012-01-18  Dirk Pranke  <dpranke@chromium.org>
1351
1352         [chromium] move Tools.gyp, switch build-webkit --chromium to All.gyp
1353         https://bugs.webkit.org/show_bug.cgi?id=76505
1354
1355         Reviewed by Tony Chang.
1356
1357         Update chromium build script to use All.gyp instead of WebKit.gyp
1358
1359         * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp: Renamed from Source/WebKit/chromium/Tools.gyp.
1360         * Scripts/webkitdirs.pm:
1361
1362 2012-01-18  Sam Weinig  <sam@webkit.org>
1363
1364         Don’t hardcode the path to ‘make‘.
1365
1366         Reviewed by Dan Bernstein.
1367
1368         * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
1369
1370 2012-01-18  Joshua Bell  <jsbell@chromium.org>
1371
1372         Adding myself to committers.py
1373         https://bugs.webkit.org/show_bug.cgi?id=76569
1374
1375         Unreviewed.
1376
1377         * Scripts/webkitpy/common/config/committers.py:
1378
1379 2012-01-18  Adam Barth  <abarth@webkit.org>
1380
1381         Up the cap on test failures tolerated by the EWS so the mac-ews can run to completion
1382         https://bugs.webkit.org/show_bug.cgi?id=76567
1383
1384         Reviewed by Eric Seidel.
1385
1386         Current the AppleMac port has slightly more failures than we allow on
1387         the EWS, causing the mac-ews to spin.  This patch ups the limit a bit,
1388         which should hopefully let the mac-ews run to completion.  Once
1389         AppleMac has fewer failures, we can lower the cap again.
1390
1391         * Scripts/webkitpy/tool/bot/layouttestresultsreader_unittest.py:
1392         (LayoutTestResultsReaderTest.test_layout_test_results):
1393         * Scripts/webkitpy/tool/steps/runtests.py:
1394         (RunTests):
1395         * Scripts/webkitpy/tool/steps/steps_unittest.py:
1396
1397 2012-01-18  Eric Carlson  <eric.carlson@apple.com>
1398
1399         Provide access to user's list of preferred languages
1400         https://bugs.webkit.org/show_bug.cgi?id=76138
1401
1402         Reviewed by Alexey Proskuryakov.
1403
1404         * GNUmakefile.am: Add JSInternalsCustom.cpp.
1405
1406 2012-01-18  Adam Barth  <abarth@webkit.org>
1407
1408         Add Chromium Linux GPU 32 to the list of allowed Chromium Linux
1409         configurations.
1410
1411         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1412         (ChromiumLinuxPort.__init__):
1413
1414 2012-01-18  Dominic Mazzoni  <dmazzoni@google.com>
1415
1416         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
1417         https://bugs.webkit.org/show_bug.cgi?id=73460
1418
1419         Reviewed by Chris Fleizach.
1420
1421         * DumpRenderTree/AccessibilityUIElement.h:
1422         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1423         (AccessibilityUIElement::AccessibilityUIElement):
1424         (AccessibilityUIElement::scrollToMakeVisibleCallback):
1425         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
1426         (AccessibilityUIElement::scrollToGlobalPointCallback):
1427         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1428         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1429         (AccessibilityUIElement::scrollToMakeVisible):
1430         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1431         (AccessibilityUIElement::scrollToGlobalPoint):
1432         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1433         (AccessibilityUIElement::scrollToMakeVisible):
1434         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1435         (AccessibilityUIElement::scrollToGlobalPoint):
1436         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1437         (AccessibilityUIElement::scrollToMakeVisible):
1438         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1439         (AccessibilityUIElement::scrollToGlobalPoint):
1440
1441 2012-01-17  Enrica Casucci  <enrica@apple.com>
1442
1443         Missing NSForegroundColorAttributeName should be treated as black instead of transparent.
1444         https://bugs.webkit.org/show_bug.cgi?id=76490
1445         <rdar://problem/9460733>
1446
1447         Added TestWebKitAPI test.
1448         
1449         Reviewed by Dan Bernstein.
1450
1451         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added new test file.
1452         * TestWebKitAPI/Tests/mac/InspectorBar.mm: Added test that calls
1453         changeAttributes to the WebView.
1454         
1455 2012-01-18  Balazs Kelemen  <kbalazs@webkit.org>
1456
1457         [Qt] Consolidate layout test crash logging
1458         https://bugs.webkit.org/show_bug.cgi?id=75088
1459
1460         Reviewed by Simon Hausmann.
1461
1462         Move backtrace generating logic into WTFReportBacktrace
1463         and add a way to deinstall signal handlers if we know
1464         that we have already printed the backtrace.
1465
1466         * DumpRenderTree/qt/main.cpp:
1467         (crashHandler):
1468         (setupSignalHandlers):
1469         (WTFCrashHook):
1470         (main):
1471         * WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
1472         (WTR::crashHandler):
1473         (WTR::setupSignalHandlers):
1474         (WTR::crashHook):
1475         (WTR::InjectedBundle::platformInitialize):
1476
1477 2012-01-18  Ilya Tikhonovsky  <loislo@chromium.org>
1478
1479         Unreviewed build fix after r105256.
1480
1481         There was a cyclic dependency between self._port and self._host assigments.
1482
1483         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1484         (PerfTestsRunner.__init__):
1485
1486 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1487
1488         Add new watchlist rule for the Qt build system
1489
1490         Reviewed by Simon Hausmann.
1491
1492         * Scripts/webkitpy/common/config/watchlist:
1493
1494 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1495
1496         [Qt] Move OTHER_FILES from WebKit.pro to Tools.pro
1497
1498         Reviewed by Simon Hausmann.
1499
1500         * Tools.pro:
1501
1502 2012-01-18  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1503
1504         [Qt] Allow building only a subset of the WebKit sub-projects
1505
1506         Fixes build-jsc, broken in r104825.
1507
1508         We no longer need the intermediate buildQMakeQtProject(), as we're the
1509         only port using qmake, and buildQMakeProject is now highly Qt specific.
1510
1511         <http://webkit.org/b/76179>
1512
1513         Reviewed by Simon Hausmann.
1514
1515         * Scripts/build-jsc:
1516         * Scripts/build-webkit:
1517         * Scripts/webkitdirs.pm:
1518         (buildQMakeProjects):
1519
1520 2012-01-18  Ryosuke Niwa  <rniwa@webkit.org>
1521
1522         run-perf-tests should generate a json file that summaries the result
1523         https://bugs.webkit.org/show_bug.cgi?id=76504
1524
1525         Reviewed by Adam Barth.
1526
1527         Add an ability to generate a json file to run-perf-test in the preparation for perf bots.
1528         New option --outout-json-path specifies the json file's path, and --source-json-path specifies
1529         another json file to be merged into the generated json file.
1530
1531         Also fixed a bug that --build wasn't set by default.
1532
1533         * Scripts/webkitpy/layout_tests/port/test.py:
1534         (TestPort.webkit_base):
1535         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
1536         (PerfTestsRunner.__init__):
1537         (PerfTestsRunner):
1538         (PerfTestsRunner._parse_args):
1539         (PerfTestsRunner.run):
1540         (PerfTestsRunner._generate_json_if_specified):
1541         (PerfTestsRunner._process_chromium_style_test_result):
1542         (PerfTestsRunner._process_parser_test_result):
1543         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
1544         (MainTest):
1545         (create_runner):
1546         (test_run_test_set):
1547         (test_run_test_set_for_parser_tests):
1548         (test_run_test_set_with_summary_json):
1549         (test_collect_tests):
1550         (test_parse_args):
1551
1552 2012-01-17  Sheriff Bot  <webkit.review.bot@gmail.com>
1553
1554         Unreviewed, rolling out r105244.
1555         http://trac.webkit.org/changeset/105244
1556         https://bugs.webkit.org/show_bug.cgi?id=76518
1557
1558         broke Chromium Mac (Requested by rolandsteiner on #webkit).
1559
1560         * DumpRenderTree/AccessibilityUIElement.h:
1561         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1562         (AccessibilityUIElement::AccessibilityUIElement):
1563         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1564         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1565         (AccessibilityUIElement::removeSelection):
1566         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1567         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1568         (AccessibilityUIElement::removeSelection):
1569
1570 2012-01-17  Dominic Mazzoni  <dmazzoni@google.com>
1571
1572         Accessibility: Chromium needs methods to scroll an object into view or to a specific location.
1573         https://bugs.webkit.org/show_bug.cgi?id=73460
1574
1575         Reviewed by Chris Fleizach.
1576
1577         * DumpRenderTree/AccessibilityUIElement.h:
1578         * DumpRenderTree/chromium/AccessibilityUIElement.cpp:
1579         (AccessibilityUIElement::AccessibilityUIElement):
1580         (AccessibilityUIElement::scrollToMakeVisibleCallback):
1581         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocusCallback):
1582         (AccessibilityUIElement::scrollToGlobalPointCallback):
1583         * DumpRenderTree/chromium/AccessibilityUIElement.h:
1584         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
1585         (AccessibilityUIElement::scrollToMakeVisible):
1586         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1587         (AccessibilityUIElement::scrollToGlobalPoint):
1588         * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
1589         (AccessibilityUIElement::scrollToMakeVisible):
1590         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1591         (AccessibilityUIElement::scrollToGlobalPoint):
1592         * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
1593         (AccessibilityUIElement::scrollToMakeVisible):
1594         (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
1595         (AccessibilityUIElement::scrollToGlobalPoint):
1596
1597 2012-01-17  Adam Barth  <abarth@webkit.org>
1598
1599         Enable tests on the mac-ews
1600         https://bugs.webkit.org/show_bug.cgi?id=76502
1601
1602         Reviewed by Simon Fraser.
1603
1604         Apple now has five machines running the mac-ews.  Lucas Forschler has
1605         asked that we try enabling tests to see if we have enough bandwidth.
1606
1607         * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
1608         (MacEWS):
1609
1610 2012-01-17  Sanjoy Das  <sanjoy@playingwithpointers.com>
1611
1612         Add a check-webkit-style exception for GDBInterface.cpp
1613         https://bugs.webkit.org/show_bug.cgi?id=76187
1614
1615         Add an exception for Source/JavaScriptCore/jit/GDBInterface.cpp so
1616         that it does not fail the style check.
1617
1618         Reviewed by Martin Robinson.
1619
1620         * Scripts/webkitpy/style/checker.py: Tell the style checker to not run
1621         readability/naming tests on
1622         Source/JavaScriptCore/jit/GDBInterface.cpp.
1623         * Scripts/webkitpy/style/checker_unittest.py:
1624         (GlobalVariablesTest.test_path_rules_specifier): Add a test for
1625         the above exception.
1626
1627 2012-01-17  Hajime Morrita  <morrita@chromium.org>
1628
1629         https://bugs.webkit.org/show_bug.cgi?id=76245
1630         [Mac][DRT] should allow query string for the file URL.
1631
1632         Reviewed by Tony Chang.
1633
1634         Taught DRT about file:// as a url instead of a file path.
1635         We are now able to use a path file:///foo/bar.html?baz as a url
1636         with which DRT opens "/foo/bar.html" with "baz" as a query string.
1637
1638         * DumpRenderTree/mac/DumpRenderTree.mm:
1639         (runTest):
1640
1641 2012-01-17  Ojan Vafai  <ojan@chromium.org>
1642
1643         Check style on test_expectations.txt files before commit
1644         https://bugs.webkit.org/show_bug.cgi?id=76484
1645
1646         Reviewed by Adam Barth.
1647
1648         Unlike other style rules, if you get the syntax of the test_expectations.txt
1649         file wrong, the layout tests won't run. Also, this check is simple and only
1650         slows down committing if you actually modify one of the test_expectations.txt files.
1651
1652         * Scripts/webkitpy/tool/steps/commit.py:
1653         (Commit._check_test_expectations):
1654         (Commit.run):
1655         * Scripts/webkitpy/tool/steps/commit_unittest.py: Added.
1656         (CommitTest):
1657         (CommitTest.test_check_test_expectations):
1658
1659 2012-01-17  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>
1660
1661         [Qt] Debug build fails with debug qt5
1662         https://bugs.webkit.org/show_bug.cgi?id=76463
1663
1664         Reviewed by Simon Hausmann.
1665
1666         Add QNetworkCookie include in order to satisfy sizeof in QTypeInfo in debug builds.
1667
1668         * QtTestBrowser/cookiejar.h:
1669
1670 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1671
1672         webkitpy: fix json import on linux
1673         https://bugs.webkit.org/show_bug.cgi?id=76481
1674
1675         Reviewed by Adam Barth.
1676
1677         For some reason, the version of simplejson autoinstalled on my linux
1678         box seems to be broken. We should really only be using
1679         simplejson if json isn't available anyway, so this change looks
1680         for json first.
1681
1682         * Scripts/webkitpy/layout_tests/layout_package/json_results_generator_unittest.py:
1683         (JSONGeneratorTest.test_test_timings_trie):
1684
1685 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1686
1687         Fix failures in test-webkitpy caused by r105177.
1688
1689         Unreviewed, build fix.
1690
1691         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1692         (ChromiumLinuxPort._determine_driver_path_statically):
1693         (ChromiumLinuxPort._static_build_path):
1694
1695 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1696
1697         webkitpy: make PortFactory.get() be fully data-driven
1698         https://bugs.webkit.org/show_bug.cgi?id=76360
1699
1700         Reviewed by Eric Seidel.
1701
1702         Refactor PortFactory.get() to be fully data-driven.
1703
1704         * Scripts/webkitpy/layout_tests/port/factory.py:
1705         (PortFactory):
1706         (PortFactory._default_port):
1707         (PortFactory.get):
1708         (PortFactory.get.in):
1709         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1710         (FactoryTest.assert_port):
1711
1712 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1713
1714         webkitpy: add determine_full_port_name(), clean up port.__init__()
1715         https://bugs.webkit.org/show_bug.cgi?id=76357
1716
1717         Reviewed by Adam Barth.
1718
1719         Each Port class is now required to implement a 'port_name' field
1720         that will match the start of all ports constructed by this class and
1721         a 'determine_full_port_name' field that will make the port
1722         fully-qualified.
1723
1724         * Scripts/webkitpy/layout_tests/port/apple.py:
1725         (ApplePort.__init__):
1726         * Scripts/webkitpy/layout_tests/port/base.py:
1727         (Port):
1728         (Port.determine_port_name):
1729         (Port.__init__):
1730         * Scripts/webkitpy/layout_tests/port/chromium.py:
1731         (ChromiumPort.for):
1732         (ChromiumPort.__init__):
1733         (ChromiumPort):
1734         (ChromiumPort._chromium_base_dir):
1735         (ChromiumPort.path_from_chromium_base):
1736         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
1737         (ChromiumGpuLinuxPort):
1738         (ChromiumGpuLinuxPort.__init__):
1739         (ChromiumGpuMacPort):
1740         (ChromiumGpuMacPort.__init__):
1741         (ChromiumGpuWinPort):
1742         (ChromiumGpuWinPort.__init__):
1743         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1744         (ChromiumLinuxPort):
1745         (ChromiumLinuxPort.determine_port_name):
1746         (ChromiumLinuxPort.__init__):
1747         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
1748         (ChromiumMacPort):
1749         (ChromiumMacPort.determine_port_name):
1750         (ChromiumMacPort.__init__):
1751         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
1752         (ChromiumWinPort):
1753         (ChromiumWinPort.determine_port_name):
1754         (ChromiumWinPort.__init__):
1755         * Scripts/webkitpy/layout_tests/port/dryrun.py:
1756         (DryRunPort):
1757         (DryRunPort.determine_port_name):
1758         (DryRunPort.__init__):
1759         * Scripts/webkitpy/layout_tests/port/efl.py:
1760         (EflPort):
1761         * Scripts/webkitpy/layout_tests/port/factory.py:
1762         (PortFactory):
1763         (PortFactory._default_port):
1764         (PortFactory.get):
1765         (PortFactory.get.in):
1766         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
1767         (FactoryTest.assert_port):
1768         * Scripts/webkitpy/layout_tests/port/google_chrome.py:
1769         (GoogleChromeLinux32Port):
1770         (GoogleChromeLinux32Port.determine_port_name):
1771         (GoogleChromeLinux64Port):
1772         (GoogleChromeLinux64Port.determine_port_name):
1773         (GoogleChromeMacPort):
1774         (GoogleChromeMacPort.determine_port_name):
1775         (GoogleChromeWinPort):
1776         (GoogleChromeWinPort.determine_port_name):
1777         * Scripts/webkitpy/layout_tests/port/gtk.py:
1778         (GtkPort):
1779         * Scripts/webkitpy/layout_tests/port/mac.py:
1780         (MacPort):
1781         (MacPort.determine_port_name):
1782         (MacPort.__init__):
1783         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
1784         (test_versions):
1785         * Scripts/webkitpy/layout_tests/port/mock_drt.py:
1786         (MockDRTPort):
1787         (MockDRTPort.determine_port_name):
1788         (MockDRTPort.__init__):
1789         (MockDRT):
1790         (MockDRT.determine_port_name):
1791         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
1792         (PortTestCase.make_port):
1793         * Scripts/webkitpy/layout_tests/port/qt.py:
1794         (QtPort):
1795         (QtPort.determine_port_name):
1796         (QtPort.__init__):
1797         * Scripts/webkitpy/layout_tests/port/test.py:
1798         (TestPort):
1799         (TestPort.determine_port_name):
1800         (TestPort.__init__):
1801         * Scripts/webkitpy/layout_tests/port/webkit.py:
1802         (WebKitPort.__init__):
1803         * Scripts/webkitpy/layout_tests/port/win.py:
1804         (WinPort):
1805         (WinPort.determine_port_name):
1806         * Scripts/webkitpy/tool/servers/rebaselineserver.py:
1807         (get_test_baselines.AllPlatformsPort.__init__):
1808         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
1809         (get_test_config.TestMacPort):
1810         (get_test_config):
1811
1812 2012-01-17  Dirk Pranke  <dpranke@chromium.org>
1813
1814         webkitpy: clean up port code in preparation for static port names
1815         https://bugs.webkit.org/show_bug.cgi?id=76356
1816
1817         Reviewed by Adam Barth.
1818
1819         To fix bug 76215 and be able to determine appropriate port names
1820         in webkitpy without actually constructing Port objects, we need
1821         to shuffle some logic in the chromium ports to be able to figure
1822         out whether we built DRT in 32 or 64-bit mode.
1823
1824         * Scripts/webkitpy/layout_tests/port/chromium.py:
1825         (ChromiumPort):
1826         (ChromiumPort._chromium_base_dir):
1827         (ChromiumPort.__init__):
1828         (ChromiumPort.path_from_chromium_base):
1829         * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
1830         (ChromiumLinuxPort):
1831         (ChromiumLinuxPort._determine_driver_path_statically):
1832         (ChromiumLinuxPort._static_build_path):
1833         (ChromiumLinuxPort._determine_architecture):
1834         (ChromiumLinuxPort.__init__):
1835         (ChromiumLinuxPort._build_path):
1836
1837 2012-01-17  Hao Zheng  <zhenghao@chromium.org>
1838
1839         Do not remove pixel failures for ref tests.
1840         https://bugs.webkit.org/show_bug.cgi?id=76243
1841
1842         Reviewed by Tony Chang.
1843
1844         'NRWT --no-pixel-tests' complains when reftests are expected to
1845         be image mismatch.
1846
1847         * Scripts/webkitpy/layout_tests/controllers/manager.py:
1848         (Manager._update_summary_with_result):
1849         * Scripts/webkitpy/layout_tests/models/test_failures.py:
1850         (is_reftest_failure):
1851         (determine_result_type):
1852
1853 2012-01-17  Balazs Ankes  <Ankes.Balazs@stud.u-szeged.hu>
1854
1855         REGRESSION(r81225): ORWT should ignore reftests
1856         https://bugs.webkit.org/show_bug.cgi?id=67936
1857
1858         Reviewed by Csaba Osztrogonác.
1859
1860         * Scripts/old-run-webkit-tests:
1861         (isUsedInReftest): $filename should be the first parameter
1862         (findTestsToRun): run test if it isn't reftest
1863
1864 2012-01-17  Gustavo Noronha Silva  <gns@gnome.org>
1865
1866         Unreviewed. Fix python unit test I broke with r105142
1867
1868         * Scripts/webkitpy/common/config/ports_unittest.py:
1869         (WebKitPortTest.test_gtk_port):
1870
1871 2012-01-17  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1872
1873         [Qt] Don't set the 'primary' flag when mocking touch-points in MiniBrowser
1874
1875         Qt 5 no longer has that flag.
1876
1877         Reviewed by Simon Hausmann.
1878
1879         * MiniBrowser/qt/MiniBrowserApplication.cpp:
1880         (MiniBrowserApplication::notify):
1881
1882 2012-01-16  Gustavo Noronha Silva  <gustavo.noronha@collabora.com>
1883
1884         GTK+ EWS needs to run update-webkitgtk-libs after applying a patch
1885         https://bugs.webkit.org/show_bug.cgi?id=75857
1886
1887         Reviewed by Adam Barth.
1888
1889         * Scripts/webkitdirs.pm:
1890         (buildAutotoolsProject): move cleaning the build to happen early,
1891         and handle --update-gtk right after that
1892         * Scripts/webkitpy/common/config/ports.py:
1893         (GtkPort.build_webkit_command): pass --update-gtk as an option to
1894         build-webkit, so that it is used in the EWS
1895
1896 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1897
1898         [Qt] Don't add _debug postfix to targets when building debug only
1899
1900         Reviewed by Simon Hausmann.
1901
1902         * qmake/mkspecs/features/default_post.prf:
1903         * qmake/mkspecs/features/functions.prf:
1904
1905 2012-01-16  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1906
1907         [Qt] Add more files to project-files for easier navigation in Qt Creator
1908
1909         Reviewed by Simon Hausmann.
1910
1911         * Tools.pro:
1912         * qmake/config.tests/README: Added.
1913
1914 2012-01-17  Philippe Normand  <pnormand@igalia.com>
1915
1916         [GTK] fast/dom/constructed-objects-prototypes.html fails
1917         https://bugs.webkit.org/show_bug.cgi?id=75432
1918
1919         Reviewed by Martin Robinson.
1920
1921         * Scripts/build-webkit: Enable video-track in GTK.
1922
1923 2012-01-16  Takashi Toyoshima  <toyoshim@chromium.org>
1924
1925         Add toyoshim as a committer.
1926         https://bugs.webkit.org/show_bug.cgi?id=76390
1927
1928         Reviewed by Hajime Morita.
1929
1930         * Scripts/webkitpy/common/config/committers.py:
1931
1932 2012-01-16  Brent Fulgham  <bfulgham@webkit.org>
1933
1934         [WinCairo] Build fix after r104155. Several debug targets were not
1935         properly including the '*Common' property sheets.
1936
1937         * DumpRenderTree/win/DumpRenderTreeLauncherDebugCairoCFLite.vsprops:
1938         * DumpRenderTree/win/ImageDiffLauncherDebugCairoCFLite.vsprops:
1939         * MiniBrowser/Configurations/MiniBrowserLauncherDebugCairoCFLite.vsprops:
1940         * WebKitTestRunner/win/WebKitTestRunnerLauncherDebugCairoCFLite.vsprops:
1941         * WinLauncher/WinLauncherLauncherDebugCairoCFLite.vsprops:
1942
1943 2012-01-16  Csaba Osztrogonác  <ossy@webkit.org>
1944
1945         REGRESSION(r105072): It broke committers_unittest.py
1946         https://bugs.webkit.org/show_bug.cgi?id=76396
1947
1948         Reviewed by Adam Barth.
1949
1950         * Scripts/webkitpy/common/config/committers_unittest.py:
1951         (CommittersTest._assert_fuzz_match): It should handle more names.
1952         (CommittersTest.test_contributors_by_fuzzy_match_with_legacy_names): Fix the testcase.
