On Thu, Oct 08, 2015 at 03:01:34PM +0000, Mattia Rizzolo wrote:
> If that fields differs again [...] it just gives up and mark the
> package as unreproducible
The last item (mark as unreproducible) does not seem right to me.
A package is said to be reproducible when you build it two times
under the "same" environment (build-depends) and the result is the same.
If we are unable to reproduce the environment to begin with, i.e. if
we don't even give the package the opportunity to show that it's
reproducible, then it is not the package's fault, and I see no reason
to mark it as unreproducible.
Reproducible-builds mailing list