distutils replacement discussion
Kevin Dangoor
discussion going on on distutils-sig about the replacement for
distutils. At this point, I don't personally want to get involved in
that but I am certainly following the discussion because Paver will
ultimately support whatever mechanism they create.
Kevin
Toshio Kuratomi
if I could get you to participate :-)
I've mentioned paver to a few people and in one of the multitude of
emails that's been flying around. I really like paver's declarative/
imperative design choices. Wish you'd join in if there's a point
where it makes sense.
-Toshio
Kevin Dangoor
> Shoot, I just caught up on my distutils threads and hopped over to see
> if I could get you to participate :-)
>
> I've mentioned paver to a few people and in one of the multitude of
> emails that's been flying around. I really like paver's declarative/
> imperative design choices. Wish you'd join in if there's a point
> where it makes sense.
I saw that you mentioned Paver over there. Thanks for spreading the
word!
The trouble is that the level of infrastructure they're talking about
doesn't interest me. I have too many things on my plate to really take
an interest in the metadata format, how Linux distro X will do its
packaging, how PyPI will catalog everything, blah blah.
I actually think that the "new world" that they're working on will
make things easier for Paver. Part of the point of their exercise is
to standardize some formats and interfaces so that more tools can
bloom. I also anticipate that it's going to be a while before we see
some real movement.
On the plus side, Paver already has a good bootstrapping mechanism
(the minilib). By adding a small file to your distribution, you will
be able to use the latest tools even before most people have them
installed.
There are a lot of higher level problems (virtualenv+PoachEggs is
still giving me pains, haven't even gotten to pytoss) that will be a
bigger boon now than spending time on that thread.
Kevin