kyk zzcl

kanyikan/kanyikan_main/run_kykzzcl_recommend.py

@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+# @Author: wangkun
+# @Time: 2023/6/21
+import argparse
+import random
+from mq_http_sdk.mq_client import *
+from mq_http_sdk.mq_consumer import *
+from mq_http_sdk.mq_exception import MQExceptionBase
+sys.path.append(os.getcwd())
+from common.common import Common
+from common.public import get_consumer, ack_message, task_fun_mq
+from common.scheduling_db import MysqlHelper
+from kanyikan.kanyikan_recommend.kanyikan_recommend_zzcl import KanyikanRecommend
+
+
+def main(log_type, crawler, topic_name, group_id, env):
+    consumer = get_consumer(topic_name, group_id)
+    # 长轮询表示如果Topic没有消息，则客户端请求会在服务端挂起3秒，3秒内如果有消息可以消费则立即返回响应。
+    # 长轮询时间3秒（最多可设置为30秒）。
+    wait_seconds = 30
+    # 一次最多消费3条（最多可设置为16条）。
+    batch = 1
+    Common.logger(log_type, crawler).info(f'{10 * "="}Consume And Ack Message From Topic{10 * "="}\n'
+                                          f'WaitSeconds:{wait_seconds}\n'
+                                          f'TopicName:{topic_name}\n'
+                                          f'MQConsumer:{group_id}')
+    Common.logging(log_type, crawler, env, f'{10 * "="}Consume And Ack Message From Topic{10 * "="}\n'
+                                           f'WaitSeconds:{wait_seconds}\n'
+                                           f'TopicName:{topic_name}\n'
+                                           f'MQConsumer:{group_id}')
+    while True:
+        try:
+            # 长轮询消费消息。
+            recv_msgs = consumer.consume_message(batch, wait_seconds)
+            for msg in recv_msgs:
+                Common.logger(log_type, crawler).info(f"Receive\n"
+                                                      f"MessageId:{msg.message_id}\n"
+                                                      f"MessageBodyMD5:{msg.message_body_md5}\n"
+                                                      f"MessageTag:{msg.message_tag}\n"
+                                                      f"ConsumedTimes:{msg.consumed_times}\n"
+                                                      f"PublishTime:{msg.publish_time}\n"
+                                                      f"Body:{msg.message_body}\n"
+                                                      f"NextConsumeTime:{msg.next_consume_time}\n"
+                                                      f"ReceiptHandle:{msg.receipt_handle}\n"
+                                                      f"Properties:{msg.properties}")
+                Common.logging(log_type, crawler, env, f"Receive\n"
+                                                       f"MessageId:{msg.message_id}\n"
+                                                       f"MessageBodyMD5:{msg.message_body_md5}\n"
+                                                       f"MessageTag:{msg.message_tag}\n"
+                                                       f"ConsumedTimes:{msg.consumed_times}\n"
+                                                       f"PublishTime:{msg.publish_time}\n"
+                                                       f"Body:{msg.message_body}\n"
+                                                       f"NextConsumeTime:{msg.next_consume_time}\n"
+                                                       f"ReceiptHandle:{msg.receipt_handle}\n"
+                                                       f"Properties:{msg.properties}")
+                # ack_mq_message
+                ack_message(log_type=log_type, crawler=crawler, recv_msgs=recv_msgs, consumer=consumer)
+
+                # 处理爬虫业务
+                task_dict = task_fun_mq(msg.message_body)['task_dict']
+                rule_dict = task_fun_mq(msg.message_body)['rule_dict']
+                task_id = task_dict['id']
+                select_user_sql = f"""select * from crawler_user_v3 where task_id={task_id}"""
+                user_list = MysqlHelper.get_values(log_type, crawler, select_user_sql, env, action="")
+                our_uid_list = []
+                for user in user_list:
+                    our_uid_list.append(user["uid"])
+                our_uid = random.choice(our_uid_list)
+                Common.logger(log_type, crawler).info(f"调度任务:{task_dict}")
+                Common.logging(log_type, crawler, env, f"调度任务:{task_dict}")
+                Common.logger(log_type, crawler).info(f"抓取规则:{rule_dict}")
+                Common.logging(log_type, crawler, env, f"抓取规则:{rule_dict}")
+                Common.logger(log_type, crawler).info(f"用户列表:{user_list}\n")
+                Common.logger(log_type, crawler).info(f'开始抓取:{task_dict["taskName"]}\n')
+                Common.logging(log_type, crawler, env, f'开始抓取:{task_dict["taskName"]}\n')
+                KanyikanRecommend.get_videoList(log_type=log_type,
+                                                crawler=crawler,
+                                                rule_dict=rule_dict,
+                                                our_uid=our_uid,
+                                                env=env)
+                Common.del_charles_files(log_type, crawler)
+                Common.logger(log_type, crawler).info('抓取一轮结束\n')
+                Common.logging(log_type, crawler, env, '抓取一轮结束\n')
+
+        except MQExceptionBase as err:
+            # Topic中没有消息可消费。
+            if err.type == "MessageNotExist":
+                Common.logger(log_type, crawler).info(f"No new message! RequestId:{err.req_id}\n")
+                Common.logging(log_type, crawler, env, f"No new message! RequestId:{err.req_id}\n")
+                continue
+
+            Common.logger(log_type, crawler).info(f"Consume Message Fail! Exception:{err}\n")
+            Common.logging(log_type, crawler, env, f"Consume Message Fail! Exception:{err}\n")
+            time.sleep(2)
+            continue
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()  ## 新建参数解释器对象
+    parser.add_argument('--log_type', type=str)  ## 添加参数,注明参数类型
+    parser.add_argument('--crawler')  ## 添加参数
+    parser.add_argument('--topic_name')  ## 添加参数
+    parser.add_argument('--group_id')  ## 添加参数
+    parser.add_argument('--env')  ## 添加参数
+    args = parser.parse_args()  ### 参数赋值，也可以通过终端赋值
+    main(log_type=args.log_type,
+         crawler=args.crawler,
+         topic_name=args.topic_name,
+         group_id=args.group_id,
+         env=args.env)

kanyikan/kanyikan_recommend/kanyikan_recommend_zzcl.py

@@ -0,0 +1,172 @@
+# -*- coding: utf-8 -*-
+# @Author: wangkun
+# @Time: 2023/6/21
+import json
+import os
+import random
+import sys
+import time
+import requests
+import urllib3
+sys.path.append(os.getcwd())
+from common.mq import MQ
+from common.common import Common
+from common.scheduling_db import MysqlHelper
+from common.public import get_config_from_mysql, download_rule
+proxies = {"http": None, "https": None}
+
+
+class KanyikanRecommend:
+    platform = "看一看"
+    strategy = "推荐抓取策略"
+
+    @classmethod
+    def repeat_video(cls, log_type, crawler, video_id, env):
+        sql = f""" select * from crawler_video where platform in ("{crawler}","{cls.platform}") and create_time>='2023-10-09' and out_video_id="{video_id}"; """
+        repeat_video = MysqlHelper.get_values(log_type, crawler, sql, env)
+        return len(repeat_video)
+
+    @classmethod
+    def get_videoList(cls, log_type, crawler, our_uid, rule_dict, env):
+        mq = MQ(topic_name="topic_crawler_etl_" + env)
+        try:
+            Common.logger(log_type, crawler).info(f"正在抓取列表页")
+            Common.logging(log_type, crawler, env, f"正在抓取列表页")
+            session = Common.get_session(log_type, crawler, env)
+            if session is None:
+                time.sleep(1)
+                cls.get_videoList(log_type, crawler, our_uid, rule_dict, env)
+            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, crawler).info("获取视频list时，session过期，随机睡眠 31-50 秒")
+                Common.logging(log_type, crawler, env, "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))
+                cls.get_videoList(log_type, crawler, our_uid, rule_dict, env)
+            elif "items" not in response.json()["data"]:
+                Common.logger(log_type, crawler).info(f"get_feeds:{response.json()}，随机睡眠 1-3 分钟")
+                Common.logging(log_type, crawler, env, 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))
+                cls.get_videoList(log_type, crawler, our_uid, rule_dict, env)
+            feeds = response.json().get("data", {}).get("items", "")
+            if feeds == "":
+                Common.logger(log_type, crawler).info(f"feeds:{feeds}")
+                Common.logging(log_type, crawler, env, f"feeds:{feeds}")
+                return
+            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, crawler).info(f"{k}:{v}")
+                    Common.logging(log_type, crawler, env, f"video_dict:{video_dict}")
+
+                    if video_dict["video_id"] == "" or video_dict["video_title"] == "" or video_dict["video_url"] == "":
+                        Common.logger(log_type, crawler).info("无效视频\n")
+                        Common.logging(log_type, crawler, env, "无效视频\n")
+                    elif download_rule(log_type=log_type, crawler=crawler, video_dict=video_dict, rule_dict=rule_dict) is False:
+                        Common.logger(log_type, crawler).info("不满足抓取规则\n")
+                        Common.logging(log_type, crawler, env, "不满足抓取规则\n")
+                    elif any(str(word) if str(word) in video_dict["video_title"] else False
+                             for word in get_config_from_mysql(log_type=log_type,
+                                                               source=crawler,
+                                                               env=env,
+                                                               text="filter",
+                                                               action="")) is True:
+                        Common.logger(log_type, crawler).info('已中过滤词\n')
+                        Common.logging(log_type, crawler, env, '已中过滤词\n')
+                    elif cls.repeat_video(log_type, crawler, video_dict["video_id"], env) != 0:
+                        Common.logger(log_type, crawler).info('视频已下载\n')
+                        Common.logging(log_type, crawler, env, '视频已下载\n')
+                    else:
+                        video_dict["out_user_id"] = video_dict["user_id"]
+                        video_dict["platform"] = crawler
+                        video_dict["strategy"] = log_type
+                        video_dict["strategy_type"] = "zzcl"
+                        video_dict["out_video_id"] = video_dict["video_id"]
+                        video_dict["width"] = video_dict["video_width"]
+                        video_dict["height"] = video_dict["video_height"]
+                        video_dict["crawler_rule"] = json.dumps(rule_dict)
+                        video_dict["user_id"] = our_uid
+                        video_dict["publish_time"] = video_dict["publish_time_str"]
+
+                        mq.send_msg(video_dict)
+                except Exception as e:
+                    Common.logger(log_type, crawler).error(f"抓取单条视频异常:{e}\n")
+                    Common.logging(log_type, crawler, env, f"抓取单条视频异常:{e}\n")
+        except Exception as e:
+            Common.logger(log_type, crawler).error(f"抓取列表页时异常:{e}\n")
+            Common.logging(log_type, crawler, env, f"抓取列表页时异常:{e}\n")
+
+
+if __name__ == "__main__":
+    print(get_config_from_mysql(log_type="recommend",
+                                source="kanyikan",
+                                env="dev",
+                                text="filter",
+                                action=""))
+    pass