[Bioperl-l] SeqIO cleanup

[Hilmar Lapp]

Jason Stajich wrote:
> 
> I'm starting to do some stringent Bio::SeqIO cleanup (working towards
> diffless round tripping that Kris Boulez suggest a while ago and Elia and
> I agree is the right thing to do here).
> 
> Couple of things I've noticed - some SeqIO implementations won't accept a
> PrimarySeq to write_seq.  This is probably bad...  How should a module
> like embl proceed if it gets a PrimarySeq in write_seq - currently it
> balks.

If a PrimarySeq does not contain enough information to write a
valid entry, then it should rise an exception. E.g. if you need
certain features (source for genbank) or annotation.

> 
> Also, some routines return a Seq object always rather than what fasta does
> with next_primary_seq and next_seq - should this be the same behavior
> everywhere?

I think we don't need two methods here. If choosing between
PrimarySeq and Seq really makes sense, this should be accomplished
by a long-planned Bio::Factory::Seq object (managed by and
inherited from SeqIO.pm).

	Hilmar
> 
> I'll have my inplace diffing script committed soon - it does depend on
> String::DiffLine to do cross-platform diffs, I can write it as a script or
> even make it into a t/t test if String::DiffLine is installed on the
> system.  Does that sound like a good idea - otherwise it can just be a
> script in scripts/ or examples/ or a new dir utilties? (can never decide
> where to put things).
> 
> -jason
> 
> Jason Stajich
> jason@chg.mc.duke.edu
> Center for Human Genetics
> Duke University Medical Center
> http://www.chg.duke.edu/
> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l

-- 
-----------------------------------------------------------------
Hilmar Lapp                              email: hilmarl@yahoo.com
GNF, San Diego, Ca. 92122                phone: +1 858 812 1757
-----------------------------------------------------------------