wangkun 1 年之前
父節點
當前提交
4775e10960
共有 3 個文件被更改,包括 30 次插入27 次删除
  1. 14 14
      main/process.sh
  2. 3 0
      xigua/logs/__init__.py
  3. 13 13
      xigua/xigua_recommend/xigua_recommend_scheduling.py

+ 14 - 14
main/process.sh

@@ -174,20 +174,20 @@ else
   echo "$(date "+%Y-%m-%d %H:%M:%S") 西瓜定向爬虫策略 进程状态正常" >> ${log_path}
 fi
 
-# 西瓜推荐榜爬虫策略
-echo "$(date "+%Y-%m-%d %H:%M:%S") 正在监测 西瓜推荐榜爬虫策略 进程状态" >> ${log_path}
-ps -ef | grep "run_xigua_recommend.py" | grep -v "grep"
-if [ "$?" -eq 1 ];then
-  echo "$(date "+%Y-%m-%d_%H:%M:%S") 异常停止,正在重启!" >> ${log_path}
-  if [ ${env} = "dev" ];then
-    cd ${piaoquan_crawler_dir} && sh main/scheduling_main.sh ./xigua/xigua_main/run_xigua_recommend.py --log_type="recommend" --crawler="xigua" --env="dev" xigua/logs/nohup-recommend.log
-  else
-    cd ${piaoquan_crawler_dir} && /usr/bin/sh main/scheduling_main.sh ./xigua/xigua_main/run_xigua_recommend.py --log_type="recommend" --crawler="xigua" --env="prod" xigua/logs/nohup-recommend.log
-  fi
-  echo "$(date "+%Y-%m-%d %H:%M:%S") 重启完成!" >> ${log_path}
-else
-  echo "$(date "+%Y-%m-%d %H:%M:%S") 西瓜推荐榜爬虫策略 进程状态正常" >> ${log_path}
-fi
+## 西瓜推荐榜爬虫策略
+#echo "$(date "+%Y-%m-%d %H:%M:%S") 正在监测 西瓜推荐榜爬虫策略 进程状态" >> ${log_path}
+#ps -ef | grep "run_xigua_recommend.py" | grep -v "grep"
+#if [ "$?" -eq 1 ];then
+#  echo "$(date "+%Y-%m-%d_%H:%M:%S") 异常停止,正在重启!" >> ${log_path}
+#  if [ ${env} = "dev" ];then
+#    cd ${piaoquan_crawler_dir} && sh main/scheduling_main.sh ./xigua/xigua_main/run_xigua_recommend.py --log_type="recommend" --crawler="xigua" --env="dev" xigua/logs/nohup-recommend.log
+#  else
+#    cd ${piaoquan_crawler_dir} && /usr/bin/sh main/scheduling_main.sh ./xigua/xigua_main/run_xigua_recommend.py --log_type="recommend" --crawler="xigua" --env="prod" xigua/logs/nohup-recommend.log
+#  fi
+#  echo "$(date "+%Y-%m-%d %H:%M:%S") 重启完成!" >> ${log_path}
+#else
+#  echo "$(date "+%Y-%m-%d %H:%M:%S") 西瓜推荐榜爬虫策略 进程状态正常" >> ${log_path}
+#fi
 
 # 西瓜搜索爬虫策略
 if [[ "$time" > "00:00:00" ]] && [[ "$time" < "00:10:00" ]]; then

+ 3 - 0
xigua/logs/__init__.py

@@ -0,0 +1,3 @@
+# -*- coding: utf-8 -*-
+# @Author: wangkun
+# @Time: 2023/5/25

+ 13 - 13
xigua/xigua_recommend/xigua_recommend_scheduling.py

@@ -9,7 +9,6 @@ import shutil
 import string
 import sys
 import time
-from datetime import date, timedelta
 from hashlib import md5
 import requests
 import urllib3
@@ -17,15 +16,13 @@ from requests.adapters import HTTPAdapter
 from selenium import webdriver
 from selenium.webdriver import DesiredCapabilities
 from selenium.webdriver.chrome.service import Service
-
-from common.public import download_rule, get_config_from_mysql
-
 sys.path.append(os.getcwd())
 from common.userAgent import get_random_user_agent
 from common.publish import Publish
 from common.common import Common
 from common.feishu import Feishu
 from common.scheduling_db import MysqlHelper
+from common.public import download_rule, get_config_from_mysql
 
