How do I split on all nonalphanumeric characters, EXCEPT the apostrophe?
re.split('\W+',text)
works, but will also split on apostrophes. How do I add an exception to this rule?
Thanks!
Add a comment
|
3 Answers
Try this:
re.split(r"[^\w']+",text)
Note the w is now lowercase, because it represents all alphanumeric characters (note that that includes the underscore). The character class [^\w'] refers to anything that's not (^) either alphanumeric (\w) or an apostrophe.
Comments
re.split(r"[^\w']+",text)
By starting a character class with ^, it inverts the definition, so [^\w'] is the inverse of [\w'], which would match an alphanumeric/underscore/apostrophe.
Comments
The answers here don't work, as 'quoted' words will not be stripped of their apostrophes.
What works for me is
re.split(r"\W'+|^'+|'+\W|'$|[^\w']+", text)
i.e. remove:
apostrophe(s) after non-word OR apostrophe(s) at line start OR apostrophe(s) before non-word OR the current solution
