If I run solved([[x, o, o], [o, o, o], [o, o, o]]) it should output true as there is only x and if I run solved([[x, o, o], [o, o, o], [o, o, x]]) it should output false as there is more than 1 x. However, running it outputs true all the time as it infers a value for C.
:- use_module(library(clpfd)).
rotate_clock(Xss, Zss) :-
transpose(Xss, Yss),
maplist(reverse, Yss, Zss).
rotate_anti(Xss, Zss) :-
maplist(reverse, Xss, Yss),
transpose(Yss, Zss).
linjmp([x, x, o | T], [o, o, x | T]).
linjmp([o, x, x | T], [x, o, o | T]).
linjmp([H|T1], [H|T2]) :- linjmp(T1,T2).
horizjmp([A|T],[B|T]) :- linjmp(A,B).
horizjmp([H|T1],[H|T2]) :- horizjmp(T1,T2).
jump(B,A) :- horizjmp(B,A).
jump(B,A) :- rotate_clock(B,BR), horizjmp(BR,BRJ), rotate_anti(BRJ, A).
num_x(A, C) :- count(A, x, C).
count([],X,0).
count([X|T],X,Y):- count(T,X,Z), Y is 1+Z.
count([_|T],X,Z):- count(T,X,Z).
sum_list([], 0).
sum_list([H|T], Sum) :-
sum_list(T, Rest),
Sum is H + Rest.
solved(A) :-
maplist(num_x, A, B),
sum_list(B, C),
C == 1.
Instead of trying to change the behavior of a language (which might be possible, but is of course a challenge), it might be better to investigate why eventually Prolog finds a C with C == 1. If we evaluate the maplist(num_x, A, B), ourselves, we see:
?- maplist(num_x, [[x, o, o], [o, o, o], [o, o, x]], B).
B = [1, 0, 1] ;
B = [1, 0, 0] ;
B = [0, 0, 1] ;
B = [0, 0, 0].
So it appears that the num_x/2 predicate, can generate multiple results, for the same list: for a list with one x, it first generates 1, and then 0.
This is confirmed if we do some tests with count/3:
?- count([x, o, o], x, C).
C = 1 ;
C = 0.
?- count([x, x, o], x, C).
C = 2 ;
C = 1 ;
C = 1 ;
C = 0.
So it appears that count/3 each time has a backtracking point where it can decide to count a given x, or not.
Indeed, if we take a look at the count/3 predicate, we see:
count([],X,0).
count([X|T],X,Y):- count(T,X,Z), Y is 1+Z.
count([_|T],X,Z):- count(T,X,Z).
So here for a non-empty list, there are two ways clauses: one where the head is equal to X, the element we want to count, in which case Y is 1+Z, but the last clause says that, regardless what the value of the head is, Prolog will not count that element. Prolog performs backtracking, and thus will eventually pick both clauses.
We can add a dif/2 to add a constraint that the head and X should be differrent, like:
count([],X,0).
count([X|T],X,Y):- count(T,X,Z), Y is 1+Z.
count([H|T],X,Z):- dif(H, X), count(T,X,Z).
So now if an X appears in the list, we will count that element.
Related
I'm trying to code a program in prolog that says true if all the paths from a to b are the same size. Example : we have a path from a to b and another from a to c to b, here it's false because there are two paths from a to b with different sizes, the first is 1 and the other is 2. They all must be the same size otherwise it's false.
I started doing this to get the length of each path, but I'm stuck here, I just need to compare if there are two same paths or not, if yes then we compare the two results if they are the same length then true otherwise false, but I don't know how to do it in Prolog :
chemin1(X, Y):-
arete(X,Y).
chemin1(X, Y):-
arete(X,Z),
chemin1(Z,Y).
chemin2(X, Y, N):-
arete(X, Y),
N is 1.
chemin2(X, Y, N):-
arete(X, Z),
N1 is 1,
chemin2(Z, Y, N2),
N is N1+N2.
I'm assuming you have an acyclic directed graph and that a path is represented by a vertex list.
% b
% / \
% a d
% \ / \
% c---e
arete(a, b).
arete(a, c).
arete(b, d).
arete(c, d).
arete(c, e).
arete(d, e).
chemin(X, X, [X]).
chemin(X, Z, [X|Xs]):- arete(X, Y), chemin(Y, Z, Xs).
Examples:
?- chemin(a, d, C).
C = [a, b, d] ;
C = [a, c, d] ;
false.
?- chemin(a, e, C).
C = [a, b, d, e] ;
C = [a, c, d, e] ;
C = [a, c, e] ;
false.
Then, all paths between two vertices X and Y are of the same size, if there are no two paths between vertices X and Y that are of different sizes.
% all_same_size(+X, +Y)
all_same_size(X, Y) :-
not( ( chemin(X, Y, Xs),
chemin(X, Y, Ys),
not( same_size(Xs, Ys) ) ) ).
same_size([], []).
same_size([_|Xs], [_|Ys]) :- same_size(Xs, Ys).
Examples:
?- all_same_size(a, d).
true.
?- all_same_size(a, e).
false.
chemin2(X0,X, N) :-
path(arete, Path, X0,X),
length(Path, N).
allequallength(X0, X) :-
setof(N, chemin2(X0,X, N), [_]).
Using path/4.
With this definition you can also ask a more general question using the facts you indicated:
arete(a, b).
arete(b, d).
arete(b, c).
arete(a, c).
?- allequallength(X0,X).
X0 = X
; X0 = a, X = b
; X0 = a, X = d
; X0 = b, X = c
; X0 = b, X = d.
I've written a tail-recursive predicate in Prolog which outputs the integers between A and B in a list K. I've used "reverse" to bring the numbers into the right order:
numbers(A,B,K) :- numbers(A,B,[],K).
numbers(Y,Y,X,K) :- !, reverse([Y|X],K).
numbers(A,B,X,K) :- A<B, C is A+1, numbers(C,B,[A|X],K).
Query:
?- numbers(3,6, K).
K=[3,4,5,6]
All works fine. What I now want to do is that I only want to have odd numbers of the range between A and B in the list K. How can I do that? Thanks in advance!
Firstly, I would try to avoid using reverse/2. If you have such a solution, it's often an indicator that there's a better way to get the answer forwards more directly. Not always, but most often. reverse/2 is probably the 2nd favorite band-aid in Prolog right behind use of the cut. :)
In many problems, an auxiliary accumulator is needed. In this particular case, it is not. Also, I would tend to use CLP(FD) operations when involving integers since it's the more relational approach to reasoning over integers. But you can use the solution below with is/2, etc, if you wish. It just won't be as general.
numbers(S, E, []) :- S #> E. % null case
numbers(X, X, [X]).
numbers(S, E, [S|T]) :-
S #< E,
S1 #= S + 1,
numbers(S1, E, T).
| ?- numbers(3, 8, L).
L = [3,4,5,6,7,8] ? ;
no
| ?- numbers(A, B, [2,3,4,5]).
A = 2
B = 5 ? ;
no
| ?-
This solution avoids reverse/2 and is tail recursive.
To update it for odd integers, the first thought is that we can easily modify the above to do every other number by just adding 2 instead of 1:
every_other_number(S, E, []) :- S #> E.
every_other_number(X, X, [X]).
every_other_number(S, E, [S|T]) :-
S #< E,
S1 #= S + 2,
every_other_number(S1, E, T).
| ?- every_other_number(3, 7, L).
L = [3,5,7] ? ;
no
| ?- every_other_number(3, 8, L).
L = [3,5,7] ? ;
no
| ?- every_other_number(4, 8, L).
L = [4,6,8] ? ;
no
| ?-
Then we can do odd numbers by creating an initial predicate to ensure the condition that the first value is odd and calling every_other_number/3:
odd_numbers(S, E, L) :-
S rem 2 #= 1,
every_other_number(S, E, L).
odd_numbers(S, E, L) :-
S rem 2 #= 0,
S1 #= S + 1,
every_other_number(S1, E, L).
| ?- odd_numbers(2, 8, L).
L = [3,5,7] ? ;
no
| ?- odd_numbers(2, 9, L).
L = [3,5,7,9] ? ;
no
| ?- odd_numbers(3, 8, L).
