[Debian-med-packaging] Bug#901073: python-biom-format autopkg test failure

Andreas Tille tille at debian.org
Fri Jun 8 23:35:55 BST 2018


Control: forwarded -1 https://github.com/biocore/biom-format/issues/777
Control: tags -1 upstream
Control: tags -1 help

Hi,

I have reported the issue upstream.  I have the vague suspicion that the
migration to numpy 1.4.x might have caused the problem since it is
connected in time to the failure of this test.  There is a bit more
detailed information than the autopkg log provides:


    $ python2.7 -m nose tests
    ...
    unzip(items) should be the inverse of zip(*items) ... ok
    
    ======================================================================
    FAIL: TableSummarizer functions as expected
    ----------------------------------------------------------------------
    Traceback (most recent call last):
      File "/tmp/tmp.ovLIvfIx3j/tests/test_cli/test_summarize_table.py", line 33, in test_default
        self.assertEqual(sorted(result), sorted(summary_default))
    AssertionError: Lists differ: [u'\n', u'\n', u'\n', u'\n', u... != ['\n', '\n', '\n', '\n', '\n',...
    
    First differing element 68:
    u','
    '-'
    
    Diff is 9507 characters long. Set self.maxDiff to None to see it.
    
    ======================================================================
    FAIL: TableSummarizer functions as expected with qualitative=True
    ----------------------------------------------------------------------
    Traceback (most recent call last):
      File "/tmp/tmp.ovLIvfIx3j/tests/test_cli/test_summarize_table.py", line 42, in test_qualitative
        self.assertEqual(sorted(result), sorted(summary_qualitative))
    AssertionError: Lists differ: [u'\n', u'\n', u'\n', u'\n', u... != ['\n', '\n', '\n', '\n', '\n',...
    
    First differing element 56:
    u','
    '.'
    
    Diff is 7916 characters long. Set self.maxDiff to None to see it.
    
    ----------------------------------------------------------------------
    Ran 308 tests in 31.762s
    
    FAILED (SKIP=1, failures=2)

With Python3 the output is slightly different but the same test is failing:

    $ python3 -m nose tests
    ...
    unzip(items) should be the inverse of zip(*items) ... ok
    
    ======================================================================
    FAIL: TableSummarizer functions as expected
    ----------------------------------------------------------------------
    Traceback (most recent call last):
      File "/tmp/tmp.ovLIvfIx3j/tests/test_cli/test_summarize_table.py", line 33, in test_default
        self.assertEqual(sorted(result), sorted(summary_default))
    AssertionError: Lists differ: ['\n'[355 chars])', ',', ',', ',', ',', ',', ',', ',', ',', ',[1687 chars] 'z'] != ['\n'[355 chars])', '-', '.', '.', '.', '.', '.', '.', '.', '.[1687 chars] 'z']
    
    First differing element 68:
    ','
    '-'
    
    Diff is 3455 characters long. Set self.maxDiff to None to see it.
    
    ======================================================================
    FAIL: TableSummarizer functions as expected with qualitative=True
    ----------------------------------------------------------------------
    Traceback (most recent call last):
      File "/tmp/tmp.ovLIvfIx3j/tests/test_cli/test_summarize_table.py", line 42, in test_qualitative
        self.assertEqual(sorted(result), sorted(summary_qualitative))
    AssertionError: Lists differ: ['\n'[293 chars] ', ',', ',', ',', ',', ',', ',', ',', ',', ',[1402 chars] 'y'] != ['\n'[293 chars] ', '.', '.', '.', '.', '.', '.', '.', '.', '.[1402 chars] 'y']
    
    First differing element 56:
    ','
    '.'
    
    Diff is 2901 characters long. Set self.maxDiff to None to see it.
    
    ----------------------------------------------------------------------
    Ran 308 tests in 2.339s
    
    FAILED (SKIP=1, failures=2)
    
Any help to fix this is welcome.  My current approach would probably be
to exclude the two failing tests and decrease the severity of the bug to
normal.

What do you think?

Kind regards

      Andreas.

-- 
http://fam-tille.de



More information about the Debian-med-packaging mailing list