[Bioperl-l] SeqI interface

[Ewan Birney]

On Thu, 8 Feb 2001, Hilmar Lapp wrote:

> Jason Stajich wrote:
> > 
> > On Thu, 8 Feb 2001, Ewan Birney wrote:
> > 
> > > On Thu, 8 Feb 2001, Jason Stajich wrote:
> > >
> > > > Shouldn't methods like primary_seq, annotation, add_SeqFeature be part of
> > > > the SeqI interface?  They are part of Seq.pm.
> > > > I'm trying to be sure that I provide a complete implementation of
> > > > Bio::SeqI in Bio::CorbaClient::Seq.
> > >
> > > primary_seq and annotation - yes
> > okay I'll include them in the SeqI definition.
> 
> Hmm. Why primary_seq() in the interface? The only reason this method
> exists in Seq.pm is that it delegates to a PrimarySeq object, right?
> Maybe I'm missing something here, maybe it's even in the IDL. In any
> way, it's trivial to implement, non-delegating classes could just
> return $self. (I assume SeqI inherits off PrimarySeqI, right?)

Oh no. The delegation model here is crucial the preventing large memory
monsters forming and in large scale world absolutely needed.

by calling primary_seq and releasing seq, the client is indicating to the
server that it can reap the otherwise memory-hogging seq object with all
it's associated crap. 

we use this basic route alot in, eg, ensembl...


 > 
> 	Hilmar
> -- 
> -------------------------------------------------------------
> Hilmar Lapp                            email: lapp@gnf.org
> GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
> -------------------------------------------------------------
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
> 

-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<birney@ebi.ac.uk>. 
-----------------------------------------------------------------