[mercury-users] impure predicates

[Henk Vandecasteele]

Fergus Henderson wrote:
> 
> On 20-Oct-1998, Ralph Becket <rwab1 at cam.sri.com> wrote:
> > Henk Vandecasteele wrote on 20 Oct:
> > >
> > > So, should the mercury-system backtrack over impure
> > > predicates which have only input-arguments?
> 
> Yes, IMHO.  The automatic pruning which the Mercury
> system does for goals which have no output variables
> is really an optimization that is only valid for
> pure (or semipure) goals; it does not make sense to
> apply this to impure goals, which might have side effects.
> 
> The fact that we currently do so is just an oversight.
> I have attached a patch which fixes this.  I have just
> committed this patch, so it will be included in the
> next ROTD snapshot as well as the next major release.

Thanks a lot for the patch, I just wondered if this patch is applicable
on rotd-1998-09-25? Or should I catch the "unstable" rotd?

Regards,

Henk