add 小年糕rule

main/process_mq.sh

@@ -64,8 +64,8 @@ elif [ ${crawler} = "xngplus" ] && [ ${log_type} = "recommend" ];then
   log_path=${piaoquan_crawler_dir}main/main_logs/process-mq-$(date +%Y-%m-%d).log
 
 else
-  piaoquan_crawler_dir=/root/piaoquan_crawler/
-  profile_path=/etc/profile
+  piaoquan_crawler_dir=/Users/tzld/Desktop/piaoquan_crawler/
+  profile_path=/.base_profile
   python=python3
   log_path=${piaoquan_crawler_dir}main/main_logs/process-mq-$(date +%Y-%m-%d).log
 
@@ -89,7 +89,7 @@ echo "$(date "+%Y-%m-%d %H:%M:%S") 代码更新完成！" >> ${log_path}
 echo "$(date "+%Y-%m-%d %H:%M:%S") 正在监测 ${crawler}_${log_type} 进程状态" >> ${log_path}
 ps -ef | grep "run_${crawler}_${log_type}.py" | grep -v "grep"
 if [ "$?" -eq 1 ];then
-  echo "$(date "+%Y-%m-%d_%H:%M:%S") 异常停止,正在重启!" >> ${log_path}
+  echo "$(date "+%Y-%m-%d %H:%M:%S") 异常停止,正在重启!" >> ${log_path}
   cd ${piaoquan_crawler_dir} && nohup ${python} -u ${path}/${path}_main/run_${crawler}_${log_type}.py --log_type="${log_type}" --crawler="${path}" --topic_name="${crawler}_${log_type}_${env}" --group_id="${crawler}_${log_type}_${env}" --env="${env}" >> ${path}/logs/${log_type}-shell.log 2>&1 &
   echo "$(date "+%Y-%m-%d %H:%M:%S") 重启完成!" >> ${log_path}
 else

main/process_offline.sh

@@ -143,7 +143,7 @@ if [[ "$time" > "07:00:00"  &&  "$time" < "08:59:59" || "$time" > "19:00:00"  &&
     echo "$(date "+%Y-%m-%d %H:%M:%S") 小年糕-rule 小程序爬虫, 异常停止, 正在重启!" >> ${log_path}
     adb forward --remove-all
     cd ${piaoquan_crawler_dir}
-    nohup python3 -u xiaoniangaoplus/xiaoniangaoplus_main/run_xngrule_recommend.py --log_type="recommend" --crawler="xiaoniangaorule" --env=${env} >>xiaoniangaoplus/logs/nohup-recommend.log 2>&1 &
+    nohup python3 -u xiaoniangaoplus/xiaoniangaoplus_main/run_xngrule_recommend.py --log_type="recommend" --crawler="xiaoniangaoplus" --env=${env} >>xiaoniangaoplus/logs/nohup-recommend.log 2>&1 &
     echo "$(date "+%Y-%m-%d %H:%M:%S") 重启完成!" >> ${log_path}
   else
     echo "$(date "+%Y-%m-%d %H:%M:%S") 小年糕-rule 程序爬虫, 进程状态正常" >> ${log_path}

xiaoniangaoplus/xiaoniangaoplus/xiaoniangao_plus_rule.py

@@ -1,18 +1,18 @@
 # -*- coding: utf-8 -*-
-# @Time: 2023/10/30
+# @Time: 2023/10/31
 import json
 import os
+import requests
 import sys
 import time
 from hashlib import md5
 
-import requests
 from appium import webdriver
 from appium.webdriver.extensions.android.nativekey import AndroidKey
+from appium.webdriver.webdriver import WebDriver
 from bs4 import BeautifulSoup
-from selenium.common.exceptions import NoSuchElementException
+from selenium.common import NoSuchElementException
 from selenium.webdriver.common.by import By
-import multiprocessing
 
 sys.path.append(os.getcwd())
 from common.common import Common
@@ -20,43 +20,31 @@ from common.mq import MQ
 from common.public import download_rule, get_config_from_mysql
 from common.scheduling_db import MysqlHelper
 
-
 def get_redirect_url(url):
     res = requests.get(url, allow_redirects=False)
     if res.status_code == 302 or res.status_code == 301:
         return res.headers['Location']
     else:
         return url
-
 class XiaoNianGaoPlusRecommend:
-    env = None
-    driver = None
-    log_type = None
+    platform = "小年糕-rule"
+    download_cnt = 0
+    element_list = []
+    i = 0
 
