[Biopython-dev] Jython and sqlite

[Tiago Antão]

On 8/30/07, Peter <biopython-dev at maubp.freeserve.co.uk> wrote:
> I'm not aware of anyone trying Biopython on Jython (or even other
> variants like Iron Python).  Trying Bio.PopGen on Jython sounds sensible
> - have you tried the Biopython unit tests to see what happens?

Nothing good. I personally use a customized version of Jython from
CVS, based on 2.3.
I tried with the supplied version on Kubuntu (Gutsy Gibbon) which is
2.1. And it doesn't even complete the import phase.
I use Jython myself a bit (Most public software that I do runs inside
Java Web Start), and, to be honest, it is still a bit behind CPython.

My main question was to see if people would be stressed if I sometimes
use old Python dialects because of the fact that I use part of the
code inside Jython. Although I suppose that most of the code that I
will be committing from now on will not be (even partially) developed
inside Jython.


> I haven't used sqllite before, but my initial reaction is wanting to use
> a Python 2.5 only module would be bad.  Have you looked at the existing
> BioSQL code in Biopython...

I had a brief look at BioSQL. It is not targeted at HapMap at all (I
was not expecting it to be).
I think I am going to subscribe to the BioSQL mailing list, throw the
question about HapMap support and wait for answers, although my guess
is that it probably won't happen. In that case probably the best
solution would be something like anydbm. Any thoughts?

Tiago


-- 
http://www.tiago.org/ps