1953
1954 2012-01-16  Mario Sanchez Prada  <msanchez@igalia.com>
1955
1956         [GTK] Include gtk+3 in the jhbuild modules
1957         https://bugs.webkit.org/show_bug.cgi?id=76284
1958
1959         Reviewed by Philippe Normand.
1960
1961         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
1962
1963 2012-01-16  Dan Winship  <danw@gnome.org>
1964
1965         Add myself as a committer
1966
1967         Unreviewed
1968
1969         * Scripts/webkitpy/common/config/committers.py:
1970
1971 2012-01-16  Raphael Kubo da Costa  <kubo@profusion.mobi>
1972
1973         Unreviewed; fix the build when --as-needed and --no-copy-dt-needed-entries are passed to the linker.
1974
1975         Some recent versions of a few Linux distributions have started passing
1976         --as-needed and --no-copy-dt-needed-entries by default to the linker,
1977         which broke the build when SHARED_CORE was turned on, as libwtf_efl.so
1978         needed a few symbols from libjavascriptcore_efl.so.
1979
1980         * DumpRenderTree/efl/CMakeLists.txt: Explicitly pass WTF_LIBRARY_NAME
1981         before JavaScriptCore_LIBRARY_NAME in DumpRenderTree_LIBRARIES so the
1982         libraries are searched in the right order by the linker.
1983
1984 2012-01-16  Philippe Normand  <pnormand@igalia.com>
1985
1986         [GTK] set audio resources path in NRTW
1987         https://bugs.webkit.org/show_bug.cgi?id=76380
1988
1989         Reviewed by Martin Robinson.
1990
1991         * Scripts/webkitpy/layout_tests/port/gtk.py:
1992         (GtkPort.setup_environ_for_server): Set $AUDIO_RESOURCES_PATH so
1993         the uninstalled audio resources are used during webaudio tests.
1994
1995 2012-01-16  Carlos Garcia Campos  <cgarcia@igalia.com>
1996
1997         [GTK] ldflags and cflags should take precedence over the existing flags in gtkdoc.py
1998         https://bugs.webkit.org/show_bug.cgi?id=76369
1999
2000         Reviewed by Xan Lopez.
2001
2002         If there's an installed version of webkit and there are new
2003         symbols in the build tree, gtkdoc-scanobj fails due to undefined
2004         references, because the env ldflags are given before the local
2005         ones.
2006
2007         * gtk/gtkdoc.py:
2008         (GTKDoc._run_gtkdoc_scangobj):
2009
2010 2012-01-15  Pablo Flouret  <pablof@motorola.com>
2011
2012         Fix compilation errors on build-webkit --debug --no-video on mac.
2013         https://bugs.webkit.org/show_bug.cgi?id=75867
2014
2015         Reviewed by Philippe Normand.
2016
2017         Add --[no-]video-track to build-webkit, since video track feature
2018         depends on video.
2019
2020         * Scripts/build-webkit:
2021
2022 2012-01-14  James Robinson  <jamesr@chromium.org>
2023
2024         [chromium] Failing webkit_unit_tests does not make the cr-linux EWS bubble red
2025         https://bugs.webkit.org/show_bug.cgi?id=76313
2026
2027         Reviewed by Adam Barth.
2028
2029         The exit code of this script should be the exit code of webkit_unit_tests. Tested manually.
2030
2031         * Scripts/run-chromium-webkit-unit-tests:
2032
2033 2012-01-14  Sheriff Bot  <webkit.review.bot@gmail.com>
2034
2035         Unreviewed, rolling out r105017.
2036         http://trac.webkit.org/changeset/105017
2037         https://bugs.webkit.org/show_bug.cgi?id=76333
2038
2039         broke the build (Requested by philn-tp on #webkit).
2040
2041         * gtk/jhbuild.modules:
2042
2043 2012-01-14  Mario Sanchez Prada  <msanchez@igalia.com>
2044
2045         [GTK] Include gtk+3 in the jhbuild modules
2046         https://bugs.webkit.org/show_bug.cgi?id=76284
2047
2048         Reviewed by Philippe Normand.
2049
2050         * gtk/jhbuild.modules: Add some gtk+3 to the jhbuild moduleset.
2051
2052 2012-01-13  Adam Barth  <abarth@webkit.org>
2053
2054         Remove --dry-run support from webkit-patch
2055         https://bugs.webkit.org/show_bug.cgi?id=76300
2056
2057         Reviewed by Dirk Pranke.
2058
2059         This is the rest of the patch from Bug 76300.  I landed it separtely
2060         because main.py had some stray conflict markers that I needed to clean
2061         up first.
2062
2063         * Scripts/webkitpy/test/main.py:
2064         (Tester.parse_args):
2065         (Tester._run_tests):
2066
2067 2012-01-13  Adam Barth  <abarth@webkit.org>
2068
2069         test-webkitpy does not run
2070         https://bugs.webkit.org/show_bug.cgi?id=76318
2071
2072         Unreviewed.
2073
2074         Looks like some conflict markers got left in the file.
2075
2076         * Scripts/webkitpy/test/main.py:
2077         (Tester.run):
2078
2079 2012-01-13  Adam Barth  <abarth@webkit.org>
2080
2081         Remove --dry-run support from webkit-patch
2082         https://bugs.webkit.org/show_bug.cgi?id=76300
2083
2084         Reviewed by Dirk Pranke.
2085
2086         We don't use this functionality for anything, and it doesn't really
2087         work.  Someone in #webkit tried using it and and got sad/confused.  We
2088         should just remove it.
2089
2090         * Scripts/webkitpy/common/checkout/scm/git.py:
2091         (Git.push_local_commits_to_server):
2092         * Scripts/webkitpy/common/checkout/scm/scm.py:
2093         (SCM.__init__):
2094         * Scripts/webkitpy/common/checkout/scm/scm_unittest.py:
2095         (_shared_test_commit_with_message):
2096         (GitSVNTest.test_commit_text_parsing):
2097         * Scripts/webkitpy/common/checkout/scm/svn.py:
2098         (SVN.commit_with_message):
2099         * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
2100         (Bugzilla.__init__):
2101         (Bugzilla.quips):
2102         (Bugzilla.authenticate):
2103         (Bugzilla.add_attachment_to_bug):
2104         (Bugzilla.add_patch_to_bug):
2105         (Bugzilla.create_bug):
2106         (Bugzilla.clear_attachment_flags):
2107         (Bugzilla.set_flag_on_attachment):
2108         (Bugzilla.obsolete_attachment):
2109         (Bugzilla.add_cc_to_bug):
2110         (Bugzilla.post_comment_to_bug):
2111         (Bugzilla.close_bug_as_fixed):
2112         (Bugzilla.reassign_bug):
2113         (reopen_bug):
2114         * Scripts/webkitpy/test/main.py:
2115         (Tester.parse_args):
2116         (Tester.run_tests):
2117
2118 2012-01-13  David Levin  <levin@chromium.org>
2119
2120         check-webkit-style: should encourage the use of Own* classes for Windows DC.
2121         https://bugs.webkit.org/show_bug.cgi?id=76227
2122
2123         Reviewed by Dirk Pranke.
2124
2125         * Scripts/webkitpy/style/checkers/cpp.py:
2126         (check_for_leaky_patterns): The new check.
2127         (process_line): Added a call to the new check.
2128         (CppChecker): Added the new error type.
2129         * Scripts/webkitpy/style/checkers/cpp_unittest.py:
2130         (CppStyleTestBase):
2131         (CppStyleTestBase.perform_leaky_pattern_check):
2132         The check for only leaky pattern errors.
2133         (LeakyPatternTest): Test cases.
2134
2135 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2136
2137         test-webkitpy: clean up handling of test directories, QueueStatusServer
2138         https://bugs.webkit.org/show_bug.cgi?id=76238
2139
2140         Reviewed by Adam Barth.
2141
2142         This patch cleans up a bunch of hacks related to how we find all
2143         of the python unit tests and set up the sys.path variable in
2144         test-webkitpy.
2145
2146         This patch changes how test-webkitpy works slightly in that we
2147         will now look for test files under Tools/Scripts and Source/WebKit2/Scripts
2148         rather than just Tools/Scripts/webkitpy and Source/WebKit2/Scripts/webkit.
2149         This greatly simplifies the mapping of paths to module names and
2150         appears to have no significant performance impact or other drawbacks.
2151
2152         There are other ways that we could remove all of the hacks and
2153         still have a more flexible mapping of paths to module names, but
2154         the ones I've thought of so far add a decent amount of complexity to
2155         the implementation and don't seem to buy us much.
2156         
2157         * QueueStatusServer/__init__.py:
2158         * Scripts/test-webkitpy:
2159         * Scripts/webkitpy/test/main.py:
2160         (Tester._modules_from_paths):
2161         (Tester._is_module):
2162
2163 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2164
2165         test-webkitpy: should support classes and individual test names as well as modules
2166         https://bugs.webkit.org/show_bug.cgi?id=76233
2167
2168         Reviewed by Adam Barth.
2169
2170         The default unittest.main() supports a list of individual
2171         TestCase classes or individual tests as well as modules. This
2172         change adds support for the same to test-webkitpy, so now you
2173         can say
2174
2175           % test-wekbitpy webkitpy.test.cat_unittest.CatTest
2176
2177         and
2178
2179           % test-wekbitpy webkitpy.test.cat_unittest.CatTest.test_basic
2180
2181         in addition to
2182
2183           % test-wekbitpy webkitpy.test.cat_unittest
2184
2185         * Scripts/test-webkitpy:
2186         * Scripts/webkitpy/test/cat_unittest.py:
2187         * Scripts/webkitpy/test/main.py:
2188         (Tester.parse_args):
2189         (Tester.run):
2190         (Tester):
2191         (Tester._find_modules):
2192         (Tester._exclude):
2193         (Tester._run_tests):
2194         (Tester._is_module):
2195         (Tester._log_exception):
2196
2197 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2198
2199         test-webkitpy: add basic code coverage support
2200         https://bugs.webkit.org/show_bug.cgi?id=76201
2201
2202         Reviewed by Adam Barth.
2203
2204         This test adds basic code coverage support to test-webkitpy; if
2205         you pass the -c flag, we will generate a new coverage file. We
2206         can start simply with this and add more features as necessary.
2207
2208         This patch requires coverage to be already installed on the
2209         machine. I am not auto-installing this for now.
2210
2211         * Scripts/webkitpy/test/main.py:
2212         (Tester.parse_args):
2213         (Tester.run_tests):
2214
2215 2012-01-13  Dirk Pranke  <dpranke@chromium.org>
2216
2217         StatusBubbleTest.test_build_bubble is failing
2218         https://bugs.webkit.org/show_bug.cgi?id=76222
2219
2220         Reviewed by Adam Barth.
2221
2222         * QueueStatusServer/handlers/statusbubble_unittest.py:
2223         (StatusBubbleTest.test_build_bubble):
2224
2225 2012-01-13  Kristóf Kosztyó  <kkristof@inf.u-szeged.hu>
2226
2227         nrwt should use multiple http shards
2228         https://bugs.webkit.org/show_bug.cgi?id=75958
2229
2230         Add --max-locked-shards option to the nrwt for manual set
2231         the maximum number of the locked shards
2232
2233         Reviewed by Dirk Pranke.
2234
2235         * Scripts/webkitpy/layout_tests/controllers/manager.py:
2236         (Manager._max_locked_shards):
2237         * Scripts/webkitpy/layout_tests/controllers/manager_unittest.py:
2238         (ShardingTests.get_shards):
2239         (ShardingTests.test_multiple_locked_shards):
2240         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2241         (parse_args):
2242
2243 2012-01-13  Raphael Kubo da Costa  <kubo@profusion.mobi>
2244
2245         prepare-ChangeLog: Look for a text editor in a few more variables.
2246         https://bugs.webkit.org/show_bug.cgi?id=76289
2247
2248         Reviewed by Ryosuke Niwa.
2249
2250         A text editor was looked for only in the CHANGE_LOG_EDITOR and
2251         CHANGE_LOG_EDIT_APPLICATION environment variables before falling back
2252         to the OS X-specific open(1) command.
2253
2254         As there does not seem to be a common way of looking for a text editor
2255         among the scripts in Tools/Scripts right now, just look at $VISUAL and
2256         $EDITOR (set much more often than the other variables in Unix systems)
2257         before falling back to open(1).
