[Bioperl-l] Perl 6 hacking was YAPC anyone?

[Chris Fields]

> Chris Fields wrote:
> > Speaking of Perl6, there was interest here at one point in getting a
> > bioperl-experimental going, which at this point in the game should
> involve
> > Perl6.  If there were enough interest in it we could probably get it set
> up
> > via CVS and moving along.  We might need to split the Perl6 stuff from
> Perl5
> > experimental modules in some way to prevent confusion (bioperl6-
> live???),
> > though I'm not up to speed Perl6-wise so I'm not sure about namespace
> > collisions and so on.
> 
> As far as I understood it, the plan is to have a very smooth migration
> path from Perl 5 to Perl 6. You start with plain old Perl 5 code. When
> new stuff is coming along, or when refactoring is done, you drop in
> 
>    use v6;
> 
> or
> 
>    use v6-pugs;
> 
> and contiue with Perl 6. See http://svn.openfoundry.org/pugs/lib/v6.pm
> or Audrey Tangs presentation at the Nordic Perl Workshop:
> http://pugs.blogs.com/talks/npw06-deploying-perl6.pdf.
> So I would argue against having a completely seperate Perl6 experimental
> repository.

Makes sense.  I know Pugs is the Perl6 implementation in Haskell but I also
know eventually Parrot will be taking over as the compiler (hopefully).
Perl6 is pretty exciting since it's built to support OOP from the ground up,
unlike the bolted-on OOP for Perl5, and has several other features that make
it very useful (the new way regexes are handled).  I just haven't had time
to play around with it seriously enough.  I may try using Pugs a bit more,
though.

So, as long as Perl5-Perl6 work together a separate repository wouldn't be
necessary.  

> > bioperl-experimental would be, like the name implies, a sort of testing
> > ground for ideas (good and bad).  It seemed like it was going to take
> off a
> > few years ago but it lost steam, I'm guess.
> >
> > As for your parsers, would you build them from the ground up (i.e. from
> > Bio::Root::Root on up)?
>
> I'm just a casual Bio::Perl user and never hacked on any internals. So I
> don't know whether the current Bio::Perl framework is a good fit.
> 
> The idea that is floating in my mind is to make a showcase of Perl 6
> parsing, by tackling the various sequences and alignment formats.
> So this would involve shopping around for the cleanest parser
> implementations and porting that to Perl6.
> 
> Which repository to use is more a question of social engineering.
> Are there more Pugs/Perl6 hackers interested in cool biological hacking,
> or biologist aching to try out Perl6?

I suppose the best way is initially to use a non-bioperl approach using
Perl6, then try working the parsers in using 'use v6-pugs;'.  Bioperl is
heavily object-oriented so the code would probably need to be refactored
from the bottom up (or top down, depending on your view) to fit Perl6.
Having a perl5->perl6 translator helps, though.  And, again, having Perl5
and Perl6 work together helps as well.

Chris

> Regards,
>    Bernhard Schmalhofer
> 
> --
> **************************************************
> Dipl.-Physiker Bernhard Schmalhofer
> Senior Developer
> Biomax Informatics AG
> Lochhamer Str. 11
> 82152 Martinsried, Germany
> Tel: +49 89 895574-839
> Fax: +49 89 895574-825
> eMail: Bernhard.Schmalhofer at biomax.com
> Website: www.biomax.com
> **************************************************