Re: question on assert and retract

[Vic Bancroft]

On Tue, 27 Aug 2002, bruno patin wrote:

> how not to add the same facts two times ? 

Consider the following use of (=..)/2 - univ for unique literals,

  %%% -*- Mode: Prolog; -*-
  %%% unique literals
  
  dynamic( ul/2 ).
  
  u_assertz( X ) :-
        X =.. [ Key | Values ],
        Z =.. [ 'ul', Key, Values ],
        Q =.. [ 'ul', Key, Priors ],
        catch( call( Q ), _, fail ),
        !,
        retract( Q ),
        assertz( Z ).
  
  u_assertz( X ) :-
        X =.. [ Head | Tail ],
        Z =.. [ 'ul', Head, Tail ],
        assertz( Z ).
  
This treats the principle functor as the key and the term list as the data for
the key.  The length of the term list can vary.  An example usage might be,

  | ?- u_assertz( toto(1) ).
  
  yes
  | ?- listing( ul/2 ).
  
  ul(toto, [1]).
  
  yes
  | ?- u_assertz( toto(2) ).
  
  yes
  | ?- listing( ul/2 ).
  
  ul(toto, [2]).
  
  yes
  | ?- u_assertz( toto(3,'go',P) ).
  
  yes
  | ?- listing( ul/2 ).
  
  ul(toto, [3, go, _]).
  
  yes

It doesn't seem to matter if some element of the term list is unbound.
It seems fairly natural to add another predicate, u_call/1, as

  u_call( X ) :-
        X =.. [ Key | Values ],
        Z =.. [ 'ul', Key, Values ],
        catch( call( Z ), _, fail ).

In the prior example, one can then do 

  | ?- u_call( toto( 3, X, 9 ) ).
  
  X = go

more,
l8r,

------------------------------------------------------------------- 
Victor Bancroft, Principal Engineer, Zvolve Systems [v]770.551.4505 
1050 Crown Pointe Pkwy, Suite 300, Atlanta GA 30338 [f]770.551.4509 
Fellow, Artificial Intelligence Center              [v]706.542-0358 
Athens, Georgia  30602, U.S.A           http://ai.uga.edu/~bancroft