Question

I’m simplifying a larger complex problem with the following…

Given three arrays of integers, what’s the most efficient way to return all the possible combinations of each of the elements? Note that each value from each array will always be given in the same position so [A,B,C] would be the same as [C,B,A]. My desired result is an array of arrays with each hash containing a single combination. For example:

Given:

var array1 = [1,2]
var array2 = [a,b]
var array3 = [foo,bar]

The result would be:

[
  [1,a,foo],
  [2,a,foo],
  [1,b,foo],
  [2,b,foo],
  [1,a,bar],
  [2,a,bar],
  [1,b,bar],
  [2,b,bar]
]

Answer 1

In Python, use itertools.product:

itertools.product(array1, array2, array3)

and wrap it in a list if you need a sequence, not just an iterable.

If you want to see how it’s done, this is the “equivalent” code given in the itertools docs:

def product(*args, **kwds):
    # product('ABCD', 'xy') --> Ax Ay Bx By Cx Cy Dx Dy
    # product(range(2), repeat=3) --> 000 001 010 011 100 101 110 111
    pools = map(tuple, args) * kwds.get('repeat', 1)
    result = [[]]
    for pool in pools:
        result = [x+[y] for x in result for y in pool]
    for prod in result:
        yield tuple(prod)

although that version of the code isn’t particularly efficient.

There is also a Python implementation in the PyPy version of itertools.