2258
2259         * Scripts/prepare-ChangeLog:
2260         (openChangeLogs):
2261
2262 2012-01-13  Gabor Rapcsanyi  <rgabor@webkit.org>
2263
2264         [GTK] gtkdoc.py ignoring LDFLAGS and CFLAGS environment variables
2265         https://bugs.webkit.org/show_bug.cgi?id=76269
2266
2267         Reviewed by Martin Robinson.
2268
2269         * gtk/gtkdoc.py:
2270         (GTKDoc._run_gtkdoc_scangobj):
2271
2272 2012-01-13  Csaba Osztrogonác  <ossy@webkit.org>
2273
2274         [Qt][WK2]REGRESSION(r104881):It broke hundreds of tests
2275         https://bugs.webkit.org/show_bug.cgi?id=76247
2276
2277         Reviewed by Kenneth Rohde Christiansen.
2278
2279         * WebKitTestRunner/TestController.cpp:
2280         (WTR::TestController::resetStateToConsistentValues): Disable mock scrollbars on Qt by default. (workaround)
2281
2282 2012-01-13  Hajime Morrita  <morrita@chromium.org>
2283
2284         [Mac][DRT] Should support --no-timeout
2285         https://bugs.webkit.org/show_bug.cgi?id=76242
2286
2287         Reviewed by Darin Adler.
2288
2289         * DumpRenderTree/mac/DumpRenderTree.mm:
2290         (initializeGlobalsFromCommandLineOptions): Added "--no-timeout" to the options list.
2291         (setWaitToDumpWatchdog): Added to encapsulate the waitToDumpWatchdog global variable.
2292         (shouldSetWaitToDumpWatchdog): Added to check the useTimeoutWatchdog flag set by the option.
2293         * DumpRenderTree/mac/DumpRenderTreeMac.h:
2294         * DumpRenderTree/mac/LayoutTestControllerMac.mm:
2295         (LayoutTestController::setWaitToDump): Now calls shouldSetWaitToDumpWatchdog() to check the flag.
2296
2297 2012-01-12  Simon Fraser  <simon.fraser@apple.com>
2298
2299         Mitigate scrollbar differences when running pixel tests
2300         https://bugs.webkit.org/show_bug.cgi?id=67217
2301
2302         Reviewed by Dan Bernstein.
2303         
2304         Enable mock scrollbars by default for tests.
2305
2306         * DumpRenderTree/mac/DumpRenderTree.mm:
2307         (resetDefaultsToConsistentValues):
2308         * WebKitTestRunner/TestController.cpp:
2309         (WTR::TestController::resetStateToConsistentValues):
2310
2311 2012-01-12  Ryosuke Niwa  <rniwa@webkit.org>
2312
2313         Need a script to run tests in PerformanceTests
2314         https://bugs.webkit.org/show_bug.cgi?id=76132
2315
2316         Reviewed by Adam Barth.
2317
2318         Add run-perf-tests to run performance tests using parser/resources/runner.js.
2319         Unfortunately, there isn't an easy way of telling which test uses which format
2320         so hard-code directory that uses Chromium perf-bot style (only inspector perf. tests for now).
2321
2322         All test outputs are re-formatted to match Chromium perf-bot style.
2323
2324         * Scripts/run-inspector-perf-tests.py:
2325         * Scripts/run-perf-tests: Added.
2326         * Scripts/webkitpy/performance_tests/perftestsrunner.py:
2327         (PerfTestsRunner):
2328         (PerfTestsRunner.__init__):
2329         (PerfTestsRunner._collect_tests):
2330         (PerfTestsRunner.run):
2331         (PerfTestsRunner._print_status):
2332         (PerfTestsRunner._run_tests_set):
2333         (PerfTestsRunner._process_chromium_style_test_result):
2334         (PerfTestsRunner._should_ignore_line_in_parser_test_result):
2335         (PerfTestsRunner._process_parser_test_result):
2336         (PerfTestsRunner._run_single_test):
2337         * Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
2338         (MainTest.TestDriver.run_test):
2339         (create_runner):
2340         (run_test):
2341         (test_run_test_set):
2342         (test_run_test_set_for_parser_tests):
2343         (test_collect_tests):
2344
2345 2012-01-12  Dirk Pranke  <dpranke@chromium.org>
2346
2347         webkitpy: clean up port unit tests in preparation for making port_name mandatory
2348         https://bugs.webkit.org/show_bug.cgi?id=76123
2349
2350         Reviewed by Adam Barth.
2351
2352         This change updates the unit tests to be more consistent in how
2353         ports are created, in preparation for making port_name a
2354         mandatory parameter to the constructor and removing any version
2355         interpretation from it.
2356
2357         There should be no functional changes in this patch, but some
2358         tests that are no longer relevant have been removed or reworked.
2359
2360         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2361         (PortTest.test_get_option__default):
2362         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2363         (ChromiumLinuxPortTest):
2364         (ChromiumLinuxPortTest.assert_architecture):
2365         (ChromiumLinuxPortTest.test_determine_architecture_fails):
2366         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2367         (ChromiumMacPortTest):
2368         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2369         (ChromiumPortTest):
2370         (ChromiumPortTest.test_all_test_configurations):
2371         (ChromiumPortTest.TestMacPort.__init__):
2372         (ChromiumPortTest.TestLinuxPort.__init__):
2373         (ChromiumPortTest.TestWinPort.__init__):
2374         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2375         (ChromiumWinTest):
2376         * Scripts/webkitpy/layout_tests/port/efl_unittest.py:
2377         (EflPortTest):
2378         * Scripts/webkitpy/layout_tests/port/gtk_unittest.py:
2379         (GtkPortTest):
2380         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2381         (MacTest):
2382         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2383         (MockDRTPortTest):
2384         (MockDRTPortTest.make_port):
2385         * Scripts/webkitpy/layout_tests/port/qt.py:
2386         (QtPort._port_flag_for_scripts):
2387         (QtPort.__init__):
2388         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2389         (QtPortTest):
2390         (QtPortTest._assert_search_path):
2391         (QtPortTest.test_baseline_search_path):
2392         (QtPortTest.test_operating_system):
2393         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2394         (WebKitPortTest):
2395         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2396         (WinPortTest):
2397         (WinPortTest._assert_search_path):
2398         (WinPortTest.test_baseline_search_path):
2399         * Scripts/webkitpy/tool/servers/rebaselineserver_unittest.py:
2400         (get_test_config.TestMacPort.__init__):
2401
2402 2012-01-12  Dirk Pranke  <dpranke@chromium.org>
2403
2404         test-webkitpy: clean up logging and provide a real --help message
2405         https://bugs.webkit.org/show_bug.cgi?id=76142
2406
2407         Reviewed by Adam Barth.
2408
2409         This test is mostly more refactoring - it calls objects in
2410         unittest directly instead of routing through unittest.main() --
2411         in order to gain more control over the logging and to provide a
2412         correct --help message. This will also give us a better
2413         foundation for adding more feautures down the road.
2414
2415         This patch adds different levels of logging controlled by the
2416         '-s', '-q', and '-v' options (see --help for details), and
2417         removes the --verbose-logging option.
2418
2419         * Scripts/test-webkitpy:
2420         * Scripts/webkitpy/test/main.py:
2421         (Tester):
2422         (Tester.clean_packages):
2423         (Tester.__init__):
2424         (Tester.parse_args):
2425         (Tester.configure):
2426         (Tester._configure_logging):
2427         (Tester.find_modules):
2428         (Tester._exclude):
2429         (Tester._modules_from_paths):
2430         (Tester.run_tests):
2431
2432 2012-01-12  Simon Hausmann  <simon.hausmann@nokia.com>
2433
2434         Make the new WTF module build on Qt
2435         https://bugs.webkit.org/show_bug.cgi?id=76163
2436
2437         Reviewed by Tor Arne Vestbø.
2438
2439         * qmake/mkspecs/modules/wtf.prf: Pull in WTF from the new location.
2440
2441 2012-01-12  Vsevolod Vlasov  <vsevik@chromium.org>
2442
2443         Unreviewed, mac DumpRenderTree fix.
2444
2445         * DumpRenderTree/mac/UIDelegate.mm:
2446         (-[UIDelegate webView:addMessageToConsole:withSource:]):
2447
2448 2012-01-12  Vsevolod Vlasov  <vsevik@chromium.org>
2449
2450         WebKitTestRunner should not print console message line number when it is equal to zero.
2451         https://bugs.webkit.org/show_bug.cgi?id=76167
2452
2453         Reviewed by Csaba Osztrogonác.
2454
2455         * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
2456         (WTR::InjectedBundlePage::willAddMessageToConsole):
2457
2458 2012-01-11  Vsevolod Vlasov  <vsevik@chromium.org>
2459
2460         Make default console messages line numbers consistent.
2461         https://bugs.webkit.org/show_bug.cgi?id=74075
2462
2463         Reviewed by Pavel Feldman.
2464
2465         Unset line numbers are not printed to console now.
2466         Added Source/WebCore/inspector to efl DumpRenderTree include path.
2467
2468         * DumpRenderTree/chromium/WebViewHost.cpp:
2469         (WebViewHost::didAddMessageToConsole):
2470         * DumpRenderTree/efl/CMakeLists.txt:
2471         * DumpRenderTree/efl/DumpRenderTreeView.cpp:
2472         (onConsoleMessage):
2473         * DumpRenderTree/gtk/DumpRenderTree.cpp:
2474         (webViewConsoleMessage):
2475         * DumpRenderTree/mac/UIDelegate.mm:
2476         (-[UIDelegate webView:addMessageToConsole:withSource:]):
2477         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
2478         (WebCore::WebPage::javaScriptConsoleMessage):
2479         * DumpRenderTree/win/UIDelegate.cpp:
2480         (UIDelegate::webViewAddMessageToConsole):
2481         * DumpRenderTree/wx/DumpRenderTreeWx.cpp:
2482         (LayoutWebViewEventHandler::OnConsoleMessageEvent):
2483
2484 2012-01-10  Dirk Pranke  <dpranke@chromium.org>
2485
2486         test-webkitpy: push more logic into webkitpy.test.main, clean up code
2487         https://bugs.webkit.org/show_bug.cgi?id=76021
2488
2489         Reviewed by Eric Seidel.
2490
2491         I plan to add more functionality to test-webkitpy, but it is
2492         difficult to hack on now; this patch is the first of two that
2493         brings test-webkitpy more inline with current coding style by
2494         pushing more logic into the webkitpy.test.main.Tester class (so
2495         that it will be testable itself).
2496         
2497         There should be no functional changes in this patch.
2498
2499         * Scripts/test-webkitpy:
2500         (_path_from_webkit_root):
2501         * Scripts/webkitpy/test/main.py:
2502         (Tester):
2503         (Tester.init):
2504         (Tester.configure_logging):
2505         (Tester.configure_logging.filter):
2506         (Tester.clean_packages):
2507         (Tester.run_tests):
2508
2509 2012-01-11  Wei Jia  <wjia@chromium.org>
2510
2511         implement layout tests for <video> with media stream
2512         https://bugs.webkit.org/show_bug.cgi?id=74882
2513
2514         Reviewed by Kent Tamura.
2515
2516         * DumpRenderTree/DumpRenderTree.gypi:
2517         * DumpRenderTree/chromium/WebUserMediaClientMock.cpp: Added.
2518         (WebKit::WebUserMediaClientMock::create):
2519         (WebKit::WebUserMediaClientMock::IsMockStream):
2520         (WebKit::WebUserMediaClientMock::requestUserMedia):
2521         (WebKit::WebUserMediaClientMock::cancelUserMediaRequest):
2522         * DumpRenderTree/chromium/WebUserMediaClientMock.h: Added.
