Question

Why doesn’t this:

def subsets(x):
    if not x:
        return [[]]
    else:
        return [x + y for x in [[], [x[0]]] for y in subsets(x[1:])]

print subsets(['a', 'b'])

produce the same output as this:

def subsets(x):
    if not x:
        return [[]]
    else:
        z = subsets(x[1:])
        return [x + y for x in [[], [x[0]]] for y in z]

print subsets(['a', 'b'])

Answer 1

The problem is that you are using the name x for two different things. When you introduce x as a variable inside a list comprehension it hides the x that was defined in the function. As a result, the final x in your list comprehension isn’t refering to the x that you had hoped for.

Renaming one of the x to something else fixes the problem:

def subsets(original_x):
    if not original_x:
        return [[]]
    else:
        return [x + y for x in [[], [original_x[0]]]
                      for y in subsets(original_x[1:])]

I would also suggest that you try to find more descriptive names for your variables. This will reduce the likelihood of accidentally reusing a name.