[Numpy-discussion] parallel compilation of numpy

David Cournapeau cournape@gmail....
Thu Feb 19 01:47:00 CST 2009


On Thu, Feb 19, 2009 at 4:26 PM, Andrew Straw <strawman@astraw.com> wrote:

> Maybe if you need a level of backward compatibility, (and really, to
> gain a decent audience for this idea, I think you do need some level of
> backward compatibility) the new tool could emit setup.py files for
> consumption by distutils as a fallback plan.

When I say I don't care about backward compatibility, it was in the
context of numpy. If I wanted to do a better tool for python, I would
have done things differently.

> Furthermore, if you're not
> opposed to dropping in your own distutils monkeypatches, like lots of
> other packages, you probably could do anything you wanted. For example,
> bypassing the build_ext command and injecting the built products into
> the distutils install command.

you've just described the scons command in numpy.distutils :)

> This reminds of Linus' criticism of svn: its goal is to be a better cvs.
> Said dripping with incredulity due to his perception of the fatal flaws
> of CVS. Well, I think (parts of) setuptools are better than distutils,
> but by being distutils+, it will always share the same flawed genetic
> material...

We should have a Linus Torvald : very few people are capable of
building a new tool which blows away the whole field, in two weeks :)
Also, svn, for all its flaws, was at least based on concepts which
were widely shared among the people working on the problem, whereas
distutils isn't (building things from a DAG is known for decades).

David


More information about the Numpy-discussion mailing list