Question

I think pat1 = ‘[ab]’ and pat2 = ‘a|b’ have the same function in Python(python2.7, windows) ‘re’ module as a regular expression pattern. But I am confused with ‘[ab]+’ and ‘(a|b)+’, do they have the same function, if not can you explain details.

'''
Created on 2012-9-4

@author: melo
'''

import re
pat1 = '(a|b)+'
pat2 = '[ab]+'
text = '22ababbbaa33aaa44b55bb66abaa77babab88'

m1 = re.search(pat1, text)
m2 = re.search(pat2, text)
print 'search with pat1:', m1.group()
print 'search with pat2:', m2.group()

m11 = re.split(pat1, text)
m22 = re.split(pat2, text)
print 'split with pat1:', m11
print 'split with pat2:', m22

m111 = re.findall(pat1, text)
m222 = re.findall(pat2, text)
print 'findall with pat1:', m111
print 'findall with pat2:', m222

output as below:

search with pat1: ababbbaa
search with pat2: ababbbaa
split with pat1: ['22', 'a', '33', 'a', '44', 'b', '55', 'b', '66', 'a', '77', 'b', '88']
split with pat2: ['22', '33', '44', '55', '66', '77', '88']
findall with pat1: ['a', 'a', 'b', 'b', 'a', 'b']
findall with pat2: ['ababbbaa', 'aaa', 'b', 'bb', 'abaa', 'babab']

why are ‘pat1’ and ‘pat2’ different and what’s their difference?
what kind of strings can ‘pat1’ actually match?

Answer 1

You have a capturing group in the first pattern.

According to the docs,

re.split()
… If capturing parentheses are used in pattern, then the text of all groups in the pattern are also returned as part of the resulting list. …

Try making the group non-capturing and see if you get what you expect:

pat1 = '(?:a|b)+'