2523         (WebKit::WebUserMediaClientMock::~WebUserMediaClientMock):
2524         (WebKit::WebUserMediaClientMock::WebUserMediaClientMock):
2525         * DumpRenderTree/chromium/WebViewHost.cpp:
2526         (WebViewHost::userMediaClient):
2527         (WebViewHost::userMediaClientMock):
2528         (WebViewHost::createMediaPlayer):
2529         (WebViewHost::mediaStreamUtil):
2530         (WebViewHost::testMediaStreamClient):
2531         * DumpRenderTree/chromium/WebViewHost.h:
2532
2533 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2534
2535         Unreviewed, build fix.
2536
2537         Fix chromium-win's version remapping to work correctly on Win 7 SP1.
2538
2539         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2540         (ChromiumWinPort.__init__):
2541
2542 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2543
2544         Sheriffbot build fix. Don't emit empty lines on IRC.
2545
2546         * Scripts/webkitpy/tool/bot/irc_command.py:
2547         (LastGreenRevision.execute):
2548
2549 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2550
2551         last-green-revision should report the revision that succeeded on all bots
2552         https://bugs.webkit.org/show_bug.cgi?id=76109
2553
2554         Reviewed by Adam Barth.
2555
2556         Add an algorithm to find the last known good revision among the specified bots. For a revision to be
2557         considered green by this algorithm, all matching builders must have a successful run at the revision
2558         or two consecutive successful runs before and after the revision.
2559
2560         Also fixed a bug in irc_command that the result wasn't posted on IRC properly and a bug in queries.py
2561         that resulted in an exception when the user didn't pass BUILDER_NAME argument.
2562
2563         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2564         (BuildBot._revisions_for_builder):
2565         (BuildBot):
2566         (BuildBot._find_green_revision):
2567         (BuildBot.last_green_revision):
2568         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
2569         (MockBuildBot.last_green_revision):
2570         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2571         (test_revisions_for_builder):
2572         (test_find_green_revision):
2573         (test_last_green_revision):
2574         * Scripts/webkitpy/tool/bot/irc_command.py:
2575         (LastGreenRevision.execute):
2576         * Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
2577         (SheriffIRCBotTest.test_lgr):
2578         * Scripts/webkitpy/tool/commands/queries.py:
2579         (LastGreenRevision):
2580
2581 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2582
2583         Unreviewed, fix build bustage on win32 introduced by r104725.
2584
2585         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2586         (MockDRTPortTest.make_port):
2587         (MockChromiumDRTTest.test_pixeltest__fails):
2588
2589 2012-01-11  Jochen Eisinger  <jochen@chromium.org>
2590
2591         [Chromium] mimic the (old) behavior of the mac port for (re)setting the color profile
2592         https://bugs.webkit.org/show_bug.cgi?id=75618
2593
2594         Eventually, we should get away from setting the display's color profile, as the mac port is doing now.
2595
2596         Reviewed by Tony Chang.
2597
2598         * DumpRenderTree/chromium/LayoutTestHelper.mm:
2599         (installLayoutTestColorProfile):
2600         (restoreUserColorProfile):
2601         (main):
2602
2603 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2604
2605         test-webkitpy: fix -v and eliminate some spurious warnings
2606         https://bugs.webkit.org/show_bug.cgi?id=76099
2607
2608         Reviewed by Adam Barth.
2609
2610         when we added support for --xml output, we broke the -v flag.
2611
2612         Also, fix the code to swallow a harmless warning from
2613         VCSUtils.pm and fix a deprecated warning in jsonchecker.py.
2614
2615         test-webkitpy is now silent again except for the one known
2616         logging issue.
2617
2618         * Scripts/webkitpy/common/checkout/checkout.py:
2619         (Checkout.commit_message_for_this_commit):
2620         * Scripts/webkitpy/common/checkout/checkout_unittest.py:
2621         (test_commit_message_for_this_commit):
2622         * Scripts/webkitpy/style/checkers/jsonchecker.py:
2623         (JSONChecker.check):
2624         (JSONChecker.line_number_from_json_exception):
2625         * Scripts/webkitpy/test/main.py:
2626         (Tester.run_tests):
2627
2628 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2629
2630         webkitpy: webkitpy.layout_tests.port tests should only use MockSystemHost, not MockHost
2631         https://bugs.webkit.org/show_bug.cgi?id=76084
2632
2633         Reviewed by Adam Barth.
2634
2635         This will help make it easier to catch layering violations. This
2636         change just refactors test code; there are no changes to
2637         production code.
2638
2639         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
2640         (PortTest.make_port):
2641         (PortTest.test_find_no_paths_specified):
2642         (PortTest.test_find_one_test):
2643         (PortTest.test_find_glob):
2644         (PortTest.test_find_with_skipped_directories):
2645         (PortTest.test_find_with_skipped_directories_2):
2646         (PortTest.test_parse_reftest_list):
2647         (PortTest.test_httpd_returns_error_code):
2648         (PortTest.test_virtual_methods):
2649         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
2650         (ChromiumGpuTest.assert_port_works):
2651         (ChromiumGpuTest._assert_baseline_path):
2652         (ChromiumGpuTest.test_graphics_type):
2653         (ChromiumGpuTest.test_default_tests_paths.test_paths):
2654         (ChromiumGpuTest.test_test_files):
2655         (ChromiumGpuTest.test_test_files.test_paths):
2656         (ChromiumGpuTest):
2657         * Scripts/webkitpy/layout_tests/port/chromium_linux_unittest.py:
2658         (ChromiumLinuxPortTest.assert_architecture):
2659         (ChromiumLinuxPortTest.test_check_illegal_port_names):
2660         (ChromiumLinuxPortTest.test_determine_architecture_fails):
2661         * Scripts/webkitpy/layout_tests/port/driver_unittest.py:
2662         (DriverTest.make_port):
2663         (DriverTest):
2664         (DriverTest.assertVirtual):
2665         (DriverTest._assert_wrapper):
2666         (DriverTest.test_virtual_driver_methods):
2667         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
2668         (MockDRTPortTest.make_port):
2669         (MockDRTPortTest.test_port_name_in_constructor):
2670         (MockDRTTest.assertTest):
2671         (MockDRTTest.test_main):
2672         (MockChromiumDRTTest.test_pixeltest__fails):
2673         * Scripts/webkitpy/layout_tests/port/qt_unittest.py:
2674         (QtPortTest._assert_search_path):
2675         * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
2676         (TestWebKitPort.__init__):
2677         (WebKitPortUnitTests.test_default_options):
2678         (test_runtime_feature_list):
2679         (test_test_expectations):
2680
2681 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2682
2683         Style skipped tests in the flakiness dashboard
2684         https://bugs.webkit.org/show_bug.cgi?id=76091
2685
2686         Reviewed by Tony Chang.
2687
2688         * TestResultServer/static-dashboards/dashboard_base.js:
2689         Removed "FLAKY" from this list since we don't ever mark individual runs
2690         as flaky. Also deleted an outdated line.
2691         * TestResultServer/static-dashboards/flakiness_dashboard.html:
2692
2693 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2694
2695         Remove gtest normalization code from the test results server
2696         https://bugs.webkit.org/show_bug.cgi?id=76089
2697
2698         Reviewed by Tony Chang.
2699
2700         The normalization now happens on the gtest side.
2701
2702         * TestResultServer/model/jsonresults.py:
2703         (JsonResults.merge):
2704         * TestResultServer/model/jsonresults_unittest.py:
2705         (JsonResultsTest.test_gtest):
2706
2707 2012-01-11  Ojan Vafai  <ojan@chromium.org>
2708
2709         Be more aggressive about deleting unnecessary results from the test-results-server json files
2710         https://bugs.webkit.org/show_bug.cgi?id=76086
2711
2712         Reviewed by Tony Chang.
2713
2714         Remove any tests that only contain skip/pass/nodata values whose runtime
2715         is less than 5 seconds.
2716
2717         * TestResultServer/model/jsonresults.py:
2718         (JsonResults._normalize_results_json):
2719         * TestResultServer/model/jsonresults_unittest.py:
2720         (JsonResultsTest.test_merge_remove_test):
2721         (JsonResultsTest.test_merge_keep_test_with_all_pass_but_slow_time):
2722
2723 2012-01-11  Ryosuke Niwa  <rniwa@webkit.org>
2724
2725         last-green-revision should give us per-bot information
2726         https://bugs.webkit.org/show_bug.cgi?id=76011
2727
2728         Reviewed by Adam Barth.
2729
2730         Rewrote last-green-revision command. Instead of finding a revision for which all bots succeeded,
2731         we report the latest green run on each bot from the last 100 runs.
2732
2733         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
2734         (BuildBot._fetch_builder_page):
2735         (BuildBot):
2736         (BuildBot._green_revision_for_builder):
2737         (BuildBot.last_green_revision):
2738         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
2739         (test_green_revision_for_builder):
2740         (test_last_green_revision):
2741         * Scripts/webkitpy/tool/bot/irc_command.py:
2742         (LastGreenRevision.execute):
2743         * Scripts/webkitpy/tool/commands/queries.py:
2744         (LastGreenRevision.execute):
2745
2746 2012-01-11  Dirk Pranke  <dpranke@chromium.org>
2747
2748         webkitpy: clean up version detection in webkitpy.layout_tests.port
2749         https://bugs.webkit.org/show_bug.cgi?id=76016
2750
2751         Reviewed by Adam Barth.
2752
2753         This change updates the code in webkitpy.layout_tests.port to
2754         use the new os_name and os_version fields on host.platform. This
2755         gets layout_tests.port out of the version parsing business. Most
2756         of the diffs are just cleaning up test code to accomodate the
2757         changes.
2758
2759         * Scripts/webkitpy/common/system/executive_mock.py:
2760         (MockExecutive):
2761         (MockExecutive.ignore_error):
2762         (MockExecutive2):
2763         (MockExecutive2.ignore_error):
2764         * Scripts/webkitpy/common/system/platforminfo_mock.py:
2765         (MockPlatformInfo.is_mac):
2766         (MockPlatformInfo.is_linux):
2767         (MockPlatformInfo.is_win):
2768         * Scripts/webkitpy/common/system/systemhost_mock.py:
2769         (MockSystemHost.__init__):
2770         * Scripts/webkitpy/layout_tests/port/apple.py:
2771         (ApplePort.__init__):
2772         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
2773         (ChromiumMacPort.__init__):
2774         (ChromiumMacPort.check_wdiff):
2775         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
2776         (ChromiumMacPortTest):
2777         (ChromiumMacPortTest.assert_name):
2778         (ChromiumMacPortTest.test_versions):
2779         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
2780         (ChromiumPortTest.test_all_test_configurations):
2781         (ChromiumPortTest.TestMacPort.__init__):
2782         (ChromiumPortTest.TestLinuxPort.__init__):
2783         (ChromiumPortTest.TestWinPort.__init__):
2784         * Scripts/webkitpy/layout_tests/port/chromium_win.py:
2785         (ChromiumWinPort.__init__):
2786         * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
2787         (ChromiumWinTest):
2788         (ChromiumWinTest.assert_name):
2789         (ChromiumWinTest.test_versions):
2790         (ChromiumWinTest.test_baseline_path):
2791         (ChromiumWinTest.test_build_path):
2792         * Scripts/webkitpy/layout_tests/port/factory.py:
2793         (PortFactory._port_name_from_arguments_and_options):
2794         (PortFactory.get):
2795         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
2796         (FactoryTest):
2797         (FactoryTest.setUp):
2798         (FactoryTest.assert_port):
2799         (FactoryTest.test_mac):
2800         (FactoryTest.test_win):
2801         (FactoryTest.test_google_chrome):
2802         (FactoryTest.test_gtk):
2803         (FactoryTest.test_qt):
2804         (FactoryTest.test_chromium_gpu):
2805         (FactoryTest.test_chromium_gpu_linux):
2806         (FactoryTest.test_chromium_gpu_mac):
2807         (FactoryTest.test_chromium_gpu_win):
2808         (FactoryTest.test_chromium_mac):
2809         (FactoryTest.test_chromium_linux):
2810         (FactoryTest.test_chromium_win):
2811         (FactoryTest.test_unknown_specified):
2812         (FactoryTest.test_unknown_default):
2813         * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
2814         (TestGoogleChromePort):
2815         (TestGoogleChromePort._verify_baseline_path):
2816         (TestGoogleChromePort._verify_expectations_overrides):
2817         (TestGoogleChromePort.test_get_google_chrome_port):
2818         * Scripts/webkitpy/layout_tests/port/mac.py:
2819         (MacPort):
2820         * Scripts/webkitpy/layout_tests/port/mac_unittest.py:
2821         (MacTest):
2822         (assert_name):
2823         (test_tests_for_other_platforms):
2824         (test_version):
2825         (test_versions):
2826         (test_is_version_methods):
2827         (test_setup_environ_for_server):
2828         (_assert_search_path):
2829         (test_show_results_html_file):
2830         * Scripts/webkitpy/layout_tests/port/port_testcase.py:
2831         (PortTestCase):
2832         (PortTestCase.make_port):
2833         * Scripts/webkitpy/layout_tests/port/win.py:
2834         (WinPort):
2835         * Scripts/webkitpy/layout_tests/port/win_unittest.py:
2836         (WinPortTest):
2837         (WinPortTest.test_show_results_html_file):
2838         (WinPortTest._assert_search_path):
2839         (WinPortTest._assert_version):
2840
2841 2012-01-11  Hugo Parente Lima  <hugo.lima@openbossa.org>
2842
2843         [Qt][WK2] Remove workaround introduced into bug 75470
2844         https://bugs.webkit.org/show_bug.cgi?id=75889
2845
2846         Reviewed by Tor Arne Vestbø.
2847
2848         * MiniBrowser/qt/qml/BrowserWindow.qml:
2849
2850 2012-01-10  Dmitry Lomov  <dslomov@google.com>
2851
2852         [Chromium] Remove obsolete references to WebWorker class.
