Merge branch 'zhangyongbak'

huanhuanxixizhufudao/huanhuanxixizhufudao_main/run_hhxxzfd_recommend.py

@@ -0,0 +1,161 @@
+# -*- coding: utf-8 -*-
+# @Author:
+# @Time: 2023/9/7
+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
+
+
+sys.path.append(os.getcwd())
+from common.common import Common
+from common.public import get_consumer, ack_message, task_fun_mq, get_rule_from_mysql
+from common.scheduling_db import MysqlHelper
+from huanhuanxixizhufudao.huanhuanxixizhufudao_recommend.huanhuanxixizhufudao_recommend import HhxxzfdRecommend
+
+
+
+def run(args1, args2, args3, args4, args5):
+    HhxxzfdRecommend.start_wechat(log_type=args1,
+                               crawler=args2,
+                               rule_dict=args3,
+                               our_uid=args4,
+                               env=args5)
+
+
+class HhxxzfdMain:
+    @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')
+                    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, 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)
+                            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, rule_dict, our_uid, env)
+                            )
+                            process.start()
+                        time.sleep(60)
+
+                    # # 抓取符合规则的视频列表
+                    # ZFQZRecommend.start_wechat(log_type=log_type,
+                    #                            crawler=crawler,
+                    #                            rule_dict=rule_dict,
+                    #                            our_uid=our_uid,
+                    #                            env=env)
+                    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
+    HhxxzfdMain.main(log_type=args.log_type,
+                              crawler=args.crawler,
+                              topic_name=args.topic_name,
+                              group_id=args.group_id,
+                              env=args.env)

huanhuanxixizhufudao/huanhuanxixizhufudao_recommend/huanhuanxixizhufudao_recommend.py

@@ -0,0 +1,312 @@
+# -*- coding: utf-8 -*-
+# @Author: wang
+# @Time: 2023/9/6
+import json
+import os
+import random
+import sys
+import time
+from hashlib import md5
+
+from appium import webdriver
+from appium.webdriver.common.touch_action import TouchAction
+from appium.webdriver.extensions.android.nativekey import AndroidKey
+from appium.webdriver.webdriver import WebDriver
+from bs4 import BeautifulSoup
+from selenium.common import NoSuchElementException
+from selenium.webdriver.common.by import By
+
+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 HhxxzfdRecommend:
+    platform = "欢欢喜喜祝福到"
+    download_cnt = 0
+    element_list = []
+    i = 0
+
+    @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/crawler/Downloads/chromedriver_v111/chromedriver"
+
+        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": "11",
+            # "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"},
+        }
+        driver = webdriver.Remote("http://localhost:4723/wd/hub", caps)
+        driver.implicitly_wait(30)
+
+        for i in range(120):
+            try:
+                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 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(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(log_type, crawler).info('打开小程序"欢欢喜喜祝福到"')
+        Common.logging(log_type, crawler, env, '打开小程序"欢欢喜喜祝福到"')
+        driver.find_elements(By.XPATH, '//*[@text="欢欢喜喜祝福到"]')[-1].click()
+        time.sleep(5)
+
+        cls.get_videoList(log_type, crawler, driver, env, rule_dict, our_uid)
+
+        time.sleep(1)
+        driver.quit()
+
+    @classmethod
+    def search_elements(cls, driver: WebDriver, xpath):
+        time.sleep(1)
+        windowHandles = driver.window_handles
+        for handle in windowHandles:
+            driver.switch_to.window(handle)
+            time.sleep(1)
+            try:
+                elements = driver.find_elements(By.XPATH, xpath)
+                if elements:
+                    return elements
+            except NoSuchElementException:
+                pass
+
+    @classmethod
+    def check_to_applet(cls, log_type, crawler, env, driver: WebDriver):
+        time.sleep(1)
+        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:
+            driver.switch_to.window(handle)
+            time.sleep(1)
+            try:
+                driver.find_element(By.XPATH, '//*[@class="nav_item active"]')
+                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)
+
+    @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)
+
+    @classmethod
+    def swipe_up(cls, driver: WebDriver):
+        cls.search_elements(driver, '//*[@class="bless--list"]')
+        size = driver.get_window_size()
+        driver.swipe(int(size["width"] * 0.5), int(size["height"] * 0.8),
+                     int(size["width"] * 0.5), int(size["height"] * 0.4), 200)
+
+    @classmethod
+    def get_video_url(cls, log_type, crawler, driver: WebDriver, video_title_element):
+        for i in range(3):
+            cls.search_elements(driver, '//*[@class="bless--list"]')
+            Common.logger(log_type, crawler).info(f"video_title_element:{video_title_element[0]}")
+            time.sleep(1)
+            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(log_type, crawler).info("点击标题")
+            video_title_element[0].click()
+            # 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[@id='video0']")
+            if video_url_elements:
+                return video_url_elements[0].get_attribute("src")
+
+    @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
+        Common.logger(log_type, crawler).info('关闭广告')
+        size = driver.get_window_size()
+        TouchAction(driver).tap(x=int(size['width'] * 0.5), y=int(size['height'] * 0.1)).perform()
+        cls.check_to_applet(log_type=log_type, crawler=crawler, env=env, driver=driver)
+        time.sleep(1)
+
+        page = 0
+        while True:
+            if cls.search_elements(driver, '//*[@class="videoplay"]') 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
+
+            cls.swipe_up(driver)
+
+            page_source = driver.page_source
+            soup = BeautifulSoup(page_source, 'html.parser')
+            soup.prettify()
+
+            video_list_elements = soup.findAll("wx-view", class_="video_item_ob")
+            # video_list_elements 有，cls.element_list 中没有的元素
+            video_list_elements = list(set(video_list_elements).difference(set(cls.element_list)))
+            # video_list_elements 与 cls.element_list 的并集
+            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")
+
+            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
+
+            for i, video_element in enumerate(video_list_elements):
+                try:
+                    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")
+
+                    video_title = video_element.find("wx-view", class_="video_title").text
+                    cover_url = video_element.find("wx-image", class_="video_bg")["src"]
+                    user_name = video_element.find("wx-text", class_="text-group_2").select_one("span:nth-of-type(2)").text.strip()
+                    out_video_id = video_element.get("id")
+
+                    video_dict = {
+                        "video_title": video_title,
+                        "video_id": out_video_id,
+                        "duration": 0,
+                        "play_str": 0,
+                        "play_cnt": 0,
+                        "like_cnt": 0,
+                        "share_cnt": 0,
+                        "user_name": user_name,
+                        "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": cover_url,
+                        "cover_url": cover_url,
+                        "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}")
+                    # 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)
+                        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_dict['video_url'] = video_url
+                        Common.logger(log_type, crawler).info(f"video_url:{video_url}")
+
+                        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)
+                        cls.swipe_up(driver)
+                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": 10, "max": 20},
+                  "share_cnt": {"min": 0, "max": 0}}
+    HhxxzfdRecommend.start_wechat("recommend", "huanhuanxixizhufudao", "dev", rule_dict1, 6267141)