Hello,
***MY SYSTEM***
I'm working on a i686 AMD Sempron 2400+ under Archlinux, I'm using gcc 4.3.2
and gprolog 1-3.0 (the lastest stable version available right now).
I would like to interface C and Prolog in order to make a solver for some
puzzle (written in C) and learn about constraint programming.
***MY ISSUE***
So I've carefully read the manual's last chapter and I've tried out the
examples in there before going to build up mine, but even some of them don't
work properly with gcc 4.3.2!
****EXAMPLE****
The last one in gprolog's manual ( section 9.5.1, you can find the source
new_main_c.c as far as I know located in ExamplesC folder).
Let's look function Main_Wrapper in there:
***CODE****
static int
Main_Wrapper(int argc, char *argv[])
{
int func;
WamWord arg[10];
char str[100];
char *sol[100];
int i, nb_sol = 0;
Bool res;
Start_Prolog(argc, argv);
func = Find_Atom("anc");
for (;;)
{
printf("\nEnter a name (or 'end' to finish): ");
scanf("%s", str);
if (strcmp(str, "end") == 0)
break;
Pl_Query_Begin(TRUE);
arg[0] = Mk_Variable();
arg[1] = Mk_String(str);
nb_sol = 0;
res = Pl_Query_Call(func, 2, arg);
while (res)
{
sol[nb_sol++] = Rd_String(arg[0]);
res = Pl_Query_Next_Solution();
}
Pl_Query_End(PL_RECOVER);
for (i = 0; i < nb_sol; i++)
printf(" solution: %s\n", sol[i]);
printf("%d solution(s)\n", nb_sol);
}
Stop_Prolog();
return 0;
}
***END CODE***
**** MY EXPLANATION ****
If I run this simple example asking for the john's ancestors list (for
further information please look at the prolog file new_main.pl), it simply
answers with "peter" while the correct answer was: peter, bob, jane, mary
and paul !!
I guess that function Pl_Query_Next_Solution() (look in the while(res) )
doesn't work properly: it always returns FALSE even if there is more than
one solution! That's awful, you can't go further than the first result in
the query because res is set equal to 0 in the while(res) at the very first
attempt (so the while condition becomes FALSE)!
The only workaround I was able to find out is just to install gcc 3.x.x and
try again and it worked flawlessly! :-)
Although the query works fine in that way, I'd still prefer using gcc 4.x.x
if I could find out how to!
Any suggestions would be nice! Thanks.
Idomeneo
p.s. sorry for my BAD English, actually I'm still learning it! Anyway, I
hope my message is still intelligible to all of yours. :-D
_______________________________________________
Bug-prolog mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-prolog
