crawler_kanyikan/main/kanyikan_recommend.py

# -*- coding: utf-8 -*-
# @Author: wangkun
# @Time: 2023/6/1
import os
import random
import shutil
import sys
import time
import requests
import urllib3
sys.path.append(os.getcwd())
from main.common import Common
from main.feishu_lib import Feishu
from main.publish import Publish
proxies = {"http": None, "https": None}


class Kanyikanrecommend:
    @classmethod
    def get_filter_word(cls, log_type, crawler):
        while True:
            filter_sheet = Feishu.get_values_batch(log_type, crawler, "rofdM5")
            if filter_sheet is None:
                Common.logger(log_type).info(f"filter_sheet:{filter_sheet}")
                time.sleep(1)
                continue
            # 敏感词库列表
            word_list = []
            for i in filter_sheet:
                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

    @classmethod
    def download_rule(cls, video_dict):
        now = int(time.time())
        publish_day = int(int(now - video_dict["publish_time_stamp"]) / (3600*24))
        if (int(video_dict["video_width"]) or int(video_dict["video_height"]) >= 0) \
            and int(video_dict["duration"]) >= 40\
            and ((publish_day >= 7 and int(video_dict["play_cnt"]) >= 2000000) or (publish_day < 7 and int(video_dict["play_cnt"]) >= 500000)):
            # and int(video_dict["publish_time_stamp"]) >= 1672502400:  # 发布时间>=2023-01-01 00:00:00
            return True
        else:
            return False

    @classmethod
    def get_videoList(cls, log_type, crawler, env):
        while True:
            for page in range(1, 101):
                Common.logger(log_type).info(f"正在抓取第{page}页")
                try:
                    session = Common.get_session(log_type)
                    if session is None:
                        time.sleep(1)
                        continue
                    url = 'https://search.weixin.qq.com/cgi-bin/recwxa/recwxavideolist?'
                    header = {
                        "Connection": "keep-alive",
                        "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/8.0.18(0x18001236) "
                                      "NetType/WIFI Language/zh_CN",
                        "Referer": "https://servicewechat.com/wxbb9a805eb4f9533c/234/page-frame.html",
                    }
                    params = {
                        'session': session,
                        "offset": 0,
                        "wxaVersion": "3.9.2",
                        "count": "10",
                        "channelid": "208",
                        "scene": '310',
                        "subscene": '1089',
                        "clientVersion": '8.0.18',
                        "sharesearchid": '0',
                        "nettype": 'wifi',
                        "switchprofile": "0",
                        "switchnewuser": "0",
                    }
                    urllib3.disable_warnings()
                    response = requests.get(url=url, headers=header, params=params, proxies=proxies, verify=False)
                    if "data" not in response.text:
                        Common.logger(log_type).info("获取视频list时，session过期，随机睡眠 31-50 秒")
                        # 如果返回空信息，则随机睡眠 31-40 秒
                        time.sleep(random.randint(31, 40))
                        continue
                    elif "items" not in response.json()["data"]:
                        Common.logger(log_type).info(f"get_feeds:{response.json()}，随机睡眠 1-3 分钟")
                        # 如果返回空信息，则随机睡眠 1-3 分钟
                        time.sleep(random.randint(60, 180))
                        continue
                    feeds = response.json().get("data", {}).get("items", "")
                    if feeds == "":
                        Common.logger(log_type).info(f"feeds:{feeds}")
                        time.sleep(random.randint(31, 40))
                        continue
                    for i in range(len(feeds)):
                        try:
                            video_title = feeds[i].get("title", "").strip().replace("\n", "") \
                                .replace("/", "").replace("\\", "").replace("\r", "") \
                                .replace(":", "").replace("*", "").replace("？", "") \
                                .replace("?", "").replace('"', "").replace("<", "") \
                                .replace(">", "").replace("|", "").replace(" ", "") \
                                .replace("&NBSP", "").replace(".", "。").replace(" ", "") \
                                .replace("'", "").replace("#", "").replace("Merge", "")
                            publish_time_stamp = feeds[i].get("date", 0)
                            publish_time_str = time.strftime("%Y/%m/%d %H:%M:%S", time.localtime(publish_time_stamp))
                            # 获取播放地址
                            if "videoInfo" not in feeds[i]:
                                video_url = ""
                            elif "mpInfo" in feeds[i]["videoInfo"]["videoCdnInfo"]:
                                if len(feeds[i]["videoInfo"]["videoCdnInfo"]["mpInfo"]["urlInfo"]) > 2:
                                    video_url = feeds[i]["videoInfo"]["videoCdnInfo"]["mpInfo"]["urlInfo"][2]["url"]
                                else:
                                    video_url = feeds[i]["videoInfo"]["videoCdnInfo"]["mpInfo"]["urlInfo"][0]["url"]
                            elif "ctnInfo" in feeds[i]["videoInfo"]["videoCdnInfo"]:
                                video_url = feeds[i]["videoInfo"]["videoCdnInfo"]["ctnInfo"]["urlInfo"][0]["url"]
                            else:
                                video_url = feeds[i]["videoInfo"]["videoCdnInfo"]["urlInfo"][0]["url"]
                            video_dict = {
                                "video_title": video_title,
                                "video_id":  feeds[i].get("videoId", ""),
                                "play_cnt":  feeds[i].get("playCount", 0),
                                "like_cnt":  feeds[i].get("liked_cnt", 0),
                                "comment_cnt":  feeds[i].get("comment_cnt", 0),
                                "share_cnt":  feeds[i].get("shared_cnt", 0),
                                "duration":  feeds[i].get("mediaDuration", 0),
                                "video_width":  feeds[i].get("short_video_info", {}).get("width", 0),
                                "video_height":  feeds[i].get("short_video_info", {}).get("height", 0),
                                "publish_time_stamp":  publish_time_stamp,
                                "publish_time_str":  publish_time_str,
                                "user_name": feeds[i].get("source", "").strip().replace("\n", ""),
                                "user_id": feeds[i].get("openid", ""),
                                "avatar_url": feeds[i].get("bizIcon", ""),
                                "cover_url": feeds[i].get("thumbUrl", ""),
                                "video_url": video_url,
                                "session": session,
                            }
                            for k, v in video_dict.items():
                                Common.logger(log_type).info(f"{k}:{v}")

