#!/usr/bin/env python3
"""Тест задачи 08: сравнение вывода solution.sh с эталоном на фикстуре и большом логе."""
import os
import random
import subprocess
import sys

HERE = os.path.dirname(os.path.abspath(__file__))
SCRIPT = os.path.join(HERE, "solution.sh")
failures = 0


def check(cond, name, extra=""):
    global failures
    if cond:
        print(f"ok   {name}")
    else:
        print(f"FAIL {name} {extra}")
        failures += 1


def run(path):
    r = subprocess.run(["bash", SCRIPT, path], capture_output=True, text=True, timeout=120)
    return r.returncode, r.stdout, r.stderr


def expected(lines):
    from collections import Counter
    ips = [l.split()[0] for l in lines if l.strip()]
    total = len(lines)
    cnt = Counter(ips)
    top = sorted(cnt.items(), key=lambda kv: (-kv[1], kv[0]))[:3]
    five = 0
    for l in lines:
        parts = l.split('"')
        if len(parts) >= 3:
            codes = parts[2].split()
            if codes and codes[0].isdigit() and 500 <= int(codes[0]) <= 599:
                five += 1
    out = [f"TOTAL {total}"]
    for ip, c in top:
        out.append(f"TOP {ip} {c}")
    out.append(f"5XX {five}")
    return "\n".join(out)


def compare(path, lines, label):
    rc, out, err = run(path)
    exp = expected(lines)
    got = out.strip("\n")
    if got != exp:
        print(f"--- ожидалось ({label}) ---\n{exp}\n--- получено ---\n{got}")
    check(rc == 0, f"{label}: код возврата 0 (получен {rc})", err[:300])
    check(got == exp, f"{label}: вывод совпадает с эталоном")


def main():
    fixture = os.path.join(HERE, "access.log")
    with open(fixture, encoding="utf-8") as f:
        lines = f.read().splitlines()
    compare(fixture, lines, "фикстура")

    # пустой файл
    empty = os.path.join(HERE, "empty.log")
    open(empty, "w").close()
    compare(empty, [], "пустой файл")

    # большой лог: 200k строк, часть 5xx, один лидер, ничья между двумя IP
    rng = random.Random(7)
    big = os.path.join(HERE, "big.log")
    ips = ["10.0.0.1"] * 50000 + ["10.0.0.2"] * 30000 + ["10.0.0.3"] * 30000
    ips += [f"192.168.1.{i}" for i in range(1, 61)] * 10
    ips += [f"172.16.0.{i}" for i in range(1, 21)] * 5
    rng.shuffle(ips)
    big_lines = []
    for ip in ips:
        code = rng.choice([200, 200, 200, 301, 404, 500, 502, 503])
        big_lines.append(f'{ip} - - [23/Sep/2026:11:00:00 +0700] "GET /p HTTP/1.1" {code} 100')
    with open(big, "w", encoding="utf-8") as f:
        f.write("\n".join(big_lines) + "\n")
    compare(big, big_lines, "большой лог 110k строк")

    print(f"\n{'FAILURES: %d' % failures if failures else 'ALL PASS'}")
    sys.exit(1 if failures else 0)


if __name__ == "__main__":
    main()