[mercury-users] number of args for bag.init exported to C

[Guillaume Yziquel]

Le Wednesday 08 Jun 2011 à 11:43:50 (+1000), Paul Bone a écrit :
> On Tue, Jun 07, 2011 at 04:38:49PM +0200, Guillaume Yziquel wrote:
>
> :- pred bag_of_ints_init(bag(int)::out) is det.
> :- pragma foreign_export("C", bag_of_ints_init(out), "no_props").
> 
> bag_of_ints_init(Bag) :-
>     bag.init(Bag).
> 
> Hope this helps.

Yes, it does. I'm now having two problems with the following code:

	:- type prop --->
		term_definition(term, term)	;
		term_typed(term, typ)		.

	:- pred typecheck(term::in, typ::out, bag(prop)::in) is semidet.
	:- pragma foreign_export("C", typecheck(in, out, in), "typecheck").
		typecheck(Term, Type, Props) :-
			promise_equivalent_solutions [Prop, RemainingProps]
			bag.member(Prop, Props, RemainingProps),
			promise_equivalent_solutions [Type]
			(	Prop = term_typed(Term, Type)
			;	typecheck(Term, Type, RemainingProps)
			).

Essentially, this looks up into the bag of props for a proposition that
refers to an explicit type annotation in my custom language.

So, first of all, I'd like to know if there is a more idiomatic way to
find an element in a bag on which to do unification. And if a bag is the
right data structure for that goal.

Secondly, I'm using promise_equivalent_solutions to be able to force the
determinism category of typecheck to semidet in order to export it to C.
I'm under the impression that I'm using this in a suboptimal manner (if
it's not outright wrong).

So what's a clean and idiomatic way to do something like the piece of
code above.

-- 
     Guillaume Yziquel
--------------------------------------------------------------------------
mercury-users mailing list
Post messages to:       mercury-users at csse.unimelb.edu.au
Administrative Queries: owner-mercury-users at csse.unimelb.edu.au
Subscriptions:          mercury-users-request at csse.unimelb.edu.au
--------------------------------------------------------------------------