[Bioperl-l] Smaller Bioperl Modules

[Chris Fields]

On Jul 23, 2008, at 8:02 PM, Hilmar Lapp wrote:

> What would be really nice to have for such purposes is a build  
> option that, given a list of modules, creates a tar (or whatever  
> else that's suitable for distribution) with those modules and all  
> other modules they depend on, recursively of course.
>
> Presumably this is what you would be looking for? It'd be great if  
> someone would like to take this on (perhaps yourself?), though we  
> should also keep in mind that in reality this is going to be a  
> rather rare need. Still, as you describe, for those that do have the  
> need it might help a lot.

I think that's a really good idea, and obviates the need to split  
everything up.  We need to come up with a scheme to tie together the  
various dependencies and tests with their modules, though, but I think  
it's doable.  Just needs someone with the time to try it out on a  
branch *cough*Nathan*cough*

> I should also note that in my observation it is not uncommon for  
> sysadmins in biology or other department to have had to install  
> bioperl already for faculty or students, so having to do this may  
> not be such an extraordinary task for many out there.
>
> 	-hilmar

Agreed, though many sysadmins are adverse to installing bioperl-live  
over (say) an RPM or off CPAN, so the latest one could get in that  
case would be 1.5.2.  Actually, some don't even consider 1.5.2  
'stable' due to our designation of the 1.5 series as 'developer';  
maybe something we need to rethink down the road (and the reason I  
proposed the core-tools-dev split).

chris