On Wed, Jan 07, 2009 at 11:14:26AM -0800, Christopher Lee wrote:
-> in the interest of making pygr easier to understand and work with,  
-> I've tried to subdivide large modules (such as seqdb) into smaller,  
-> more coherent pieces.  For example, I recently moved all the BLAST  
-> support out of seqdb.py into a new module blast.py.
-> 
-> Perhaps we should also move annotation functionality out of seqdb into  
-> a new module (annotate.py or annotation.py?).  AnnotationDB doesn't  
-> depend on seqdb; I don't think the prospective annotation.py would  
-> even need to import seqdb...  Their association is purely conceptual.
-> 
-> One consideration: backwards compatibility for unpickling and thus  
-> pygr.Data.  Unpickling looks for classes based on the module location  
-> saved when the object was originally pickled, so to maintain the  
-> ability to unpickle (e.g. retrieve from pygr.Data) existing annotation  
-> resources, we'd have seqdb import a few key classes (AnnotationDB and  
-> kin) from annotation.py, so that those names continue to be accessible  
-> from the seqdb namespace as before.
-> 
-> What do you think?  Does this seem worthwhile, and should we include  
-> this in the 0.8 release?

Yes, that makes sense!  I'm agnostic about the release timing.

How about adding __all__ attributes to modules, too?  Then we can know
what's supposed to be exported.

cheers,
--titus
-- 
C. Titus Brown, [email protected]

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"pygr-dev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/pygr-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to