[Bioperl-l] Found bug in fasta.pm

Hilmar Lapp hlapp@gmx.net
Thu, 09 Nov 2000 09:12:52 -0800

Aaron J Mackey wrote:
> I'm not sure why SeqIO/fasta.pm was migrated to a read-line-by-line
> architecture (thus necessitating the nitty-gritty mentioned below), but

It wasn't. It reads whole records, at least it tries to, and in the past I
advised against suggestions to migrate to line-by-line reading. The winner
in your code sample is setting $/ to "\n>" instead of ">" only, and as
someone reluctant to delve into those cryptic perl built-in variables ;-) I
wasn't aware that you can specify multi-char strings. If no-one else does
I'll take care of simplifying the code by exploiting this. The fasta module
needs an amendment anyway to correctly cope with empty sequences (which we
agreed upon some time ago that we'd like to have it).


Hilmar Lapp                                email: hlapp@gmx.net
GNF, San Diego, Ca. 92122                  phone: +1 858 812 1757