On Thu, 2020-04-30 at 16:11 +0200, Dominique Leuenberger / DimStar wrote:
Review of individual patches for given software packages are the package maintainer's task. Package maintainers are likely to work with git or some other VC system on their package anyway, and if they are, they are served better by the VC than by the patch listing in the spec file.
Considering that the review team every now and then still declines patches that are obviously wrong: what does that tell us about the maintainers? They're human. They can make errors. Luckily, this is not the common case. But having the patches visible does help.
I appreciate that the review team does this, but I'd never have
expected that, and I believe it can by no means be a requirement for
the distro-level review (in the sense that a reviewer could be
reprimanded for having approved a package with a patch that others
found "obviously wrong" later).
I guess it all depends on the number and complexity of the patches. The
kernel is one obvious, extreme example. Other packages, where we
basically take upstream releases with just a few trivial patches, would
be the opposite extreme.
Regards,
Martin
--
Dr. Martin Wilck