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

Re: OTB



Hello Bas,

On Tue, Dec 29, 2015 at 10:13 PM, Sebastiaan Couwenberg <sebastic@xs4all.nl> wrote:
On 29-12-15 21:42, Rashad Kanavath wrote:
> Thanks a lot for sponsoring OTB package!. I had comment on bug tracker [1]
> which says splitting of shared libraries is not good for OTB. In OTB the
> version of all .so from source package changes altogether and it should be
> put in a single package.
>
> Should I go back to single libotb  and libotb-dev ?

I don't recommend switching back to bundling libraries.
 
Thanks. 

> I thought it might be easier for other projects to have separated packages.
> But comments from [1]  says that each minor release of package must result
> in all libotb* packages to pass through the NEW queue.

Whether a SONAME bump changes the package name for one or many packages
doesn't make a difference with respect to the NEW queue. Just one
renamed package is enough to have to pass the NEW queue again.

> Any thoughts ?
>
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=809312#15

ITK & VTK are not good examples of well managed packages in my opinion,
so I wouldn't look at them for inspiration.

Okay. The comments were saying it causing a burden on co-maintainers and keeping them away. That is why I asked. if the packaging need to be redone. I must know. I did seperate because it can be helpful in others. I also listed that this method was using in qgis for example. which I think has a similar situation.

I was thinking if the SONAME  have simply the major version instead of major.minor that might be a good.  I can check with this with upstream if that is feasible. 

The NEW queue is not a bad thing that one should optimize its packaging
for to avoid.

Okay. I don't know much about NEW queue and it whole process. To put it simple, the packages are checked manually to keep debian clean legally and otherwise. 

Kind Regards,

Bas

--
 GPG Key ID: 4096R/6750F10AE88D4AF1
Fingerprint: 8182 DE41 7056 408D 6146  50D1 6750 F10A E88D 4AF1




--
Regards,
   Rashad

Reply to: