[Bioperl-l] Package status for 1.5.2

Chris Fields cjfields at uiuc.edu
Mon Sep 25 20:33:53 UTC 2006


> -----Original Message-----
> From: bioperl-l-bounces at lists.open-bio.org [mailto:bioperl-l-
> bounces at lists.open-bio.org] On Behalf Of Sendu Bala
> Sent: Monday, September 25, 2006 1:21 PM
> To: bioperl-l at bioperl.org
> Subject: Re: [Bioperl-l] Package status for 1.5.2
> 
> Sendu Bala wrote:
> > To what extent do we want a unified release for all the various
> > packages? As I see it, here is the current status:
> >
> > Core
> > ----
> > P1 and P2 bugs remain in bugzilla, but nothing that need prevent the
> > release. These only need to be resolved before 1.6.
> >
> > Test suite passes satisfactorily, though see
> > http://www.bioperl.org/wiki/Release_1.5.2 for some minor issues (again,
> > nothing that prevents release).
> >
> > Installation issues (ie. PPM, Windows) are being investigated by Nathan
> > and Chris. Are they serious enough to prevent release? (Minimal Perl
> > version will be 5.6.1)
> >
> > Conclusion: ok to be tagged bioperl-release-1-5-2-rc1 and released for
> > testing.
> 
> Sorry, I had forgotten about the Bio::DB::SeqFeature failures on Windows
> which need to be resolved before an RC is worth releasing. I've pushed
> back RC1 release date to October 2nd; hopefully we can resolve the issue.
> 
> It will also give me time to resolve the bioperl-run problems that I
> see, so I can do a unified release.

I don't think you need to have all of the non-core distributions ready at
the same time, especially if you want to have a few release candidates for
the core before a final one.  I don't see the harm in getting it out there. 

Anyway, I have worked out the BioDBSeqFeature_BDB.t issues on WinXP and have
committed a fix; these now pass (though I can't get it to find my WinXP
MySQL installation).  Essentially it was a path name issue; WinXP can't use
names with '::' in paths, which the BDB adaptor was trying to make.  I added
a workaround to check for $^O and modify the name of the tempfiles created.


There is still a problem with the way Bioperl sets up tempfiles on WinXP (it
creates them w/o problems, but runs into issues deleting them).  I'll draft
up a quick test script and submit it to bugzilla.  My guess is that
something changed in later versions of ActivePerl due to heightened
(paranoiac?) security concerns regarding Windows.  There will probably need
to be some updating of Root::IO to deal with this.  I'll see if Nathan Haigh
can confirm this on his WinXP installation, so I know it's not just me ;>

It's not a release killer, though, so I wouldn't worry too much about it.

Christopher Fields
Postdoctoral Researcher - Switzer Lab
Dept. of Biochemistry
University of Illinois Urbana-Champaign




More information about the Bioperl-l mailing list