Check typos in domain names based on Levenshtein distance

domains_typos.py

import pandas as pd
import Levenshtein as lev
import numpy as np

white_domains = [
    'gmail.com',
    'yahoo.com',
    'icloud.com',
    'mail.ru',
    'yandex.ru',
]

df = pd.DataFrame()
df['email'] = ['yandex.ru', 'yandax.ru', 'mail.ru', 'maik.ru']

df['typo_in_email_domain_flag'] = df['email'].apply(lambda x: min([
  i for i in [
   lev.distance(x, d) for d in white_domains   
  ] if i != 0  
]) < 3)