Bug#850478: git-pbuilder deliberately deletes source.changes

[Guido Günther]

On Sat, Jan 07, 2017 at 09:30:57AM -0800, Russ Allbery wrote:
> Mattia Rizzolo <mattia at debian.org> writes:
> > On Fri, Jan 06, 2017 at 06:18:14PM -0800, Russ Allbery wrote:
> 
> >> Yeah, it does that because I didn't know how to do better (and this
> >> started life as my personal script for my workflow, and at the time all
> >> *_source.changes files were garbage).
> 
> > umh, $pkg_$version_source.changes? :)
> 
> Okay, how do I get $pkg and $version?  :)  git-pbuilder right now doesn't
> know any of that stuff.  I suppose I could try to parse debian/changelog
> using dpkg-parsechangelog or something....

gbp buildpackage could pass this in env vars too. In fact I already
looked how this could be done and wonder if we should do at least that
for stretch? If so I can fix this up in gbp and git-pbuilder until
pbuilder is fixed.

Slightly related: What would be nice if gbp would know which changes
file got created by git-pbuilder / pdebuild so it doesn't have to guess
which one to use for an upload:

    https://github.com/agx/git-buildpackage/blob/master/examples/gbp-posttag-push#L147

> But if pdebuild stops generating the bad *_source.changes file, then I can
> just delete all that code and everything becomes easier.
> 
> >> >     Now, I'm of the view that dpkg-source -b should be used instead,
> >> >     which is what sbuild uses to create the dsc. This also has the
> >> >     advantage of not generating .buildinfo files (no annoying
> >> >     debian/files lingering after the build, either). Then the only
> >> >     _source.changes generated by pbuilder would be if the user requested
> >> >     it, and therefore having it deleted by git-pbuilder would be wrong.
> >> 
> >> Yeah, this seems reasonable to me.  Definitely happy to change
> >> git-pbuilder once pdebuild is fixed to not produce the spurious and
> >> useless *_changes.file.
> 
> > Indeed, is this a suggestion for pdebuild?
> 
> Yeah, I think you might not have gotten the original message, since I
> think the X-Debugs-Cc may have been in the wrong spot?

Cheers,
 -- Guidox