[Bioperl-l] Cleanup of BioPerl distribution website
Chris Fields
cjfields at uiuc.edu
Sun Sep 17 13:56:54 UTC 2006
On Sep 17, 2006, at 5:09 AM, Nathan S. Haigh wrote:
...
> I don't have any immediate problems about the changes either. I've
> seen
> a few posts recently about installing Bioperl on Windows - how soon
> will
> 1.5.2 been released? It's not too difficult to generate a new ppd file
> etc so I could make a barebones ppd for 1.5.1?
>
> I know when I made the ppd file for 1.5 I included a lot of prereqs to
> ensure that most of bioperl would work without the need to manually
> install the modules later once the user found out that something
> didn't
> work. Personally, despite the need to download and install a lot more
> packages in one sitting, I thought this was important since Windows
> users that install bioperl are probably (or more likely) not from a
> programming background (no offence intended if your a whiz programmer
> working in Windows! :o) ). Therefore, their first experience of
> bioperl
> would get off to a better start if everything worked out of the hat
> after its installation, despite having a longer/bigger install.
> What do
> you think?
I'm not sure how the new GUI version of PPM (PPM4) affects
installation. If we keep the prereqs in we might want to remove the
version requirements; a previous poster stated that when they
attempted installation of Bioperl 1.4 it wanted to downgrade the
dependencies, likely to the versions listed in the ppd. I have never
seen it do that before, so it could be something to do with the new
PPM version.
http://bioperl.org/pipermail/bioperl-l/2006-September/023002.html
>
> What is the state of play with regards to tracking dependencies? I've
> just noticed that Makefile.PL has a lot more packages in %packages, is
> this a complete list of prereqs? If so, could they be added to
> PREREQ_PM
> in the WriteMakefile sub in order to make it easier for generating
> a ppd
> with a complete prereqs list?
>
> Nath
Many of those are included with current versions of ActivePerl, like
XML::Simple, libwww-perl, etc. We are also planning on having a
minimal requirement of perl 5.6.1 (Features parsing requires
recursive regexes, which aren't found before 5.6.1). All the tests
pass so far with ActivePerl. I haven't tried anything with CygWin,
though.
Christopher Fields
Postdoctoral Researcher
Lab of Dr. Robert Switzer
Dept of Biochemistry
University of Illinois Urbana-Champaign
More information about the Bioperl-l
mailing list