normalize_to_ascii.py

import string
import unicodedata

valid_filename_chars = "-_.() %s%s" % (string.ascii_letters, string.digits)

def clean(filename, whitelist=valid_filename_chars, replace=' '):
    for r in replace:
        filename = filename.replace(r,'_')

    # keep only valid ascii chars
    cleaned_filename = unicodedata.normalize('NFKD', filename).encode('ASCII', 'ignore').decode()

    # keep only whitelisted chars
    return ''.join(c for c in cleaned_filename if c in whitelist)