Python Split list into sublists based on the first element in the sublists

Question

I want to split a list that looks something like this:

list = [5, a, b, c, d, e, 2, a, b, 4, a ,b ,c ,d , ...]

into this:

list  = [ [5, a, b, c, d, e], [2, a, b] , [4, a ,b ,c ,d] ...]

The first element/number is variable, so no pattern to split it in even chunks. The chunks size or length should be based on that first element of the chunk. Also the alphabetic letters are just placeholders to make the example more readable, in reality the alphabetic letters are floats and numbers.

So the big list really looks something like this:

list = [5, 7, 3.2, 3.1, 4.6, 3, 2, 5.1, 7.1, 4, 5.12 ,3.4 ,4.8 ,12.1 , ...]

Since you haven't made any attempt, here's a solution with pandas for you. `x = pd.Series(lst); result = x.groupby(x.astype(str).str.isdigit().cumsum()).apply(list).tolist()` Good luck! — cs95, Dec 16 '17 at 22:24
I disagree with the duplicate target. The answers from there aren't directly applicable here. — vaultah, Dec 16 '17 at 22:28

Stefan Pochmann · Accepted Answer · 2017-12-17T00:31:09.353

Simple way, read each chunk length n and include the next n values:

it = iter(l)
[[n, *islice(it, n)] for n in it]

Python 2 version:

it = iter(l)
[[n] + list(islice(it, n)) for n in it]

Or without islice:

it = iter(l)
[[n] + [next(it) for _ in range(n)] for n in it]

Demo:

>>> from itertools import islice
>>> l = [5, 'a', 'b', 'c', 'd', 'e', 2, 'f', 'g', 4, 'h' ,'i' ,'j' ,'k']
>>> it = iter(l)
>>> [[n, *islice(it, n)] for n in it]
[[5, 'a', 'b', 'c', 'd', 'e'], [2, 'f', 'g'], [4, 'h', 'i', 'j', 'k']]

Vasilis G. · Answer 2 · 2017-12-16T22:41:00.353

2

You can try this one:

l = [5, 'a', 'b', 'c', 'd', 'e', 2, 'a', 'b', 4, 'a' ,'b' ,'c' ,'d']

pos = [(index, i) for index, i in enumerate(l) if type(i)==int]
l = [l[p[0]:p[0]+p[1]+1] for p in pos]
print(l)

Output:

[[5, 'a', 'b', 'c', 'd', 'e'], [2, 'a', 'b'], [4, 'a', 'b', 'c', 'd']]

edited Dec 16 '17 at 22:41

answered Dec 16 '17 at 22:25

Vasilis G.

7,556
4
19
29

1

The last list does *not* contain the last item. But nevertheless, please do not feed trolls... – Willem Van Onsem Dec 16 '17 at 22:31
Doesn't work if there are ints among the "placeholders". Like there are in their explicit example. – Stefan Pochmann Dec 16 '17 at 23:04

score 1 · Answer 3 · answered Dec 17 '17 at 03:25

Could also trying something simple like this:

l = [5, 'a', 'b', 'c', 'd', 'e', 2, 'f', 'g', 4, 'h' ,'i' ,'j' ,'k']

numbers = [i for i, e in enumerate(l) if isinstance(e, int)]

result = [l[s:e] for s, e in zip(numbers[:-1], numbers[1:])] + [l[numbers[-1]:]]

print(result)
# [[5, 'a', 'b', 'c', 'd', 'e'], [2, 'f', 'g'], [4, 'h', 'i', 'j', 'k']]

Python Split list into sublists based on the first element in the sublists

3 Answers3