On Wed, Sep 14, 2016 at 12:56:52AM +0200, Matthias Klose wrote:
> sorry, I don't share you view that this might be funny. You are known to make
> issues RC issues like the missing build-indep/build-arch targets. I think the
> RC severity of such reports is at least questionable.
No, that's not true and you are a little bit confused.
If you really refer to packages missing build-indep/build-arch targets,
that was a decision from dpkg author, not mine.
> This is very much different than adopting to be compatible with a
> new upstream software.
The issue about missing build-indep/build-arch targets is exactly
adopting a package to be compatible with a new dpkg-dev. This was policy
since a long time, but dpkg-dev had a workaround for buggy packages not
to fail. The workaround was removed and as a result lots of packages
started to FTBFS.
Moreover, most of the bugs were submitted by Niels, not by me. If you
have seen my name associated with such kind of bugs, maybe it's
because I submitted patches for quite a bunch of them.
OTOH, if you refer to packages which fail with "dpkg-buildpackage -A",
I did not make the issue RC on my own either, there is a decision from
the Release Managers about it in Bug #830997, so this is a release goal
for stretch now.
In this case, it is also adopting packages to be compatible with the
software which has been running in the buildds for a lot of time
("dpkg-buildpackage -B" in the traditional buildds and "dpkg-buildpackage -A"
in the "Arch: all" autobuilder already for a year now).
Maybe this being RC is questionable in Ubuntu where I heard somewhere
that to create arch-independent packages you make a normal "dpkg-buildpackage"
and then discard the arch-dependent packages, but that's not how
our "Arch: all" autobuilder works.
If you really think this is questionable, please tell it to the
Release Managers in Bug #830997, not to me.
> Your behavior [...]
Sorry, I am completely lost here. You started by saying I was doing
issues RC on my own. You cited a wrong example, and I clarified the
other example that maybe you had in mind that was also wrong.
If you want to rant me about what I do, please start again and stick
to the facts.
BTW: I said "funny" because I believe you forgot the </joke> tag.
If your point was that we can't make package maintainers responsible
in *exclusive* for the breakage of their packages by other packages
entering testing, I fully agree.
(I really think this was your point in the idea that you expressed
naively or sarcastically of making a test rebuild before the package
enters testing, as I am quite sure that you know perfectly that such
idea would not work).
But if your point is that package maintainers should not be bothered
at *all* with this, sorry, I disagree. It's stretch what we will release
as the next stable, and we want packages in stretch to be buildable in
stretch, so I would expect at least *some* interest on the side of the
maintainers themselves about their packages meeting this release
criteria, we can't consider this to be a responsability of the Release
Managers alone.
Thanks.