-    def __init__(self, log_type, crawler, env, rule_dict, our_uid):
-        self.mq = None
-        self.platform = "小年糕-rule"
-        self.download_cnt = 0
-        self.element_list = []
-        self.count = 0
-        self.swipe_count = 0
-        self.log_type = log_type
-        self.crawler = crawler
-        self.env = env
-        self.rule_dict = rule_dict
-        self.our_uid = our_uid
-        if self.env == "dev":
+    @classmethod
+    def start_wechat(cls, log_type, crawler, env, rule_dict, our_uid):
+        if env == "dev":
             chromedriverExecutable = "/Users/tzld/Downloads/chromedriver_v111/chromedriver"
         else:
             chromedriverExecutable = "/Users/tzld/Downloads/chromedriver_v111/chromedriver"
 
-        Common.logger(self.log_type, self.crawler).info("启动微信")
-        Common.logging(self.log_type, self.crawler, self.env, '启动微信')
-        # 微信的配置文件
+        Common.logger(log_type, crawler).info("启动微信")
+        Common.logging(log_type, crawler, env, '启动微信')
         caps = {
             "platformName": "Android",
             "devicesName": "Android",
-            # "platformVersion": "13",
+            # "platformVersion": "11",
             # "udid": "emulator-5554",
             "appPackage": "com.tencent.mm",
             "appActivity": ".ui.LauncherUI",
@@ -74,302 +62,293 @@ class XiaoNianGaoPlusRecommend:
             "chromedriverExecutable": chromedriverExecutable,
             "chromeOptions": {"androidProcess": "com.tencent.mm:appbrand0"},
         }
-        self.driver = webdriver.Remote("http://localhost:4723/wd/hub", caps)
-        self.driver.implicitly_wait(30)
+        driver = webdriver.Remote("http://localhost:4723/wd/hub", caps)
+        driver.implicitly_wait(30)
 
         for i in range(120):
             try:
-                if self.driver.find_elements(By.ID, "com.tencent.mm:id/f2s"):
-                    Common.logger(self.log_type, self.crawler).info("微信启动成功")
-                    Common.logging(self.log_type, self.crawler, self.env, '微信启动成功')
+                if driver.find_elements(By.ID, 'com.tencent.mm:id/f2s'):
+                    Common.logger(log_type, crawler).info("微信启动成功")
+                    Common.logging(log_type, crawler, env, '微信启动成功')
                     break
-                elif self.driver.find_element(By.ID, "com.android.systemui:id/dismiss_view"):
-                    Common.logger(self.log_type, self.crawler).info("发现并关闭系统下拉菜单")
-                    Common.logging(self.log_type, self.crawler, self.env, '发现并关闭系统下拉菜单')
-                    self.driver.find_element(By.ID, "com.android.system:id/dismiss_view").click()
+                elif driver.find_element(By.ID, "com.android.systemui:id/dismiss_view"):
+                    Common.logger(log_type, crawler).info("发现并关闭系统下拉菜单")
+                    Common.logging(log_type, crawler, env, '发现并关闭系统下拉菜单')
+                    driver.find_element(By.ID, "com.android.system:id/dismiss_view").click()
                 else:
                     pass
             except NoSuchElementException:
                 time.sleep(1)
 
-        Common.logger(self.log_type, self.crawler).info("下滑，展示小程序选择面板")
-        # Common.logging(self.log_type, self.crawler, self.env, '下滑，展示小程序选择面板')
-        size = self.driver.get_window_size()
-        self.driver.swipe(int(size['width'] * 0.5), int(size['height'] * 0.2),
-                          int(size['width'] * 0.5), int(size['height'] * 0.8), 200)
+        Common.logger(log_type, crawler).info("下滑，展示小程序选择面板")
+        Common.logging(log_type, crawler, env, '下滑，展示小程序选择面板')
+        size = driver.get_window_size()
+        driver.swipe(int(size['width'] * 0.5), int(size['height'] * 0.2),
+                     int(size['width'] * 0.5), int(size['height'] * 0.8), 200)
         time.sleep(1)
-        Common.logger(self.log_type, self.crawler).info('打开小程序"小年糕+"')
-        # Common.logging(self.log_type, self.crawler, self.env, '打开小程序"小年糕+"')
-        self.driver.find_elements(By.XPATH, '//*[@text="小年糕+"]')[-1].click()
+        Common.logger(log_type, crawler).info('打开小程序"小年糕+"')
+        Common.logging(log_type, crawler, env, '打开小程序"小年糕+"')
+        driver.find_elements(By.XPATH, '//*[@text="小年糕+"]')[-1].click()
         time.sleep(5)
-        self.get_videoList()
+
+        cls.get_videoList(log_type, crawler, driver, env, rule_dict, our_uid)
+
         time.sleep(1)
-        self.driver.quit()
+        driver.quit()
 
