really stop "really" commits (really!)

[Stanislav Ochotnicky]

Quoting Lukas Zapletal (2013-12-17 11:41:29)
> On Mon, Dec 16, 2013 at 03:10:08AM -0700, T.C. Hollingsworth wrote:
> > > I do commit locally
> > > although I probably don't want push the snapshot sources, because I update
> > > them later, when time comes.
> 
> +1
> 
> > This should happen rarely enough that having to use `git commit
> > --no-verify` to bypass it wouldn't be too much trouble?
> 
> I am all for these kinds of checks and I appreciate your effort. These
> are all good ideas, but please let's do not do this on the git level.
> This is not appropriate place when we already have a nice layer for this
> kinds of tests: fedpkg.

I am sure I won't be the only one who doesn't use fedpkg for git manipulation
(commit, push, pull etc). So higher level will not work because by the time I
get to "fedpkg build" it's already too late.

> Every time a maintainer want to build a package using fedpkg, we can add
> this kind of hooks and verify what is necessary in the similar form you
> recommend (with the option to skip). Let's build a new command fedpkg
> check to do explicit checks as well which can help when fixing those
> mistakes.

Just out of curiosity: How many times have you run "fedpkg lint"? Zero? Once to
see what it does? The whole point of the hook was so that you wouldn't forget to
run appropriate commands. Adding another will make the problem worse, not better

> We do not want to get to the state when packages being committed are
> automatically being built in koji (using git hooks). Because what you
> recommend here effectively leads to this "happy" ending - you might
> unintentionally seed this kind of approach. I was already there and
> believe me - this is not what we want to do. :)

Hold your horses

-- 
Stanislav Ochotnicky <sochotnicky at redhat.com>
Software Engineer - Developer Experience

PGP: 7B087241
Red Hat Inc.                               http://cz.redhat.com