[Biopython-dev] TravisCI BioSQL - TypeError: unhashable type: 'bytearray'
Wibowo Arindrarto
w.arindrarto at gmail.com
Fri Oct 3 13:34:11 UTC 2014
Hi Peter, everyone,
Peter: ah yes, I meant this email to be sent to the whole group :).
The replies are below:
On Fri, Oct 3, 2014 at 11:59 AM, Peter Cock <p.j.a.cock at googlemail.com> wrote:
> Off list by mistake? Feel free to include this in any reply to the list.
>
> I like the fact that TravisCI tends to use "latest and greatest" so it
> warns us of cutting edge problems.
>
> Our buildbot buildslaves tend to be more "old and stable" and thus
> help catch any regressions where something breaks under an older
> library.
>
> Peter
>
>
> On Fri, Oct 3, 2014 at 10:53 AM, Wibowo Arindrarto
> <w.arindrarto at gmail.com> wrote:
>> Hi Peter,
>>
>> I took a brief look at our test dependencies last night and saw that the
>> mysql python connector was updated sometime last week and it seemed to be
>> updated around Sept 24th:
>> https://dev.mysql.com/doc/relnotes/connector-python/en/.
>>
>> As a side point, should we also pin external dependencies to specific
>> versions in our builds?
>>
>> Cheers,
>> Bow
>>
>> On Oct 3, 2014 11:47 AM, "Peter Cock" <p.j.a.cock at googlemail.com> wrote:
>>>
>>> Hi all,
>>>
>>> As noted by Bow and Evan, something appears to have changed
>>> on TravisCI with MySQL which is breaking our BioSQL tests,
>>> https://github.com/biopython/biopython/issues/364
>>>
>>> Can anyone reproduce this or spot what was changed on Travis?
>>> e.g. an update to MySQL itself, or the mysql Python bindings, or
>>> a Travis configuration change?
>>>
>>> Thanks,
>>>
>>> Peter
>>>
>>>
>>> e.g.
>>> https://travis-ci.org/biopython/biopython/jobs/36919122
>>>
>>> test_BioSQL_MySQLdb ... ok
>>> test_BioSQL_mysql_connector ... FAIL
>>> test_BioSQL_psycopg2 ...
>>>
>>> /home/travis/build/biopython/biopython/build/lib.linux-x86_64-2.6/BioSQL/BioSeqDatabase.py:121:
>>> BiopythonWarning: Your BioSQL PostgreSQL schema includes some rules
>>> currently required for bioperl-db but which may cause problems loading
>>> data using Biopython (see BioSQL Bug 2839). If you do not use BioPerl,
>>> please remove these rules. Biopython should cope with the rules
>>> present, but with a performance penalty when loading new records.
>>> "new records.", BiopythonWarning)
>>> ok
>>> test_BioSQL_sqlite3 ... ok
>>>
>>> ...
>>>
>>> ======================================================================
>>> ERROR: GenBank file to BioSQL, then again to a new namespace,
>>> protein_refseq2.
>>> ----------------------------------------------------------------------
>>> Traceback (most recent call last):
>>> File "/home/travis/build/biopython/biopython/Tests/common_BioSQL.py",
>>> line 634, in test_protein_refseq2
>>> self.trans("GenBank/protein_refseq2.gb", "gb")
>>> File "/home/travis/build/biopython/biopython/Tests/common_BioSQL.py",
>>> line 667, in trans
>>> self.assertTrue(compare_records(original_records, biosql_records))
>>> File "/home/travis/build/biopython/biopython/Tests/seq_tests_common.py",
>>> line 357, in compare_records
>>> if not compare_record(old_r, new_r):
>>> File "/home/travis/build/biopython/biopython/Tests/seq_tests_common.py",
>>> line 284, in compare_record
>>> if not compare_features(old.features, new.features):
>>> File
>>> "/home/travis/build/biopython/biopython/build/lib.linux-x86_64-2.6/BioSQL/BioSeq.py",
>>> line 545, in __get_features
>>> self._primary_id)
>>> File
>>> "/home/travis/build/biopython/biopython/build/lib.linux-x86_64-2.6/BioSQL/BioSeq.py",
>>> line 226, in _retrieve_features
>>> qualifiers.setdefault(qv_name, []).append(qv_value)
>>> TypeError: unhashable type: 'bytearray'
>>> _______________________________________________
>>> Biopython-dev mailing list
>>> Biopython-dev at mailman.open-bio.org
>>> http://mailman.open-bio.org/mailman/listinfo/biopython-dev
More information about the Biopython-dev
mailing list