138 lines
5.2 KiB
Python
138 lines
5.2 KiB
Python
"""斗鱼 Web Cookie 补齐逻辑。"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
from collections.abc import Callable
|
|
|
|
import requests
|
|
from loguru import logger
|
|
|
|
|
|
class CookieEnricher:
|
|
"""补齐登录后 Web 侧需要的 Cookie。"""
|
|
|
|
CSRF_API = "https://www.douyu.com/japi/carnival/nc/common/generateCsrf"
|
|
CSRF_REFERER = "https://www.douyu.com/pages/live-peace-handbook/web/shop?ditchname=pass0&roomId=9263298"
|
|
ACF_CCN_API = "https://www.douyu.com/curl/csrfNlApi/getCsrfCookie"
|
|
ACF_CCN_REFERER = "https://www.douyu.com/pages/ord-task-center?clientType=web&panelSource=1&rid=0"
|
|
TIMEOUT = (5, 10)
|
|
RETRIES = 3
|
|
|
|
def __init__(
|
|
self,
|
|
session: requests.Session,
|
|
request_func: Callable[..., requests.Response],
|
|
ensure_not_stopped: Callable[[], None],
|
|
sleep_interruptible: Callable[[float], None],
|
|
):
|
|
self.session = session
|
|
self.request = request_func
|
|
self.ensure_not_stopped = ensure_not_stopped
|
|
self.sleep_interruptible = sleep_interruptible
|
|
|
|
def enrich_with_retry(self, max_attempts: int = RETRIES) -> None:
|
|
"""独立重试补齐 cvl_csrf_token 和 acf_ccn,不触发整条登录链路重跑。"""
|
|
last_error: Exception | None = None
|
|
for attempt in range(1, max_attempts + 1):
|
|
self.ensure_not_stopped()
|
|
try:
|
|
logger.info(f"补CK尝试 {attempt}/{max_attempts}...")
|
|
self.generate_csrf_cookie()
|
|
self.generate_acf_ccn_cookie()
|
|
logger.info("补CK完成")
|
|
return
|
|
except Exception as e:
|
|
last_error = e
|
|
logger.warning(f"补CK失败 {attempt}/{max_attempts}: {e}")
|
|
if attempt < max_attempts:
|
|
self.sleep_interruptible(1)
|
|
raise ValueError(f"已重试 {max_attempts} 次仍未补齐CK: {last_error}") from last_error
|
|
|
|
def generate_csrf_cookie(self) -> str:
|
|
"""访问 generateCsrf 接口,从 Set-Cookie 中同步 cvl_csrf_token。"""
|
|
logger.info("补齐CSRF Cookie...")
|
|
headers = {
|
|
'Accept': 'application/json, text/plain, */*',
|
|
'Accept-Language': 'zh-CN,zh;q=0.9,en-US;q=0.6,en;q=0.5',
|
|
'Origin': 'https://www.douyu.com',
|
|
'Referer': self.CSRF_REFERER,
|
|
'Sec-Fetch-Dest': 'empty',
|
|
'Sec-Fetch-Mode': 'cors',
|
|
'Sec-Fetch-Site': 'same-origin',
|
|
# 覆盖登录接口的默认表单头,尽量贴近浏览器抓包。
|
|
'Content-Type': None,
|
|
'X-Requested-With': None,
|
|
}
|
|
response = self.request(
|
|
'post',
|
|
self.CSRF_API,
|
|
headers=headers,
|
|
timeout=self.TIMEOUT,
|
|
max_retries=1,
|
|
)
|
|
response.raise_for_status()
|
|
|
|
body = response.text.strip()
|
|
if not body:
|
|
raise ValueError("生成CSRF失败: 响应为空")
|
|
|
|
try:
|
|
payload = response.json()
|
|
except json.JSONDecodeError as exc:
|
|
preview = body[:200].replace("\n", "\\n")
|
|
raise ValueError(f"生成CSRF失败: 响应不是有效 JSON: {preview}") from exc
|
|
|
|
if payload.get('error') != 0:
|
|
raise ValueError(f"生成CSRF失败: {payload.get('msg', '未知错误')}")
|
|
|
|
cookies = self.session.cookies.get_dict()
|
|
csrf_token = cookies.get('cvl_csrf_token', '')
|
|
if not csrf_token:
|
|
raise ValueError("生成CSRF失败: 响应没有 cvl_csrf_token")
|
|
|
|
logger.info("CSRF Cookie已补齐")
|
|
return csrf_token
|
|
|
|
def generate_acf_ccn_cookie(self) -> str:
|
|
"""访问 getCsrfCookie 接口,从 Set-Cookie 中同步 acf_ccn。"""
|
|
logger.info("补齐 acf_ccn Cookie...")
|
|
headers = {
|
|
'Accept': 'application/json, text/plain, */*',
|
|
'Accept-Language': 'zh-CN,zh;q=0.9',
|
|
'Cache-Control': 'no-cache',
|
|
'Pragma': 'no-cache',
|
|
'Priority': 'u=1, i',
|
|
'Referer': self.ACF_CCN_REFERER,
|
|
'Sec-CH-UA': '"Google Chrome";v="149", "Chromium";v="149", "Not)A;Brand";v="24"',
|
|
'Sec-CH-UA-Mobile': '?0',
|
|
'Sec-CH-UA-Platform': '"macOS"',
|
|
'Sec-Fetch-Dest': 'empty',
|
|
'Sec-Fetch-Mode': 'cors',
|
|
'Sec-Fetch-Site': 'same-origin',
|
|
# 该接口抓包没有 Origin、表单 Content-Type 和 X-Requested-With。
|
|
'Origin': None,
|
|
'Content-Type': None,
|
|
'X-Requested-With': None,
|
|
}
|
|
response = self.request(
|
|
'get',
|
|
self.ACF_CCN_API,
|
|
headers=headers,
|
|
timeout=self.TIMEOUT,
|
|
max_retries=1,
|
|
)
|
|
response.raise_for_status()
|
|
|
|
cookies = self.session.cookies.get_dict()
|
|
acf_ccn = cookies.get('acf_ccn', '') or response.cookies.get('acf_ccn', '')
|
|
if acf_ccn and not cookies.get('acf_ccn'):
|
|
# 极少数情况下响应 Cookie 未合入 get_dict,手动补到斗鱼域名下。
|
|
self.session.cookies.set('acf_ccn', acf_ccn, domain='.douyu.com', path='/')
|
|
|
|
if not acf_ccn:
|
|
raise ValueError("补齐 acf_ccn 失败: 响应没有 acf_ccn")
|
|
|
|
logger.info("acf_ccn Cookie已补齐")
|
|
return acf_ccn
|