[R6RS] Draft of arithmetic SRFI

[William D Clinger]

Mike Sperber wrote:
> Assume that FX+ takes an additional argument---a binary procedure that
> gets called on overflow:

[Scheme code omitted]

> Why would this be worse than your bummed code?

It wouldn't be.  When I said "error-signalling semantics", I was
talking about a semantics that signals an error in the sense of
R5RS or raises an exception in the sense of SRFI-34.  Calling a
third argument would work just fine.

Will