Prolog: DRY way of creating lists declaratively - prolog

Question

I'm doing some experiments with Prolog and having difficulties with the following rule:
row(Row, Matrix, [R1,R2,R3,R4]) :-
cell(1, Row, Matrix, R1),
cell(2, Row, Matrix, R2),
cell(3, Row, Matrix, R3),
cell(4, Row, Matrix, R4).
This rule extracts one row from a matrix, given its row number. For example,
row(2, [1,2,3,4,5,6,7,8], X)
X = [5,6,7,8]
What nags me is that there is lots of repetition in that code. After finishing with 4x4 matrices, I will have to deal with 9x9 ones. And the code can get very non DRY.
Is there a way to extract that repetition out?
Thanks.
Edit: The complete code giving me trouble is here: https://github.com/kikito/7-languages-in-7-weeks/blob/master/3-prolog/day-3/sudoku-refactor.pl

Answer 1

After writing my first answer, I realized that you can also simplify your program using findall
row(Row, Matrix, L) :- findall(X,cell(_,Row,Matrix,X),L).

Answer 2

I would think about changing the representation to a list of lists rather than a flat list, then selecting a row becomes very easy. You can just use the built-in nth1/3:
:- use_module(library(lists)). % I use Sicstus Prolog
row(N,M,X) :- nth1(N,M,X).
cell(R,C,M,X) :- nth1(R,M,Y), nth1(C,Y,X).
column(N,M,X) :- findall(Y,(nth1(_,M,Z), nth1(N,Z,Y)),X).
m([[1,2,3,4],[5,6,7,8],[9,10,11,12],[13,14,15,16]]).
example(Row,Cell,Column) :- m(M), row(2,M,Row), cell(2,3,M,Cell), column(2,M,Column).
%| ?- example(A,B,C).
%A = [5,6,7,8],
%B = 7,
%C = [2,6,10,14] ? ;
%no