crawler_kanyikan/main/kanyikan_moment.py

# -*- coding: utf-8 -*-
# @Author: wangkun
# @Time: 2022/6/10
import os
import random
import sys
import time
import requests
import urllib3
sys.path.append(os.getcwd())
from main.feishu_lib import Feishu
from main.common import Common
from main.kanyikan_moment_publish import Publish
proxies = {"http": None, "https": None}


class Moment:
    # 过滤词库
    @classmethod
    def sensitive_words(cls):
        word_list = []
        # 从云文档读取所有敏感词，添加到词库列表
        lists = Feishu.get_values_batch("moment", "kanyikan", "rofdM5")
        for i in lists:
            for j in i:
                # 过滤空的单元格内容
                if j is None:
                    pass
                else:
                    word_list.append(j)
        return word_list

    # 朋友圈视频 ID
    @classmethod
    def moment_videoids(cls):
        try:
            videoid_list = []
            # 从云文档读取所有敏感词，添加到词库列表
            lists = Feishu.get_values_batch("moment", "kanyikan", "iK58HX")
            for i in lists:
                for j in i:
                    # 过滤空的单元格内容
                    if j is None:
                        pass
                    else:
                        videoid_list.append(j)
            return videoid_list
        except Exception as e:
            Common.logger("moment").error("获取朋友圈视频ID异常:{}", e)
            return "t3256lo1cmk"

    # 抓取基础规则
    @staticmethod
    def download_rule(d_duration, d_width, d_height, d_play_cnt, d_like_cnt, d_share_cnt):
        """
        抓取基础规则
        :param d_duration: 时长
        :param d_width: 宽
        :param d_height: 高
        :param d_play_cnt: 播放量
        :param d_like_cnt: 点赞量
        :param d_share_cnt: 分享量
        :return: 满足规则，返回 True；反之，返回 False
        """
        if int(float(d_duration)) >= 60:
            if int(d_width) >= 0 or int(d_height) >= 0:
                if int(d_play_cnt) >= 100000:
                    if int(d_like_cnt) >= 0:
                        if int(d_share_cnt) >= 0:
                            return True
                        else:
                            return False
                    else:
                        return False
                else:
                    return False
            return False
        return False

    # 获取推荐视频列表
    @classmethod
    def get_recommend(cls):
        url = "https://search.weixin.qq.com/cgi-bin/recwxa/snsgetvideoinfo?"
        headers = {
            "content-type": "application/json",
            "Accept-Encoding": "gzip,compress,br,deflate",
            "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/8.0.20(0x18001442) NetType/WIFI Language/zh_CN",
            "Referer": "https://servicewechat.com/wxbb9a805eb4f9533c/236/page-frame.html"
        }
        time.sleep(1)
        videoid = random.choice(cls.moment_videoids())
        # Common.logger("moment").info("videoid:{}", videoid)
        params = {
            "vid": videoid,
            "openid": "1924336296754305",
            "model": "iPhone 11<iPhone12,1>14.7.1",
            "sharesearchid": "8406805193800900989",
            "shareOpenid": "oh_m45YffSEGxvDH--6s6g9ZkPxg",
        }
        try:
            urllib3.disable_warnings()
            r = requests.get(url=url, headers=headers, params=params, proxies=proxies, verify=False)
            # Common.logger("moment").info("response:{}", r.json())
            if "rec_video_list" not in r.json()["data"]:
                Common.logger("moment").warning("该视频无推荐视频列表:{}", videoid)
            else:
                feeds = r.json()["data"]["rec_video_list"]
                for i in range(len(feeds)):
                    # video_id
                    if "vid" in feeds[i]:
                        video_id = feeds[i]["vid"]
                    else:
                        video_id = 0

                    # video_title
                    if "title" in feeds[i]:
                        video_title = feeds[i]["title"].strip().replace("\n", "") \
                                .replace("/", "").replace("\\", "").replace("\r", "") \
                                .replace(":", "").replace("*", "").replace("？", "") \
                                .replace("?", "").replace('"', "").replace("<", "") \
                                .replace(">", "").replace("|", "").replace(" ", "") \
                                .replace("&NBSP", "").replace(".", "。").replace(" ", "") \
                                .replace("小年糕", "").replace("#", "").replace("Merge", "")
                    else:
                        video_title = 0

                    # video_play_cnt
                    if "played_cnt" in feeds[i]:
                        video_play_cnt = feeds[i]["played_cnt"]
                    else:
                        video_play_cnt = 0

                    # video_comment_cnt
                    if "comment_cnt" in feeds[i]:
                        video_comment_cnt = feeds[i]["comment_cnt"]
                    else:
                        video_comment_cnt = 0

                    # video_liked_cnt
                    if "liked_cnt" in feeds[i]:
                        video_liked_cnt = feeds[i]["liked_cnt"]
                    else:
                        video_liked_cnt = 0

                    # video_share_cnt
                    if "shared_cnt" in feeds[i]:
                        video_share_cnt = feeds[i]["shared_cnt"]
                    else:
                        video_share_cnt = 0

                    # video_duration
                    if "duration" in feeds[i]:
                        video_duration = feeds[i]["duration"]
                    else:
                        video_duration = 0

                    # video_width / video_height
                    if "width" in feeds[i] or "height" in feeds[i]:
                        video_width = feeds[i]["width"]
                        video_height = feeds[i]["height"]
                    else:
                        video_width = 0
                        video_height = 0

                    # video_send_time
                    if "upload_time" in feeds[i]:
                        video_send_time = feeds[i]["upload_time"]
                    else:
                        video_send_time = 0

                    # user_name
                    if "user_info" not in feeds[i]:
                        user_name = 0
                    elif "nickname" not in feeds[i]["user_info"]:
                        user_name = 0
                    else:
                        user_name = feeds[i]["user_info"]["nickname"].strip().replace("\n", "")

                    # user_id
                    if "user_info" not in feeds[i]:
                        user_id = 0
                    elif "openid" not in feeds[i]["user_info"]:
                        user_id = 0
                    else:
                        user_id = feeds[i]["user_info"]["openid"]

                    # head_url
                    if "user_info" not in feeds[i]:
                        head_url = 0
                    elif "headimg_url" not in feeds[i]["user_info"]:
                        head_url = 0
                    else:
                        head_url = feeds[i]["user_info"]["headimg_url"]

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

                    # video_url
                    if "play_info" not in feeds[i]:
                        video_url = 0
                    elif "items" not in feeds[i]["play_info"]:
                        video_url = 0
                    else:
                        video_url = feeds[i]["play_info"]["items"][-1]["play_url"]

                    Common.logger("moment").info("video_id:{}", video_id)
                    Common.logger("moment").info("video_title:{}", video_title)
                    Common.logger("moment").info("user_name:{}", user_name)
                    Common.logger("moment").info("video_play_cnt:{}", video_play_cnt)
                    Common.logger("moment").info("video_liked_cnt:{}", video_liked_cnt)
                    Common.logger("moment").info("video_share_cnt:{}", video_share_cnt)
                    Common.logger("moment").info("video_duration:{}", video_duration)
                    Common.logger("moment").info("video_width * video_height:{}*{}", video_width, video_height)
                    Common.logger("moment").info("video_url:{}", video_url)

                    # 过滤无效视频
                    if video_id == 0 or video_title == 0 or video_duration == 0 or video_send_time == 0 or user_id == 0\
                            or head_url == 0 or cover_url == 0 or video_url == 0:
                        Common.logger("moment").warning("无效视频")
                    # 抓取基础规则
                    elif cls.download_rule(
                            d_duration=video_duration, d_width=video_width, d_height=video_height,
                            d_play_cnt=video_play_cnt, d_like_cnt=video_liked_cnt,
                            d_share_cnt=video_share_cnt) is False:
                        Common.logger("moment").info("不满足基础规则:{}", video_title)
                    elif int(video_send_time) < 1659283200:
                        Common.logger("moment").info('发布时间{}<2022-08-01', video_send_time)
                    # 过滤词库
                    elif any(word if word in video_title else False for word in cls.sensitive_words()) is True:
                        Common.logger("moment").info("视频已中过滤词：{}".format(video_title))
                    # 从已下载视频表去重:https://w42nne6hzg.feishu.cn/sheets/shtcngRPoDYAi24x52j2nDuHMih?sheet=20ce0c
                    elif video_id in [j for m in Feishu.get_values_batch("moment", "kanyikan", "20ce0c") for j in m]:
                        Common.logger("moment").info("该视频已下载：{}", video_title)
                    # 从feeds视频表去重:https://w42nne6hzg.feishu.cn/sheets/shtcngRPoDYAi24x52j2nDuHMih?sheet=tGqZMX
                    elif video_id in [j for n in Feishu.get_values_batch("moment", "kanyikan", "tGqZMX") for j in n]:
                        Common.logger("moment").info("该视频已在moment_feeds中:{}", video_title)
                    else:
                        Common.logger("moment").info("该视频未下载，添加至moment_feeds中:{}", video_title)
                        # 看一看+工作表，插入首行
                        Feishu.insert_columns("moment", "kanyikan", "tGqZMX", "ROWS", 1, 2)
                        # 获取当前时间
                        get_feeds_time = int(time.time())
                        # 准备写入云文档的数据
                        values = [[time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(get_feeds_time)),
                                   "朋友圈",
                                   video_id,
                                   video_title,
                                   video_play_cnt,
                                   video_comment_cnt,
                                   video_liked_cnt,
                                   video_share_cnt,
                                   video_duration,
                                   str(video_width)+"*"+str(video_height),
                                   time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(video_send_time)),
                                   user_name,
                                   user_id,
                                   head_url,
                                   cover_url,
                                   video_url]]
                        time.sleep(1)
                        Feishu.update_values("moment", "kanyikan", "tGqZMX", "A2:P2", values)

        except Exception as e:
            Common.logger("moment").error("获取视频列表异常:{}", e)

    # 下载/上传视频
    @classmethod
    def download_publish(cls, env):
        try:
            moment_feeds = Feishu.get_values_batch("moment", "kanyikan", "tGqZMX")
            for i in range(1, len(moment_feeds) + 1):
                time.sleep(1)
                # download_push_time = moment_feeds[i][0]
                download_video_id = moment_feeds[i][2]
                download_video_title = moment_feeds[i][3]
                download_video_play_cnt = moment_feeds[i][4]
                download_video_comment_cnt = moment_feeds[i][5]
                download_video_like_cnt = moment_feeds[i][6]
                download_video_share_cnt = moment_feeds[i][7]
                download_video_duration = moment_feeds[i][8]
                download_video_resolution = moment_feeds[i][9]
                download_video_send_time = moment_feeds[i][10]
                download_user_name = moment_feeds[i][11]
                download_user_id = moment_feeds[i][12]
                download_head_url = moment_feeds[i][13]
                download_cover_url = moment_feeds[i][14]
                download_video_url = moment_feeds[i][15]

                Common.logger("moment").info("正在判断第{}行，视频:{}", i, download_video_title)

                # 发布时间的时间戳格式（秒为单位）
                v_send_time = int(time.mktime(time.strptime(download_video_send_time, "%Y/%m/%d %H:%M:%S")))
                # 抓取时间的时间戳格式（秒为单位）
                # v_push_time = int(time.mktime(time.strptime(download_push_time, "%Y/%m/%d %H:%M:%S")))

