[Biopython-dev] BioSQL Loader.py patch

[Peter (BioPython Dev)]

Leighton Pritchard wrote:
> Hi all,
> 
> I'm not sure if you want this here, or on the BioPython BugZilla, but

Using bugzilla is probably a better idea as:
(a) It will look after patches rather than fighting email attachments
(b) Its easier for the developers to see in one place what is 
outstanding/in need of attention.

> I've written a patch that modifies BioSQL/Loader.py to load db_xref
> qualifier/value pairs, and also contains a patch to correct bug 1921
> caused by the attempted default insertion of an invalid taxon_id.

I had noticed bug 1921 when you logged it, but having never dabbled with 
mySQL I didn't want to touch it.

> The .diff is attached.  I've been living with the patched code for a
> month without any issues, so it's been stable as far as I've needed it
> to be (and has worked with all bacterial GenBank .gbk files and the
> BioPython GenBank parser).

Out of interest, are you running the CVS GenBank parser?

> There are two sections of the patched code that print/write to stdout.
> Is this an acceptable way of reporting to the user in a BioPython style?

Good question.  I have seen some parts of the code simply using "print" 
to output warning messages.  Some of the PDB code explicitly directs its 
warnings to std error.

Anyone?

Peter