[mercury-users] Loop optimisation and predicate specialisation and stuff.

[Peter Moulder]

Andrew Bromage <bromage at cs.mu.OZ.AU> writes:

> however there is a semantic "gotcha".
> 
> Suppose:
> 
> 	f(A) = 100/A.

Yes, but this is just a quality-of-implementation issue.
Implementations are allowed to be incomplete 
(*see (mercury_ref)Semantics::, and Fergus' (?) recent post describing 
the semantics of error/1).

For example, the current Mercury implementation doesn't terminate the
program with error for

	pred_that_fails,
	pred_that_calls_error

but does for

	pred_that_calls_error,
	pred_that_fails

whereas a more complete implementation wouldn't terminate for either.

(The following has been described as the most redundant phrase on
usenet, but: please correct me if I'm wrong.)

pjm.