[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#976386: transition: octave



Le lundi 21 décembre 2020 à 10:35 +0100, Sebastian Ramacher a écrit :
> On 2020-12-20 19:30:13, Sébastien Villemot wrote:
> > Le dimanche 20 décembre 2020 à 19:24 +0100, Paul Gevers a écrit :
> > > On 20-12-2020 19:15, Sébastien Villemot wrote:
> > > > I’m not familiar with britney’s output, but my impression is that the
> > > > problem is related to src:plplot. The version in testing currently
> > > > builds a binary package named octave-plplot, which has been dropped in
> > > > the version currently in unstable (because it was not possible to build
> > > > it against octave 6).
> > > > 
> > > > Maybe britney needs to be given some hint to force the transition.
> > > 
> > > plplot is part of the gnat-10 transition, which means the transitions
> > > now got entangled. I'm not following close enough to know what that
> > > means now.
> > 
> > My understanding is that the gnat-10 transition is not a blocking
> > factor, since gnat-10 is already in testing (as part of src:gcc-10).
> > But maybe I got it wrong.
> 
> The blocking issues is indeed plplot. The version in unstable dropped
> the octave package, but the version in testing still has it. So it's
> part of the octave transition. libplplotada3-dev in unstable depends on
> gnat and gnat-10, so is unsable to migrate because gnat in testing still
> depends on gnat-9 but gnat-9 and gnat-10 are conflicting with each
> other.
> 
> To unblock the situation, I currently see the following options:
> * Upload plplot 5.15.0+dfsg-16 together with the change to drop
>   octave-plplot to testing-pu [1]. Then plplot would no longer be
> involved
>   in the octave transition and octave would be able to migrate.
> * Remove plplot and reverse dependencies from testing.

> [1] 
> https://www.debian.org/doc/manuals/developers-reference/pkgs.en.html#t-p-u

Thanks for your analysis. I’m putting Rafael Laboissière (plplot
maintainer and member of the Debian Octave Group) in CC, so that he can
decide what is the best course of action. My impression is that an
upload to testing-proposed-updates would be the least disrupting
solution.

-- 
⢀⣴⠾⠻⢶⣦⠀  Sébastien Villemot
⣾⠁⢠⠒⠀⣿⡁  Debian Developer
⢿⡄⠘⠷⠚⠋⠀  https://sebastien.villemot.name
⠈⠳⣄⠀⠀⠀⠀  https://www.debian.org

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: