Regex problem with specific string in python

You could use

(?<=\S)(?=[A-Z])|(.{2}\.)

Which needs to be replaced with a function, see a demo on regex101.com.

import re

data = """
Aaa Bbb
AaaBbbCcc
OneTwost.Three
"""

rx = re.compile(r'(?<=\S)(?=[A-Z])|(.{2}\.)')

def replacer(match):
    if match.group(1):
        return " {} ".format(match.group(1))
    return r' '

data = rx.sub(replacer, data)
print(data)

Which yields

Aaa Bbb
Aaa Bbb Ccc
One Two st. Three

Based on what you said you want and the fact that you said, "I have the string":

I have the string

'Aaa Bbb', 'AaaBbbCcc' ,'OneTwost.Three'

These should do it.

Input:

>>> import re
>>> string = """'Aaa Bbb', 'AaaBbbCcc' ,'OneTwost.Three'"""

Output:

>>> re.sub(r'((?<![\',\s])[A-Z]+|[\S]{2}\.)', r' \1', string)
"'Aaa Bbb', 'Aaa Bbb Ccc' ,'One Two st. Three'"

.

Edit

Input (Acting on string and new variable string_1 which removes the ''s)

>>> import re
>>> string = """'Aaa Bbb', 'AaaBbbCcc' ,'OneTwost.Three'"""
>>> string_1 = """Aaa Bbb, AaaBbbCcc ,OneTwost.Three"""

Output

>>> re.sub(r'(?<!^)(?<!,)(?<!\s)(?<!\')([A-Z]+|[\S]{2}\.)', r' \1', string)
"'Aaa Bbb', 'Aaa Bbb Ccc' ,'One Two st. Three'"


>>> re.sub(r'(?:(?<!^)(?<!,)(?<!\s)(?<!\'))([A-Z]+|[\S]{2}\.)', r' \1', 
string)
"'Aaa Bbb', 'Aaa Bbb Ccc' ,'One Two st. Three'"


>>> re.sub(r'(?<!^)(?<!,)(?<!\s)(?<!\')([A-Z]+|[\S]{2}\.)', r' \1', string_1)
'Aaa Bbb, Aaa Bbb Ccc ,One Two st. Three'


>>> re.sub(r'(?:(?<!^)(?<!,)(?<!\s)(?<!\'))([A-Z]+|[\S]{2}\.)', r' \1', string_1)
'Aaa Bbb, Aaa Bbb Ccc ,One Two st. Three'

.

Explanation of the First:

• Made it a string as your quote suggested
• Using re.sub in this situation with the raw_string (r) option to allow for printing of dynamic/changing/variable capturing functionality and will return an edited string
• With the first "(" I'm setting it up to capture everything in the subsequent query
• With "(?<![\',\s])" I'm saying make sure that what follows which I am trying to capture is not preceded by a " ' " or "whitespace"
• With "[A-Z]+" positioned where it is, I am saying capture any group of capital letters (BUT NOTE: This will also match ABC, SDZ, FFRD, ZXF, etc. but will not capture any lowercase letters or other symbols)
• With "|" I'm telling the re engine, "OR" capture this next query
• And with "[\S]{2}\." I'm saying capture if you find any 2 "non-whitespace characters" followed by a "."
• The final ")" ends the capture group directive
• .
• With the second argument "r' \1'" I'm saying print the first group you capture (in this case I only have 1 capture group anyway) and place a single space in front of it

Edit: Slight Explanation of the Following 2 which can act on string_1

• I swear, re.sub's behavior with lookarounds is just wonky. Given your comment below. Through each of the (?<!YOUR_IGNORED_CHARACTER), I'm telling re.sub to essentially not capture if the capital letters are preceded by the designated character. (?<!^), however, means do not capture if the capture group occurs at the beginning of the line.

• Note also, in the string for this example I've removed the ' from the one you had given.