🚀 go-pugleaf

In article <1988Nov22.191600.26206@gpu.utcs.toronto.edu> sarathy@gpu.utcs.utoronto.ca (Rajiv Sarathy) writes:
>The example on page 99 of the edition I have of C & M says that
>?- display(a+b*c*c).
>	should return
>+(a,*(*(b,c),c))
>yes

I don't know what you mean by "return", but that is indeed what should come
out on your terminal.

>	However, the implementation of C-Prolog I'm using (on a SUN 3/280, if
>it matters) merely returns X.

Transcript, please!  (Easy way to do that: use the Unix 'script' command.
Which is why C Prolog doesn't implement log/0 and nolog/0.)  Last time I
used C Prolog, display/1 worked fine.

>Does someone have the definition of the "display" predicate?

If you have a copy of C Prolog, you have the source code, so YOU have
the definition.  If you have a copy of the DEC-10 Prolog library, look
in WRITE.PL.  Otherwise:

	display(Term) :-
		telling(CurrentOutput),
		tell(user),
		display_1(Term),
		tell(CurrentOutput).

	display_1(Term) :-		% for a COMPOUND term,
		nonvar(Term),
		functor(Term, F, N),
		N > 0,
		!,
		write(F), put("("),	% <fn>(<arg 1>,...,<arg N>)
		display_1(1, N, Term).
	display_1(Term) :-		% for a SIMPLE term,
		write(Term).		% (variable, number, or atom)

	display_1(I, N, Term) :-
		arg(I, Term, Arg),
		display_1(Arg),
		(   I < N ->
		    put(","),
		    J is I+1,
		    display_1(J, N, Term)
		;   put(")")
		).

There are some fine points that this doesn't get exactly right, but
it's pretty close.  (One of the fine points is that write/1 may round
floating point numbers off; display/1 should write all the digits that
make sense.)