                # 过滤空行及空标题视频
                if download_video_id is None\
                        or download_video_id == ""\
                        or download_video_title is None\
                        or download_video_title == "":
                    Common.logger("moment").warning("标题为空或空行，删除")
                    # 删除行或列，可选 ROWS、COLUMNS
                    Feishu.dimension_range("moment", "kanyikan", "tGqZMX", "ROWS", i + 1, i + 1)
                    return
                # # 视频的抓取时间小于 2 天
                # elif int(time.time()) - v_push_time > 172800:
                #     Common.logger("moment").info("抓取时间超过2天:{}", download_video_title)
                #     # 删除行或列，可选 ROWS、COLUMNS
                #     Feishu.dimension_range("tGqZMX", "ROWS", i + 1, i + 1)
                #     return
                # 视频发布时间不小于 2021-06-01 00:00:00
                elif v_send_time < 1622476800:
                    Common.logger("moment").info(
                        "发布时间小于2021年6月:{}，{}", download_video_title, download_video_send_time)
                    # 删除行或列，可选 ROWS、COLUMNS
                    Feishu.dimension_range("moment", "kanyikan", "tGqZMX", "ROWS", i + 1, i + 1)
                    return
                # 从已下载视频表中去重
                elif download_video_id in [j for m in Feishu.get_values_batch(
                        "moment", "kanyikan", "20ce0c") for j in m]:
                    Common.logger("moment").info("视频已下载:{}", download_video_title)
                    # 删除行或列，可选 ROWS、COLUMNS
                    Feishu.dimension_range("moment", "kanyikan", "tGqZMX", "ROWS", i + 1, i + 1)
                    return
                else:
                    Common.logger("moment").info("开始下载视频:{}", download_video_title)
                    # 下载封面
                    Common.download_method(log_type="moment", text="cover",
                                           d_name=str(download_video_title), d_url=str(download_cover_url))
                    # 下载视频
                    Common.download_method(log_type="moment", 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_video_resolution) + "\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" +
                                  "KANYIKAN_MOMENT")
                    Common.logger("moment").info("==========视频信息已保存至info.txt==========")

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

                    # 保存视频 ID 到云文档:https://w42nne6hzg.feishu.cn/sheets/shtcngRPoDYAi24x52j2nDuHMih?sheet=20ce0c
                    Common.logger("moment").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("moment", "kanyikan", "20ce0c", "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)),
                               "朋友圈",
                               str(download_video_id),
                               str(download_video_title),
                               our_video_link,
                               download_video_play_cnt,
                               download_video_comment_cnt,
                               download_video_like_cnt,
                               download_video_share_cnt,
                               download_video_duration,
                               str(download_video_resolution),
                               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("moment", "kanyikan", "20ce0c", "F2:W2", values)

                    # 保存视频信息到监控表
                    Common.logger("moment").info("添加视频到监控表:{}", download_video_title)
                    # 插入空行
                    time.sleep(1)
                    Feishu.insert_columns("moment", "monitor", "6fed97", "ROWS", 1, 2)
                    # 视频信息写入监控表
                    values = [[time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(int(upload_time))),
                               str(download_video_id),
                               download_video_title,
                               our_video_link,
                               download_video_duration,
                               str(download_video_send_time),
                               download_video_play_cnt]]
                    time.sleep(1)
                    Feishu.update_values("moment", "monitor", "6fed97", "F2:L2", values)

                    # 删除行或列，可选 ROWS、COLUMNS
                    Feishu.dimension_range("moment", "kanyikan", "tGqZMX", "ROWS", i + 1, i + 1)
                    return
        except Exception as e:
            Common.logger("moment").error("下载视频异常:{}", e)
            # 删除行或列，可选 ROWS、COLUMNS
            Feishu.dimension_range("moment", "kanyikan", "tGqZMX", "ROWS", 2, 2)

    # 执行下载/上传
    @classmethod
    def run_download_publish(cls, env):
        try:
            while True:
                if len(Feishu.get_values_batch("moment", "kanyikan", "tGqZMX")) == 1:
                    break
                else:
                    cls.download_publish(env)
        except Exception as e:
            Common.logger("moment").error("执行下载/上传异常:{}", e)


if __name__ == "__main__":
    kuaishou = Moment()
    kuaishou.run_download_publish("dev")

    pass