# download a csv file from the mixpanel portal here:
# https://mixpanel.com/project/2667276/view/3204508/app/boards#id=5168849&edited-bookmark=KdaZirXsTSd3

import re
import csv
import sys
from collections import defaultdict

# Check if filename was provided as command line argument
if len(sys.argv) != 2:
    print("Usage: python check_new_users.py <csv_filename>")
    sys.exit(1)

GOOD = 0
BAD = 0
# Load emails from CSV file using command line argument
emails = []
domains = defaultdict(int)

with open(sys.argv[1]) as csvfile:
    reader = csv.reader(csvfile)
    next(reader)  # Skip header row if present
    emails = [row[1] for row in reader]  # Get second column

regexes = {
    "regex1": r"[a-z]{7}[0-9]{4}[a-z]{1}@(outlook|hotmail)\.com",
    "regex2": r"[a-z]{5}[0-9]{5}@(outlook|hotmail)\.com",
}

legitimate_outlook = [ e for e in emails if "@outlook.com" in e ]

for label, regex in regexes.items():
    GOOD = 0
    BAD = 0

    for email in emails:
        if re.match(regex, email):
            BAD += 1
            if email in legitimate_outlook:
                legitimate_outlook.remove(email)
        else:
            GOOD += 1

    print(f"{str(regex)} GOOD: {GOOD}, BAD: {BAD}")

print(f"\nLegitimate outlook emails: {len(legitimate_outlook)}\n")

for email in emails:
    try:
        domains[email.split("@")[1]] += 1
    except Exception as e:
        domains["phone"] += 1

# Sort domains by value (count) in descending order and get top 20
sorted_domains = dict(sorted(domains.items(), key=lambda x: x[1], reverse=True)[:10])
for domain, count in sorted_domains.items():
    print(f"{domain:26} {count:8}")