How to find all possible uniform substrings of a string?

Question

I have a string like

aaabbbbcca

And I'd like to parse all possible uniform substrings from that. So my expected substrings for this string are

['a', 'aa', 'aaa', 'b', 'bb', 'bbb', 'bbbb', 'c', 'cc', 'a']

I tried the following

import re

print(re.findall(r"([a-z])(?=\1*)", "aaabbbbcca"))
# Output: ['a', 'a', 'a', 'b', 'b', 'b', 'b', 'c', 'c', 'a']

Is it possible trough regular expressions? If yes, then how?

Answer 1

You can use a regex to find streaks of the same character, and then some Python on top to build the smaller streaks.

import re

s = 'aaabbbbcca'
matches = (m.group() for m in re.finditer(r'([a-z])\1*', s))
result = [m[:i] for m in matches for i in range(1, len(m) + 1)]

There's also an itertools solution.

from itertools import groupby                                               
s = 'aaabbbbcca'                                                            
matches = (''.join(g) for _, g in groupby(s))                               
result = [m[:i] for m in matches for i in range(1, len(m) + 1)]  

Answer 2

Using two itertools functions:

from itertools import groupby, accumulate

s = 'aaabbbbcca'

print([a for _, g in groupby(s) for a in accumulate(g)])

Or just with basics:

s = 'aaabbbbcca'

a = ''
print([a := a * (c in a) + c for c in s])

Output for both:

['a', 'aa', 'aaa', 'b', 'bb', 'bbb', 'bbbb', 'c', 'cc', 'a']

Answer 3

I think this particular problem can be solved with a regex. The answer is based on this answer, where parts of numbers are extracted. The explanation is the same as in the other answer. Each match creates an empty group and a group within the lookahead. The lookahead captures sequences of a, b or c of at least length 1. Afterward, we simply create a list of strings that are in the second group.

import re 

s = "aaabbbbcca"
matches = re.finditer(r'(?=(a{1,}|b{1,}|c{1,}))',s)
results = [match.group(1) for match in matches]
print(results)

Output:

['aaa', 'aa', 'a', 'bbbb', 'bbb', 'bb', 'b', 'cc', 'c', 'a']

The values of the output are the same as requested, but not the exact same order.