> Right, I'll do some test with the hardcoded values you mentioned.
> Btw, I don't think -0.0 is worth it, that value is a typical
> arithmetic result and not something you typically get from input data.

I was suggesting -0.0 in the hope that it might make the patch a bit
simpler, but I was probably wrong: you still have to test the float's
sign to distinguish between the two values.

> Also, while interning in python is certainly possible, perhaps it
> would make sense to implement such in c.

I think it makes sense in the float implementation if it doesn't
significantly decrease performance. I suggest you post the benchmark
numbers (see http://hg.python.org/benchmarks/ ).

Adding a separate builtin/stdlib function would be overkill IMO.