Ricardo Wurmus <ricardo.wur...@mdc-berlin.de> skribis:
> Ludovic Courtès <l...@gnu.org> writes:
>
>> Ricardo Wurmus <ricardo.wur...@mdc-berlin.de> skribis:
>>
>>> From e8cffdb9e7a2447657b3f700b2f0beb00ad0f98e Mon Sep 17 00:00:00 2001
>>> From: Ricardo Wurmus <ricardo.wur...@mdc-berlin.de>
>>> Date: Wed, 3 Jun 2015 17:44:20 +0200
>>> Subject: [PATCH] gnu: Add BioPerl.
>>>
>>> * gnu/packages/bioinformatics.scm (bioperl-minimal): New variable.
>>
>> [...]
>>
>>> +(define-public bioperl-minimal
>>> + (package
>>> + (name "bioperl-minimal")
>>
>> So is it meant to be used as a Perl library, as a standalone executable,
>> or both?
>
> It's a library and a collection of perl executables using the library,
> so I suppose "bioperl-minimal" would be appropriate?
Yeah that’s fine with me.
>>> + (native-search-paths (list (search-path-specification
>>> + (variable "PERL5LIB")
>>> + (files '("lib/perl5/site_perl")))))
>>
>> Why is it needed? At first sight it looks wrong because PERL5LIB is
>> “owned” by Perl itself.
>>
>> If there’s an executable that needs to find the libs listed in
>> ‘propagated-inputs’, the best way would be to use ‘wrap-program’, I
>> think.
>
> That's for the executables to find the library contained in this package
> as well as the propagated Perl libraries. There are *many* executables,
> which made me shy away from wrapping each of them to be run with the
> PERL5LIB variable set.
There are two cases:
1. When using BioPerl as a library, users will also have Perl
installed, so ‘guix package’ will report the right value for
PERL5LIB. No problem here.
2. When using just the executables, Perl might be missing from the
profile. ‘wrap-program’ looks best to me for that, even if there
are many executables.
WDYT?
Thanks,
Ludo’.
