[Bioperl-l] BioPerl CPAN

Heikki Lehvaslaiho heikki at ebi.ac.uk
Tue Jul 1 16:59:38 EDT 2003


Nat,

Sounds great. I'd like to try this. 

I can volunteer to do the cpan admin to start with if no-one else will.

First, let's discuss what exactly we want to do here and how:

1. Where can be get more information how CPAN/PAUSE works?

http://www.cpan.org/modules/04pause.html gives some guidelines but not
enough.

Are there any written guides to the whole system which are essential
reading?


2. Jarkko, how do we make sure that the new effort does not conflict
with the current bioperl bundle in CPAN? We will be using the same Bio
name space, aren't we?


3. How do we split up the current bioperl-live?

Should there be a relatively massive bioperl-core or should everything
be maintained as small 6-20 module chunks?

For example, the smallest core chunk I can think of consists of 
- Bio::PrimarySeq
- Bio::Seq 
+ Bio::SeqFeature::*
- Bio::Range
+ Bio::Tools::Translate
? Bio::SeqUtils
+ corresponding I files
+ t/files


  A thought: should we try to clean up the name space and break the
backward compatibility at the same time:
Bio::PrimarySeq should become Bio::Seq::Primary?

Next set could be:
- Bio::SeqIO
+ Bio::Factory::*

+ Bio::SeqIO::* components
- Bio::Seq::RichSeq


after that:
- Bio::DB::* 

and:
- Bio::SimpleAlign
- Bio::LocatableSeq
- Bio::AlignIO
*+ Bio::Align::*

and:
- Bio::SearchIO...


CPAN supports dependencies so we do not have to be too particular how we
split these up, but I do not think it makes sense to go down to single
single module level most of the time? 

Or would it make sense to have all those above and some more as one
package covering all basic sequence management classes. And then start
building on top of that using sets which are less loosely connected to
core?

One option is to rely heavily on bundles which pull together really fine
grained subsets of bioperl?



Opinions?

	-Heikki


On Tue, 2003-07-01 at 15:32, Nathan (Nat) Goodman wrote:
> Hi Folks
> 
> I had a very positive response from Jarkko Hietaniemi, the CPAN master
> librarian, about the idea of setting up a BioPerl repository within CPAN.
> 
> I floated two ideas past Jarkko. One was to set up a separate category in
> the CPAN module list and then for us to use CPAN "as is".  The other was to
> use the CPAN/PAUSE software to create our own CPAN-like system that we would
> administer on our own.
> 
> Jarkko encouraged the first option, saying, "Setting up a new ‘BioPerl’
> category is definitely not a problem.  I think we could enable one of you
> bioperl guys to act as a ‘CPAN admin’ meaning that you can accept new users
> and new modules."
> 
> I think this is the best option, also.
> 
> Perhaps the next step would be to propose a short list of modules for an
> initial test of the concept and to use the experience to derive principles
> along which to populate the repository. Heikki also suggested we use the
> opportunity to try out Module::Build instead of ExtUtils::MakeMaker.
> 
> I am happy to “volunteer” the Biblio modules I’ve been working on as one
> test case for both the CPAN idea and Module::Build.  It would be good to get
> a couple of other examples, including at least one module that’s already in
> the standard release to see how painful it will be to excise code from the
> main body.
> 
> Do people agree?  Who else wants to work on this?
> 
> Best,
> Nat
> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
-- 
______ _/      _/_____________________________________________________
      _/      _/                      http://www.ebi.ac.uk/mutations/
     _/  _/  _/  Heikki Lehvaslaiho    heikki_at_ebi ac uk
    _/_/_/_/_/  EMBL Outstation, European Bioinformatics Institute
   _/  _/  _/  Wellcome Trust Genome Campus, Hinxton
  _/  _/  _/  Cambs. CB10 1SD, United Kingdom
     _/      Phone: +44 (0)1223 494 644   FAX: +44 (0)1223 494 468
___ _/_/_/_/_/________________________________________________________



More information about the Bioperl-l mailing list