[mercury-users] Type error in higher-order predicate and existential type

[Nicholas Nethercote]

On Tue, 6 Nov 2007, Peter Ross wrote:

>> Using a higher-order predicate, I get an type error and I don't
>> understand where it comes from. Here is a much simplified version of the
>> predicate I wrote :
>>
>> :- pred higher_order_pred(pred(T,T),T).
>> :- mode higher_order_pred(in(pred(in, out) is semidet),out) is semidet.
>> higher_order_pred(HOPred,Result) :- call(HOPred,5,Result).
>>
> For the line
>
> call(HOPred, 5, Result)
>
> the first argument to the higher-order pred is an int, but you said in
> the type declaration that it can be any T thus the error.
>
> :- pred higher_order_pred(pred(int,int),int).
>
> fixes the error.
>
> Another explanation is that T in the type declaration can be any T,
> but the body of the predicate restricts the set of allowable T's to be
> int only.

Here's what may be an easier way to think about it -- what would happen you 
passed in an HOPred with the type 'pred(float,float)'?

Nick
--------------------------------------------------------------------------
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
--------------------------------------------------------------------------