-    def search_elements(self, xpath):
+    @classmethod
+    def search_elements(cls, driver: WebDriver, xpath):
         time.sleep(1)
-        windowHandles = self.driver.window_handles
+        windowHandles = driver.window_handles
         for handle in windowHandles:
-            self.driver.switch_to.window(handle)
+            driver.switch_to.window(handle)
             time.sleep(1)
             try:
-                elements = self.driver.find_elements(By.XPATH, xpath)
+                elements = driver.find_elements(By.XPATH, xpath)
                 if elements:
                     return elements
             except NoSuchElementException:
                 pass
 
-    def check_to_applet(self, xpath):
+    @classmethod
+    def check_to_applet(cls, log_type, crawler, env, driver: WebDriver):
         time.sleep(1)
-        webViews = self.driver.contexts
-        self.driver.switch_to.context(webViews[-1])
-        windowHandles = self.driver.window_handles
+        webViews = driver.contexts
+        Common.logger(log_type, crawler).info(f"webViews:{webViews}")
+        Common.logging(log_type, crawler, env, f"webViews:{webViews}")
+        driver.switch_to.context(webViews[1])
+        windowHandles = driver.window_handles
         for handle in windowHandles:
-            self.driver.switch_to.window(handle)
+            driver.switch_to.window(handle)
             time.sleep(1)
             try:
-                self.driver.find_element(By.XPATH, xpath)
-                Common.logger(self.log_type, self.crawler).info("切换到WebView成功\n")
-                Common.logging(self.log_type, self.crawler, self.env, '切换到WebView成功\n')
+                driver.find_element(By.XPATH, '//*[@class="tab-bar--tab tab-bar--tab-selected"]')
+                Common.logger(log_type, crawler).info("切换到小程序成功\n")
+                Common.logging(log_type, crawler, env, '切换到小程序成功\n')
                 return
             except NoSuchElementException:
                 time.sleep(1)
 
-    def repeat_video(self, video_id):
-        sql = f""" select * from crawler_video where platform in ("众妙音信", "刚刚都传", "吉祥幸福", "知青天天看", "zhufuquanzi", "祝福圈子", "haitunzhufu", "海豚祝福", "小年糕") and out_video_id="{video_id}"; """
-        repeat_video = MysqlHelper.get_values(self.log_type, self.crawler, sql, self.env)
+    @classmethod
+    def repeat_video(cls, log_type, crawler, video_id, env):
+        sql = f""" select * from crawler_video where platform in ("{crawler}","{cls.platform}") and out_video_id="{video_id}"; """
+
+        repeat_video = MysqlHelper.get_values(log_type, crawler, sql, env)
         return len(repeat_video)
 
-    def swipe_up(self):
-        self.search_elements('//*[@class="list-list--list"]')
-        size = self.driver.get_window_size()
-        self.driver.swipe(int(size["width"] * 0.5), int(size["height"] * 0.8),
-                          int(size["width"] * 0.5), int(size["height"] * 0.442), 200)
-        self.swipe_count += 1
+    @classmethod
+    def swipe_up(cls, driver: WebDriver):
+        cls.search_elements(driver, '//*[@class="videoplay"]')
+        size = driver.get_window_size()
+        driver.swipe(int(size["width"] * 0.7), int(size["height"] * 0.8),
+                     int(size["width"] * 0.7), int(size["height"] * 0.4), 200)
 
-    def get_video_url(self, video_title_element):
+    @classmethod
+    def get_video_url(cls, log_type, crawler, driver: WebDriver, video_title_element):
         for i in range(3):
-            self.search_elements('//*[@class="list-list--list"]')
-            Common.logger(self.log_type, self.crawler).info(f"video_title_element:{video_title_element[0]}")
+            cls.search_elements(driver, '//*[@class="list-list--list"]')
+            Common.logger(log_type, crawler).info(f"video_title_element:{video_title_element[0]}")
             time.sleep(1)
-            Common.logger(self.log_type, self.crawler).info("滑动标题至可见状态")
-            self.driver.execute_script("arguments[0].scrollIntoView({block:'center',inline:'center'});",
-                                       video_title_element[0])
+            Common.logger(log_type, crawler).info("滑动标题至可见状态")
+            driver.execute_script("arguments[0].scrollIntoView({block:'center',inline:'center'});", video_title_element[0])
             time.sleep(3)
-            Common.logger(self.log_type, self.crawler).info("点击标题")
+            Common.logger(log_type, crawler).info("点击标题")
             video_title_element[0].click()
-            self.check_to_applet(xpath=r'//wx-video[@class="dynamic-index--video-item dynamic-index--video"]')
-            Common.logger(self.log_type, self.crawler).info("点击标题完成")
-            time.sleep(10)
-            video_url_elements = self.search_elements(
-                '//wx-video[@class="dynamic-index--video-item dynamic-index--video"]')
+            # driver.execute_script("arguments[0].click();", video_title_element[0])
+            Common.logger(log_type, crawler).info("点击标题完成")
+            time.sleep(1)
+            video_url_elements = cls.search_elements(driver, '//wx-video[@class="dynamic-index--video-item dynamic-index--video"]')
             if video_url_elements:
                 return video_url_elements[0].get_attribute("src")
 
