[Biopython-dev] BioSQL buglets

[Peter]

On Wed, Nov 12, 2008 at 10:57 AM, Cymon Cox <cymon.cox at googlemail.com> wrote:
> All,
>
> Selects on the seqfeature_qualifier_value and dbxref tables were not being
> ordered by rank. This caused multiple qualifier values to be out of order
> which in turn caused the tests to fail - see comment in
> http://bugzilla.open-bio.org/show_bug.cgi?id=2616
>
> This also solves a TODO in the test_BioSQL_SeqIO.py:
>
>  85 +#TODO - Pin down the "Duplicate entry" IntegrityError from this:
>  86 +#    ("genbank",False, 'GenBank/cor6_6.gb', 6),
>
> This test now works and Ive generated new output.
>
> In test_BioSQL.py create_database(), postgres returns an error string that
> 'find's on index 0 when the the database doesnt exist. The comparision
> therefore needs to be >= 0 rather than >0.
>
> All tests now pass OK with postgresql/psycopg2.
> Patch attached.
>
> Cheers, C.

Excellent - that patch made perfect sense and I've checked it in
(almost as is - I tweaked the find index bit slightly).  Thank you!

At this rate you'll be co-opted as an official maintainer for the
BioSQL module ;)

Peter

P.S. It might have been better to upload the patch to Bug 2616 (or a
new Bug) rather than sending it to everyone on the mailing list.