prologclp

How can I display the results of an SWI-Prolog program in a table? (CLP problem)


I am trying to solve a CLP problem in SWI-Prolog. The task is very similar to the zebra problem. There are a total of 25 variables with the domains between 1 to 5. So the related variables will get the same tags. (The program is written in hungarian.)

I want the output to show not only the labels assigned to the variables, but the related variables in a table. Is there a way to do this?

% Constraint Logic Programming
:- use_module(library(clpfd)).
:- use_module(library(lists)).
% Your program goes here

egyetemista(X, All):-
    All = [Egressy,Fenyvesi,Gallyas,Jeney,Vadkerti,
            Edina, Frida, Gabriella, Jozsef, Vince,
            Budapest,Debrecen,Miskolc,Pecs,Szeged,
            Biologia,Informatika,Jog,Kemia,Magyar],
    All ins 1..5,
    all_different([Egressy,Fenyvesi,Gallyas,Jeney,Vadkerti]),
    all_different([Edina, Frida, Gabriella, Jozsef, Vince]),
    all_different([Budapest,Debrecen,Miskolc,Pecs,Szeged]),
    all_different([Biologia,Informatika,Jog,Kemia,Magyar]),

    Fenyvesi #= Jog,
    Fenyvesi #\= Debrecen,
    Fenyvesi #\= Jozsef,
    Fenyvesi #\= Vince,

    Jozsef #\= Gallyas,
    Jozsef #\= Biologia,
    Jozsef #= Budapest,

    Vadkerti #= Gabriella,
    Vadkerti #\= Kemia,
    Vadkerti #\= Szeged,
    Gabriella #\= Kemia,
    Gabriella #\= Szeged,
    Kemia #= Szeged,

    Jeney #= Pecs,
    Jeney #\= Vince,

    Frida #= Magyar,

    Edina #= Egressy #\ Edina #= Miskolc,

    Informatika #\= Edina,
    Informatika #\= Frida,
    Informatika #\= Gabriella,

    labeling([], All),

    %Szak:
    nth0(N, All, Szeged),
    nth0(N, All, X).

    %egyetemista(X, All)

If you run the program like this, the output is: All = [1, 2, 3, 4, 5, 2, 4, 5, 1, 3, 1, 5, 2, 4, 3, 5, 1, 2, 3, 4], X = 3

Which means that the label assigned to the variable 'Szeged' is 3, and the associated variables are obtained by substituting variables from the All list that also have a label of 3 on the output. So, for example, the first row of the table could be: 'gallyas vince szeged kemia'

Thank you so much in advance.


Solution

  • Prolog cannot map from a variable back to its name in the source code, so you need to keep a mapping yourself. A simple way to do this is like this:

    egyetemista(X, AllPairs):-
        AllPairs = [
            egressy-Egressy, fenyvesi-Fenyvesi, gallyas-Gallyas, jeney-Jeney, vadkerti-Vadkerti,
            edina-Edina, frida-Frida, gabriella-Gabriella, jozsef-Jozsef, vince-Vince,
            budapest-Budapest, debrecen-Debrecen, miskolc-Miskolc, pecs-Pecs, szeged-Szeged,
            biologia-Biologia, informatika-Informatika, jog-Jog, kemia-Kemia, magyar-Magyar
        ],
        pairs_values(AllPairs, All),
        All ins 1..5,
        % ... the rest of your definition unchanged
    

    The result visible in a call of this is a list of Name-Value pairs:

    ?- egyetemista(X, All).
    X = 3,
    All = [egressy-1, fenyvesi-2, gallyas-3, jeney-4, vadkerti-5, edina-2, frida-4, gabriella-5, ... - ...|...] ;
    X = 3,
    All = [egressy-1, fenyvesi-2, gallyas-3, jeney-4, vadkerti-5, edina-2, frida-4, gabriella-5, ... - ...|...] ;
    X = 3,
    All = [egressy-1, fenyvesi-2, gallyas-3, jeney-4, vadkerti-5, edina-2, frida-4, gabriella-5, ... - ...|...] ;
    X = 3,
    All = [egressy-1, fenyvesi-2, gallyas-3, jeney-4, vadkerti-5, edina-2, frida-4, gabriella-5, ... - ...|...] .  % etc.
    

    You can then filter out only those entries that have their value equal to X, and print those entries in whatever format you like.