[613] in athena10
Re: debathena-pidgin-wrapper pre-hardy
daemon@ATHENA.MIT.EDU (Greg Price)
Wed Nov 12 01:34:13 2008
Date: Wed, 12 Nov 2008 01:33:24 -0500
From: Greg Price <price@MIT.EDU>
To: Tim Abbott <tabbott@mit.edu>
Cc: athena10@mit.edu
Message-ID: <20081112063324.GO20760@vinegar-pot.mit.edu>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <alpine.DEB.2.00.0811072234580.28546@vinegar-pot.mit.edu>
Yeah, I think we want a minimal-work solution; people running dapper,
feisty, or etch can just do without having a Debathena wrapper for
their pidgin unless someone emerges who cares about it.
I think that means for debathena-workstation to depend on
debathena-pidgin-wrapper only on the releases that have a 'pidgin'
package.
I don't know of a way to express this in an equivs package, so the way
I'd implement this is to turn debathena-workstation into a regular
source package with a trivial CDBS rules file, and then add a substvar
and a few lines in rules to depend on debathena-pidgin-wrapper
conditionally.
I'm not really enamored of equivs packages in the first place, so this
change wouldn't bother me.
Then AFAIK it should be possible to install debathena-cluster-software
on mega-man, which means that anything that goes wrong in it will
swiftly come to affect me personally. =)
Greg
On Fri, Nov 07, 2008 at 10:35:15PM -0500, Tim Abbott wrote:
> On Fri, 7 Nov 2008, Greg Price wrote:
>
> >On Fri, Nov 07, 2008 at 09:27:45PM -0500, Tim Abbott wrote:
> >>One option is to make debathena-pidgin-wrapper actually act as
> >>debathena-pidgin-or-gaim-wrapper, and have it declare all the Debathena
> >>dependencies on either pidgin or gaim, depending on the release (removing
> >>the direct dependencies on them from debathena-cluster-software).
> >
> >I think this is a little more complicated. debathena-pidgin-wrapper
> >contains pidgin.debathena, which seems to know about various pathnames
> >that look like the rename would have affected them. So we'd have to
> >write and test a gaim/pidgin agnostic version.
>
> Yeah, I wasn't meaning to suggest that there wasn't work involved in
> making debathena-pidgin-wrapper act as debathena-pidgin-or-gaim-wrapper.
>
> -Tim Abbott
