[mercury-users] type error with existentially typed data structure

[Peter Ross]

On Thu, Sep 06, 2001 at 05:06:39PM +0200, Peter Ross wrote:
> With the attached file I get the following type error, and I am a bit
> lost as to why it occurs.
> 
> petri.m:125: In clause for predicate `petri:bind_env/4':
> petri.m:125:   in unification of variable `Env'
> petri.m:125:   and term `BindEnv(Token)':
> petri.m:125:   type error in argument(s) of higher-order term (with arity 1).
> petri.m:125:   Functor (BindEnv) has type `((func T) = int)',
> petri.m:125:   expected type was `((func V_9) = int)';
> petri.m:125:   argument 1 (Token) has type `T',
> petri.m:125:   expected type was `V_9'.
> For more information, try recompiling with `-E'.

Problem found.  Don't use the automatically generated accessor functions
with existentially quantified data structures.

> 
> :- pred bind_env(arc_place_transition::in, environment::out,
> 		petri::in, petri::out) is nondet.
> 
> bind_env(ArcPlaceTransition, Env, Petri0, Petri) :-
> 	Place = get_place(Petri0, ArcPlaceTransition ^ from),
> 	BindEnv = ArcPlaceTransition ^ bind,
> 	set__member(Token, Place ^ tokens),
> 	Env = BindEnv(Token),
> 	Petri = Petri0.
> 

If you rewrite this as

bind_env(ArcPlaceTransition, Env, Petri0, Petri) :-
    ArcPlaceTransition = arc_place_transition(BindEnv, From),
    Place = get_place(Petri0, ArcPlaceTransition ^ from),
    set__member(Token, Place ^ tokens),
    Env = BindEnv(Token),
    Petri = Petri0.

You get the more usual error message of the higher order term having the
wrong inst.

Pete
--------------------------------------------------------------------------
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------