 
 class XiguarecommendScheduling:
@@ -618,30 +615,33 @@ class XiguarecommendScheduling:
                 "channelId": "94349543909",
                 "count": "9",
                 "maxTime": str(int(time.time())),
-                "request_from": "701",
+                # "maxTime": "1683190690",
                 "queryCount": str(queryCount),
+                "_signature": signature,
+                "request_from": "701",
                 "offset": "0",
+                "referrer:": "https://open.weixin.qq.com/",
                 "aid": "1768",
-                "msToken": "j0KQp7ejmMFXXXGniwo32qrgVtLD_a7pAhJ4zzoyD_zTXzjdbNH-G0vJi5niZ0FnS98FnahfvUYf7bm5SpHDsdIx0bLc1DnHcdz8ppI0As0P-T5OmE46H7ejeJTyQHE=",
-                "X-Bogus": "DFSzswVYG40ANtawttFukY/F6qxk",
-                "_signature": signature,
+                "msToken": "XDpSA6_ZPP-gAkkBV-_WRQvNpG20uUUGPwf3E-S-txhznjBcXNbK2sbOuSpF3U7Jki6R9HwLDPeW4Gj7n6PURPTKrKLEs8J-ieFrwXDvMp2DX94ZoMua",
+                # "X-Bogus": "DFSzswVOx7bANt0TtCAcOFm4pIkR",
             }
             headers = {
+                'referer': 'https://www.ixigua.com/',
+                'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/112.0.0.0 Safari/537.36',
                 'authority': 'www.ixigua.com',
                 'accept': 'application/json, text/plain, */*',
                 'accept-language': 'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6',
                 'cache-control': 'no-cache',
-                'cookie': 'MONITOR_WEB_ID=67cb5099-a022-4ec3-bb8e-c4de6ba51dd0; passport_csrf_token=72b2574f3c99f8ba670e42df430218fd; passport_csrf_token_default=72b2574f3c99f8ba670e42df430218fd; sid_guard=c7472b508ea631823ba765a60cf8757f%7C1680867422%7C3024002%7CFri%2C+12-May-2023+11%3A37%3A04+GMT; odin_tt=b893608d4dde2e1e8df8cd5d97a0e2fbeafc4ca762ac72ebef6e6c97e2ed19859bb01d46b4190ddd6dd17d7f9678e1de; SEARCH_CARD_MODE=7168304743566296612_0; support_webp=true; support_avif=false; csrf_session_id=a5355d954d3c63ed1ba35faada452b4d; __ac_signature=_02B4Z6wo00f01G-ByvwAAIDBF08h-UIi.zRvoc5AAH.FLUld9yCjyqpKhLRWUia0dPU0ewdqWjxfXn--vkuavRTCjuIdXseqrNPgbp-ltXUK41RCbVx2UZm8ohx3riithUoZowB5XVCC9bot92; ixigua-a-s=1; s_v_web_id=verify_lhoket5d_0qlKZtzS_YZkf_4Uaj_82mX_j6lRT4PcYJ7A; msToken=j0KQp7ejmMFXXXGniwo32qrgVtLD_a7pAhJ4zzoyD_zTXzjdbNH-G0vJi5niZ0FnS98FnahfvUYf7bm5SpHDsdIx0bLc1DnHcdz8ppI0As0P-T5OmE46H7ejeJTyQHE=; tt_scid=RNxY3L30Tje39GrbDEFHI9xj-6QlojhexB2MwVj.jvAC4gib1X0k5lxTt74CXRcOeba2; ttwid=1%7CHHtv2QqpSGuSu8r-zXF1QoWsvjmNi1SJrqOrZzg-UCY%7C1685006419%7Cadadba21478a3551bc8364ebc49cbb7b6b775479d7c0dad3c65c9812722f4cf7',
+                # 'cookie': 'ttcid=5d8f917a525e46759dc886296bf1111b69; MONITOR_WEB_ID=ad1c8360-d4c9-4fa2-a801-d9fd68dfc1b2; s_v_web_id=verify_lh8vaa6v_VI4RQ0ET_nVbq_4PXw_8mfN_7Xp6wdLOZi08; passport_csrf_token=0e7c6992cb6170c9db034c3696191fff; passport_csrf_token_default=0e7c6992cb6170c9db034c3696191fff; odin_tt=b102690fef38bf07c400e3c69cdc27627701802bdd816fa827e3721c33607c4d2c0cbef09fe99c7d370e4a9e9e11c263; sid_guard=8dec4ecbe52cbdcff99dafe622b586b4%7C1683189144%7C3024002%7CThu%2C+08-Jun-2023+08%3A32%3A26+GMT; uid_tt=1dccbeaf685e24afd018fec335f3151d; uid_tt_ss=1dccbeaf685e24afd018fec335f3151d; sid_tt=8dec4ecbe52cbdcff99dafe622b586b4; sessionid=8dec4ecbe52cbdcff99dafe622b586b4; sessionid_ss=8dec4ecbe52cbdcff99dafe622b586b4; sid_ucp_v1=1.0.0-KGVhZTIxYjFlNzRlZTNhZjk5MjNlNzk2NGRhOWJlYzZiNGI5NzBhMzYKFQiu3d-eqQIQmNvNogYYGCAMOAhACxoCaGwiIDhkZWM0ZWNiZTUyY2JkY2ZmOTlkYWZlNjIyYjU4NmI0; ssid_ucp_v1=1.0.0-KGVhZTIxYjFlNzRlZTNhZjk5MjNlNzk2NGRhOWJlYzZiNGI5NzBhMzYKFQiu3d-eqQIQmNvNogYYGCAMOAhACxoCaGwiIDhkZWM0ZWNiZTUyY2JkY2ZmOTlkYWZlNjIyYjU4NmI0; support_webp=true; support_avif=true; csrf_session_id=9dd5d8287d4f075ae24ff163cd22e51f; msToken=XDpSA6_ZPP-gAkkBV-_WRQvNpG20uUUGPwf3E-S-txhznjBcXNbK2sbOuSpF3U7Jki6R9HwLDPeW4Gj7n6PURPTKrKLEs8J-ieFrwXDvMp2DX94ZoMua; ixigua-a-s=1; tt_scid=UTduWO4ij7cX6YKx23sDuV4zjvFkGFtFk5ZBhEnd1lJ1EZBykStzU7tbWQOSzGdE0fc6; ttwid=1%7C4zaTJmlaHpEa8rAB-KjREdxT3sNBUJWrAzRJnNvqExQ%7C1683198318%7Cffc2eef612caab19a0db93b4cec27e21a6230f9b82ab4bf5b1c6193d082baab1',
                 'pragma': 'no-cache',
-                'referer': 'https://www.ixigua.com/',
-                'sec-ch-ua': '"Microsoft Edge";v="113", "Chromium";v="113", "Not-A.Brand";v="24"',
+                'sec-ch-ua': '"Chromium";v="112", "Google Chrome";v="112", "Not:A-Brand";v="99"',
                 'sec-ch-ua-mobile': '?0',
                 'sec-ch-ua-platform': '"macOS"',
                 'sec-fetch-dest': 'empty',
                 'sec-fetch-mode': 'cors',
                 'sec-fetch-site': 'same-origin',
-                'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/113.0.0.0 Safari/537.36 Edg/113.0.1774.35',
-                'x-secsdk-csrf-token': '00010000000109233e2aa164da5098d92d53a87b346eb042a041e5a93d29fcb47e03f5d4a0e7176258649d64b4ae',
+                # 'tt-anti-token': '95Ny0vj4Q-90dd9b91193b34ce554cc2861439b9629d897723f4d33719b9747d7d18a2ff7c',
+                # 'x-secsdk-csrf-token': '000100000001ecb8f07e247a89e289b3ab55f3c967a8e88f88aa0addb1ddca9d3e36f35d7999175be79b8699c881'
             }
             urllib3.disable_warnings()
             s = requests.session()