"Loren P Meissner" <[log in to unmask]> wrote:
...
> Keep the old ATAN2 and create a new one with a different name?
Might be worth a try if you think that's more likely to pass the
committee than changing the wording that describes the existing
one. Frankly I see no real cause to keep the old one. Examples
of code supposedly dependent on the old behavior seem more
than a little contrived. The recommended new behavior would
be immediately useful to many.
In fact, I kind of like the idea of overloading ATAN so that if
two arguments are passed the functionality of ATAN2 is put
into play automatically (with the improved branch cuts, of course).
--
J. Giles
|