Flattening a list which consist of integer and list

Question

Assuming input:

[1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]

Expected output:

[1,1,1,1,2,2,3,3,4,5,6,6,9]

How do I flatten the list without removing the duplicates?

My current situation

def flatten(lst):
    nlist = []
    for item in lst:
        nlist = nlist + [item]
    return nlist

My initial thought was that to re-add the elements into a new list to get the expected output. However it did not went well, I am getting

What i get:

[1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]

I am using IDLE 3.3, and I am totally a newbie, if it is possible please tell me how to define it manually instead of using built in functions, meaning using recursive or iterative method. thanks guys!!

You realise that link is also a duplicate. And if you look at that, it is also a duplicate of at least 4 others. It's duplicates all the way down. — ebarr, Apr 08 '14 at 09:26

thefourtheye · Accepted Answer · 2016-06-09T03:42:31.390

You can recursively flatten the data like this

>>> def rec(current_item):
...     if type(current_item) == list:
...         for items in current_item:
...             for item in rec(items):
...                 yield item
...     elif type(current_item) == int:
...         yield current_item

and then sort it like this

>>> sorted(rec([1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]))
[1, 1, 1, 1, 2, 2, 3, 3, 4, 5, 6, 6, 9]

score 2 · Answer 2 · answered Apr 08 '14 at 10:44

From funcy module (https://github.com/Suor/funcy) you can pick flatten function.

In this case, provided that funcy is available on your host, the following code should work as expected:

from funcy import flatten

nlist = [1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]
flat_list = flatten(nlist)

print(nlist)
# [1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]

print(sorted(flat_list))
# [1, 1, 1, 1, 2, 2, 3, 3, 4, 5, 6, 6, 9]

score 0 · Answer 3 · answered Apr 08 '14 at 10:06

A python 2.* only solution would be using the ast module from the compiler package (not available in python 3 anymore). It fits perfectly with this particular example:

import compiler

a = [1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]
print sorted(compiler.ast.flatten(a))
# [1, 1, 1, 1, 2, 2, 3, 3, 4, 5, 6, 6, 9]

score 0 · Answer 4 · edited May 23 '17 at 10:28

In python 2.x, you could use the flatten method in the compiler.ast module ("Deprecated since version 2.6: The compiler package has been removed in Python 3.") as follows:

from compiler.ast import flatten

l = [1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]
flattened = flatten(l)
sorted_list = sorted(flattened)

In python 3, to flatten an arbitrarily nested list, you can use the following code, as stated here:

def flatten(l):
    result = []
    for element in l:
        if hasattr(element, "__iter__") and not isinstance(element, str):
            result.extend(flatten(element))
        else:
            result.append(element)
    return result

score 0 · Answer 5 · answered Apr 08 '14 at 11:34

0

How about using regular expressions?

>>> import re
>>> l = [1, [3, 6], 9, [2, [1, 3]], [4, [1], 5], [6], 1, [[2]]]
>>> l2 = map(int,re.sub('[\[\]\s]','',str(l)).split(','))
>>> l2.sort()
>>> l2
[1, 1, 1, 1, 2, 2, 3, 3, 4, 5, 6, 6, 9]
>>>

answered Apr 08 '14 at 11:34

Harpal

12,057
18
61
74

Haha, I am too newbie to understand your 3rd step, thanks by the way :) – shawnngtq Apr 08 '14 at 14:12

Flattening a list which consist of integer and list

5 Answers5