[BioPython] Updates to biopython-corba

[Brad Chapman]

Hello all;
    I'm not positive if anyone is following the ol' corba stuff, but I 
just made a huge update and rearrangement in cvs, so I thought I would 
let everyone know, just in case.
    The biggest change is that I moved everything to live under the 
'BioCorba' namespace and kind of expanded the goals of this module. 
Andrew's comments to my original message asking about the namespace 
made me start thinking about the fact that biopython-corba could also 
serve as a "gateway" to other corba servers besides just the 
Biopython/Bioperl/Biojava one. I thought this could kind of give 
Biopython some uniqueness over Bioperl and Biojava, in that users 
could work with Biopython objects, and also use these objects to query 
remote servers through corba interfaces. I thought this was a way I 
could help suppliment the stuff that you all are doing with the main 
module with the stuff that I know how to do (sort of :-) and am 
interested in.
    So at any rate, with this goal in mind biopython-corba also now 
has a little test client which connects with the EMEST database at EBI 
(http://industry.ebi.ac.uk/~muilu/EST/Database/index.html). The test 
client connects with their remote server and does database "searching" 
through their corba interface. I think it's pretty neat to be able to 
connect to their database halfway across the world (at least from me) 
and be able to do queries through function calls directly into their 
database and get back results that don't need any parsing. And it all 
goes pretty fast, too. My eventual plan was to build a client 
interface that kind of automates the querying to directly give you Seq 
objects from info in the databases. The EMEST stuff only works with 
omniORB right now, but I'll get Fnorb going on it soon...
    Other news is that we can now connect without problem to the 4 
bioperl corba servers using Fnorb (omniORB causes their server to seg 
fault and die). It looks like we are well on our way to some serious 
corba interoperability!
    Also, biopython-corba also has a Disutils build, you can get it 
installed by just setting variables in the configuration file, and 
then running 'python setup.py install.' Distutils is very smooth, and 
I managed to hook into it to do the idl file compilation automatically 
during install without too many problems (although now there is a lot 
of installation code :-). 
    So that is all the news that's fit to print on the corba front...

Brad