[Bioperl-l] happy bioperl ;)

Ewan Birney birney@ebi.ac.uk
Thu, 1 Mar 2001 23:31:37 +0000 (GMT)


On Thu, 1 Mar 2001, Hilmar Lapp wrote:

[back from pub....]


> Jason Stajich wrote:
> > 
> > All tests for me pass on multiple architectures.  I suspect there may be
> > some Index.t and Tempfile bugs still lurking, but I leave it up to Hilmar
> > on whether or not they will get fixed for this release or if they'll come
> > out in the soon-to-follow point release.  I'd like to get back to the
> > mantra of "release early, release often" which can provide some quick
> > turnaround to major issues.
> 
> Release early, but not premature. I'd like to be able to keep the
> proud feeling for a few days before the blame floods in ;)
> 
> There may issues lurking with Index.t and tempfile, but these are
> platform specific. If something's going to stop me from branching
> tonight, it has to be either a showstopper report on Unix, or my
> inability to figure out how to branch off in cvs.

cvs tag -b <branch-tag-name> from memory.

There is a school of thought that says for each branch you should do a

cvs tag <snapshot-of-where-branch-started-tag>
cvs tag -b <branch-tag>

I'm so-so with this. 


Only advice: double/triple check that you have cvs committ'd everything
and that you are cvs update'd correctly wrt to the repository. I have
previously put in a branch before some "final commits"and then had to
refix down the branch. Yuk.

> 
> Putting together the CHANGES may also cost me some time, I'll have to
> check what Ewan already added (ideally everything :-).
> 
> Your mantra is BTW also the reason that I'm not going to further
> extend the SeqFeature::Gene::* interfaces with respect to regulatory
> elements for the 0.7 release, contrary to what I initially intended.
> This needs both more experience and discussion of real code. So, I'll
> just mark the promoter() methods as experimental and possible to
> change. I'm just mentioning this here so that people don't think I've
> simply forgotten it.
> 

This sounds very sane!


> > 
> > All tests pass on:
> >  x86 FreeBSD4.2 perl 5.005_03, Alpha linux perl 5.005_03, Sparc Solaris
> >  2.7/2.8 perl 5.6.0, x86 linux perl 5.005_03.   WinNT activestate 5.6.0.
> > 
> > We need to make sure MANIFEST file is up to date (I think we can
> > regenerate with a simple find command), and that all mailing list info is
> > up to date.
> 
> Yeah, interesting that the old one still pop up. I thought some time
> ago someone went over this with a search/replace.
> 
> > 
> > As for biocorba:
> > 
> > There are some stubborn BioCORBA issues that I am tracking down.
> > BioCORBA could branch in a week or just fix the bugs in a point release
> > because I honestly think it is a ORBit bug.
> > 
> 
> It might be nice if we could indeed do a sync'ed release, and
> cross-reference the other packages in the release notes. It's up to
> you how good you feel with a point release, but I don't see why it's
> bad.
> 
> 	Hilmar
> -- 
> -------------------------------------------------------------
> Hilmar Lapp                            email: lapp@gnf.org
> GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
> -------------------------------------------------------------
> 

-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<birney@ebi.ac.uk>. 
-----------------------------------------------------------------