[BioPython] idl/corba implementation ?'s

[Brad Chapman]

> I was thinking about the same problem Brad, and I was wondering if
> the following IDL / extension might be interesting:
> 
>   // needs exceptions
>   interface BioEnv {
> 	Seq  get_Seq_file(in string filename);
>        PrimarySeq get_PrimarySeq_file(in string filename);
>        PrimarySeqIterator get_PrimarySeqIterator_file(in string 
filename);
>  // using jeff's database names
> 	PrimarySeqDB get_PrimarySeqDB(in string name);
>        SeqDB        get_SeqDB(in string name);
>    };

This looks great! I just updated the biopython corba server to support 
this new interface (it won't be a big deal to add exceptions in since 
there isn't really any code to need exceptions in biopython yet :).
    So now the Biopython corba server is a full-fledged server which 
publishes an IOR for the BioEnv interface and has all of the methods 
and interfaces implemented. Of course, they don't do anything more 
than return an item of the appropriate type, but they all seem to work 
from a quick test so any of the other languages should be able to 
write clients to the biopython server to test it.

> The IOR for this BioEnv object we could define as being in the file
> pointed by environment variable
> 
> BIOCORBA_BIOENV_IOR_FILE
>
> This way when a bioperl/biojava/biopython script starts up it can 
just
> say "get me the environment" and from then onwards not know/care 
where
> this was actually being executed...
>
> I think this is sorta cool ;)

I agree! I wish there was an even more flexible method since the 
environmental variable can't carry over into a new x-term window...
    But the biopython server now sets the environmental variable, and 
I have a backup for reading a file from a default location and also a 
way to set the filename directly from python code when you start up 
the server, so it's as flexible as possible.

    Well, I guess the next step is to try and get an actual 
functioning server set up in the code. Jeff has already kindly 
supplied some starting code in the current biopython module 
(Bio/Sequence.py), and I know there was a lot of discussion about this 
earlier on the list. Is anyone still interested in trying to get this 
together? It would have immediate usefulness to both Biopythoners and 
now, thanks to the idl and corba, to Bioperlers and Biojavaers as 
well. Any thoughts/comments from anyone?

Brad