Bioperl: "port" to NT

Osborne, Brian
Fri, 11 Dec 1998 12:23:30 -0500

To the group,

The NT port proceeds.Here's what I've done :

Install the "full.exe" from Cygnus. This creates
a bash shell on NT, with a tiny set of Unix utilities
plus gcc. Very easy (

Install Activestate version of perl. As mentioned, it's
based on 5.003_07. It does not contain or associated
files. Not hard ( Makes C:\Perl.

Download the Bioperl .04 *tar.gz file (I'll check the *.zip
file). Ignore 'perl' and so on. Rather copy the 
Bio\ directory tree to the pre-existing C:\Perl\lib directory.
This was Ewan's suggestion. Essentially ignore anything that
would require compiling for the moment, just deal with the 
pure Perl.

Test with "perl -e use Bio::Seq" or "perl -e use Bio::Object;",etc.
Comment out all lines 'use'. That's in and Change 'require 5.004' to 'require 5.003'.
That's in and (BTW, I see that and say 'require 5.002' but I didn't change that of course).

Test a little script that creates a sequence object and does
some simple things to the sequence (called "sliceDNA"). Works fine.
(BTW, on my machine, "#!/usr/bin/perl" needs to be "#!perl".

Important detail : Perl on NT means one of 2 things (at least!).
Either, through a Unix shell as described or through NT's "Command
Prompt". The latter requires a little more messing around
in the "Environent" in order to associate the "pl" suffix with 
the executable Perl (else one always needs to do "perl <some script>".
But I've done this as well when I installed the Activestate
package a while back. So if I take sliceDNA, rename
it "", it works as well.

Nice job you all! What next?

Brian O.

Brian Osborne
Cadus Pharmaceutical Corporation
777 Old Saw Mill River Rd.
Tarrytown NY USA
TEL 914 467 6291
FAX 914 345 3565

=========== Bioperl Project Mailing List Message Footer =======
Project URL:
For info about how to (un)subscribe, where messages are archived, etc: