#!/usr/bin/env python3
"""Extract the IANA registered top-level-domain (TLD) list into tables.

Source : https://data.iana.org/TLD/tlds-alpha-by-domain.txt
Snapshot: Version 2026102600, Last Updated 2026-10-26 07:07:01 UTC
Format : '# Version ...' comment header, then one lowercase TLD per line
         (1437 names, no blank lines, 9528 bytes).

Outputs (same directory):
  tlds.csv          per-TLD detail: tld,first_letter,length
  tlds_by_letter.tsv summary table: letter<TAB>count (no header)
"""
import hashlib
import pathlib

HERE = pathlib.Path(__file__).resolve().parent
SRC = HERE / "tlds-alpha-by-domain.txt"
MD5 = "8a32a215f1ad7eeb691d0731403b3208"

def main():
    data = SRC.read_bytes()
    actual = hashlib.md5(data).hexdigest()
    assert actual == MD5, f"source md5 mismatch: {actual} != {MD5}"
    rows = [ln.lower() for ln in data.decode("ascii").splitlines()
            if ln and not ln.startswith("#")]
    assert len(rows) == 1437, f"expected 1437 TLDs, got {len(rows)}"
    assert all(" " not in t for t in rows), "non-lower/space in names"

    with open(HERE / "tlds.csv", "w", encoding="ascii") as f:
        f.write("tld,first_letter,length\n")
        for t in rows:
            f.write(f"{t},{t[0]},{len(t)}\n")

    counts = {}
    for t in rows:
        counts[t[0]] = counts.get(t[0], 0) + 1
    with open(HERE / "tlds_by_letter.tsv", "w", encoding="ascii") as f:
        for letter in sorted(counts):
            f.write(f"{letter}\t{counts[letter]}\n")

    print(f"MD5 ok | TLDs={len(rows)} | by-letter rows={len(counts)}")

if __name__ == "__main__":
    main()