LongArticlesJob/coldStartTasks/crawler/toutiao/detail_page_recommendation.py

"""
@author: luojunhui
"""
from __future__ import annotations

import json
import requests
from tenacity import retry

from applications import log
from applications.utils import proxy, request_retry
from .use_js import call_js_function

retry_desc = request_retry(retry_times=3, min_retry_delay=2, max_retry_delay=30)


@retry(**retry_desc)
def get_associated_recommendation(article_id: str, cookie: str):
    """
    toutiao related recommendation
    """
    ms_token = "-aYwLj97uyCi3oghPfhz2nXaekLoFR5YnYUBA5SuyQZae_NLllO4zC30-CeVLth0A6Hmm7MuGr4_IN9MjHUn8wkq-UQKXJxoGmIAokpUsPsOLjdQKffe-cGWCiZ6xqgh7XE%3D"
    query_params = [
        0,
        1,
        14,
        "min_behot_time=0&channel_id=91558184576&category=pc_profile_channel&disable_raw_data=true&client_extra_params=%7B%22playparam%22%3A%22codec_type%3A0%2Cenable_dash%3A1%2Cunwatermark%3A1%22%2C%22group_id%22%3A%22{}%22%7D&aid=24&app_name=toutiao_web&msToken={}".format(
            article_id, ms_token, ms_token),
        "",
        "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/133.0.0.0 Safari/537.36"
    ]
    a_bogus = call_js_function(query_params)
    url = f"https://www.toutiao.com/api/pc/list/feed?min_behot_time=0&channel_id=91558184576&category=pc_profile_channel&disable_raw_data=true&client_extra_params=%7B%22playparam%22%3A%22codec_type%3A0%2Cenable_dash%3A1%2Cunwatermark%3A1%22%2C%22group_id%22%3A%22{article_id}%22%7D&aid=24&app_name=toutiao_web&msToken={ms_token}&a_bogus={a_bogus}"
    headers = {
        'accept': 'application/json, text/plain, */*',
        'accept-language': 'zh',
        'referer': 'https://www.toutiao.com/video/{}/'.format(article_id),
        'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/133.0.0.0 Safari/537.36',
        'Cookie': cookie
    }
    try:
        response = requests.get(url, headers=headers, proxies=proxy())
        response.raise_for_status()
        return response.json()
    except requests.exceptions.RequestException as e:
        log(
            task="toutiao account crawler",
            function="get_toutiao_account_video_list",
            message=f"API请求失败: {e}",
            data={"account_id": article_id},
        )
    except json.JSONDecodeError as e:
        log(
            task="toutiao account crawler",
            function="get_toutiao_account_video_list",
            message=f"响应解析失败: {e}",
            data={"account_id": article_id},
        )
    return None