Re: Dynamic Allocation seems to crash gprolog (follow-up)

[Daniel Diaz]

Leonardo, there is a bug in your program that makes it create a lot of terms in 
the heap (compound
terms).
I have tried your program with SWI-Prolog, Yap and SICStus Prolog. In all 
systems an overflow
occurs.

Sorry to not enough time to help you to find the problem...

To makes the program under those systems I have defined:

% not all systems accept several declarations separated by comma

:- dynamic(parent/2).
:- dynamic(root/1).
:- dynamic(is_parent/2).
:- dynamic(is_root/1).

:- dynamic(a/1).
new_atom(X) :-
        (retract(a(I)) ; I = 0), !,
        I1 is I+1,
        asserta(a(I1)),
        name(I1,LCodes),
        atom_codes(IA,LCodes),
        atom_concat('tree_', IA, X).




for(I,I,J):-
        I =< J.

for(X,I,J) :-
        I < J,
        I1 is I+1,
        for(X,I1,J).


For yap, I had to rename path/1 in my_path/1 (path/1 is a built-in).

"Leonardo B. Lopes" wrote:
> 
> Dear Daniel and Friends,
> 
>         This is an update on the message I sent earlier. I upgraded to
> 1.2.8 and the problem still persisted... I appreciate any help.
> 
> Thanks!

-- 
Daniel Diaz   address@hidden
http://pauillac.inria.fr/~diaz
gprolog --version 2>&1 | sed -n -e 's/By //p'