Regex [] vs () in Python with respect to re.split()

Question

What is the difference between [,.] and (,|.) when used as a pattern in re.split(pattern,string)? Can some please explain with respect to this example in Python:

import re
regex_pattern1 = r"[,\.]"
regex_pattern2 = r"(,|\.)"
print(re.split(regex_pattern1, '100,000.00')) #['100', '000', '00']
print(re.split(regex_pattern2, '100,000.00'))) #['100', ',', '000', '.', '00']

The first one is a _character class_ and the second is a _capture group_ — Gilles Quénot, Jul 06 '20 at 01:06
You may also want to check: [Using alternation or character class for single character matching?](https://stackoverflow.com/q/4724588/8967612) — 41686d6564 stands w. Palestine, Jul 06 '20 at 01:17
This was closed as a duplicate of a page that did not explain `split`'s behaviour at all. — ikegami, Jul 06 '20 at 18:55
Sorry I guess my title wasn't clear, what I wanted to ask was why character class and capture group works diffrently with respect to re.split() — Vishwa Swaroop, Jul 06 '20 at 20:04

ikegami · Accepted Answer · 2020-07-06T18:52:38.070

[,\.] is equivalent to ,|\..^[1]

(,|\.) is equivalent to ([,\.]).

() creates a capture, and re.split returns captured text as well as the text separated by the pattern.

>>> import re
>>> re.split(r'([,\.])', '100,000.00')
['100', ',', '000', '.', '00']
>>> re.split(r'(,|\.)', '100,000.00')
['100', ',', '000', '.', '00']
>>> re.split(r',|\.', '100,000.00')
['100', '000', '00']
>>> re.split(r'(?:,|\.)', '100,000.00')
['100', '000', '00']
>>> re.split(r'[,\.]', '100,000.00')
['100', '000', '00']

You might sometime need (?:,|\.) to limit what is considered the operands of | when you embed it in a larger pattern, though.

The `.` does not need to be escaped in a character class, so it can be `[,.]` — dawg, Jul 06 '20 at 01:19

Regex [] vs () in Python with respect to re.split()

1 Answers1