[Bioperl-l] Re: External dependencies and Macs

[Marc Colosimo]

I am also trying to use BioPerl on a Mac and have found several things difficult. I for
one really hate down loading a package, setting it up and running it to find out that
i'm missing X number of things. I don't think it matters which platform any one is on, I
think to some extent everyone has this same problem. As for external 'C' libraries, I
think that considerable thought should be put in before using them. Most of us on this
list probably can build or tweak with the C stuff to get it to work. However, I think my
time is better spent using BioPerl than to get it to work.

I have three suggestions, two  of which I hate to suggest. 1) there should be a bioperl
script for installing the package (using PERL itself). I think there is a UNIX install
script and until MAC OS X come rolling out I can't use it. Plus, the user should have
PERL. 2)  This was already, suggested - include the externals (PERL modules) in the
BioPerl distribution. And 3) change the 'C' libraries over to use JAVA (BioJava anyone).
The last one I think is sort of evil, but JAVA is pretty standard over systems and the
compiled code is fast (not as fast a being native).

One other thing (This should be  one it's own), line endings are a pain to handle. The
Mac uses a different code for \n than PC/UNIX (actually so does the PC vs UNIX but it
probably does not show up as a problem).  If I down load a fasta file (large one with
many sequences in it) as a UNIX file (I could have it translated when decompressing, but
that messes up stuff for MAME :). Stuff that looks for end of line or \n don't find it
and end up choking. I don't know if there is a fix for this, but it is something to
think about when getting data from the web.

Marc