Thank you. That was exactly the information I needed.
I already knew that tail-recursive procedures usually
out-performed non-tail-recursive procedures because
linear iterative processes were usually more space-
and time-efficient than linear recursive processes,
but I had just wanted to know if there was a way to
write Oz procedures that computed linear recursive
processes similarly to the corresponding Scheme
procedures in SICP. Your response answered my
question.
Cheers,
Ben
--- Raphael Collet <[EMAIL PROTECTED]> wrote:
> Benjamin L. Russell wrote:
> > Thank you; that makes sense. So, if I understand
> > correctly, essentially, it's syntactic sugar for
> the
> > equivalent tail-recursive procedure.
> >
> > But that brings up the question of whether I can
> make
> > a procedure explicitly non-tail-recursive. E.g.,
> if
> > the following piece of code
> >
> >>Z = X|{SMerge Xr Ys}
> >
> > is always automatically converted to the following
> > piece of tail-recursive code
> >
> >>local Temp in
> >> Z = X|Temp
> >> {SMerge Xr Ys Temp}
> >>end
>
> This conversion is perfectly correct in Oz. We
> chose this one instead
> of the non-tail-recursive one (below), because tail
> recursion performs
> better in most cases (in terms of time and space
> performance).
>
> local Temp in
> {SMerge Xr Ys Temp}
> Z = X|Temp
> end
>
> > then how can I write the code so that it runs in a
> > non-tail-recursive manner?
>
> Don't nest expressions inside lists or records, but
> evaluate them
> explicitly before making the list or record. For
> instance, the
> expression X|{SMerge Xr Ys} can be written
>
> local Zr={SMerge Xr Ys} in X|Zr end
>
> raph
>
>
_________________________________________________________________________________
> mozart-users mailing list
> [email protected]
>
http://www.mozart-oz.org/mailman/listinfo/mozart-users
>
_________________________________________________________________________________
mozart-users mailing list
[email protected]
http://www.mozart-oz.org/mailman/listinfo/mozart-users
