xiaoniangao_plus_by_ljh

xiaoniangaoplus/xiaoniangaoplus/xiaoniangao_plus_scheduling2.py

@@ -0,0 +1,359 @@
+# -*- coding: utf-8 -*-
+# @Author: luojunhui
+# @Time: 2023/9/27
+import json
+import os
+import sys
+import time
+from hashlib import md5
+
+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.webdriver.common.by import By
+import multiprocessing
+
+sys.path.append(os.getcwd())
+from common.common import Common
+from common.mq import MQ
+from common.public import download_rule, get_config_from_mysql
+from common.scheduling_db import MysqlHelper
+
+
+class XiaoNianGaoPlusRecommend:
+    env = None
+    driver = None
+    log_type = None
+
+    def __init__(self, log_type, crawler, env, rule_dict, our_uid):
+        self.mq = None
+        self.platform = "小年糕"
+        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":
+            chromedriverExecutable = "/Users/a123456/Downloads/chromedriver_V111/chromedriver"
+        else:
+            chromedriverExecutable = "/Users/a123456/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
+        caps = {
+            "platformName": "Android",
+            "devicesName": "Android",
+            "platformVersion": "13",
+            # "udid": "emulator-5554",
+            "appPackage": "com.tencent.mm",
+            "appActivity": ".ui.LauncherUI",
+            "autoGrantPermissions": "true",
+            "noReset": True,
+            "resetkeyboard": True,
+            "unicodekeyboard": True,
+            "showChromedriverLog": True,
+            "printPageSourceOnFailure": True,
+            "recreateChromeDriverSessions": True,
+            "enableWebviewDetailsCollection": True,
+            "setWebContentsDebuggingEnabled": True,
+            "newCommandTimeout": 6000,
+            "automationName": "UiAutomator2",
+            "chromedriverExecutable": chromedriverExecutable,
+            "chromeOptions": {"androidProcess": "com.tencent.mm:appbrand0"},
+        }
+        self.driver = webdriver.Remote("http://localhost:4723/wd/hub", caps)
+        self.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')
+                    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()
+                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)
+        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()
+        time.sleep(5)
+        self.get_videoList()
+        time.sleep(1)
+        self.driver.quit()
+
+    def search_elements(self, xpath):
+        time.sleep(1)
+        windowHandles = self.driver.window_handles
+        for handle in windowHandles:
+            self.driver.switch_to.window(handle)
+            time.sleep(1)
+            try:
+                elements = self.driver.find_elements(By.XPATH, xpath)
+                if elements:
+                    return elements
+            except NoSuchElementException:
+                pass
+
+    def check_to_applet(self, xpath):
+        time.sleep(1)
+        webViews = self.driver.contexts
+        self.driver.switch_to.context(webViews[-1])
+        windowHandles = self.driver.window_handles
+        for handle in windowHandles:
+            self.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')
+                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)
+        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
+
+    def get_video_url(self, 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]}")
+            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])
+            time.sleep(3)
+            Common.logger(self.log_type, self.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"]')
+            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]
+
+    def get_video_info(self, video_element):
+        try:
+            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_user_id = md5(user_name.encode('utf8')).hexdigest()
+
+            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())}"
+            }
+            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}")
+            # Common.logger(self.log_type, self.crawler).info(f"==========分割线==========\n")
+
+            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')
+                # self.swipe_up()
+                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')
+                # self.swipe_up()
+                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}")
+                    # continue
+                    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)
+                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
+                video_dict['video_url'] = video_url
+                Common.logger(self.log_type, self.crawler).info(f"video_url:{video_url}")
+
+                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)
+                # self.swipe_up()
+        except Exception as e:
+            Common.logger(self.log_type, self.crawler).error(f"抓取单条视频异常:{e}\n")
+            # Common.logging(self.log_type, self.crawler, self.env, f"抓取单条视频异常:{e}\n")
+
+    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
+
+        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
+
+        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)
+
+
+def run():
+    rule_dict1 = {"period": {"min": 365, "max": 365},
+                  "duration": {"min": 30, "max": 1800},
+                  "favorite_cnt": {"min": 0, "max": 0},
+                  "videos_cnt": {"min": 10000, "max": 0},
+                  "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)