Feedback on Ubuntu-specific debian/rules for building sysvinit

Olivier Gayot olivier.gayot at canonical.com
Fri May 12 17:12:34 BST 2023


Hi Joost,

On 5/12/23 06:50, Joost van Baal-Ilić wrote:
> I do not oversee the complete picture here, but let me add this:
> 
> "Bug #904302: any use of dpkg's vendor-specific patch series feature is a bug
> for packages in the Debian archive and such feature will be forbidden in the
> Debian archive after the release of Debian Buster."
> 
> See https://bugs.debian.org/904302 ,
> https://lists.debian.org/msgid-search/875zx0dei6.fsf@err.no .
> 
> Some people felt very strongly that a build's result should be predictable, and
> should therefore be as independent as possible of the system the build is
> executed on.  Doing hidden / tricky things makes software more non free.  I
> believe that was the gist of the thinking there.  Anyway, a valid argument in
> my opinion.
> 
> Consequence here would be to keep all machinery on Ubuntu's side.

Thanks for pointing me to the relevant discussions. I think it's overall 
a sound approach.

If you would like to drop the Ubuntu-specific code from debian/rules, 
please go ahead and we will add back what's missing on Ubuntu's side.

In the near future, I plan to submit a set of changes that I think can 
improve debian/rules in sysvinit. Hopefully, this should make it overall 
better and should allow us on Ubuntu to minimize our delta.

Are pull-requests on salsa (against master) the preferred approach?

Thanks,
Olivier



More information about the Debian-init-diversity mailing list