Don't use regular expressions at all, first convert allow_characters to a set and then use ''.join() with a generator expression that strips out the unwanted characters. Assuming the string you are transforming is called s:

allow_char_set = set(allow_characters) s = ''.join(c for c in s if c in allow_char_set) 

That being said, here is how this might look with regex:

s = re.sub(r'[^#.\-_a-zA-Z0-9]+', '', s) 

You could convert your allow_characters string into this regex, but I think the first solution is significantly more straightforward.

Don't use regular expressions at all, first convert allow_characters to a set and then use ''.join() with a generator expression that strips out the unwanted characters. Assuming the string you are transforming is called s:

allow_char_set = set(allow_characters) s = ''.join(c for c in s if c in allow_char_set) 

That being said, here is how this might look with regex:

s = re.sub(r'[^#.\-_a-zA-Z0-9]+', '', s) 

You could convert your allow_characters string into this regex, but I think the first solution is significantly more straightforward.

Edit: As pointed out by DSM in comments, str.translate() is often a very good way to do something like this. In this case it is slightly complicated but you can still use it like this:

import string allow_characters = "#.-_abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789" all_characters = string.maketrans('', '') delete_characters = all_characters.translate(None, allow_characters) s = s.translate(None, delete_characters)