[m-dev.] is this a solver bug?

[Ralph Becket]

Peter Schachte, Thursday,  7 July 2005:
> Hi all,
> 
> I'm enclosing a test file that gives a mode error when compiling, but
> I don't think there's a mode error, or at least I don't think the
> error is what is claimed.  It involves impurity, existential types and
> automatically-generated solver type cast functions.  The code in
> question is:
> 
> 	:- pragma promise_pure(delayed/2).
> 	delayed(F, V0) = V :-
> 		impure Ref = 'representation of any dvar/1'(V0),
> 		impure new_reference(unbound([func1(F,Ref)]), DV0),
> 		impure V = 'representation to any dvar/1'(DV0).

I would say the problem was that func1 here should be 'new func1'
because this is a construction of a value with an existentially
quantified type.

> 
> The error message I get is:
> 
> 	bug.m:047: In clause for `delayed(in, in) = out':
> 	bug.m:047:  in argument 1 of call to function `bug.representation to
> 	any dvar/1/1':
> 	bug.m:047:   mode error: variable `DV0' has instantiatedness `free',
> 	bug.m:047:   expected instantiatedness was `ground'.

This error message is not very helpful, agreed...

-- Ralph
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------