如何递归嵌套python列表？

Question

list0 = [[]]
list0[0].append([])
list0[0][0].append([])
list0[0][0][0].append(["I'm in deep!"])
print(list0)

How do I achieve the nested effect above in a loop? Like how I can append lists to a list in a loop.

def list_deepener(layers):
    list1 = [[]]
    count = 0
    while count != layers:
        count += 1
        y = "list1" + ("[0]" * count)
        y.append([])
        print(x)
list_deepener(5)

I tried this but I couldn't think of any way to convert the string into code which would allow me to alter list1.

Answer 1

Suppose that we have mylist = [] to begin with, and we want to add lists to a certain depth. The trick is to use a separate variable to keep track of where we are in the nesting process , and update that variable as we add each layer. Thus:

>>> mylist = []
>>> current = mylist
>>> for i in range(5):
...     to_add = []
...     current.append(to_add)
...     current = to_add
... 
>>> mylist # 6 pairs of brackets: the initial list, plus 5 added nested lists
[[[[[[]]]]]]

Answer 2

There's many ways to solve this issue, but when you find yourself constructing code in strings and looking for ways to execute that, you're on the wrong path.

A recursive solution:

def nest(x, levels):
    assert levels >= 0
    if levels == 0:
        return [x]
    else:
        return [nest(x, levels-1)]


print(nest("I'm in deep", 5))

An iterative solution:

def nest(x, levels):
    assert levels >= 0
    result = [x]
    for _ in range(levels):
        result = [result]
    return result


print(nest("I'm in deep", 5))

Answer 3

Another way to think about this is recursively. Consider what you want in the deepest part of your list (your 'base' case') and then layer on top for each iteration.

def deepen(layers: int) -> list:
    if layers <= 1:  # Base case, i.e. 'what happens at the bottom'
        return ["Deeper than deep"]
    else:  # Iterative case, i.e. 'what happens at each layer'
        return [deepen(layers - 1)]

This gives you the correct result:

>>> def deepen(layers: int) -> list:
...     if layers <= 1:
...         return ["Deeper than deep"]
...     else:
...         return [deepen(layers - 1)]
...
>>> deepen(5)
[[[[['Deeper than deep']]]]]

Answer 4

You could pass a range and initial string to functools.reduce for a simple one-liner:

from functools import reduce

layers = 4

reduce(lambda a, _: [a], range(layers), "I'm in deep!")
# [[[["I'm in deep!"]]]]