-    def parse_detail(self, index):
-        page_source = self.driver.page_source
-        soup = BeautifulSoup(page_source, 'html.parser')
-        soup.prettify()
-        video_list = soup.findAll(name="wx-view", attrs={"class": "expose--adapt-parent"})
-        element_list = [i for i in video_list][index:]
-        return element_list[0]
+    @classmethod
+    def get_videoList(cls, log_type, crawler, driver: WebDriver, env, rule_dict, our_uid):
+        mq = MQ(topic_name="topic_crawler_etl_" + env)
+        driver.implicitly_wait(20)
+        # 鼠标左键点击， 1为x坐标， 2为y坐标
+        cls.check_to_applet(log_type=log_type, crawler=crawler, env=env, driver=driver)
+        time.sleep(1)
 
-    def get_video_info_2(self, video_element):
-        Common.logger(self.log_type, self.crawler).info(f"本轮已抓取{self.download_cnt}条视频\n")
-        # Common.logging(self.log_type, self.crawler, self.env, f"本轮已抓取{self.download_cnt}条视频\n")
-        if self.download_cnt >= int(self.rule_dict.get("videos_cnt", {}).get("min", 10)):
-            self.count = 0
-            self.download_cnt = 0
-            self.element_list = []
-            return
-        self.count += 1
-        Common.logger(self.log_type, self.crawler).info(f"第{self.count}条视频")
-        # Common.logging(self.log_type, self.crawler, self.env, f"第{self.count}条视频")
-        # 标题
-        video_title = video_element.find("wx-view", class_="dynamic--title").text
-        # 播放量字符串
-        play_str = video_element.find("wx-view", class_="dynamic--views").text
-        info_list = video_element.findAll("wx-view", class_="dynamic--commerce-btn-text")
-        # 点赞数量
-        like_str = info_list[1].text
-        # 评论数量
-        comment_str = info_list[2].text
-        # 视频时长
-        duration_str = video_element.find("wx-view", class_="dynamic--duration").text
-        user_name = video_element.find("wx-view", class_="dynamic--nick-top").text
-        # 头像 URL
-        avatar_url = video_element.find("wx-image", class_="avatar--avatar")["src"]
-        # 封面 URL
-        cover_url = video_element.find("wx-image", class_="dynamic--bg-image")["src"]
-        play_cnt = int(play_str.replace("+", "").replace("次播放", ""))
-        duration = int(duration_str.split(":")[0].strip()) * 60 + int(duration_str.split(":")[-1].strip())
-        if "点赞" in like_str:
-            like_cnt = 0
-        elif "万" in like_str:
-            like_cnt = int(like_str.split("万")[0]) * 10000
-        else:
-            like_cnt = int(like_str)
-        if "评论" in comment_str:
-            comment_cnt = 0
-        elif "万" in comment_str:
-            comment_cnt = int(comment_str.split("万")[0]) * 10000
-        else:
-            comment_cnt = int(comment_str)
-        out_video_id = md5(video_title.encode('utf8')).hexdigest()
-        out_video_id = out_video_id+"rule"
-        out_user_id = md5(user_name.encode('utf8')).hexdigest()
-        out_user_id = out_user_id+"rule"
-        video_dict = {
-            "video_title": video_title,
-            "video_id": out_video_id,
-            "duration_str": duration_str,
-            "duration": duration,
-            "play_str": play_str,
-            "play_cnt": play_cnt,
-            "like_str": like_str,
-            "like_cnt": like_cnt,
-            "comment_cnt": comment_cnt,
-            "share_cnt": 0,
-            "user_name": user_name,
-            "user_id": out_user_id,
-            'publish_time_stamp': int(time.time()),
-            'publish_time_str': time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(int(time.time()))),
-            "avatar_url": avatar_url,
-            "cover_url": cover_url,
-            "session": f"xiaoniangao-{int(time.time())}"
-        }
-        video_percent = '%.3f' % (like_cnt / play_cnt)
-        if play_cnt < 100000:
-            if float(video_percent) <= 0.017:
-                Common.logger(self.log_type, self.crawler).info(f"不符合条件：点赞/播放-{video_percent},播放量-{play_cnt}\n")
+        page = 0
+        while True:
+            if cls.search_elements(driver, '//*[@class="list-list--list"]') is None:
+                Common.logger(log_type, crawler).info("窗口已销毁\n")
+                Common.logging(log_type, crawler, env, '窗口已销毁\n')
+                cls.i = 0
+                cls.download_cnt = 0
+                cls.element_list = []
                 return
-        for k, v in video_dict.items():
-            Common.logger(self.log_type, self.crawler).info(f"{k}:{v}")
-        Common.logging(self.log_type, self.crawler, self.env, f"video_dict:{video_dict}")
 
-        if video_title is None or cover_url is None:
-            Common.logger(self.log_type, self.crawler).info("无效视频\n")
-            Common.logging(self.log_type, self.crawler, self.env, '无效视频\n')
-            # self.swipe_up()
-            time.sleep(0.5)
-        elif download_rule(log_type=self.log_type,
-                           crawler=self.crawler,
-                           video_dict=video_dict,
-                           rule_dict=self.rule_dict) is False:
-            Common.logger(self.log_type, self.crawler).info("不满足抓取规则\n")
-            Common.logging(self.log_type, self.crawler, self.env, "不满足抓取规则\n")
-            # self.swipe_up()
-            time.sleep(0.5)
-        elif any(str(word) if str(word) in video_dict["video_title"] else False
-                 for word in get_config_from_mysql(log_type=self.log_type,
-                                                   source=self.crawler,
-                                                   env=self.env,
-                                                   text="filter",
-                                                   action="")) is True:
-            Common.logger(self.log_type, self.crawler).info('已中过滤词\n')
-            Common.logging(self.log_type, self.crawler, self.env, '已中过滤词\n')
-            time.sleep(0.5)
-        elif self.repeat_video(out_video_id) != 0:
-            Common.logger(self.log_type, self.crawler).info('视频已下载\n')
-            Common.logging(self.log_type, self.crawler, self.env, '视频已下载\n')
-            time.sleep(5)
-        else:
-            video_title_element = self.search_elements(f'//*[contains(text(), "{video_title}")]')
-            if video_title_element is None:
-                Common.logger(self.log_type, self.crawler).warning(
-                    f"未找到该视频标题的element:{video_title_element}")
-                Common.logging(self.log_type, self.crawler, self.env,
-                               f"未找到该视频标题的element:{video_title_element}")
-                return
-            Common.logger(self.log_type, self.crawler).info("点击标题，进入视频详情页")
-            Common.logging(self.log_type, self.crawler, self.env, "点击标题，进入视频详情页")
-            video_url = self.get_video_url(video_title_element)
-            video_url = get_redirect_url(video_url)
-            if video_url is None:
-                Common.logger(self.log_type, self.crawler).info("未获取到视频播放地址\n")
-                self.driver.press_keycode(AndroidKey.BACK)
-                time.sleep(5)
-                return
+            cls.swipe_up(driver)
 
-            video_mid_elements = self.search_elements("//wx-view[@class='bar--navBar-content-capsule']")
-            mid = 0
-            if video_mid_elements:
-                mid = int(video_mid_elements[0].get_attribute("data-mid"))
-            video_dict['profile_id'] = mid
-            video_dict['video_url'] = video_url
-            Common.logger(self.log_type, self.crawler).info(f"video_url:{video_url}")
+            page_source = driver.page_source
+            soup = BeautifulSoup(page_source, 'html.parser')
+            soup.prettify()
 
-            video_dict["platform"] = self.crawler
-            video_dict["strategy"] = self.log_type
-            video_dict["out_video_id"] = video_dict["video_id"]
-            video_dict["crawler_rule"] = json.dumps(self.rule_dict)
-            video_dict["user_id"] = self.our_uid
-            video_dict["publish_time"] = video_dict["publish_time_str"]
-            self.mq.send_msg(video_dict)
-            # print(video_dict)
-            self.download_cnt += 1
-            self.driver.press_keycode(AndroidKey.BACK)
-            time.sleep(5)
+            video_list_elements = soup.findAll("wx-view", class_="expose--adapt-parent")
+            video_list_elements = list(set(video_list_elements).difference(set(cls.element_list)))
+            cls.element_list = list(set(video_list_elements) | set(cls.element_list))
+            Common.logger(log_type, crawler).info(f"正在抓取第{page + 1}页，共:{len(video_list_elements)}条视频")
+            Common.logging(log_type, crawler, env, f"正在抓取第{page + 1}页，共:{len(video_list_elements)}条视频")
 
-    def get_video_info(self, video_element):
-        try:
-            self.get_video_info_2(video_element)
-        except Exception as e:
-            Common.logger(self.log_type, self.crawler).error(f"抓取单条视频异常:{e}\n")
+            if len(video_list_elements) == 0:
+                for i in range(10):
+                    Common.logger(log_type, crawler).info(f"向上滑动第{i + 1}次")
+                    cls.swipe_up(driver)
+                    time.sleep(0.5)
+                continue
 
-    def get_videoList(self):
-        self.mq = MQ(topic_name="topic_crawler_etl_" + self.env)
-        self.driver.implicitly_wait(20)
-        # 切换到 web_view
-        self.check_to_applet(xpath='//*[@class="tab-bar--tab tab-bar--tab-selected"]')
-        print("切换到 webview 成功")
-        time.sleep(1)
-        page = 0
-        if self.search_elements('//*[@class="list-list--list"]') is None:
-            Common.logger(self.log_type, self.crawler).info("窗口已销毁\n")
-            Common.logging(self.log_type, self.crawler, self.env, '窗口已销毁\n')
-            self.count = 0
-            self.download_cnt = 0
-            self.element_list = []
-            return
+            for i, video_element in enumerate(video_list_elements):
+                try:
 
-        print("开始获取视频信息")
-        for i in range(50):
-            print("下滑{}次".format(i))
-            element = self.parse_detail(i)
-            self.get_video_info(element)
-            self.swipe_up()
-            time.sleep(1)
-            if self.swipe_count > 100:
-                return
+                    Common.logger(log_type, crawler).info(f"本轮已抓取{cls.download_cnt}条视频\n")
+                    Common.logging(log_type, crawler, env, f"本轮已抓取{cls.download_cnt}条视频\n")
+                    if cls.download_cnt >= int(rule_dict.get("videos_cnt", {}).get("min", 10)):
+                        cls.i = 0
+                        cls.download_cnt = 0
+                        cls.element_list = []
+                        return
+                    cls.i += 1
+                    Common.logger(log_type, crawler).info(f"第{cls.i}条视频")
+                    Common.logging(log_type, crawler, env, f"第{cls.i}条视频")
 
