[mercury-users] Pred defns

[Thomas Charles CONWAY]

Peter Phelan, you write:
	[message rearranged]
> I'm confused by pred definitions in Mercury.
> 
> It's something to do with input/output, but this is also
> difficult to understand.

What seems to be confusing you is DCG syntax. DCGs are a
layer of syntactic sugar which was developed to help with
writing parsers. The fact that they are useful for I/O and
other situations where you want to thread an argument is
coincidental.

> The following is snipped from calculator.m in samples.
> 
> :- pred fullexpr(expr::out, list(char)::in, list(char)::out)
> is semidet.
> fullexpr(X) -->
>  expr(X),
>  ['\n'].
> 
> How come the declarative definition takes just one argument,
> whilst there are 3 in the type declaration?
> 

Notice that the form of the definition is
	Head --> Body
not
	Head :- Body.

The use of --> inicates that this definition is a DCG definition.
It is expanded by the compiler as I explained breifly in a previous
message, and is fully documented in the language reference manual.

The definition for fullexpr HAS three arguments - two of them
(the last two) are made implicit by the use of DCG notation.

If we do the DCG expantion by hand on your code, we get:

fullexpr(X, DCG0, DCG2) :-
	expr(X, DCG0, DCG1),
	DCG1 = ['\n'|DCG2].

Thomas
-- 
Thomas Conway    || conway at cs.mu.oz.au
AD DEUM ET VINUM || Nail here [] for new monitor.