fixedpoint cmp conundrum

[Robert Brewer]

I described a problem about list.index() and fixedpoint:
> > >>> import fixedpoint
> > >>> a = fixedpoint.FixedPoint(3, 2)
> > >>> bunch = [None, 12, a]
> 
> ...
> 
> > >>> bunch.index(a)
> > Traceback (most recent call last):
> >   File "<interactive input>", line 1, in ?
> >   File "C:\Python23\Lib\fixedpoint.py", line 316, in __cmp__
> >     xn, yn, p = _norm(self, other, FixedPoint=type(self))
> >   File "C:\Python23\Lib\fixedpoint.py", line 482, in _norm
> >     y = FixedPoint(y, x.p)
> >   File "C:\Python23\Lib\fixedpoint.py", line 255, in __init__
> >     raise TypeError("can't convert to FixedPoint: " + `value`)
> > TypeError: can't convert to FixedPoint: None
> >
> >
> > Mmrph! I could fix this by:
> > 
> > 1. Iterating through "bunch" by hand and trapping TypeError,
> > 2. Overriding or hacking the __cmp__ function in fixedpoint, or
> > 3. Overriding or hacking the _norm function in fixedpoint
> >
> > The question is: given that there are multiple places to "fix" this,
> > where should I do so?

and Tim himself replied:
> 4. Define an __eq__ method, which returns False if its arguments
>    are non-comparable.

...to which I can only say:

Thanks! :) Nailed it.


Robert Brewer
MIS
Amor Ministries
fumanchu at amor.org