Re: Best place for .3 MPI man pages
On Thu, 2007-09-13 at 08:43 -0400, Adam C Powell IV wrote:
> Thank you Fabrice for the reply and sorry about the delay in following
> up.
>
> On Sat, 2007-09-08 at 13:22 +0200, Fabrice Lorrain wrote:
> > Adam C Powell IV a écrit :
> > > Greetings,
> > >
> >
> > Hello Adam
> >
> > > For years now, there has been a conflict between mpi-doc and lam4-dev
> > > because they both provide .3 MPI API man pages, such as
> > > MPI_Comm_set_name.3.gz .
> >
> > Playing with apt-file and grep it seems that MPI_Comm_set_name.3.gz is
> > the only file causing problem :
> >
> > $ apt-file show bla > /tmp/bla with bla={mpi-doc,lam4-dev}
> > $ for i in `grep man /tmp/mpi-doc | cut -d'/' -f 5`; do echo -n "$i : ";
> > grep -c $i /tmp/lam4-dev_man;done | grep ": 1"
>
> Thank you for looking into that! It seemed to me that there might be a
> bunch, but you're right, that's the only one.
>
> mpi-doc would be incomplete without it, so I don't want to leave it out.
> That leaves two options: a diversion, which I don't know how to do and
> seems a bit extreme for this situation, or alternatives, which would
> require cooperation from the lam maintainer.
>
> I think alternatives is the sensible way to do this, so unless other
> ideas come from this list first, I'll take care of that in my next
> upload, eliminate the Conflict, and file a bug against lam4-dev.
Just uploaded mpich 1.2.7-4 which implemented this, and filed bug 445291
against lam4-dev. When the maintainer adds an alternatives symlink,
I'll change mpi-doc to conflict only with the old lam4-dev.
Thanks,
-Adam
--
GPG fingerprint: D54D 1AEE B11C CE9B A02B C5DD 526F 01E8 564E E4B6
Engineering consulting using open source tools
http://www.opennovation.com/
Reply to: