"""抓取快手验证页的网络协议 + JS bundle，评估"扣 JS 纯算"可行性。

落盘：
- out/captcha_net/timeline.json —— 所有 request(method/url/post_data head/headers)
  与 response(status/content-type/url) 时间线
- out/captcha_net/js_*.js       —— 页面加载的所有外部脚本源码

重点看：
1. challenge 拉图请求（哪个端点、什么参数、返回什么）；
2. kSecretApiVerify 的请求体是明文距离/轨迹，还是加密 blob；
3. 构造 verify 体的 JS 能否单独扒出来 + 是否依赖 DOM/window（补环境成本）。

用法：
  uv run python -m tools.captcha_net --mobile 17666663175
  uv run python -m tools.captcha_net --url 'https://...captcha.html?key=...'
"""

from __future__ import annotations

import argparse
import json
import re
import subprocess
import sys
import time
from pathlib import Path

ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
    sys.path.insert(0, str(ROOT))

OUT_DIR = ROOT / "out" / "captcha_net"


def _fresh_error_url(mobile: str, device_profile: str, base_url: str) -> str:
    cmd = [
        sys.executable, "-m", "tools.sms_login_cli",
        "--mobile", mobile, "--base-url", base_url,
        "--code", "000000", "--transport", "okhttp4-android10",
    ]
    print("[net] 取新鲜 error_url ...")
    proc = subprocess.run(cmd, capture_output=True, text=True, cwd=str(ROOT))
    out = proc.stdout + proc.stderr
    m = re.search(r'https://app\.m\.kuaishou\.com/verify/captcha\.html\?[^"\s\\]+', out)
    if not m:
        print(out[-1500:])
        raise SystemExit("[net] 未提取到 error_url")
    return m.group(0)


def capture(error_url: str, *, wait: int) -> None:
    import shutil
    import subprocess as sp
    import tempfile

    from core.captcha_assist import _find_system_chromium, _free_local_port, _wait_for_cdp_endpoint
    from playwright.sync_api import sync_playwright

    OUT_DIR.mkdir(parents=True, exist_ok=True)
    for f in OUT_DIR.glob("js_*.js"):
        f.unlink()

    browser_path = _find_system_chromium()
    port = _free_local_port()
    profile_dir = tempfile.mkdtemp(prefix="ksjsb-net-")
    proc = sp.Popen(
        [
            browser_path,
            f"--remote-debugging-port={port}",
            "--remote-debugging-address=127.0.0.1",
            f"--user-data-dir={profile_dir}",
            "--no-first-run", "--no-default-browser-check",
            "--window-size=430,920", "about:blank",
        ],
        stdout=sp.DEVNULL, stderr=sp.DEVNULL,
    )
    timeline: list[dict] = []
    js_urls: list[str] = []
    try:
        endpoint = _wait_for_cdp_endpoint(port, timeout=20)
        with sync_playwright() as pw:
            browser = pw.chromium.connect_over_cdp(endpoint, timeout=20000)
            context = browser.contexts[0]
            page = context.pages[0] if context.pages else context.new_page()

            def on_request(req: object) -> None:
                url = str(getattr(req, "url", "") or "")
                method = str(getattr(req, "method", "") or "")
                post = getattr(req, "post_data", None)
                if isinstance(post, (bytes, bytearray)):
                    post = post.decode("utf-8", "replace")
                headers = dict(getattr(req, "headers", {}) or {})
                if any(k in url for k in ("captcha", "sliding", "kSecret", "verify", "/zt/", "/wd/")):
                    timeline.append({
                        "kind": "req", "method": method, "url": url,
                        "post_head": (post or "")[:600],
                        "header_keys": sorted(headers),
                    })
                if url.endswith(".js") or ".js?" in url:
                    js_urls.append(url)

            def on_response(resp: object) -> None:
                url = str(getattr(resp, "url", "") or "")
                if any(k in url for k in ("captcha", "sliding", "kSecret", "verify", "/zt/", "/wd/")):
                    headers = dict(getattr(resp, "headers", {}) or {})
                    timeline.append({
                        "kind": "resp", "url": url,
                        "status": int(getattr(resp, "status", 0) or 0),
                        "content_type": headers.get("content-type", ""),
                    })

            page.on("request", on_request)
            page.on("response", on_response)
            print(f"[net] goto {error_url[:90]}...")
            page.goto(error_url, wait_until="domcontentloaded", timeout=30000)
            time.sleep(min(max(wait, 2), 12))

            # 列出页面实际加载的脚本 src，逐个落盘
            scripts = page.evaluate(
                "Array.from(document.scripts).map(s=>s.src).filter(Boolean)"
            ) or []
            # 合并 on_request 捕获到的 .js（含动态注入的）
            all_js = list(dict.fromkeys([*scripts, *js_urls]))
            print(f"[net] 脚本数={len(all_js)}，开始下载源码")
            from curl_cffi.requests import get as cffi_get
            for idx, src in enumerate(all_js):
                try:
                    r = cffi_get(src, timeout=20, headers={"Referer": error_url})
                    body = r.content or b""
                except Exception as exc:  # noqa: BLE001
                    print(f"  [{idx}] 下载失败 {src[:80]}: {exc.__class__.__name__}")
                    continue
                name = f"js_{idx:02d}_{re.sub(r'[^a-z0-9]+','_',src.lower())[-40:]}.js"
                (OUT_DIR / name).write_bytes(body)
                print(f"  [{idx}] {len(body)}B  {src[:90]}")
    finally:
        if proc.poll() is None:
            proc.terminate()
            try:
                proc.wait(timeout=3)
            except sp.TimeoutExpired:
                proc.kill()
        shutil.rmtree(profile_dir, ignore_errors=True)

    (OUT_DIR / "timeline.json").write_text(
        json.dumps(timeline, ensure_ascii=False, indent=2), encoding="utf-8"
    )
    print("\n=== 协议时间线（captcha 相关）===")
    for t in timeline:
        if t["kind"] == "req":
            print(f"  REQ {t['method']} {t['url'][:95]}")
            if t.get("post_head"):
                print(f"        post: {t['post_head'][:300]}")
        else:
            print(f"  RESP[{t['status']}] {t.get('content_type','')} {t['url'][:95]}")
    print(f"\n[net] 落盘目录: {OUT_DIR}")


def main() -> int:
    ap = argparse.ArgumentParser()
    ap.add_argument("--mobile", default="17666663175")
    ap.add_argument("--url", default="")
    ap.add_argument("--device-profile", default="out/sms_device_latest.json")
    ap.add_argument("--base-url", default="https://az1-api.ksapisrv.com")
    ap.add_argument("--wait", type=int, default=8)
    args = ap.parse_args()
    error_url = args.url or _fresh_error_url(args.mobile, args.device_profile, args.base_url)
    print(f"[net] error_url = {error_url}")
    capture(error_url, wait=args.wait)
    return 0


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