[Bioperl-l] PrimarySeqI and IdentifiableI

Hilmar Lapp hlapp@gnf.org
Thu, 5 Sep 2002 18:37:30 -0700


I apologize for my ignorance if this has all been thrashed out and I 
just forgot.

Right now PrimarySeqI is-a IdentifiableI, which means every 
Bio::PrimarySeqI compliant object is, in fact must be, 
Bio::IdentifiableI compliant at the same time.

Bio::Seq as an example theoretically is but practically is not 
because it doesn't implement a number of methods from IdentifiableI.

There are at least 4 methods to implement.

This prompted me to wonder whether we should really mandate that 
every implementor of PrimarySeqI needs to implement IdentifiableI at 
the same time. Alternatively, they go separate, and a particular 
implementation can choose whether it implements IdentifiableI too or 
not. Analogous for DescribableI.

I'm actually leaning towards separating them. Someone may want to 
have a very bare-bones sequence object that isn't even Identifiable 
nor Describable. This may not be LSID compatible though (anyone got 
thoughts on that?).

Please voice your opinion if you want to weigh in. Core people?

I'll have to fix Bio::Seq anyway (I guess there's little doubt it 
should be Identifiable), but if we decide to separate IdentifiableI 
(and DescribableI) from PrimarySeqI, I'll fix that right away, too.

	-hilmar

--
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------