Split string with multiple delimiters and keep the string in "groups"

[DreamingInsanity]

Putting a title on this was hard so hopefully I can do a better job of explaining it here.

Let's say I had a string like this: .nnddd9999999999ddnn. (A) or like this: ''0000aaa0000'' (B)

I would like to be able to split those strings with multiple delimiters, but keep the strings "intact" like this: string A would become:
['.', 'nn', 'ddd', '9999999999', 'ddd', 'nn', '.']
string B would become:
['\'\'', '0000', 'aaa', '0000', '\'\'']

As you can see, each string is split with multiple delimiters, but the "chains of characters" are intact rather than it appearing like:
['.', 'n', 'n', 'd', 'd', 'd', '9', '9', '9'...]

The closest I have gotten is using re.split with a delimiter like (.|n|d|9) but that produces an array like:
['.', '', 'n', '', 'n', '', 'd', ''...]

How could I get this to work? Is using re.split even the best way?

[anbu23]

>>> import re
>>> regex = re.compile(r'((.)\2*)')
>>>
>>> matches = regex.finditer(".nnddd9999999999ddnn.")
>>> s = []
>>> for match in matches:
...     s.append(match.group(0))
...
>>> print(s)
['.', 'nn', 'ddd', '9999999999', 'dd', 'nn', '.']

[bowlofred]

Is there really something separating each character, or are you just looking for any repeated character? (You use the term "delimiter", but I don't see any).

If you really just want repeated characters, you could do the following. The grouping is a bit odd, but you can pull the repeated strings out of the first part of each tuple.

>>> s = ".nnddd9999999999ddnn."
>>> re.findall(r"((.)\2+)", s)
[('nn', 'n'), ('ddd', 'd'), ('9999999999', '9'), ('dd', 'd'), ('nn', 'n')]

Or if you really just have a few characters and you want all the strings of them, you could do what you did earlier, but add the repetition (+) operator to them:

>>> re.findall(r"(\.+|n+|d+|9+)",s)
['.', 'nn', 'ddd', '9999999999', 'dd', 'nn', '.']

[DreamingInsanity]

Is there really something separating each character, or are you just looking for any repeated character? (You use the term "delimiter", but I don't see any).

Repeated characters is correct - I used the term 'delimiter' because the way I was imagining it was each unique character acts as it's own delimiter, because it separates the other characters, if that makes any sense.

>>> re.findall(r"(\.+|n+|d+|9+)",s)
['.', 'nn', 'ddd', '9999999999', 'dd', 'nn', '.']

Thanks for the solution - I prefer this as it's pretty easy to understand whats going on.

[DeaD_EyE]

You can do it in a complete different way with more_itertools.split_when, which works also with other types as str.

The generator function split_when takes an iterable (the str e.g.) and a predicate function.
The generator calls the predicate function with last_elent and current_element.

import more_itertools


def predicate(last, current):
    return last != current


for unique in more_itertools.split_when(".nnddd9999999999ddnn.", predicate):
    print("".join(unique))
    # all elements in unique, are separated
    # the join converts it back to a str.