L = [3,5,7] ? ;
no
| ?-
This could be a solution, using mod/2 operator.
numbers(A,B,K) :-
B1 is B+1,
numbers(A,B1,[],K).
numbers(Y,Y1,X,K) :-
Y = Y1,
reverse(X,K).
numbers(A,B,X,K) :-
A<B,
C is A+1,
C1 is mod(C,2),
(C1 = 0 ->
numbers(C,B,[A|X],K)
; numbers(C,B,X,K)).
Another possibility is to use DCG :
numbers(A,B,K) :-
phrase(odd(A,B), K).
odd(A,B) --> {A > B, !}, [].
odd(A,B) --> {A mod2 =:= 0, !, C is A+1}, odd(C,B).
odd(A,B) --> {C is A+2}, [A], odd(C, B).
Very, VERY new to Prolog here. My function needs to compare two lists of equal length by taking the larger number into a new list (e.g. larger([3, 12, 5], [6, 3, 11], X) returns X = [6, 12, 11].) This is what I have, but it is not getting me what I need:
larger([],[],[]).
larger([H|T],[E|A],X):- H > E, larger([T],[A],[H|X]).
larger([H|T],[E|A],X):- H < E, larger([T],[A],[E|X]).
Any help is much appreciated.
The other answer is OK, this is a slightly different approach.
Two clauses should be enough:
larger([], [], []).
larger([X|Xs], [Y|Ys], [Z|Zs]) :-
/* Z is the larger number from (X, Y) */
larger(Xs, Ys, Zs).
How you do the part in the comments depends on your exact problem statement and maybe the implementation. At least SWI-Prolog and GNU-Prolog both have an arithmetic function max() that you can use like this in the above:
larger([], [], []).
larger([X|Xs], [Y|Ys], [Z|Zs]) :-
Z is max(X, Y),
larger(Xs, Ys, Zs).
This is arguably nicer than the solution with three clauses because it won't leave behind unnecessary choice points. Like the other solution, it will work fine as long as the two lists have numbers in them.
This would be identical to using a maplist, for example like this:
larger(Xs, Ys, Zs) :-
maplist(max_number, Xs, Ys, Zs).
max_number(X, Y, Z) :- Z is max(X, Y).
You're not far.
Try with
larger([], [], []).
larger([H | T], [E | A], [H | X]) :-
H > E,
larger(T, A, X).
larger([H | T], [E | A], [E | X]) :-
H =< E,
larger(T, A, X).
If I'm not wrong, there are three errors in your code.
(1) you have to translate the bigger head value (H or E) in the third argument of larger/3, not in the recursive call
% ------- H added here ---v
larger([H | T], [E | A], [H | X]) :-
H > E,
larger(T, A, X).
% not here ----^
(2) T and A, the tails in [H|T] and [E|A], are lists, so you have to pass they recursively as T and A, not as [T] and [A]
larger([H | T], [E | A], [H | X]) :-
H > E,
larger(T, A, X).
% not larger([T], [A], X)
(3) if you have the cases H > E and H < E, your code fail when H is equal to E; one solution is H > E and H =< E; the secon case cover H equal to E.
It is folk knowledge that append(X,[Y],Z) finds the last element
Y of the list Z and the remaining list X.
But there is some advantage of having a customized predicate last/3,
namely it can react without leaving a choice point:
?- last([1,2,3],X,Y).
X = 3,
Y = [1,2]
?- append(Y,[X],[1,2,3]).
Y = [1,2],
X = 3 ;
No
Is there a way to realize a different implementation of
append/3 which would also not leave a choice point in the
above example?
P.S.: I am comparing:
/**
* append(L1, L2, L3):
* The predicate succeeds whenever L3 unifies with the concatenation of L1 and L2.
*/
% append(+List, +List, -List)
:- public append/3.
append([], X, X).
append([X|Y], Z, [X|T]) :- append(Y, Z, T).
And (à la Gertjan van Noord):
/**
* last(L, E, R):
* The predicate succeeds with E being the last element of the list L
* and R being the remainder of the list.
*/
% last(+List, -Elem, -List)
:- public last/3.
last([X|Y], Z, T) :- last2(Y, X, Z, T).
% last2(+List, +Elem, -Elem, -List)
:- private last2/4.
last2([], X, X, []).
last2([X|Y], U, Z, [U|T]) :- last2(Y, X, Z, T).
