Builtin Float Epsilon? (was: Re: Does python suck or I am just stupid? )

[Gerrit Holl]

djw schreef op zaterdag 22 februari om 19:43:06 +0000:
> >
> >Please see 
> >http://www.python.org/cgi-bin/faqw.py?req=show&file=faq04.098.htp
> >
> >(Then change the u==x comparison from a strict equality check, as
> >prescribed therein.)
> >
> >-Peter
> 
> I've always wondered, given this fact, why is there an equivalence 
> operator for floats? (or, maybe the == operator should do the epsilon 
> range check for you somehow). It seems to be strange to have an valid 
> operator for a type that you should never in practice use. Maybe you 
> should be able to write something like:
> 
> sys.floatepsilon = 0.000000001 # or maybe some nice default value
> 
> ...
> u = 1.0000000000001
> v = 1.0
> 
> if u == v: # uses the system float epsilon test instead of "true" test
> 	...
> 
> Seems like this would keep more people from running into this 'problem'.

I think this is enough:

  9 >>> class Myfloat(float):
  9 ...  epsilon = 0.001
  9 ...  def __cmp__(self, other):
  9 ...   if abs(self-other)<self.epsilon: return 0
  9 ...   else: return float.__cmp__(self, other)
  9 ...
 10 >>> Myfloat(10) == Myfloat(10)
True
 11 >>> Myfloat(10) == Myfloat(11)
False
 12 >>> Myfloat(10) == Myfloat(10.01)
False
 13 >>> Myfloat(10) == Myfloat(10.0001)
True

yours,
Gerrit.

-- 
Asperger Syndroom - een persoonlijke benadering:
	http://people.nl.linux.org/~gerrit/
Het zijn tijden om je zelf met politiek te bemoeien:
	http://www.sp.nl/