2853         https://bugs.webkit.org/show_bug.cgi?id=76020
2854
2855         Reviewed by David Levin.
2856
2857         * DumpRenderTree/chromium/TestWebWorker.h: Removed.
2858         * DumpRenderTree/chromium/WebViewHost.cpp:
2859         * DumpRenderTree/chromium/WebViewHost.h:
2860
2861 2012-01-11  Csaba Osztrogonác  <ossy@webkit.org>
2862
2863         [Qt] Some css3 filter tests are failing after r104698
2864         https://bugs.webkit.org/show_bug.cgi?id=76055
2865
2866         Reviewed by Simon Hausmann.
2867
2868         * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
2869         (LayoutTestController::overridePreference): Add WebKitAcceleratedCompositingEnabled preference.
2870
2871 2012-01-11  Tony Chang  <tony@chromium.org>
2872
2873         Don't remove 0 byte files when running build-webkit --chromium
2874         https://bugs.webkit.org/show_bug.cgi?id=75997
2875
2876         Reviewed by Csaba Osztrogonác.
2877
2878         If QTDIR is set, isQt() is true even when running build-webkit --chromium.
2879
2880         * Scripts/webkitdirs.pm:
2881         (determineIsQt):
2882
2883 2012-01-10  David Kilzer  <ddkilzer@apple.com>
2884
2885         Add TestWebKitAPI tests for KURL
2886         <http://webkit.org/b/75774>
2887
2888         Reviewed by Adam Barth.
2889
2890         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add a
2891         WebCore folder and KURL.cpp source file to the project.
2892         * TestWebKitAPI/Tests/WebCore/KURL.cpp: Added.
2893         (TestWebKitAPI::WebCore::SetUp): Initialize the main thread
2894         during setup since this test doesn't go through WebKit or
2895         WebKit2.
2896         (TestWebKitAPI::TEST_F): Add two basic tests.
2897
2898 2012-01-10  David Kilzer  <ddkilzer@apple.com>
2899
2900         Extract common code for comparing WTF::String objects into WTFStringUtilities.h
2901
2902         Reviewed by Adam Barth.
2903
2904         Needed for: <http://webkit.org/b/75774> Add TestWebKitAPI tests for KURL
2905
2906         * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Add
2907         WTFStringUtilities.h to the project.
2908         * TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Update to use
2909         WTFStringUtilities.h.
2910         * TestWebKitAPI/WTFStringUtilities.h: Added.
2911         (WTF::operator<<):
2912
2913 2012-01-10  Vincent Scheib  <scheib@chromium.org>
2914
2915         Clarify run-webkit-tests options reset-results and new-baseline
2916         https://bugs.webkit.org/show_bug.cgi?id=75993
2917
2918         Reviewed by Dirk Pranke.
2919
2920         * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
2921         (parse_args):
2922
2923 2012-01-10  Simon Fraser  <simon.fraser@apple.com>
2924
2925         Disabled mock scrollbars should draw differently
2926         https://bugs.webkit.org/show_bug.cgi?id=75995
2927
2928         Reviewed by James Robinson.
2929
2930         When the scrollbar is disabled, paint the entire track of
2931         mock scrollbars with a lighter gray, and hide the thumb.
2932
2933         * DumpRenderTree/mac/DumpRenderTree.mm:
2934         (-[DRTMockScroller drawKnob]):
2935         (-[DRTMockScroller drawRect:]):
2936
2937 2012-01-10  James Robinson  <jamesr@chromium.org>
2938
2939         [chromium] Run webkit_unit_tests as part of webkit-patch build-and-test
2940         https://bugs.webkit.org/show_bug.cgi?id=75988
2941
2942         Reviewed by Adam Barth.
2943
2944         * Scripts/webkitpy/common/config/ports.py:
2945         (WebKitPort.run_webkit_unit_tests_command):
2946         (WebKitPort):
2947         (ChromiumPort.run_webkit_unit_tests_command):
2948         (ChromiumPort):
2949         * Scripts/webkitpy/common/config/ports_mock.py:
2950         (MockPort.run_webkit_unit_tests_command):
2951         * Scripts/webkitpy/tool/commands/download_unittest.py:
2952         (DownloadCommandsTest.test_build_and_test):
2953         (test_land):
2954         (test_land_red_builders):
2955         * Scripts/webkitpy/tool/steps/runtests.py:
2956         (RunTests.run):
2957         * Scripts/webkitpy/tool/steps/runtests_unittest.py:
2958
2959 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
2960
2961         webkitpy: add os_name, os_version to platforminfo
2962         https://bugs.webkit.org/show_bug.cgi?id=75931
2963
2964         Reviewed by Eric Siedel.
2965
2966         As a first step in cleaning up the version parsing logic in
2967         webkitpy.layout_tests.port, this adds common routines for
2968         getting the os_name and os_version fields to the PlatformInfo
2969         class. Nothing uses them yet but I've added FIXME's to some of
2970         the code that needs to be deleted in the port files.
2971
2972         * Scripts/webkitpy/common/system/platforminfo.py:
2973         (PlatformInfo.__init__):
2974         (PlatformInfo):
2975         (PlatformInfo._determine_os_name):
2976         (PlatformInfo._determine_mac_version):
2977         (PlatformInfo._determine_linux_version):
2978         (PlatformInfo._determine_win_version):
2979         (PlatformInfo.display_name):
2980         (PlatformInfo.total_bytes_memory):
2981         (PlatformInfo._compute_bytes_from_vm_stat_output):
2982         (PlatformInfo.free_bytes_memory):
2983         * Scripts/webkitpy/common/system/platforminfo_unittest.py: Added.
2984         (TestPlatformInfo):
2985         (TestPlatformInfo.make_info):
2986         (TestPlatformInfo.make_info.mock_run_command):
2987         (TestPlatformInfo.setUp):
2988         (TestPlatformInfo.tearDown):
2989         (TestPlatformInfo.test_basic):
2990         (TestPlatformInfo.integration_test_basic):
2991         (TestPlatformInfo.test_display_name_mac):
2992         (TestPlatformInfo.test_display_name_win32):
2993         (TestPlatformInfo.test_memory_mac):
2994         (TestPlatformInfo.test_memory_win32):
2995         (TestPlatformInfo.test_determine_os_name):
2996         (TestPlatformInfo.test_determine_mac_version):
2997         (TestPlatformInfo.test_determine_linux_version):
2998         (TestPlatformInfo.test_determine_win_version_from_tuple):
2999         (TestPlatformInfo.test_determine_win_version_from_cmd):
3000         * Scripts/webkitpy/layout_tests/port/apple.py:
3001         (ApplePort.__init__):
3002         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
3003         (ChromiumMacPort.__init__):
3004         * Scripts/webkitpy/layout_tests/port/mac.py:
3005         (MacPort):
3006         * Scripts/webkitpy/layout_tests/port/win.py:
3007         (WinPort):
3008
3009 2012-01-10  Adam Roben  <aroben@apple.com>
3010
3011         Make it possible to type data: URLs into MiniBrowser on Windows
3012
3013         Fixes <http://webkit.org/b/75084> Crash when trying to navigate to a data: URL in
3014         MiniBrowser on Windows
3015
3016         Reviewed by Darin Adler.
3017
3018         * MiniBrowser/win/BrowserWindow.cpp:
3019         (BrowserWindow::handleMessage): Instead of checking for an "http://" prefix when deciding
3020         whether to prepend "http://", just look for whether the typed URL contains a colon. If it
3021         does, we assume it already has a scheme and don't modify it.
3022
3023 2012-01-10  Adam Roben  <aroben@apple.com>
3024
3025         Make MiniBrowser automatically escape invalid URL characters typed in its URL field
3026
3027         Fixes <http://webkit.org/b/75086> Crash when typing a data: URL containing double-quotes in
3028         MiniBrowser on Windows
3029
3030         Reviewed by Darin Adler.
3031
3032         * MiniBrowser/win/BrowserView.cpp:
3033         (BrowserView::goToURL): Pass the typed string through
3034         CFURLCreateStringByAddingPercentEscapes before trying to create a URL from it.
3035
3036 2012-01-10  Ben Wells  <benwells@chromium.org>
3037
3038         Adding myself (benwells) to committers.py
3039         https://bugs.webkit.org/show_bug.cgi?id=75971
3040
3041         Reviewed by Ojan Vafai.
3042
3043         * Scripts/webkitpy/common/config/committers.py:
3044
3045 2012-01-10  Mario Sanchez Prada  <msanchez@igalia.com>
3046
3047         AX: support helpText() in DumpRenderTree
3048         https://bugs.webkit.org/show_bug.cgi?id=40193
3049
3050         Reviewed by Chris Fleizach.
3051
3052         * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
3053         (AccessibilityUIElement::helpText): Implemented.
3054
3055 2012-01-10  Sheriff Bot  <webkit.review.bot@gmail.com>
3056
3057         Unreviewed, rolling out r104572.
3058         http://trac.webkit.org/changeset/104572
3059         https://bugs.webkit.org/show_bug.cgi?id=75967
3060
3061         It broke zillion tests (Requested by Ossy on #webkit).
3062
3063         * DumpRenderTree/qt/EventSenderQt.cpp:
3064         (EventSender::EventSender):
3065         (EventSender::sendOrQueueEvent):
3066         (EventSender::replaySavedEvents):
3067         (EventSender::eventFilter):
3068         * DumpRenderTree/qt/EventSenderQt.h:
3069
3070 2012-01-10  Hugo Parente Lima  <hugo.lima@openbossa.org>
3071
3072         [Qt] fast/events/dont-loose-last-event test fails on WK1
3073         https://bugs.webkit.org/show_bug.cgi?id=73894
3074
3075         Reviewed by Simon Hausmann.
3076
3077         Empty the entire event queue when requested besides using
3078         sendEvent instead of postEvent to make sure all events were
3079         delivered.
3080         This commit also removes not working DRT code to handle drag and drop,
3081         for more refs see: https://bugs.webkit.org/show_bug.cgi?id=31332
3082
3083         * DumpRenderTree/qt/EventSenderQt.cpp:
3084         (EventSender::EventSender):
3085         (EventSender::sendOrQueueEvent):
3086         (EventSender::replaySavedEvents):
3087         * DumpRenderTree/qt/EventSenderQt.h:
3088
3089 2012-01-10  No'am Rosenthal  <noam.rosenthal@nokia.com>
3090
3091         [Qt] Enable CSS_FILTERS in Qt build
3092         https://bugs.webkit.org/show_bug.cgi?id=75777
3093
3094         Add the CSS_FILTERS feature flag to Qt.
3095
3096         Reviewed by Kenneth Rohde Christiansen.
3097
3098         * qmake/mkspecs/features/features.prf:
3099
3100 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
3101
3102         [Qt] Unreviewed -Werror build fix.