One way to do it is to use foldl/4 with the appropriate help predicate:
swap(A, B, B, A).
list_front_last([X|Xs], F, L) :-
is_list(Xs),
foldl(swap, Xs, F, X, L).
This should be it:
?- list_front_last([a,b,c,d], F, L).
F = [a, b, c],
L = d.
?- list_front_last([], F, L).
false.
?- list_front_last([c], F, L).
F = [],
L = c.
?- Ys = [y|Ys], list_front_last(Ys, F, L).
false.
Try to see if you can leave out the is_list/1 from the definition.
As I posted:
append2(Start, End, Both) :-
% Preventing unwanted choicepoint with append(X, [1], [1]).
is_list(Both),
is_list(End),
!,
append(Start, End, Both),
!.
append2(Start, End, Both) :-
append(Start, End, Both),
% Preventing unwanted choicepoint with append(X, Y, [1]).
(End == [] -> ! ; true).
Result in swi-prolog:
?- append2(Y, [X], [1,2,3]).
Y = [1, 2],
X = 3.
I have a predicate variablize/3 that takes a list and replaces each item, in turn, with a variable, example:
% ?- variablize([a,b,c], X, L).
% L = [[X, b, c], [a, X, c], [a, b, X]]
Now I am trying to extend this predicate to accept a list of variables, example:
% ?- variablize([a,b,c], [X,Y], L).
% L = [[X, Y, c], [X, b, Y], [a, X, Y]]
My code so far is:
replace_at([_|Tail], X, 1, [X|Tail]).
replace_at([Head|Tail], X, N, [Head|R]) :- M is N - 1, replace_at(Tail, X, M, R).
replace_each([], _, _, [], _).
replace_each([_|Next], Orig, X, [Res|L], N) :-
replace_at(Orig, X, N, Res),
M is N + 1,
replace_each(Next, Orig, X, L, M).
variablize(I, X, L) :- replace_each(I, I, X, L, 1).
Any pointers? Do I extend replace_at/4 to have a list of indexes that should be skipped?
A simplified, builtin based way of implementing variablize/3
variablize(I, X, L) :-
bagof(R, U^select(U, I, X, R), L).
put in evidence that instead of select/4 we could have a distribute/3 that applies replacements of elements of X, when X becomes a list. select/4 can be implemented in this way
myselect(B, I, X, R) :-
append(A, [B|C], I), append(A, [X|C], R).
and this form is convenient because we have the part to the right of input list I, where I suppose you need to distribute remaining variables. Then a recursion on X elements should do:
distribute(I, [X|Xs], L) :-
append(A, [_|C], I),
distribute(C, Xs, R),
append(A, [X|R], L).
distribute(I, [], I).
distribute/3 behaves this way:
?- distribute([a,b,c,d],[1,2],X).
X = [1, 2, c, d] ;
X = [1, b, 2, d] ;
X = [1, b, c, 2] ;
X = [a, 1, 2, d] ;
X = [a, 1, c, 2] ;
X = [a, b, 1, 2] ;
false.
thus
variablize_l(I, X, L) :-
bagof(R, distribute(I, X, R), L).
give us:
?- variablize_l([a,b,c],[X,Y],L).
L = [[X, Y, c], [X, b, Y], [a, X, Y]].
edit
I initially wrote this way, for here the evidence of separating the distribution phase from list construction:
replace_v([_|T], X, [X|T]).
replace_v([L|T], X, [L|R]) :-
replace_v(T, X, R).
variablize(I, X, L) :-
bagof(E, replace_v(I, X, E), L).
variablize(L1,L2,L) :-
append(L1,L2,L3),
length(L1,Len1),
length(L2,Len2),
findall(L4,(combination(L3,Len1,L4),var_count(L4,Len2)),L).
combination(X,1,[A]) :-
member(A,X).
combination([A|Y],N,[A|X]) :-
N > 1,
M is N - 1,
combination(Y,M,X).
combination([_|Y],N,A) :-
N > 1,
combination(Y,N,A).
var_count([],0).
var_count([V|R],N) :-
var(V),
var_count(R,N1),
N is N1 + 1,
!.
var_count([A|R],N) :-
var_count(R,N).