On Fri, 2009-11-13 at 11:16 -0800, Per Bothner wrote:
> On 11/13/2009 10:54 AM, Thomas Bushnell BSG wrote:
> > So "passing" an arity check doesn't mean that the number of arguments
> > you pass is correct, as Dybvig's already noted.
> >
> > Also, "failing" an arity check doesn't mean that you will get any kind
> > of error or exception as a result.
>
> Arity is useful as *part of* an implemention's error-handling.
> I.e. first you check if there are too few or too many arguments,
> and *then* check the argument types.

Of course, many implementations have this thing under the hood called
"arity", and they check it for various purposes.  But "arity" is not
something the Scheme specification knows about, unless it is simply what
is defined by the defining lambda expression.  (And, if that's what it
means, then my two-value technique is fully sufficient.)

Thomas