normalize_names_matcher.py

import re
from metaphone import doublemetaphone
from unidecode import unidecode

nonalphanumeric_re = re.compile(r"[^\w ]+")
whitespace_re = re.compile(r" +")

def normalize_simple(name: str) -> str:
    return whitespace_re.sub(" ", nonalphanumeric_re.sub(" ", name.lower()))

def normalize_unidecode(name: str) -> str:
    return normalize_simple(unidecode(name))

def normalize_unidecode_metaphone(name: str) -> str:
    return " ".join(doublemetaphone(part)[0]
                    for part in normalize_unidecode(name).split(" ")