[Bioperl-l] Using a bioperl cvs checkout
Allen Day
allenday at ucla.edu
Tue Jan 11 18:28:43 EST 2005
in a calling perl script you can add a line like:
use lib 'path/to/bioperl-live';
and it will take precedence over PERL5LIB / PERLLIB / @INC / etc.
or you can do it like this (my preferred method for using an alternate lib
for a one-off:
perl -Ipath/to/bioperl-live path/to/myscript.pl
this basically puts the '-I' argument into the 0th slot of @INC so it gets
used first. you can give multiple '-I' args if needed.
-Allen
On Tue, 11 Jan 2005, Barry Moore wrote:
> Hilmar (or others)-
>
> The bioperl cvs documentation was great, and I've managed to checked out
> bioperl-live, bioperl-db and bioperl-run from anonymous cvs into a
> directory off of my home. Now I've got a couple of questions about how
> to best utilize this code from cvs. I've got an existing installation
> of bioperl 1.4 which I probably don't need to duplicate, but I'm unsure
> of what is the best way to utilize the code from cvs. I see that the
> cvs checkout comes with Makefile.PL etc. Should I run make process on
> the cvs checkout and let it install everything into my standard perl
> library location, or should I keep my cvs checkouts seperate and tell
> perl where it is? I don't have a developer account on bioperl cvs, so I
> won't be commiting (or even changing my local copy) at this point, but I
> might as well do things the right way and from reading 'Open Source
> Development with CVS' it seems like I ought to be using the cvs checkout
> without 'installing' it or moving it anywhere. If I keep them seperate
> the bioperl cvs docs suggest to export PERL5LIB='$HOME/src/bioperl' . If
> I do that I think perl will see two copies of the bioperl modules when I
> run a script (the cvs copy and the installed 1.4 copy). How do I know
> which copy of the modules a script will be using? I don't want to
> completely do away with the system installation of bioperl 1.4 because
> another user is using that.
>
> Barry
>
> Hilmar Lapp wrote:
>
> > Did the test script that you ran come with bioperl? Bio::DB::BioDB
> > comes with bioperl-db, and is not needed for anything else. Also,
> > bioperl-db is not included in the bioperl 1.4 distribution. If you
> > want it, you do need to obtain from CVS at this point. Let me know if
> > you have problems with that.
> >
> > Also, if you do want to use bioperl-db I do recommend you obtain
> > bioperl 1.4 from the CVS branch as well, or otherwise wait for the 1.5
> > release. The 1.4.0 release has problems in the interpro and GO
> > ontology parsers, and 1.4.1 was never released in anticipation of 1.5.
> >
> > -hilmar
> >
> > On Jan 10, 2005, at 5:10 PM, Barry Moore wrote:
> >
> >> I've just installed bioperl 1.4 (bioperl-core, bioperl-run and
> >> bioperl-db) on a new system (Debian woody). I run a test script that
> >> works fine on my old system and get an error that BioDB.pm can't be
> >> found. Sure enough BioDB.pm isn't on my new system, but it is on the
> >> old (also bioperl 1.4 Debian woody). I look in cvs and BioDB.pm is
> >> there, but I look in the distribution downloaded from bioperl.org and
> >> it seems to be missing BioDB.pm and several other files? I can get
> >> the files from cvs, but is this an error in the distribution file?
> >>
> >> Barry
> >>
> >> --
> >> Barry Moore
> >> Dept. of Human Genetics
> >> University of Utah
> >> Salt Lake City, UT
> >>
> >>
> >>
> >> _______________________________________________
> >> Bioperl-l mailing list
> >> Bioperl-l at portal.open-bio.org
> >> http://portal.open-bio.org/mailman/listinfo/bioperl-l
> >>
>
>
More information about the Bioperl-l
mailing list