Re: Multiple return values

(Mars Saxman) marssaxman@sprynet.com writes:
> I think it is an artefact of the origin of the "function" concept. The
> idea of a function in maths is that you give it some parameters and it
> performs some calculation on them. The result of the calculation is
> the return value. It doesn't make sense to have, for example, an
> arctangent return more than one value.


Pieter Schoenmakers <tiggr@es.ele.tue.nl> wrote:
>What about a function R x R -> R x R?


Most mathematicians would contend that this returns a single value,
and that value is an ordered pair of real numbers. Of course,
multiple arguments are just a convenience anyway; one could do just
fine talking about f((a,b,c)) instead of f(a,b,c). The situation is
really symmetric wrt inputs and outputs (especially for relations of
which functions are just a special case); it is the typical
mathematical and PL notation that makes them look different.


IMO, the ideal syntax is:


f(a, b, c) -> (c, d)


though many people prefer assignment-type syntax. The advantage is
that:


f(a) -> a


makes sense read left to right, while:


a := f(a)


looks suspiciously recursive. Using '=' for assignment qualifies as
"not even wrong", IMNSHO :-)


--
Tim Hollebeek
Electron Psychologist
Princeton University
email: tim@wfn-shop.princeton.edu
--

Return to the comp.compilers page.
Search the comp.compilers archives again.