Fergus' hack (was Re: [mercury-users] polymorphism)

[Fergus Henderson]

Bart Demoen wrote:
> 
> :- type ab ---> (a ; b).
> 
> :- inst a = bound(a).
> :- inst b = bound(b).
> 
> :- pred f(ab,int,int).
> :- mode f(in(a),in,out).
> :- mode f(in(b),out,in).
>
> f(Z,X,Y) :-
>         (Z = a ->
>             Y is  1
>         ;   X is  7
>         ).
...
> fergushack.m:032: In clause for `f(in(fergushack:a), in, out)':
> fergushack.m:032:   mode mismatch in if-then-else.
> fergushack.m:032:   `Y' :: unique(1), free.

The mode system is not sophisticated enough to figure out that
the condition of the if-then-else cannot fail in this mode.
The mode system keeps track of which goals cannot succeed,
but it does not analyse which goals cannot fail.
(The determinism system does that -- but determinism analysis
comes *after* mode analysis.)

If you write the code slightly differently, i.e.

	f(Z,X,Y) :-
	        ( Z = a, Y is  1
	        ; Z = b, X is  7
	        ).

then it will work fine.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>   |  "I have always known that the pursuit
WWW: <http://www.cs.mu.oz.au/~fjh>   |  of excellence is a lethal habit"
PGP: finger fjh at 128.250.37.3         |     -- the last words of T. S. Garp.