Fix #368: Tokenizer handled pattern 'unicode close quote, period' incorrectly.

spacy/en/language_data.py

@@ -102,8 +102,8 @@ TOKENIZER_PREFIXES = r''', " ( [ { * < $ £ “ ' `` ` # US$ C$ A$ a- ‘ .... .
 
 
 TOKENIZER_SUFFIXES = (r''', \" \) \] \} \* \! \? % \$ > : ; ' ” '' 's 'S ’s ’S ’''' 
-                    '''\.\. \.\.\. \.\.\.\. (?<=[a-z0-9)\]"'%\)])\. '''
-                    '''(?<=[0-9])km''').strip().split()
+                      r'''\.\. \.\.\. \.\.\.\. (?<=[a-z0-9)\]”"'%\)])\. '''
+                      r'''(?<=[0-9])km''').strip().split()
 
 
 TOKENIZER_INFIXES = (r'''\.\.\.+ (?<=[a-z])\.(?=[A-Z]) (?<=[a-zA-Z])-(?=[a-zA-z]) '''