combination function in python

[Alex Martelli]

mensanator at aol.com <mensanator at aol.com> wrote:

> > You could take it up with the gmpy author and
> > induce him to get gmpy included in the standard distro if you are so
> > inclined.
> 
> Alex Martelli knows more about that subject than I and
> it would be pointless for me to bug him about it.

gmpy is LGPL (because so is the underlying GMP) and therefore very
unlikely to be included in the Python distribution, whose core
committers, collectively, choose laxer, looser licensing labels.

Of course, there being no "secret algorithm" for the computation of
binary coefficients in GMP, reimplementing comb "from scratch"
(_without_ peeking at GMP's sources, so it can get a normal Python
license) as a Python extension with similar performance to gmpy.comb is
a SMOP (for which I'm not the best candidate, since I may in the past
have peeked at GMP's sources, and although I don't consciously recall
anything about the binary-coefficients computation [except that there
was nothing particularly exciting about it:-)] we surely don't want to
risk the FSF suing the PSF, now do we...).


Alex