r'(?u)[^a-zA-Z\W\d_]'

[jacklong]

# -*-coding: utf-8-*-

import re
import codecs

# To match Unicode Characters.
pattern = re.compile(r'[^a-zA-Z\W\d_]+',re.UNICODE)

with codecs.open('file', 'r', 'utf8') as f:
for line in f:
match = pattern.search(line)
# Yeah, it only matches the first group.
if match:
print match.group()