The difference between using argv and args is evident when
Prolog is invoked with numbers as arguments.
The objects returned by unix(argv(_)) are Prolog objects; that
is, if the command line argument is a number, then it will be returned
as a number. Thus:
% prolog 1
.
.
.
| ?- unix(argv([1])).
yes
| ?- unix(argv(['1'])).
no
| ?- unix(args(['1'])).
yes
% prolog 1 6.999999
.
.
.
| ?- unix(argv(X)).
X = [1, 6.9999999E+00]
| ?- unix(args(X)).
X = ['1', '6.999999']
So if your program treats the command line argument as a number, use
the form with argv, but if it is to be treated as a string, use
args. For example if the program is called with a number and
performs some arithmetic operation on the argument then displays the
result, use argv.
| ?- [user].
| runtime_entry(start) :-
unix(argv([A])),
Y is A+1,
display(Y).
| ^D
% user compiled in module user, 0.083 sec 8 bytes
yes
| ?- runtime_entry(start).
46
yes