# -*- coding: utf-8 -*-
"""
在线抓取**全新**样本（2026-10-12 起，只使用本脚本产出的数据）。

抓什么:
  1. prehandle -> sess / comm_captcha_cfg.tdc_path
  2. 连抓 N 份 tdc.js（同一个 URL **每次请求都返回不同构建**，账本 #29）
  3. 每份都算出: XTEA 密钥、eks、37 模块的字符串指纹、槽位序

产出目录: har/online/<时间戳>/
用法:
    .venv/bin/python scripts/fetch_online_samples.py --n 4
"""
from __future__ import annotations

import argparse
import hashlib
import json
import random
import sys
import time
from pathlib import Path

ROOT = Path(__file__).resolve().parents[1]
sys.path.insert(0, str(ROOT))
sys.path.insert(0, str(ROOT / "tools"))

from tcaptcha.client import TencentCaptchaClient  # noqa: E402
from tcaptcha.crypto import extract_eks  # noqa: E402
from tcaptcha.tdc_key import extract_xtea_key_pure  # noqa: E402

OUT = ROOT / "har" / "online"


def main() -> int:
    ap = argparse.ArgumentParser(description="在线抓取全新 tdc.js 样本")
    ap.add_argument("--n", type=int, default=4, help="抓几份 tdc.js")
    ap.add_argument("--aid", default="2039519451", help="雨云 aid（宽松，适合取样）")
    ap.add_argument("--tag", default=None, help="输出子目录名（默认时间戳）")
    ap.add_argument("--gap", type=float, default=1.0, help="两次抓取间隔秒")
    args = ap.parse_args()

    tag = args.tag or time.strftime("%Y%m%d-%H%M%S")
    d = OUT / tag
    d.mkdir(parents=True, exist_ok=True)
    print("输出目录:", d)

    client = TencentCaptchaClient(aid=args.aid,
                                  entry_url="https://app.rainyun.com/account/reward/earn#")
    try:
        tkid = str(random.randint(100000000, 999999999))
        ch, cfg, sid = client.prehandle(tkid)
        print("prehandle: sess=%s sid=%s kind=%s" % (ch.sess[:16], sid, ch.kind))
        tdc_path = cfg.get("tdc_path", "")
        if not tdc_path:
            print("❌ prehandle 没给 tdc_path")
            return 1
        print("tdc_path:", tdc_path[:90])
        (d / "meta.json").write_text(json.dumps({
            "aid": args.aid, "sess": ch.sess, "sid": sid,
            "tdc_path": tdc_path, "kind": ch.kind,
            "fetched_at": time.strftime("%Y-%m-%d %H:%M:%S"),
        }, ensure_ascii=False, indent=1), encoding="utf-8")

        seen = set()
        got = 0
        for i in range(args.n * 3):          # 允许重试，直到拿到 n 份不同构建
            if got >= args.n:
                break
            body = client.fetch_tdc(tdc_path)
            md5 = hashlib.md5(body.encode("utf-8")).hexdigest()
            if md5 in seen:
                time.sleep(args.gap)
                continue
            seen.add(md5)
            got += 1
            name = "tdc_%02d_%s.js" % (got, md5[:8])
            (d / name).write_text(body, encoding="utf-8")

            info = {"md5": md5, "bytes": len(body), "index": got}
            try:
                info["key"] = extract_xtea_key_pure(body)["key"]
            except Exception as e:
                info["key_error"] = "%s: %s" % (type(e).__name__, str(e)[:120])
            info["eks"] = extract_eks(body)[:40]
            print("  [%d] %s  %6d B  md5=%s  key=%s"
                  % (got, name, len(body), md5[:12], info.get("key")))
            (d / (name[:-3] + ".json")).write_text(
                json.dumps(info, ensure_ascii=False, indent=1), encoding="utf-8")
            time.sleep(args.gap)
        print()
        print(">>> 拿到 %d 份互不相同的新构建 -> %s" % (got, d))
        return 0 if got else 1
    finally:
        client.close()


if __name__ == "__main__":
    raise SystemExit(main())