[Bioperl-l] bug #1546 and BPlite/BPpsilite

[Jason Stajich]

WRT buggy code - Do you mean the parser or the running of BLAST from
withing Bioperl?

If in regard to StandAloneBlast, I totally agree.  I frequently ask for
someone to propose a new rewrite of this module which is more robust,
scalable to WU-BLAST, but so far have had no takers.  I don't personally
run BLAST that way so I've not had an itching desire to rewrite it.

As for buggy parsing, you're using BPpsilite which has been superceeded by
new code in Bio::SearchIO::blast on the main trunk (but not necessarily in
1.2.3).  I do not have any plans to personally maintain any of the
BPlite/BPpsilite/BPbl2seq modules as all of my own development effort is
focused on Bio::SearchIO.  Others are welcome to maintain these - we'll
get you an account to do this if someone volunteers.


As for the problems you are specifically reporting, hopefully someone will
have a chance to look at it soon.

-jason

On Tue, 28 Oct 2003, Michael Prentiss wrote:

> Jason,
> I did not think about the attachment.  I included the inputs, stderr,
> the code, and an example of the problem.
> I found this part of BioPerl really buggy, and the errors were transient
> !?!?.   Is this some sort of memory problem
> because perl is not careful enough with memory allocation?
> Thanks for you help, I am rather confused.
> Michael Prentiss
>

--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu