crawler_gzh/main/gzh_recommend.py

# -*- coding: utf-8 -*-
# @Author: wangkun
# @Time: 2022/8/1
# import time
# import base64
import json
import os
import time
# import urllib.parse
import requests
import urllib3
from main.common import Common
from main.feishu_lib import Feishu
from main.publish import Publish

proxies = {"http": None, "https": None}


class Recommend:
    # 获取 token，保存至飞书云文档
    @classmethod
    def get_token(cls, log_type):
        # charles 抓包文件保存目录
        charles_file_dir = "../crawler-kanyikan-recommend/chlsfiles/"

        if int(len(os.listdir(charles_file_dir))) == 1:
            Common.logger(log_type).info("未找到chlsfile文件，等待60s")
            time.sleep(60)
        else:
            try:
                # 目标文件夹下所有文件
                all_file = sorted(os.listdir(charles_file_dir))

                # 获取到目标文件
                old_file = all_file[-1]

                # 分离文件名与扩展名
                new_file = os.path.splitext(old_file)

                # 重命名文件后缀
                os.rename(os.path.join(charles_file_dir, old_file),
                          os.path.join(charles_file_dir, new_file[0] + ".txt"))

                with open(charles_file_dir + new_file[0] + ".txt", encoding='utf-8-sig', errors='ignore') as f:
                    contents = json.load(f, strict=False)
                # 定义需要返回的列表
                request_info = []
                for content in contents:
                    if "mp.weixin.qq.com" in content['host']:
                        if content["path"] == r"/mp/getappmsgext":
                            headers = content["request"]["header"]["headers"]
                            title = content["request"]["body"]["text"].split("title=")[-1].split("&ct=")[0]
                            vid = content["request"]["body"]["text"].split("vid=")[-1].split("&is_pay_subscribe")[0]
                            request_info.append(title)
                            request_info.append(vid)
                            for h in headers:
                                if h["name"] == "cookie" and "pass_ticket" in h["value"]:
                                    pass_ticket = h["value"].split("pass_ticket=")[-1]
                                    # print(f"pass_ticket:{pass_ticket}")
                                    request_info.append(pass_ticket)

                                if h["name"] == "referer":
                                    __biz = h["value"].split("__biz=")[-1].split("&mid=")[0]
                                    # print(f"__biz:{__biz}")
                                    request_info.append(__biz)

                                if h["name"] == "cookie" and "appmsg_token" in h["value"]:
                                    appmsg_token = h["value"].split("appmsg_token=")[-1]
                                    # print(f"appmsg_token:{appmsg_token}")
                                    request_info.append(appmsg_token)
                                if h["name"] == "cookie" and "wap_sid2" in h["value"]:
                                    wap_sid2 = h["value"].split("wap_sid2=")[-1]
                                    # print(f"wap_sid2:{wap_sid2}")
                                    request_info.append(wap_sid2)
                            return request_info
            except Exception as e:
                Common.logger(log_type).error("获取session异常，30s后重试：{}", e)
                time.sleep(30)
                cls.get_token(log_type)

    # 获取推荐列表
    @classmethod
    def get_recommend(cls, log_type):
        try:
            params = cls.get_token(log_type)
            if params is None:
                Common.logger(log_type).info("未获取到token等信息，30s后重试")
                time.sleep(30)
                cls.get_recommend(log_type)
            else:
                title = params[0]
                vid = params[1]
                __biz = params[2]
                appmsg_token = params[3]
                pass_ticket = params[4]
                wap_sid2 = params[5]

                url = "https://mp.weixin.qq.com/mp/getappmsgext?"
                headers = {
                    "content-type": "application/x-www-form-urlencoded; charset=UTF-8",
                    "accept": "*/*",
                    "x-requested-with": "XMLHttpRequest",
                    "accept-language": "zh-cn",
                    "accept-encoding": "gzip, deflate, br",
                    "origin": "https://mp.weixin.qq.com",
                    "user-agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 14_7_1 like Mac OS X) AppleWebKit/605.1.15 "
                                  "(KHTML, like Gecko) Mobile/15E148 MicroMessenger/8.0.26(0x18001a29)"
                                  " NetType/WIFI Language/zh_CN",
                    "referer": "https://mp.weixin.qq.com/s?"
                               "__biz=Mzg5MDY2NzY5Nw=="
                               "&mid=2247484710"
                               "&idx=1"
                               "&sn=657a341da42ed071aaa4d3ce853f64f2"
                               "&chksm=cfd852f8f8afdbeef513340dec8702433bd78137e7b4afb665d1de5014dc6837ed4dcc979684"
                               "&sessionid=1659509075"
                               "&channel_session_id="
                               "&scene=136"
                               "&subscene="
                               "&exptype="
                               "&reloadid=1659509075"
                               "&reloadseq=2"
                               "&related_video_source=10"
                               "&ascene=1"
                               "&devicetype=iOS14.7.1"
                               "&version=18001a29"
                               "&nettype=WIFI"
                               "&abtest_cookie=AAACAA%3D%3D"
                               "&lang=zh_CN"
                               "&session_us=gh_7364edd0ca9f"
                               "&fontScale=100"
                               "&exportkey=AdT9lhjADCG9r69d1meNZ5c%3D"
                               "&pass_ticket=" + pass_ticket +
                               "&wx_header=3",
                }
                query_string = {
                    "f": "json",
                    "mock": "",
                    "uin": "777",
                    "key": "777",
                    "pass_ticket": pass_ticket,
                    "wxtoken": "",
                    "devicetype": "iOS14.7.1",
                    "clientversion": "18001a29",
                    "__biz": __biz,
                    "appmsg_token": appmsg_token,
                    "x5": "0",
                    # "f": "json"
                }
                cookies = {
                    "appmsg_token": appmsg_token,
                    "devicetype": "iOS14.7.1",
                    "lang": "zh_CN",
                    "pass_ticket": pass_ticket,
                    "rewardsn": "",
                    "version": "18001a29",
                    "wap_sid2": wap_sid2,
                    "wxtokenkey": "777",
                    "wxuin": "2010747860"
                }
                form = {
                    "r": "0.13440037781889225",
                    "__biz": __biz,
                    "appmsg_type": "9",
                    "mid": "2247484710",
                    "sn": "657a341da42ed071aaa4d3ce853f64f2",
                    "idx": "1",
                    "scene": "136",
                    "title": title,
                    "ct": "1654824718",
                    "abtest_cookie": "AAACAA==",
                    "devicetype": "iOS14.7.1",
                    "version": "18001a29",
                    "is_need_ticket": "0",
                    "is_need_ad": "1",
                    "comment_id": "0",
                    "is_need_reward": "0",
                    "both_ad": "0",
                    "reward_uin_count": "0",
                    "send_time": "",
                    "msg_daily_idx": "1",
                    "is_original": "0",
                    "is_only_read": "1",
                    "req_id": "0314yH9rphN660ejUCz1hRVD",
                    "pass_ticket": pass_ticket,
                    "is_temp_url": "0",
                    "item_show_type": "5",
                    "tmp_version": "1",
                    "more_read_type": "0",
                    "appmsg_like_type": "2",
                    "related_video_sn": "",
                    "related_video_num": "5",
                    "vid": vid,
                    "is_pay_subscribe": "0",
                    "pay_subscribe_uin_count": "0",
                    "has_red_packet_cover": "0",
                    "album_id": "1296223588617486300",
                    "album_video_num": "5",
                    "cur_album_id": "",
                    "is_public_related_video": "0",
                    "encode_info_by_base64": "0",
                    "exptype": ""
                }
                urllib3.disable_warnings()
                response = requests.post(url=url, headers=headers, cookies=cookies, params=query_string, data=form,
                                         verify=False)

                if "related_tag_video" not in response.json():
                    Common.logger(log_type).warning("response:{}\n", response.text)
                elif len(response.json()["related_tag_video"]) == 0:
                    Common.logger(log_type).warning("response:{}\n", response.text)
                    time.sleep(10)
                    cls.get_recommend(log_type)
                else:
                    feeds = response.json()["related_tag_video"]
                    for m in range(len(feeds)):
                        # video_title
                        if "title" not in feeds[m]:
                            video_title = 0
                        else:
                            video_title = feeds[m]["title"]
                            # video_title = base64.b64decode(video_title).decode("utf-8")

                        # video_id
                        if "vid" not in feeds[m]:
                            video_id = 0
                        else:
                            video_id = feeds[m]["vid"]

                        # play_cnt
                        if "read_num" not in feeds[m]:
                            play_cnt = 0
                        else:
                            play_cnt = feeds[m]["read_num"]

                        # like_cnt
                        if "like_num" not in feeds[m]:
                            like_cnt = 0
                        else:
                            like_cnt = feeds[m]["like_num"]

                        # duration
                        if "duration" not in feeds[m]:
                            duration = 0
                        else:
                            duration = feeds[m]["duration"]

                        # video_width / video_height
                        if "videoWidth" not in feeds[m] or "videoHeight" not in feeds[m]:
                            video_width = 0
                            video_height = 0
                        else:
                            video_width = feeds[m]["videoWidth"]
                            video_height = feeds[m]["videoHeight"]

                        # send_time
                        if "pubTime" not in feeds[m]:
                            send_time = 0
                        else:
                            send_time = feeds[m]["pubTime"]

                        # user_name
                        if "srcDisplayName" not in feeds[m]:
                            user_name = 0
                        else:
                            user_name = feeds[m]["srcDisplayName"]
                            # user_name = base64.b64decode(user_name).decode("utf-8")

                        # user_id
                        if "srcUserName" not in feeds[m]:
                            user_id = 0
                        else:
                            user_id = feeds[m]["srcUserName"]

                        # head_url
                        if "head_img_url" not in feeds[m]:
                            head_url = 0
                        else:
                            head_url = feeds[m]["head_img_url"]

                        # cover_url
                        if "cover" not in feeds[m]:
                            cover_url = 0
                        else:
                            cover_url = feeds[m]["cover"]

                        # video_url
                        if "url" not in feeds[m]:
                            video_url = 0
                        else:
                            video_url = feeds[m]["url"]

                        # 下载链接
                        download_url = cls.get_url(log_type, video_url)

                        Common.logger(log_type).info("video_title:{}", video_title)
                        Common.logger(log_type).info("video_id:{}", video_id)
                        Common.logger(log_type).info("play_cnt:{}", play_cnt)
                        Common.logger(log_type).info("like_cnt:{}", like_cnt)
                        Common.logger(log_type).info("duration:{}", duration)
                        Common.logger(log_type).info("video_width:{}", video_width)
                        Common.logger(log_type).info("video_height:{}", video_height)
                        Common.logger(log_type).info("send_time:{}", send_time)
                        Common.logger(log_type).info("user_name:{}", user_name)
                        Common.logger(log_type).info("user_id:{}", user_id)
                        Common.logger(log_type).info("head_url:{}", head_url)
                        Common.logger(log_type).info("cover_url:{}", cover_url)
                        Common.logger(log_type).info("video_url:{}", video_url)
                        Common.logger(log_type).info("download_url:{}", download_url)

                        if video_id == 0 or video_title == 0 or duration == 0 or video_url == 0:
                            Common.logger(log_type).info("无效视频\n")
                        elif str(video_id) in [x for y in Feishu.get_values_batch(log_type, "gzh", "fCs3BT") for x in
                                               y]:
                            Common.logger(log_type).info("该视频已下载\n")
                        elif str(video_id) in [x for y in Feishu.get_values_batch(log_type, "gzh", "zWKFGb") for x in
                                               y]:
                            Common.logger(log_type).info("该视频已在feeds中\n")
                        else:
                            Feishu.insert_columns(log_type, "gzh", "zWKFGb", "ROWS", 1, 2)
                            get_feeds_time = int(time.time())
                            values = [[time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(get_feeds_time)),
                                       "推荐榜",
                                       video_title,
                                       str(video_id),
                                       play_cnt,
                                       like_cnt,
                                       duration,
                                       str(video_width) + "*" + str(video_height),
                                       time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(send_time)),
                                       user_name,
                                       user_id,
                                       head_url,
                                       cover_url,
                                       video_url,
                                       download_url
                                       ]]
                            time.sleep(1)
                            Feishu.update_values(log_type, "gzh", "zWKFGb", "D2:T2", values)
                            Common.logger(log_type).info("添加至recommend_feeds成功\n")

        except Exception as e:
            Common.logger(log_type).error("get_recommend异常:{}", e)

    # 获取视频下载链接
    @classmethod
    def get_url(cls, log_type, url):
        try:
            payload = {}
            headers = {
                'Cookie': 'rewardsn=; wxtokenkey=777'
            }
            urllib3.disable_warnings()
            response = requests.get(url=url, headers=headers, data=payload, verify=False)
            response_list = response.text.splitlines()
            video_url_list = []
            for m in response_list:
                if "mpvideo.qpic.cn" in m:
                    video_url = m.split("url: '")[1].split("',")[0].replace(r"\x26amp;", "&")
                    video_url_list.append(video_url)
            video_url = video_url_list[0]
            return video_url
        except Exception as e:
            Common.logger(log_type).error("get_url异常:{}", e)

    # 下载/上传
    @classmethod
    def download_publish(cls, log_type, env):
        try:
            recommend_feeds_sheet = Feishu.get_values_batch(log_type, "gzh", "zWKFGb")
            for i in range(1, len(recommend_feeds_sheet)):
                download_video_title = recommend_feeds_sheet[i][5]
                download_video_id = recommend_feeds_sheet[i][6]
                download_video_play_cnt = recommend_feeds_sheet[i][7]
                download_video_like_cnt = recommend_feeds_sheet[i][8]
                download_video_duration = recommend_feeds_sheet[i][9]
                download_width_height = recommend_feeds_sheet[i][10]
                download_video_send_time = recommend_feeds_sheet[i][11]
                download_user_name = recommend_feeds_sheet[i][12]
                download_user_id = recommend_feeds_sheet[i][13]
                download_head_url = recommend_feeds_sheet[i][14]
                download_cover_url = recommend_feeds_sheet[i][15]
                download_video_url = recommend_feeds_sheet[i][17]
                download_video_comment_cnt = 0
                download_video_share_cnt = 0

                Common.logger(log_type).info("正在判断第{}行", i + 1)
                Common.logger(log_type).info("download_video_title:{}", download_video_title)
                Common.logger(log_type).info("download_video_id:{}", download_video_id)
                Common.logger(log_type).info("download_video_play_cnt:{}", download_video_play_cnt)
                Common.logger(log_type).info("download_video_duration:{}", download_video_duration)
                Common.logger(log_type).info("download_video_send_time:{}", download_video_send_time)
                Common.logger(log_type).info("download_video_url:{}\n", download_video_url)
                # Common.logger(log_type).info("download_video_like_cnt:{}", download_video_like_cnt)
                # Common.logger(log_type).info("download_width_height:{}", download_width_height)
                # Common.logger(log_type).info("download_user_name:{}", download_user_name)
                # Common.logger(log_type).info("download_user_id:{}", download_user_id)
                # Common.logger(log_type).info("download_head_url:{}", download_head_url)
                # Common.logger(log_type).info("download_cover_url:{}", download_cover_url)

                # 过滤空行
                if download_video_id is None or download_video_title is None or download_video_play_cnt is None:
                    Common.logger(log_type).warning("空行，略过\n")
                # # 过滤敏感词
                # elif any(word if word in download_video_title else False for word in
                #          cls.sensitive_words(log_type)) is True:
                #     Feishu.dimension_range(log_type, "music_album", "69UxPo", "ROWS", i + 1, i + 1)
                #     Common.logger(log_type).info("视频已中敏感词，删除成功\n")
                #     return
                # # 下载规则
                # elif cls.download_rule(download_video_share_cnt, download_video_play_cnt) is False:
                #     Feishu.dimension_range(log_type, "music_album", "69UxPo", "ROWS", i + 1, i + 1)
                #     Common.logger(log_type).info("不满足下载规则，删除成功\n")
                #     return
                # 时长小于 60s，删除
                elif int(download_video_duration) < 60:
                    Feishu.dimension_range(log_type, "gzh", "zWKFGb", "ROWS", i + 1, i + 1)
                    Common.logger(log_type).info("时长{}<60，删除成功\n", download_video_duration)
                    return
                # 已下载视频表去重
                elif str(download_video_id) in [n for m in Feishu.get_values_batch(log_type, "gzh", "fCs3BT")
                                                for n in m]:
                    Feishu.dimension_range(log_type, "gzh", "zWKFGb", "ROWS", i + 1, i + 1)
                    Common.logger(log_type).info("该视频在公众号中已下载，删除成功\n")
                    return
                # 看一看已下载表去重
                elif str(download_video_id) in [n for m in Feishu.get_values_batch(log_type, "kanyikan", "20ce0c")
                                                for n in m]:
                    Feishu.dimension_range(log_type, "gzh", "zWKFGb", "ROWS", i + 1, i + 1)
                    Common.logger(log_type).info("该视频在看一看中已下载，删除成功\n")
                    return
                else:
                    # 下载封面
                    Common.download_method(log_type=log_type, text="cover",
                                           d_name=str(download_video_title), d_url=str(download_cover_url))
                    # 下载视频
                    Common.download_method(log_type=log_type, text="video",
                                           d_name=str(download_video_title), d_url=str(download_video_url))
                    # 保存视频信息至 "./videos/{download_video_title}/info.txt"
                    with open("./videos/" + download_video_title + "/" + "info.txt",
                              "a", encoding="UTF-8") as f_a:
                        f_a.write(str(download_video_id) + "\n" +
                                  str(download_video_title) + "\n" +
                                  str(download_video_duration) + "\n" +
                                  str(download_video_play_cnt) + "\n" +
                                  str(download_video_comment_cnt) + "\n" +
                                  str(download_video_like_cnt) + "\n" +
                                  str(download_video_share_cnt) + "\n" +
                                  str(download_width_height) + "\n" +
                                  str(int(time.mktime(
                                      time.strptime(download_video_send_time, "%Y/%m/%d %H:%M:%S")))) + "\n" +
                                  str(download_user_name) + "\n" +
                                  str(download_head_url) + "\n" +
                                  str(download_video_url) + "\n" +
                                  str(download_cover_url) + "\n" +
                                  "gzh")
                    Common.logger(log_type).info("==========视频信息已保存至info.txt==========")

                    # 上传视频
                    Common.logger(log_type).info("开始上传视频:{}".format(download_video_title))
                    our_video_id = Publish.upload_and_publish(log_type, env, "play")
                    our_video_link = "https://admin.piaoquantv.com/cms/post-detail/" + str(our_video_id) + "/info"
                    Common.logger(log_type).info("视频上传完成:{}", download_video_title)

                    # 保存视频 ID 到云文档
                    Common.logger(log_type).info("保存视频ID至云文档:{}", download_video_title)
                    # 视频ID工作表，插入首行
                    Feishu.insert_columns(log_type, "gzh", "fCs3BT", "ROWS", 1, 2)
                    # 视频ID工作表，首行写入数据
                    upload_time = int(time.time())
                    values = [[time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(upload_time)),
                               "推荐榜",
                               str(download_video_title),
                               str(download_video_id),
                               our_video_link,
                               download_video_play_cnt,
                               download_video_like_cnt,
                               download_video_duration,
                               str(download_width_height),
                               str(download_video_send_time),
                               str(download_user_name),
                               str(download_user_id),
                               str(download_head_url),
                               str(download_cover_url),
                               str(download_video_url)]]
                    time.sleep(1)
                    Feishu.update_values(log_type, "gzh", "fCs3BT", "D2:W2", values)

                    # 删除行或列，可选 ROWS、COLUMNS
                    Feishu.dimension_range(log_type, "gzh", "zWKFGb", "ROWS", i + 1, i + 1)
                    Common.logger(log_type).info("视频:{}，下载/上传成功\n", download_video_title)
                    return

        except Exception as e:
            Common.logger(log_type).error("download_publish异常:{}", e)

    # 执行下载/上传
    @classmethod
    def run_download_publish(cls, log_type, env):
        try:
            while True:
                recommend_feeds_sheet = Feishu.get_values_batch(log_type, "gzh", "zWKFGb")
                if len(recommend_feeds_sheet) == 1:
                    Common.logger(log_type).info("下载/上传完成")
                    break
                else:
                    cls.download_publish(log_type, env)
        except Exception as e:
            Common.logger(log_type).error("run_download_publish异常:{}", e)


if __name__ == "__main__":
    Recommend.get_recommend("recommend")
    # Recommend.download_publish("recommend")
    # Recommend.run_download_publish("recommend", "dev")
    # token = Recommend.get_token("recommend")
    # print(token)