-        print("下滑完成")
-        # time.sleep(100)
-        Common.logger(self.log_type, self.crawler).info("已抓取完一组，休眠 5 秒\n")
-        Common.logging(self.log_type, self.crawler, self.env, "已抓取完一组，休眠 5 秒\n")
-        time.sleep(5)
+                    video_title = video_element.find("wx-view", class_="dynamic--title").text
+                    # 播放量字符串
+                    play_str = video_element.find("wx-view", class_="dynamic--views").text
+                    info_list = video_element.findAll("wx-view", class_="dynamic--commerce-btn-text")
+                    # 点赞数量
+                    like_str = info_list[1].text
+                    # 评论数量
+                    comment_str = info_list[2].text
+                    # 视频时长
+                    duration_str = video_element.find("wx-view", class_="dynamic--duration").text
+                    user_name = video_element.find("wx-view", class_="dynamic--nick-top").text
+                    # 头像 URL
+                    avatar_url = video_element.find("wx-image", class_="avatar--avatar")["src"]
+                    # 封面 URL
+                    cover_url = video_element.find("wx-image", class_="dynamic--bg-image")["src"]
+                    play_cnt = int(play_str.replace("+", "").replace("次播放", ""))
+                    duration = int(duration_str.split(":")[0].strip()) * 60 + int(duration_str.split(":")[-1].strip())
+                    if "点赞" in like_str:
+                        like_cnt = 0
+                    elif "万" in like_str:
+                        like_cnt = int(like_str.split("万")[0]) * 10000
+                    else:
+                        like_cnt = int(like_str)
+                    if "评论" in comment_str:
+                        comment_cnt = 0
+                    elif "万" in comment_str:
+                        comment_cnt = int(comment_str.split("万")[0]) * 10000
+                    else:
+                        comment_cnt = int(comment_str)
+                    out_video_id = md5(video_title.encode('utf8')).hexdigest()
+                    out_video_id = out_video_id + "rule"
+                    out_user_id = md5(user_name.encode('utf8')).hexdigest()
+                    out_user_id = out_user_id + "rule"
+                    video_dict = {
+                        "video_title": video_title,
+                        "video_id": out_video_id,
+                        "duration_str": duration_str,
+                        "duration": duration,
+                        "play_str": play_str,
+                        "play_cnt": play_cnt,
+                        "like_str": like_str,
+                        "like_cnt": like_cnt,
+                        "comment_cnt": comment_cnt,
+                        "share_cnt": 0,
+                        "user_name": user_name,
+                        "user_id": out_user_id,
+                        'publish_time_stamp': int(time.time()),
+                        'publish_time_str': time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(int(time.time()))),
+                        "avatar_url": avatar_url,
+                        "cover_url": cover_url,
+                        "session": f"xiaoniangao-{int(time.time())}"
+                    }
+                    video_percent = '%.3f' % (like_cnt / play_cnt)
+                    if play_cnt < 100000:
+                        if float(video_percent) <= 0.017:
+                            Common.logger(log_type, crawler).info(
+                                f"不符合条件：点赞/播放-{video_percent},播放量-{play_cnt}\n")
+                            continue
+                    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}")
+                    # Common.logger(log_type, crawler).info(f"==========分割线==========\n")
+
+                    if video_title is None or cover_url is None:
+                        Common.logger(log_type, crawler).info("无效视频\n")
+                        Common.logging(log_type, crawler, env, '无效视频\n')
+                        cls.swipe_up(driver)
+                        time.sleep(0.5)
+                    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")
+                        cls.swipe_up(driver)
+                        time.sleep(0.5)
+                    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')
+                        cls.swipe_up(driver)
+                        time.sleep(0.5)
+                    elif cls.repeat_video(log_type, crawler, out_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')
+                        cls.swipe_up(driver)
+                        time.sleep(5)
+                    else:
+                        video_title_element = cls.search_elements(driver, f'//*[contains(text(), "{video_title}")]')
+                        if video_title_element is None:
+                            Common.logger(log_type, crawler).warning(f"未找到该视频标题的element:{video_title_element}")
+                            Common.logging(log_type, crawler, env, f"未找到该视频标题的element:{video_title_element}")
+                            continue
+                        Common.logger(log_type, crawler).info("点击标题，进入视频详情页")
+                        Common.logging(log_type, crawler, env, "点击标题，进入视频详情页")
+                        video_url = cls.get_video_url(log_type, crawler, driver, video_title_element)
+                        video_url = get_redirect_url(video_url)
 
+                        if video_url is None:
+                            Common.logger(log_type, crawler).info("未获取到视频播放地址\n")
+                            driver.press_keycode(AndroidKey.BACK)
+                            time.sleep(5)
+                            continue
+                        video_mid_elements = cls.search_elements("//wx-view[@class='bar--navBar-content-capsule']")
+                        mid = 0
+                        if video_mid_elements:
+                            mid = int(video_mid_elements[0].get_attribute("data-mid"))
+                        video_dict['profile_id'] = mid
+                        video_dict['video_url'] = video_url
+                        Common.logger(log_type, crawler).info(f"video_url:{video_url}")
 
-def run():
+                        video_dict["platform"] = crawler
+                        video_dict["strategy"] = log_type
+                        video_dict["out_video_id"] = video_dict["video_id"]
+                        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)
+                        cls.download_cnt += 1
+                        driver.press_keycode(AndroidKey.BACK)
+                        time.sleep(5)
+                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")
+            Common.logger(log_type, crawler).info("已抓取完一组，休眠 5 秒\n")
+            Common.logging(log_type, crawler, env, "已抓取完一组，休眠 5 秒\n")
+            time.sleep(5)
+            page += 1
+
+
+
+if __name__ == "__main__":
     rule_dict1 = {"period": {"min": 0, "max": 365},
                   "duration": {"min": 0, "max": 1800},
                   "favorite_cnt": {"min": 0, "max": 0},
-                  "videos_cnt": {"min": 1, "max": 0},
+                  "videos_cnt": {"min": 10, "max": 20},
                   "share_cnt": {"min": 0, "max": 0}}
-    XiaoNianGaoPlusRecommend("recommend", "xiaoniangao", "dev", rule_dict1, 6267141)
-
-
-if __name__ == "__main__":
-    process = multiprocessing.Process(
-        target=run
-    )
-    process.start()
-    while True:
-        if not process.is_alive():
-            print("正在重启")
-            process.terminate()
-            time.sleep(60)
-            os.system("adb forward --remove-all")
-            process = multiprocessing.Process(target=run)
-            process.start()
-        time.sleep(60)
+    XiaoNianGaoPlusRecommend.start_wechat("recommend", "xiaoniangao", "dev", rule_dict1, 6267141)

