I need a regular expression that will deal with apostrophes when extracting words from text

Question

Anybody know a way of dealing with apostrophes when extracting words from text using a regular expression?

>>> import re
>>> s = re.compile(r"\b[A-Za-z0-9_\-]+\b")
>>> s.findall("I don't know Sally's 'special' friend.")
['I', 'don', 't', 'know', 'Sally', 's', 'special', 'friend']

Desired result:

['I', "don't", 'know', 'Sally', 'special', 'friend']

This discussion covers how to find whole words but doesn't deal with apostrophes.

http://stackoverflow.com/a/28058612/1400768 – nhahtdh Sep 01 '15 at 05:53 — nhahtdh, Sep 01 '15 at 05:53

vks · Answer 1 · 2015-09-01T06:00:31.693

3

s = re.compile(r"(?:^|(?<=\s))[A-Za-z0-9_'\-]+(?=\s|$|\b)")

Use this instead of \b.lookarounds will work for you.See demo.

https://regex101.com/r/sS2dM8/25

edited Sep 01 '15 at 06:00

answered Sep 01 '15 at 05:53

vks

67,027
10
91
124

1

Thanks. That produces a result that is pretty close: `['I', "don't", 'know', "Sally's", "'special'"]` – Bill Sep 01 '15 at 05:56
Although it loses the last word ('friend')! – Bill Sep 01 '15 at 05:59
@Bill https://regex101.com/r/sS2dM8/26 now it wont lose – vks Sep 01 '15 at 06:00

I need a regular expression that will deal with apostrophes when extracting words from text

1 Answers1