3103
3104         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3105         (MiniBrowserApplication::MiniBrowserApplication): Re-order initialization to match
3106         declaration.
3107
3108 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
3109
3110         [Qt] Remove dependency to uitools
3111         https://bugs.webkit.org/show_bug.cgi?id=75952
3112
3113         Reviewed by Tor Arne Vestbø.
3114
3115         QUiLoader is used for creating QLabel/QProgressBar widgets for
3116         a layout test, to verify some widget embedding feature. Unfortunately
3117         in Qt 5, QUiLoader is part of qttools, which depends on webkit, which
3118         depends on qttools. Since we don't really _need_ QUiLoader for our tests,
3119         let's break the circular dependency.
3120
3121         * DumpRenderTree/qt/DumpRenderTree.pro:
3122         * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
3123         (WebCore::WebPage::createPlugin):
3124         * QtTestBrowser/QtTestBrowser.pro:
3125         * QtTestBrowser/launcherwindow.h:
3126         * QtTestBrowser/webpage.cpp:
3127         (WebPage::createPlugin):
3128         * qmake/mkspecs/features/minimal_qt.prf:
3129         * qmake/mkspecs/features/uitools.prf: Removed.
3130
3131 2012-01-10  Alexander Færøy  <alexander.faeroy@nokia.com>
3132
3133         [Qt] Don't expose m_windowOptions as part of the public interface in MiniBrowserApplication
3134         https://bugs.webkit.org/show_bug.cgi?id=75894
3135
3136         Reviewed by Tor Arne Vestbø.
3137
3138         * MiniBrowser/qt/MiniBrowserApplication.h:
3139         (MiniBrowserApplication::windowOptions):
3140         * MiniBrowser/qt/main.cpp:
3141         (main):
3142
3143 2012-01-10  Simon Hausmann  <simon.hausmann@nokia.com>
3144
3145         [Qt] DRT crashes with Qt 5
3146         https://bugs.webkit.org/show_bug.cgi?id=75951
3147
3148         Reviewed by Ossy.
3149
3150         * DumpRenderTree/qt/fonts.conf: Add missing last-resort font fallback for Qt 5 where
3151         Qt relies on font-config to define the default font family.
3152
3153 2012-01-10  Csaba Osztrogonác  <ossy@webkit.org>
3154
3155         [Qt] REGRESSION(r100130): Clean build feature is broken.
3156         https://bugs.webkit.org/show_bug.cgi?id=75863
3157
3158         Reviewed by Tor Arne Vestbø.
3159
3160         * Scripts/webkitdirs.pm:
3161         (buildQMakeProject):
3162
3163 2012-01-10  Zeno Albisser  <zeno@webkit.org>
3164
3165         [Qt][WK2] Fix keyboard shortcuts in MiniBrowser.
3166         https://bugs.webkit.org/show_bug.cgi?id=75885
3167
3168         In QML events are propagated through parents. But since the
3169         WebView may consume key events, a shortcut might never reach
3170         the top QtQuickItem.
3171         Therefore an application wide event handling function in C++
3172         needs to take care of shortcuts for MiniBrowser.
3173
3174         Reviewed by Kenneth Rohde Christiansen.
3175
3176         * MiniBrowser/qt/BrowserWindow.cpp:
3177         (BrowserWindow::reload):
3178         (BrowserWindow::focusAddressBar):
3179         * MiniBrowser/qt/BrowserWindow.h:
3180         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3181         (MiniBrowserApplication::notify):
3182         * MiniBrowser/qt/qml/BrowserWindow.qml:
3183
3184 2012-01-10  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
3185
3186         Add --efl alias for --platform=efl to run-webkit-test
3187         https://bugs.webkit.org/show_bug.cgi?id=75937
3188
3189         Reviewed by Andreas Kling.
3190         
3191         * Scripts/run-webkit-tests: Add --efl alias.
3192         * Scripts/webkitpy/layout_tests/run_webkit_tests.py: Add --efl alias.
3193         (parse_args):
3194
3195 2012-01-09  Dan Bernstein  <mitz@apple.com>
3196
3197         -[DOMRange renderedImageForcingBlackText:] fails with non-user-selectable text
3198         https://bugs.webkit.org/show_bug.cgi?id=75920
3199
3200         Reviewed by Darin Adler.
3201
3202         * TestWebKitAPI/Tests/mac/RenderedImageFromDOMRange.mm:
3203         (TestWebKitAPI::TEST): Added a test for a range in a
3204         -webkit-user-select: none; block.
3205
3206 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
3207
3208         webkitpy.layout_tests.port.mock_drt_unittest.MockChromiumDRTTest has been failing on cr-win
3209         https://bugs.webkit.org/show_bug.cgi?id=75884
3210
3211         Reviewed by Ryosuke Niwa.
3212
3213         Fix regression introduced (revealed?) in r104340. Chromium file         
3214         url syntax is different on win32 and the test results weren't           
3215         expecting that.                                                         
3216
3217         * Scripts/webkitpy/layout_tests/models/test_expectations.py:
3218         (TestExpectations._report_errors):
3219         * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
3220         (test_parse_error_nonfatal):
3221         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
3222         (MockDRTTest.input_line):
3223         (MockChromiumDRTTest.test_pixeltest__fails):
3224         * Scripts/webkitpy/layout_tests/port/test.py:
3225         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
3226         (TestRebaseliner.make_rebaseliner):
3227         (TestRebaseliner.test_one_platform):
3228         (TestRebaseliner.test_all_platforms):
3229         (TestRebaseliner.test_png_file_with_comment):
3230         (TestRebaseliner.test_png_file_with_comment_remove_old_checksum):
3231         (TestRebaseliner.test_png_file_with_comment_as_duplicate):
3232         (TestRealMain.test_all_platforms):
3233
3234 2012-01-09  Dirk Pranke  <dpranke@chromium.org>
3235
3236         Fix regression in test-webkitpy introduced by removing the webkitpy/python24 directory.
3237
3238         Unreviewed, build fix.
3239
3240         test-webkitpy had a built-in self-test routine to verify that
3241         deleting the .pyc files worked correctly. When I removed the
3242         python24 directory in r104482, I broke the self-test.
3243
3244         I have deleted the self-test for now. Since I am reworking
3245         test-webkitpy I will make sure the new version is adequately
3246         tested.
3247
3248         * Scripts/test-webkitpy:
3249         (configure_logging):
3250         (_clean_pyc_files):
3251         (_clean_packages):
3252         (init):
3253
3254 2012-01-09  Justin Novosad  <junov@chromium.org>
3255
3256         [Chromium] remove all references to the legacy accelerated 2d Canvas
3257         implementation
3258         https://bugs.webkit.org/show_bug.cgi?id=75108
3259
3260         Purging an old settings flag that is no longer referenced
3261
3262         Reviewed by Darin Fisher.
3263
3264         * DumpRenderTree/chromium/DumpRenderTree.cpp:
3265         (main):
3266         * DumpRenderTree/chromium/TestShell.cpp:
3267         (TestShell::TestShell):
3268         (TestShell::resetWebSettings):
3269         * DumpRenderTree/chromium/TestShell.h:
3270         (TestShell::setAccelerated2dCanvasEnabled):
3271         * DumpRenderTree/chromium/WebPreferences.cpp:
3272         (WebPreferences::reset):
3273         (WebPreferences::applyTo):
3274         * DumpRenderTree/chromium/WebPreferences.h:
3275
3276 2012-01-09  Adam Roben  <aroben@apple.com>
3277
3278         Convert Cygwin paths to Windows paths before passing them to DRT
3279
3280         Fixes <http://webkit.org/b/64468> <rdar://problem/10663409> WIN: DumpRenderTree hangs under
3281         NRWT
3282
3283         I couldn't find a way to test this because I couldn't figure out how to override the
3284         cygpath function as it is used by WebKitDriver.
3285
3286         Reviewed by Dirk Pranke.
3287
3288         * Scripts/webkitpy/layout_tests/port/webkit.py:
3289         (WebKitDriver._command_from_driver_input): On Cygwin, convert the test's path to a
3290         Windows-style path before passing it to DRT.
3291
3292 2012-01-06  Dirk Pranke  <dpranke@chromium.org>
3293
3294         webkitpy: revamp version checking
3295         https://bugs.webkit.org/show_bug.cgi?id=75765
3296
3297         Reviewed by Adam Barth.
3298
3299         The version-checking code in webkitpy/python24/versioning seems really
3300         awkward and overdesigned, especially since we don't support python 2.4
3301         any more.
3302
3303         In addition, I am tired of getting warnings about Python 2.6 being "a
3304         newer release than what is supported" :).
3305
3306         I'm taking a page from the depot_tools python code and just creating a
3307         new webkitpy.common.version_check module that will print a message on
3308         stderr and call sys.exit() if we're running an unsupported version.
3309
3310         * Scripts/check-webkit-style:
3311         * Scripts/new-run-webkit-httpd:
3312         * Scripts/new-run-webkit-tests:
3313         * Scripts/rebaseline-chromium-webkit-tests:
3314         * Scripts/test-webkitpy:
3315         (configure_logging):
3316         (init):
3317         * Scripts/webkit-patch:
3318         (main):
3319         * Scripts/webkitpy/common/system/executive.py:
3320         (Executive._child_process_encoding):
3321         (Executive._should_encode_child_process_arguments):
3322         * Scripts/webkitpy/common/version_check.py: Copied from Tools/Scripts/rebaseline-chromium-webkit-tests.
3323         * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
3324         * Scripts/webkitpy/python24/__init__.py: Removed.
3325         * Scripts/webkitpy/python24/versioning.py: Removed.
3326         * Scripts/webkitpy/python24/versioning_unittest.py: Removed.
3327
3328 2012-01-09  Adam Barth  <abarth@webkit.org>
3329
3330         Remove Chromium Mac CG from the flakiness dashboard
3331         https://bugs.webkit.org/show_bug.cgi?id=75873
3332
3333         Reviewed by Ojan Vafai.
3334
3335         This configuration no longer exists.
3336
3337         * TestResultServer/static-dashboards/builders.js:
3338         * TestResultServer/static-dashboards/flakiness_dashboard.html:
3339         * TestResultServer/static-dashboards/flakiness_dashboard_tests.js:
3340         (testOverrideJustBuildType):
3341         (testPlatformAndBuildType):
3342
3343 2012-01-06  Adam Roben  <aroben@apple.com>
3344
3345         Print an error message when NRWT can't run Apache
3346
3347         Fixes <http://webkit.org/b/75712> NRWT mysteriously exits when Apache returns an error code
3348
3349         Reviewed by Dirk Pranke.
3350
3351         * Scripts/webkitpy/layout_tests/port/base.py:
3352         (Port.check_httpd): Print an error message when Apache returns an error code instead of
3353         failing silently.
3354
3355         * Scripts/webkitpy/layout_tests/port/base_unittest.py:
3356         (PortTest.test_check_httpd_success): Test that we don't print anything when we successfully
3357         run Apache.
3358         (PortTest.test_httpd_returns_error_code): Test that we print an error message when Apache
3359         fails.
3360
3361 2012-01-09  Gabor Rapcsanyi  <rgabor@webkit.org>
3362
3363         [Gtk][ARM] Cross compiler buildbot is failing after r103040
3364         https://bugs.webkit.org/show_bug.cgi?id=75846
3365
3366         Reviewed by Philippe Normand.
3367
3368         * Scripts/webkitdirs.pm:
3369         (runAutogenForAutotoolsProjectIfNecessary):
3370
3371 2012-01-09  Alexander Færøy  <ahf@0x90.dk>
3372
3373         [Qt] Add support for dynamically enabling and disabling touch mocking in the Qt MiniBrowser
3374         https://bugs.webkit.org/show_bug.cgi?id=75807
3375
3376         This patch adds a new icon to navigationbar of the Qt MiniBrowser
3377         which allows you to dynamically toggle whether touch mocking is
3378         enabled or disabled.
