[Biopython-dev] [Bug 3043] test_NCBI_BLAST_tools fails

[bugzilla-daemon at portal.open-bio.org]

http://bugzilla.open-bio.org/show_bug.cgi?id=3043





------- Comment #3 from biopython-bugzilla at maubp.freeserve.co.uk  2010-04-06 04:45 EST -------
(In reply to comment #2)
> 
> Uhm, how does a test failing on some user's machine alert us to update the
> wrappers? Especially with new users, it's more likely that they will conclude
> that Biopython is buggy, and stop using it. It's better to execute such a test
> only if a user or developer specifically asks for it.
> 

I was expecting that if a user running a new BLAST+ runs the unit test and
hits this issue, they'd report the issue to us. Looking at this afresh, the
assert wasn't really a good long term solution (but it was very helpful in
checking the wrappers had full coverage).  To address this issue in the short
term, I've made this abort the test with a missing external dependency error
(so that run_tests.py will skip it) with what I hope is a clear and not too
scary error message.

In this particular case, those extra arguments *may* only these because your
copy of BLAST 2.2.22+ has been compiled in debug mode. They are not present
on my install of BLAST 2.2.22+

We don't have "full" test suite do we? That would be useful, and we could do
things like require larger test files to be present (which we can track in the
repository but not ship) or generated, or assume a particular BLAST database
will be installed.

Can we close this bug?


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.