Accepted coinmp 1.8.4+dfsg-2 (source) into unstable

[Rene Engelhard]

Hi,

Am 21.11.24 um 06:57 schrieb Pierre Gruet:
> Le 21/11/2024 à 00:18, Rene Engelhard a écrit :
>> Hi,
>>
>> Am 20.11.24 um 22:25 schrieb Rene Engelhard:
>>> Hi,
>>>
>>> Am 20.11.24 um 17:19 schrieb Debian FTP Masters:
>>>>  coinmp (1.8.4+dfsg-2) unstable; urgency=medium
>>>>  .
>>>>    * Team upload
>>>>    * Upload to unstable to start the library transition
>>>
>>> Just that it is questionable whether one needed one at all as other distros apparently still get a .so.1 out of 1.8.4.
>>>
>>> However they did that.... I wonder whether something is fishy here...
>>>
>>>
>>> That's why I didn't upload it (yet) to sid.. If you asked me I would have said that.
>>>
>
> If I understand correctly, you think better coordination would have been useful there. Sorry for not doing so, next time I will.
>
And that one needs to find out why other distros happily still get a so.1 out of this. I fear so.0 is somehow wrong.


And we become the only distro having this is not ideal(tm)


> Sorry, I am not sure to get the problem. Skipping autoreconf has been a current practice for coinor packages for many years, see coinor-cbc, coinor-cgl, coinor-osi, ... for instance. And this has not revealed to be problematic.
>
Hrrm.

>>
>> Maybe the person who did all uploads so far (and this package was introduced by me initially solely
>> for libreoffice, and has no other r-dep) should have been asked...
>
> I admit this. Still, we had the upload of 1.8.4+dfsg-1 by Andreas in experimental, 
which I noticed back then, but it just cleaned up packaging, does some questionable things IMHO, but otherwise OK. But that doesn't say anything about experimental->unstable
> no open bug about the packaging, 

What bug should that be? The changelog and that the SONAME goes backwards whereas other distros don't have that (easily lookable by looking at them) is enough warning signs, IMHO to not just upload..

 > I commit to monitor the transition and fix things if problems show up, which I don't expect.

I would expect so :)


Still, by pesonal opinion is that we should only upload this until we know why the SONAME changed and whether we really want it and this upload reverted.

Your and the other Science people opinion might differ, but...


Regards,


Rene