3379
3380         Reviewed by Tor Arne Vestbø.
3381
3382         * MiniBrowser/qt/MiniBrowser.qrc:
3383         * MiniBrowser/qt/MiniBrowserApplication.cpp:
3384         (MiniBrowserApplication::notify):
3385         * MiniBrowser/qt/MiniBrowserApplication.h:
3386         (WindowOptions::WindowOptions):
3387         (WindowOptions::touchMockingEnabled):
3388         (WindowOptions::setTouchMockingEnabled):
3389         * MiniBrowser/qt/icons/touch.png: Added.
3390         * MiniBrowser/qt/qml/BrowserWindow.qml:
3391
3392 2012-01-09  Carlos Garcia Campos  <cgarcia@igalia.com>
3393
3394         Unreviewed. Fix make distcheck issues.
3395
3396         * MiniBrowser/gtk/GNUmakefile.am:
3397
3398 2012-01-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3399
3400         Unreviewed, rolling out r104421.
3401         http://trac.webkit.org/changeset/104421
3402         https://bugs.webkit.org/show_bug.cgi?id=75816
3403
3404         Need to rebaseline some tests on Linux (Requested by noamr on
3405         #webkit).
3406
3407         * qmake/mkspecs/features/features.prf:
3408
3409 2012-01-08  Adam Barth  <abarth@webkit.org>
3410
3411         Remove support for Chromium Mac CG from garden-o-matic
3412         https://bugs.webkit.org/show_bug.cgi?id=75814
3413
3414         Reviewed by Eric Seidel.
3415
3416         This configuration no longer exists.  This patch is mostly a matter of
3417         updating config.js and removing the examples from the unit tests.
3418
3419         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders_unittests.js:
3420         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
3421         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures.js:
3422         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:
3423         * BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js:
3424
3425 2012-01-08  No'am Rosenthal  <noam.rosenthal@nokia.com>
3426
3427         [Qt] Enable CSS_FILTERS in Qt build
3428         https://bugs.webkit.org/show_bug.cgi?id=75777
3429
3430         Add the CSS_FILTERS feature flag to Qt.
3431
3432         Reviewed by Kenneth Rohde Christiansen.
3433
3434         * qmake/mkspecs/features/features.prf:
3435
3436 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3437
3438         Categorize bots by ports instead of core/non-core separation
3439         https://bugs.webkit.org/show_bug.cgi?id=75766
3440
3441         Reviewed by Eric Seidel.
3442
3443         Rename categorizes so as to preserve the original ordering of bots per Eric's suggestion
3444         (Chromium still comes before GTK and Qt).
3445
3446         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3447         (loadBuilderConfig):
3448         * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
3449
3450 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3451
3452         Remove the concept of core builders from webkitpy
3453         https://bugs.webkit.org/show_bug.cgi?id=75809
3454
3455         Reviewed by Ryosuke Niwa.
3456
3457         Treat all bots equally since core/non-core sepration has been removed from build.webkit.org,
3458         and remove the relevant code.
3459
3460         * Scripts/webkitpy/common/net/buildbot/buildbot.py:
3461         (BuildBot.__init__):
3462         (BuildBot._matches_regexps):
3463         (BuildBot._builder_statuses_with_names_matching_regexps):
3464         (BuildBot.builder_statuses):
3465         (BuildBot.failure_map):
3466         (BuildBot._latest_builds_from_builders):
3467         (BuildBot.last_green_revision):
3468         * Scripts/webkitpy/common/net/buildbot/buildbot_mock.py:
3469         (MockBuildBot.builder_statuses):
3470         * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
3471         (test_status_parsing):
3472         (test_last_green_revision):
3473         (test_last_green_revision.mock_builds_from_builders):
3474
3475 2012-01-08  Adam Barth  <abarth@webkit.org>
3476
3477         Remove Chromium Mac CG concept from webkitpy
3478         https://bugs.webkit.org/show_bug.cgi?id=75810
3479
3480         Reviewed by Ryosuke Niwa.
3481
3482         This configuration no longer exists.  We can remove all the supporting
3483         code from webkitpy.
3484
3485         * Scripts/webkitpy/common/checkout/baselineoptimizer_unittest.py:
3486         (BaselineOptimizerTest.test_move_baselines):
3487         (BaselineOptimizerTest.test_chromium_covers_mac_win_linux):
3488         (BaselineOptimizerTest.test_chromium_mac_redundant_with_apple_mac):
3489         (BaselineOptimizerTest.test_common_directory_includes_root):
3490         (BaselineOptimizerTest.test_complex_shadowing):
3491         * Scripts/webkitpy/common/config/build.py:
3492         (_should_file_trigger_build):
3493         * Scripts/webkitpy/common/config/build_unittest.py:
3494         (ShouldBuildTest):
3495         * Scripts/webkitpy/layout_tests/port/builders.py:
3496         * Scripts/webkitpy/layout_tests/port/chromium.py:
3497         (ChromiumPort):
3498         (ChromiumPort._generate_all_test_configurations):
3499         * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
3500         (_default_tests_paths):
3501         (ChromiumGpuLinuxPort.tests):
3502         * Scripts/webkitpy/layout_tests/port/chromium_gpu_unittest.py:
3503         (ChromiumGpuTest.integration_test_chromium_gpu_mac):
3504         (ChromiumGpuTest.assert_port_works):
3505         (ChromiumGpuTest.test_baseline_paths):
3506         (ChromiumGpuTest.test_graphics_type):
3507         (ChromiumGpuTest.test_default_tests_paths):
3508         (ChromiumGpuTest.test_test_files):
3509         * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
3510         (ChromiumMacPort):
3511         (ChromiumMacPort.__init__):
3512         (ChromiumMacPort.baseline_search_path):
3513         * Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
3514         (ChromiumMacPortTest.test_graphics_type):
3515         * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
3516         (ChromiumPortTest.test_all_test_configurations):
3517         * Scripts/webkitpy/layout_tests/port/dryrun.py:
3518         * Scripts/webkitpy/layout_tests/port/factory.py:
3519         (PortFactory._port_name_from_arguments_and_options):
3520         (PortFactory.get):
3521         * Scripts/webkitpy/layout_tests/port/factory_unittest.py:
3522         (FactoryTest.test_chromium_gpu_mac):
3523         (FactoryTest.test_chromium_mac):
3524         * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
3525         (GetGoogleChromePortTest._verify_expectations_overrides):
3526         * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
3527         (MockChromiumDRTTest.test_chromium_parse_options):
3528         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests.py:
3529         * Scripts/webkitpy/to_be_moved/rebaseline_chromium_webkit_tests_unittest.py:
3530         (test_url_fetcher):
3531         (test_zip_factory):
3532         (test_archive):
3533         * Scripts/webkitpy/tool/commands/rebaseline_unittest.py:
3534         (TestRebaseline.test_rebaseline_expectations):
3535
3536 2012-01-08  Ryosuke Niwa  <rniwa@webkit.org>
3537
3538         Categorize bots by ports instead of core/non-core separation
3539         https://bugs.webkit.org/show_bug.cgi?id=75766
3540
3541         Reviewed by Adam Barth.
3542
3543         Get rid of the concept of core/non-core builders, and categorize build bots by ports instead.
3544
3545         * BuildSlaveSupport/build.webkit.org-config/master.cfg:
3546         (loadBuilderConfig):
3547         * BuildSlaveSupport/build.webkit.org-config/templates/root.html:
3548
3549 2012-01-08  Sheriff Bot  <webkit.review.bot@gmail.com>
3550
3551         Unreviewed, rolling out r104403.
3552         http://trac.webkit.org/changeset/104403
3553         https://bugs.webkit.org/show_bug.cgi?id=75803
3554
3555         It broke all tests on Qt5 (Requested by Ossy_weekend on
3556         #webkit).
3557
3558         * qmake/mkspecs/features/features.prf:
3559
3560 2012-01-08  No'am Rosenthal  <noam.rosenthal@nokia.com>
3561
3562         [Qt] Enable CSS_FILTERS in Qt build
3563         https://bugs.webkit.org/show_bug.cgi?id=75777
3564
3565         Add the CSS_FILTERS feature flag to Qt.
3566
3567         Reviewed by Kenneth Rohde Christiansen.
3568
3569         * qmake/mkspecs/features/features.prf:
3570
3571 2012-01-04  Kentaro Hara  <haraken@chromium.org>
3572
3573         Rewrite the CSS parser of prepare-ChangeLog with unittests.
3574         https://bugs.webkit.org/show_bug.cgi?id=75202
3575
3576         Reviewed by David Kilzer.
3577
3578         The current CSS parser can just parse simple CSSes like
3579
3580             foo bar baz {
3581                 property1: value;
3582                 property2: value;
3583             }
3584
3585         , and cannot parse comments nor a CSS in which a selector and {
3586         appears in different lines. This patch rewrites the CSS parser
3587         (i.e. get_selector_line_ranges_for_css()) so that it can parse more CSSes
3588         shown in css_unittests.css.
3589
3590         Test: Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests.css
3591
3592         * Scripts/prepare-ChangeLog:
3593         (get_selector_line_ranges_for_css):
3594         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
3595         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests-expected.txt: Added.
3596         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/css_unittests.css: Added.
3597         (element1):
3598         (element2):
3599         (element3):
3600         (element4.p):
3601         (element5.p.q.r.s):
3602         (element6#p):
3603         (element7 element8):
3604         (element9.p element10.q):
3605         (element11#p element12#q):
3606         (element13, element14):
3607         (.p):
3608         (#p):
3609         (.p element15 #q element16.r element17#s):
3610         (element18:target):
3611         (element19):
3612         (element20):
3613         (element21):
3614         (element22):
3615
3616 2012-01-04  Kentaro Hara  <haraken@chromium.org>
3617
3618         The Perl parser of prepare-ChangeLog can parse here-documents
3619         https://bugs.webkit.org/show_bug.cgi?id=73208
3620
3621         Reviewed by David Kilzer.
3622
3623         Currently prepare-ChangeLog cannot recognize here-documents in Perl,
3624         which results in wrong subroutine names in ChangeLogs.
3625         With this patch, prepare-ChangeLog can judge whether the line of "}"
3626         is the end of a subroutine or a line inside a here-document.
3627
3628         Test: Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl
3629
3630         * Scripts/prepare-ChangeLog:
3631         (get_function_line_ranges_for_perl):
3632         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests-expected.txt:
3633         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/perl_unittests.pl:
3634         Added test cases for here-documents.
3635         (func7):
3636         (func8):
3637         (func9):
3638         (func10):
3639         (func11):
3640         (func12):
3641         (func13):
3642         (func14):
3643         (func15):
3644         (func16):
3645
3646 2012-01-08  Kentaro Hara  <haraken@chromium.org>
3647
3648         Add unittests for the JavaScript parser of prepare-ChangeLog
3649         https://bugs.webkit.org/show_bug.cgi?id=75201
3650
3651         Reviewed by David Kilzer.
3652
3653         javascript_unittests.js is a unittest for get_function_line_ranges_for_javascript()
3654         of prepare-ChangeLog.
3655
3656         * Scripts/prepare-ChangeLog:
3657         (get_function_line_ranges_for_javascript): Before this patch, a string found was just
3658         omitted from parsing. Thus,
3659
3660             str = "foo"
3661             function func() {
3662             }
3663
3664         was recognized as equivalent to
3665
3666             str = function func() {
3667             }
3668
3669         This patch replaces a string with a dummy identifier 'string_appeared_here'
3670         to tell the parser that a string appeared there.
3671
3672         * Scripts/webkitperl/prepare-ChangeLog_unittest/parser_unittests.pl:
3673         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests-expected.txt: Added.
3674         * Scripts/webkitperl/prepare-ChangeLog_unittest/resources/javascript_unittests.js: Added.
3675         (func1):
3676         (func2):
3677         (func3):
3678         (func4):
3679         (func5):
3680         (func6):
3681         (func7):
3682         (func8):
3683         (func9):
3684         (func10):
3685         (func11):
3686         (func12):
3687         (funcOverloaded):
3688         (Func1.prototype.get x1):
3689         (Func1.prototype.get x2):
3690         (Func1.prototype.set x1):
3691         (Func1.prototype.set x3):
3692         (Func2.prototype.func13):