Float compression [Re: floating point math results question]

[Jason Orendorff]

PEP 239 - Adding a Rational Type to Python
  http://python.sourceforge.net/peps/pep-0239.html

The PEP itself leaves too many things undiscussed;
but I think the idea is:

  type(150/181) is rational  #not int or float
  str(150/181) == "150/181"


Johnathan Hogg wrote:
> A more interesting problem is an efficient algorithm for refactoring an
> arbitrary rational number into a sequence of the form:
> 
>     ... + 100.k[2] + 10.k[1] + k[0] + k[-1]/10 + k[-2]/100 + ...
> 
> where 0 <= k[i] < 10, which you would want for printing out the
> representation as a normal floating point number. [Of course noting that
> many rationals are non-terminating as floats.]

For the ordinary case where the desired precision is small,
you can just:  str(float(num)/float(denom))

## Jason Orendorff    http://www.jorendorff.com/