Question

I am trying to use recursion to find the depth of an “expression”, i.e., how many layers of nested tuples there are: For example,

depth(('+', ('expt', 'x', 2), ('expt', 'y', 2))) => 2

depth(('/', ('expt', 'x', 5), ('expt', ('-', ('expt', 'x', 2), 1), ('/', 5, 2)))) => 4

Basically, I figured that I need to check (working from out to in) for each element being an instance of a tuple, and then if so, recursively call the depth function. But I need to find a way of figuring out which set of recursive calls has the greatest depth, and that’s where I’m stuck. Here’s what I have so far:

def depth3(expr):
    if not isinstance(expr, tuple):
        return 0
    else:
        for x in range(0, len(expr)):
            # But this doesn't take into account a search for max depth
            count += 1 + depth(expr[x])
    return count

Thoughts on a good way to approach this?

Answer 1

You’re on the right track, but instead of finding the “total” depth with count += 1 + depth(expr[x])
, use max to find the maximum:

def depth(expr):
    if not isinstance(expr, tuple):
        return 0
    # this says: return the maximum depth of any sub-expression + 1
    return max(map(depth, expr)) + 1

print depth(("a", "b"))
# 1
print depth(('+', ('expt', 'x', 2), ('expt', 'y', 2)))
# 2
print depth(('/', ('expt', 'x', 5), ('expt', ('-', ('expt', 'x', 2), 1), ('/', 5, 2)))) 
# 4