Remove accentuation from a utf-8/latin-1 "string" (which should actually a byte-string) to ascii

asciify.py

import unicodedata

def asciify(string, encoding='latin-1'):
  """Given an string with bytes coming from a DB or other ill-developed data
  extraction, cleanup and return an ASCII string free of accentuations.

  >>> asciify('Ba\xc3\xba')
  'Bau'
  >>> asciify('Bau')
  'Bau'
  >>> asciify('Baú', encoding='utf-8')
  'Bau'
  """
  bs = bytes(string, encoding)

  ns = bs.decode('utf-8')
  ns = unicodedata.normalize('NFKD', ns)
  ns = ns.encode('ascii', 'ignore')

  return str(ns, 'ascii')

if __name__ == "__main__":
    import doctest
    doctest.testmod(verbose=True)