ドメインが書かれたテキストファイルから同じTLDの数を数えるやつ

README.md

これを元にしました https://misskey.04.si/notes/9ipkkt4ygz

a.py

import sys
from publicsuffix2 import fetch, get_tld
from pprint import pprint

psl_file = fetch()
length = 9000
headers = {
    "Content-Type": "application/json",
}
hosts = {}

cnt = 0

with open(sys.argv[1]) as f:
    for domain in f:
        domain = str.strip(domain)
        cnt += 1
        tld = get_tld(domain, psl_file=psl_file)
        print(f"domain:{domain} tld:{tld}")
        if tld in hosts:
            hosts[tld] += 1
        else:
            hosts[tld] = 1

hosts = sorted(hosts.items(), key=lambda x: x[1], reverse=True)
with open("result.csv", "w") as f:
    for tld, count in hosts:
        f.write(f"{tld},{count}\n")
pprint(hosts)
print(f"total server:{cnt}")