Python2 distutils question: how to best "autoconfig"...?

[Fredrik Lundh]

Alex Martelli wrote:
> The specific issue is "how many bits of significance
> does a Python float have".  (Or is there somewhere in
> the Python headers a #define for this that I missed...?)

why not just do things on the Python side?  if not else,
it's a sure way to avoid optimization head-aches:

def get_base_and_digits():
    def precise(x):
        return x + 1.0 - x - 1.0 == 0.0
    a = b = c = 1.0
    base = digits = 0
    while precise(a):
        a = a + a
    while not base:
        b = b + b
        base = a + b - a
    while precise(c):
        digits = digits + 1
        c = c * base
    return base, digits

>>> print get_base_and_digits()
(2.0, 53)

(if anyone runs Python on a box that returns anything
but (2.0, 53), let me know!)

</F>