DYNAMIC unification is possible Sort Of

[kingchris]

No question here just something I found useful.

I asked a question a while back
Is DYNAMIC unification possible!

During some heavy prolog writing recently I have found the following works for me.

Code: Select all

predicates
    unification:(integer) procedure.
    unification_string:(string) procedure.
clauses
    unification_string(string1):-!,
        stdio::write("4. Found ",string1,"\n").
    
    unification_string(string2):-!,
        stdio::write("5. Found ",string2,"\n").
 
    unification_string(String):-!,
        stdio::write("6. Found ",String,"\n").
 
    unification(var1):-!,
        stdio::write("1. Found ",var1,"\n").
 
    unification(var2):-!,
        stdio::write("2. Found ",var2,"\n").
 
    unification(Var):-
        stdio::write("3. Found ",Var,"\n").
 
   test():-
        unification(89),
        unification(90),
        unification(900),
        unification(7890),
        var2 := 7890,
        unification(7890),
        unification_string("Fred"),
        unification_string("Bob"),
        unification_string("Anywhere"),
        string2:= "Anywhere",
        unification_string("Anywhere").

• 3. Found 89
  1. Found 90
  2. Found 900
  3. Found 7890
  2. Found 7890
  4. Found Fred
  5. Found Bob
  6. Found Anywhere
  5. Found Anywhere

A simple trick. By using a fact variable as our unification target then changing its value we can change what it finds.

This may be documented elsewhere but I have yet to come across it.

I find that it simplifies state machine parsing and a few other things for me.