On 31/07/07, Andy Lester <[EMAIL PROTECTED]> wrote: > > On Jul 31, 2007, at 1:32 PM, peter baylies wrote: > > > That may not be a bad idea, but I think there's a bug in that code -- > > take, for example, the case where x and y both equal approximately a > > million (or more). > > > > Maybe you wanted this instead: > > > > return (fabs(x - y) <= EPSILON) ? 1 : 0; > > The REAL question, I believe, is for comparisons to zero. Last ptc > and I talked about this, there were no places where the comparison in > question isn't against 0.0.
I was really wondering what a good general solution was. I remember discussing this with you on #parrot and you mentioned a name for a macro to handle the comparison, but I didn't understand what the implementation would be from the name (I did ask, but I think you'd gone offline). Since we're only comparing against zero, would something like: fabs(x) <= DBL_EPSILON ? 1: 0 do the job? We're still running into the problem of "what if x is float or double?". I have a horrible feeling I missed your point when we were discussing things on #parrot; could you explain how you would do it again? Thanks! Paul