                            if video_dict["video_id"] == "" \
                                    or video_dict["video_title"] == ""\
                                    or video_dict["video_url"] == "":
                                Common.logger(log_type).info("无效视频\n")
                            elif cls.download_rule(video_dict) is False:
                                Common.logger(log_type).info("不满足抓取规则\n")
                            elif any(str(word) if str(word) in video_title else False for word in cls.get_filter_word(log_type, crawler)) is True:
                                Common.logger(log_type).info("视频已中过滤词\n")
                            elif video_dict["video_id"] in [j for i in Feishu.get_values_batch(log_type, crawler, "ho98Ov") for j in i]:
                                Common.logger(log_type).info("视频已下载\n")
                            elif video_dict["video_id"] in [j for i in Feishu.get_values_batch(log_type, crawler, "20ce0c") for j in i]:
                                Common.logger(log_type).info("视频已下载\n")
                            else:
                                cls.download_publish(log_type, crawler, video_dict, env)
                        except Exception as e:
                            Common.logger(log_type).error(f"抓取单条视频异常:{e}\n")
                except Exception as e:
                    Common.logger(log_type).error(f"抓取第{page}页时异常:{e}\n")

    @classmethod
    def download_publish(cls, log_type, crawler, video_dict, env):
        Common.download_method(log_type, "video", video_dict["video_title"], video_dict["video_url"])
        try:
            if os.path.getsize(f"./videos/{video_dict['video_title']}/video.mp4") == 0:
                # 删除视频文件夹
                shutil.rmtree(f"./videos/{video_dict['video_title']}")
                Common.logger(log_type).info("视频size=0，删除成功\n")
                return
        except FileNotFoundError:
            # 删除视频文件夹
            shutil.rmtree(f"./videos/{video_dict['video_title']}")
            Common.logger(log_type).info("视频文件不存在，删除文件夹成功\n")
            return
        Common.download_method(log_type, "cover", video_dict["video_title"], video_dict["cover_url"])
        with open(f"./videos/{video_dict['video_title']}/info.txt", "a", encoding="utf8") as f_a2:
            f_a2.write(str(video_dict['video_id']) + "\n" +
                       str(video_dict['video_title']) + "\n" +
                       str(video_dict['duration']) + "\n" +
                       str(video_dict['play_cnt']) + "\n" +
                       str(video_dict['comment_cnt']) + "\n" +
                       str(video_dict['like_cnt']) + "\n" +
                       str(video_dict['share_cnt']) + "\n" +
                       f'{video_dict["video_width"]}*{video_dict["video_height"]}' + "\n" +
                       str(video_dict["publish_time_stamp"]) + "\n" +
                       str(video_dict["user_name"]) + "\n" +
                       str(video_dict["avatar_url"]) + "\n" +
                       str(video_dict["video_url"]) + "\n" +
                       str(video_dict["cover_url"]) + "\n" +
                       f"kanyikan-recommend-{int(time.time())}")
        Common.logger("recommend").info("==========视频信息已保存至info.txt==========")

        # 上传视频
        our_video_id = Publish.upload_and_publish(log_type=log_type,
                                                  crawler=crawler,
                                                  strategy="推荐抓取策略",
                                                  our_uid="recommend",
                                                  env=env,
                                                  oss_endpoint="out")
        if env == "dev":
            our_video_link = f"https://testadmin.piaoquantv.com/cms/post-detail/{our_video_id}/info"
        else:
            our_video_link = f"https://admin.piaoquantv.com/cms/post-detail/{our_video_id}/info"
        if our_video_id is None:
            try:
                # 删除视频文件夹
                shutil.rmtree(f"./videos/{video_dict['video_title']}")
                return
            except FileNotFoundError:
                return

        # 保存视频信息到云文档:
        Feishu.insert_columns(log_type, crawler, "20ce0c", "ROWS", 1, 2)
        # 看一看+ ，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(video_dict["video_id"]),
                   str(video_dict["video_title"]),
                   our_video_link,
                   video_dict["play_cnt"],
                   video_dict["comment_cnt"],
                   video_dict["like_cnt"],
                   video_dict["share_cnt"],
                   video_dict["duration"],
                   f'{video_dict["video_width"]}*{video_dict["video_height"]}',
                   video_dict["publish_time_str"],
                   video_dict["user_name"],
                   video_dict["user_id"],
                   video_dict["avatar_url"],
                   video_dict["cover_url"],
                   video_dict["video_url"]]]
        time.sleep(0.5)
        Feishu.update_values(log_type, crawler, "20ce0c", "F2:Z2", values)
        Common.logger(log_type).info("视频信息保存至云文档成功\n")


if __name__ == "__main__":
    print(Kanyikanrecommend.get_filter_word("recommend", "kanyikan"))
    print(int(time.mktime(time.strptime("2021-06-01 00:00:00", "%Y-%m-%d %H:%M:%S"))))
    pass