[Debian-med-packaging] Finishing seqan
Michael Crusoe
michael.crusoe at gmail.com
Wed Feb 17 07:46:22 UTC 2016
Agreed that consistent names are good.
How about the following?
source package "seqan1" produces binary package "libseqan1-dev" (version
1.4.2) which provides and conflicts with "seqan-dev" for backwards
compatibility. Existing reverse build-dependencies are updated to use the
new name on a rolling basis.
source package "seqan2" produces binary package "libseqan2-dev",
"seqan-apps", "libseqan2-docs" (version 2.1.x)
As for "mason2" that will be confusing given the current unrelated package
"mason".
Mie, 17 feb. 2016, 08:34, Andreas Tille <andreas at an3as.eu> a scris:
> On Wed, Feb 17, 2016 at 07:28:39AM +0000, Michael Crusoe wrote:
> > Yeah, I thought we had agreed to keeping the existing source package as
> > 'seqan' and make a new tree for version 2.x as 'seqan2' so we didn't have
> > to change any reverse build dependencies.
>
> I'm not against this agreement for the binaries but in this case we
> should also keep the name for the source package to keep source and
> binary names the same.
>
> On the other hand the number of dependencies is limited and a fresh
> upload of these should be easy. I do not mind what way you prefer but
> the way should be consistent in itself.
>
> Kind regards
>
> Andreas.
>
> > Mie, 17 feb. 2016, 08:23, Andreas Tille <andreas at an3as.eu> a scris:
> >
> > > On Tue, Feb 16, 2016 at 01:40:43PM -0800, Kevin Murray wrote:
> > > > - seqan1: The previous 1.4.2 version of seqan. Builds:
> > > > - seqan-dev: the 1.4.2 development headers only
> > >
> > > Wait a moment:
> > >
> > > seqan*1* builds seqan**-dev
> > >
> > > > - seqan: The current (2.0.1 in git, soon to be 2.1.0) version of
> seqan.
> > > Builds:
> > > > - libseqan2-dev: the 2.x headers
> > >
> > > seqan** builds libseqan*2*-dev
> > >
> > > ? This does not sound sensible to me.
> > >
> > > > - seqan-apps: misc applications not listed below
> > > > - mason2: the Mason simulator
> > >
> > > ?? Again mason*2* - but we do not have any mason**
> > >
> > > > - (at a later date) seqan-doc: the sphinx API documentation.
> > > > - (at a later date) seqan-examples: the tutorials, cookbook and
> > > example/demo
> > > > code
> > >
> > > Please get the numbering straigt. I'd suggest to add '1' to any
> package
> > > that comes from source seqan1.
> > >
> > > Kind regards
> > >
> > > Andreas.
> > >
> > > --
> > > http://fam-tille.de
> > >
> > > _______________________________________________
> > > Debian-med-packaging mailing list
> > > Debian-med-packaging at lists.alioth.debian.org
> > >
> > >
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging
> > >
> > --
> > Michael R. Crusoe CWL Community Engineer crusoe at ucdavis.edu
> > <mcrusoe at msu.edu>
> > Common Workflow Language project University of California, Davis
> > https://impactstory.org/MichaelRCrusoe http://twitter.com/biocrusoe
>
> --
> http://fam-tille.de
>
> _______________________________________________
> Debian-med-packaging mailing list
> Debian-med-packaging at lists.alioth.debian.org
>
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging
>
--
Michael R. Crusoe CWL Community Engineer crusoe at ucdavis.edu
<mcrusoe at msu.edu>
Common Workflow Language project University of California, Davis
https://impactstory.org/MichaelRCrusoe http://twitter.com/biocrusoe
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/debian-med-packaging/attachments/20160217/1e3acb94/attachment.html>
More information about the Debian-med-packaging
mailing list