On Wed, Mar 18, 2009 at 10:51:36AM -0700, Christopher Lee wrote:
-> > Specific issues
-> > ---------------
-> >
-> > I don't like some aspects of the 'ifile' handling in SequenceFileDB.
-> > In particular, I dislike the part where it re-opens the file to
-> > re-read it and then closes 'ifile'; this may not be expected behavior
-> > ('ifile' itself is only used to grab the filename!) and I don't like
-> > closing user-passed-in file handles. Plus, pulling the filename out
-> > of a file handle disallows StringIO and other "file-like" handles such
-> > as sockets... Is the ifile mechanism actually used anywhere?
->
-> This was used by some coordinator module functionality that is made
-> obsolete by the much more powerful capabilities of pygr.Data. We can
-> certainly get rid of this ifile mechanism.
Removed.
-> > The 'set_seqtype' function on SequenceDB depends on things being a
-> > FASTA file
-> > instead of just iterating. Why?? Can we remove that?
[ ... ]
-> Conclusion: we could again get rid of read_fasta_one_line().
Removed.
-> > Also, 'set_seqtype' returns the sequence type as well as setting it on
-> > the db object. This behavior isn't used anywhere in the code base.
-> > Can
-> > we trash it?
->
-> Sounds fine to me.
Removed.
-> > In fact, it seems like 'set_seqtype' is used only within SequenceDB
-> > and its children. Why not rename it to be private, i.e. prepend it
-> > with '_'?
->
-> Fine by me.
done.
-> > I don't know what idFilter does in _store_seqlen_dict, and it seems
-> > like it would be easier to just implement that in your own
-> > seqfmt-based reader, anyway. Can we trash it? It's not used anywhere
-> > in the code base.
->
-> [ ... ] We could get rid of idFilter in _store_seqlen_dict.
Done; I removed the 'ifile' mechanism from there, too, since it's no
longer needed.
-> > The SliceDB, VirtualSeq, and VirtualSeqDB classes are untested and
-> > unused in the rest of the code base. I'd appreciate some guidance on
-> > what they do, and perhaps some tests to get started. (@CTB issue)
->
-> I don't think anyone has used these classes in at least a couple of
-> years. We could delete them.
Removed.
All changes tested & pushed to my github pygr repo, branch seqdb_review.
cheers,
--titus
--
C. Titus Brown, c...@msu.edu
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"pygr-dev" group.
To post to this group, send email to pygr-dev@googlegroups.com
To unsubscribe from this group, send email to
pygr-dev+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/pygr-dev?hl=en
-~----------~----~----~----~------~----~------~--~---
