On 2 June 2017 at 16:27, Donald Stufft <firstname.lastname@example.org> wrote:
So my preference is that everything goes through the sdist step as I thinkAgreed. That's the ideal workflow. The only reason we don't do it nowis because... well, I'm not quite sure. I think it's to do with thingslike setuptools_scm not generating suitable "temporary versionnumbers" to allow us to work properly with installs that assume thatname/version uniquely identifies the code.
that is most likely to provide consistent builds everywhere both from a VCS
checkout and from a sdist that was released to PyPI.
I’m pretty sure the only reason we don’t do it now is because nobody has had the time to make it happen yet. The problems before weren’t from going via sdist, they were from trying to modify our copy tree implementation to filter out .tox, .git, etc. I don’t think we’ve ever tried going via sdist (other than there is an open PR for it, but it ended up stalling https://github.com/pypa/pip/pull/3722
). Essentially, volunteer time is finite :(