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

Re: Please binNMU polybori against libm4ri-0.0.20140914




Jonathan Wiltshire wrote:

> This is a library transition. We stopped accepting library transitions
> some weeks ago. Do you have a Really Good Reason (tm) for your upload?

I have some; 
- libm4ri and libm4rie change their SONAME at every release. However,
  the interface to the library hasn't changed for a long time. Those
  libraries should move to a saner SONAME handling.
  Upstream is aware of that:
  https://bitbucket.org/malb/m4ri/issue/51/move-to-saner-soname
  https://bitbucket.org/malb/m4rie/issue/16/move-to-saner-soname
  but nothing has been done yet.
  So, so to speak, it isn't a real library transition. At least, I
  (wrongly) didn't consider it as one.

- Except for libm4rie, polybori is the only reverse build dependency of
  libm4ri-dev. It is used only in libpolybori-groebner-0.8-3, which
  hasn't any reverse dependencies. It rebuilds just fine with the new
  libm4ri. So it's pretty safe.
But is it enough?

If it is really too late, then I apologize, and kindly ask you to forget
my request. But if it is possible to make an exception and let m4ri(e)
migrate and binNMU polybori, I would be grateful.

Thanks for your work.

Best wishes,

Cédric

Attachment: signature.asc
Description: Digital signature


Reply to: