RESOLVED FIXED76396
REGRESSION(r105072): It broke committers_unittest.py
https://bugs.webkit.org/show_bug.cgi?id=76396
Summary REGRESSION(r105072): It broke committers_unittest.py
Csaba Osztrogonác
Reported 2012-01-16 11:35:47 PST
We have one more Dan as committer, and unit test can't handle this situation: http://trac.webkit.org/changeset/105072/trunk/Tools/Scripts/webkitpy/common/config/committers.py Traceback (most recent call last): File "/ramdisk/qt-linux-release/build/Tools/Scripts/webkitpy/common/config/committers_unittest.py", line 170, in test_contributors_by_fuzzy_match_with_legacy_names self._assert_fuzz_match('Dan', 'Dan Bernstein', 0) File "/ramdisk/qt-linux-release/build/Tools/Scripts/webkitpy/common/config/committers_unittest.py", line 108, in _assert_fuzz_match self.assertEqual(([contributor.full_name for contributor in contributors], distance), (expected_names, expected_distance)) AssertionError: (['Dan Winship', 'Dan Bernstein'], 0) != (['Dan Bernstein'], 0)
Attachments
Patch (2.40 KB, patch)
2012-01-16 11:39 PST, Csaba Osztrogonác
no flags
Csaba Osztrogonác
Comment 1 2012-01-16 11:39:44 PST
Adam Barth
Comment 2 2012-01-16 11:50:17 PST
Comment on attachment 122668 [details] Patch These tests seem poorly designed...
Eric Seidel (no email)
Comment 3 2012-01-16 11:57:43 PST
rniwa is the man behind these tests iirc.
Csaba Osztrogonác
Comment 4 2012-01-16 11:59:53 PST
Ooops, it isn't Qt specific, it was copy/paste error :)
Dan Winship
Comment 5 2012-01-16 12:04:16 PST
Oh, awesome. I broke the build on my first commit. :-O (FWIW, I even ran committers_unittest.py, and it exited with status 0 without printing anything, so I figured I was good... guess I should have looked more carefully at the code.)
Csaba Osztrogonác
Comment 6 2012-01-16 12:05:08 PST
Comment on attachment 122668 [details] Patch Committed r105082
Csaba Osztrogonác
Comment 7 2012-01-16 12:10:35 PST
(In reply to comment #5) > Oh, awesome. I broke the build on my first commit. :-O > > (FWIW, I even ran committers_unittest.py, and it exited with status 0 without printing anything, so I figured I was good... guess I should have looked more carefully at the code.) Not a problem, the test was bad now. (Otherwise we usually run Tools/Scripts/test-webkitpy instead of individual test)
Eric Seidel (no email)
Comment 8 2012-01-16 12:11:49 PST
(In reply to comment #5) > Oh, awesome. I broke the build on my first commit. :-O > > (FWIW, I even ran committers_unittest.py, and it exited with status 0 without printing anything, so I figured I was good... guess I should have looked more carefully at the code.) Despite the existence of those __main__ blocks, that's not really a sanctioned way to run the tests. test-webkitpy is your best bet.
Note You need to log in before you can comment on or make changes to this bug.