xiaoniangaoplus/xiaoniangaoplus_main/run_xngrule_recommend.py

@@ -1,159 +1,142 @@
 # -*- coding: utf-8 -*-
-# @Author: luojunhui
-# @Time: 2023/9/27
+# @Time: 2023/10/24
 import argparse
 import random
+import multiprocessing
+
 from mq_http_sdk.mq_client import *
 from mq_http_sdk.mq_consumer import *
 from mq_http_sdk.mq_exception import MQExceptionBase
-import multiprocessing
 
 
 sys.path.append(os.getcwd())
-from common.public import get_consumer, ack_message, task_fun_mq, get_rule_from_mysql
 from common.common import Common
+from common.public import get_consumer, ack_message, task_fun_mq
 from common.scheduling_db import MysqlHelper
 from xiaoniangaoplus.xiaoniangaoplus.xiaoniangao_plus_rule import XiaoNianGaoPlusRecommend
 
 
 
 def run(args1, args2, args3, args4, args5):
-    XiaoNianGaoPlusRecommend(
-        log_type=args1,
-        crawler=args2,
-        env=args3,
-        rule_dict=args4,
-        our_uid=args5
-    )
-
-
-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:
-                xng_play_start_time = int(time.time())
-                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')
-                new_r = get_rule_from_mysql(task_id=task_id, log_type=log_type, crawler=crawler, env=env)
-                # Common.logger(log_type, crawler).info(f'rule_dict:{new_r}\n')
-                r_d = {}
-                for item in new_r:
-                    for k, val in item.items():
-                        r_d[k] = val
-                Common.logger(log_type, crawler).info(f"抓取规则:{r_d}")
-                Common.logging(log_type, crawler, env, f"抓取规则:{r_d}")
-                process = multiprocessing.Process(
-                    target=run,
-                    args=(log_type, crawler, env, r_d, our_uid)
-                )
-                process.start()
-                print("进程开始")
-
-                while True:
-                    if not process.is_alive():
-                        print("正在重启")
-                        process.terminate()
-                        os.system("adb forward --remove-all")
+    XiaoNianGaoPlusRecommend.start_wechat(log_type=args1,
+                               crawler=args2,
+                               rule_dict=args3,
+                               our_uid=args4,
+                               env=args5)
+
+
+class XngRuleMain:
+    @classmethod
+    def main(cls, 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.logging(log_type, crawler, env, 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')
+
+                    process = multiprocessing.Process(
+                        target=run,
+                        args=(log_type, crawler, rule_dict, our_uid, env)
+                    )
+                    process.start()
+                    print("进程开始")
+
+                    for i in range(10):
+                        if not process.is_alive():
+
+                            print("进程异常，准备重启")
+                            process.terminate()
+                            os.system("adb forward --remove-all")
+                            time.sleep(60)
+                            process = multiprocessing.Process(
+                                target=run,
+                                args=(log_type, crawler, rule_dict, our_uid, env)
+                            )
+                            process.start()
                         time.sleep(60)
-                        new_r = get_rule_from_mysql(task_id=task_id, log_type=log_type, crawler=crawler, env=env)
-                        r_d = {}
-                        for item in new_r:
-                            for k, val in item.items():
-                                r_d[k] = val
-                        Common.logger(log_type, crawler).info(f'抓取规则:{r_d}')
-                        Common.logging(log_type, crawler, env, f"抓取规则:{r_d}")
-                        process = multiprocessing.Process(target=run, args=(log_type, crawler, env, r_d, our_uid))
-                        process.start()
-                    time.sleep(60)
-                # XiaoNianGaoPlusRecommend.start_wechat(log_type=log_type,
-                #                                       crawler=crawler,
-                #                                       rule_dict=rule_dict,
-                #                                       our_uid=our_uid,
-                #                                       env=env)git
-                # Common.del_logs(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')
-                # xng_play_end_time = int(time.time())
-                # xng_play_duration = xng_play_start_time - xng_play_end_time
-                # Common.logger(log_type, crawler).info(f"duration {xng_play_duration}")
-                # Common.logging(log_type, crawler, env, f"duration {xng_play_duration}")
-        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")
+
+
+                    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
 
-            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)
+
+# 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
+#     XngRuleMain.main(log_type=args.log_type,
+#                               crawler=args.crawler,
+#                               topic_name=args.topic_name,
+#                               group_id=args.group_id,
+#                               env=args.env)