video_rewriting/data_channel/kuaishou.py

import random
import time
import requests
import json
import urllib3
from requests.adapters import HTTPAdapter

from common import Feishu, Material, Common
from common.sql_help import sqlCollect
from data_channel.data_help import dataHelp


class KS:

    @classmethod
    def get_ks_url(cls, task_mark, url_id, number, mark, feishu_id, cookie_sheet, channel_id, name):
        list = []
        pcursor = ""
        url = "https://www.kuaishou.com/graphql"

        for i in range(3):
            cookie = Material.get_cookie_data(feishu_id, cookie_sheet, channel_id)
            time.sleep(random.randint(5, 10))
            payload = json.dumps({
                "operationName": "visionProfilePhotoList",
                "variables": {
                    "userId": url_id,
                    "pcursor": pcursor,
                    "page": "profile"
                },
                "query": "fragment photoContent on PhotoEntity {\n  __typename\n  id\n  duration\n  caption\n  originCaption\n  likeCount\n  viewCount\n  commentCount\n  realLikeCount\n  coverUrl\n  photoUrl\n  photoH265Url\n  manifest\n  manifestH265\n  videoResource\n  coverUrls {\n    url\n    __typename\n  }\n  timestamp\n  expTag\n  animatedCoverUrl\n  distance\n  videoRatio\n  liked\n  stereoType\n  profileUserTopPhoto\n  musicBlocked\n  riskTagContent\n  riskTagUrl\n}\n\nfragment recoPhotoFragment on recoPhotoEntity {\n  __typename\n  id\n  duration\n  caption\n  originCaption\n  likeCount\n  viewCount\n  commentCount\n  realLikeCount\n  coverUrl\n  photoUrl\n  photoH265Url\n  manifest\n  manifestH265\n  videoResource\n  coverUrls {\n    url\n    __typename\n  }\n  timestamp\n  expTag\n  animatedCoverUrl\n  distance\n  videoRatio\n  liked\n  stereoType\n  profileUserTopPhoto\n  musicBlocked\n  riskTagContent\n  riskTagUrl\n}\n\nfragment feedContent on Feed {\n  type\n  author {\n    id\n    name\n    headerUrl\n    following\n    headerUrls {\n      url\n      __typename\n    }\n    __typename\n  }\n  photo {\n    ...photoContent\n    ...recoPhotoFragment\n    __typename\n  }\n  canAddComment\n  llsid\n  status\n  currentPcursor\n  tags {\n    type\n    name\n    __typename\n  }\n  __typename\n}\n\nquery visionProfilePhotoList($pcursor: String, $userId: String, $page: String, $webPageArea: String) {\n  visionProfilePhotoList(pcursor: $pcursor, userId: $userId, page: $page, webPageArea: $webPageArea) {\n    result\n    llsid\n    webPageArea\n    feeds {\n      ...feedContent\n      __typename\n    }\n    hostName\n    pcursor\n    __typename\n  }\n}\n"
            })
            headers = {
                'accept': '*/*',
                'content-type': 'application/json',
                'Origin': 'https://www.kuaishou.com',
                'Cookie': cookie,
                'Accept-Language': 'zh-CN,zh;q=0.9',
                'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36',
                'Referer': f'https://www.kuaishou.com/profile/{url_id}',
                'Accept-Encoding': 'gzip, deflate, br',
                'Connection': 'keep-alive'
            }
            urllib3.disable_warnings()
            s = requests.session()
            s.mount('http://', HTTPAdapter(max_retries=3))
            s.mount('https://', HTTPAdapter(max_retries=3))
            # response = requests.request("POST", url, headers=headers, data=payload, timeout=10)

            response = s.post(url=url, headers=headers, data=payload, verify=False, timeout=10)
            response.close()
            if response.status_code != 200:
                return list
            elif "visionProfilePhotoList" not in response.json()["data"]:
                Feishu.bot(mark, '机器自动改造消息通知', f'快手cookie过期，请及时更换', name)
                time.sleep(900)
                continue
            elif "feeds" not in response.json()["data"]["visionProfilePhotoList"]:
                Feishu.bot(mark, '机器自动改造消息通知', f'快手cookie过期，请及时更换', name)
                time.sleep(900)
                continue
            elif len(response.json()["data"]["visionProfilePhotoList"]["feeds"]) == 0:
                Feishu.bot(mark, '机器自动改造消息通知', f'快手cookie使用频繁无法获取到数据，请及时更换', name)
                time.sleep(900)
                continue
            pcursor = response.json()['data']['visionProfilePhotoList']['pcursor']
            feeds = response.json()['data']['visionProfilePhotoList']['feeds']
            for i in range(len(feeds)):
                try:
                    video_id = feeds[i].get("photo", {}).get("videoResource").get("h264", {}).get("videoId", "")
                except KeyError:
                    video_id = feeds[i].get("photo", {}).get("videoResource").get("hevc", {}).get("videoId", "")
                status = sqlCollect.is_used(task_mark, video_id, mark, channel_id)
                if status:
                    cover_url = feeds[i].get('photo', {}).get('coverUrl', "")
                    video_url = feeds[i].get('photo', {}).get('photoUrl', "")
                    duration = dataHelp.video_duration(video_url)
                    if int(duration) >= 45:
                        all_data = {"video_id": video_id, "cover": cover_url, "video_url": video_url}
                        list.append(all_data)
                        if len(list) == int(number):
                            Common.logger("log").info(f"获取快手视频总数:{len(list)}\n")
                            return list
        return list