|
|
@@ -0,0 +1,346 @@
|
|
|
+# -*- coding: utf-8 -*-
|
|
|
+# @Time: 2023/10/31
|
|
|
+import json
|
|
|
+import os
|
|
|
+from datetime import date, timedelta
|
|
|
+
|
|
|
+import requests
|
|
|
+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 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
|
|
|
+
|
|
|
+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:
|
|
|
+ platform = "小年糕-offline"
|
|
|
+ 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/tzld/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/ky9'):
|
|
|
+ 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"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)
|
|
|
+ driver.find_elements(By.XPATH, '//*[@text="我"]')[-1].click()
|
|
|
+ time.sleep(5)
|
|
|
+ driver.find_elements(By.XPATH, '//*[@text="关注"]')[-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="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)
|
|
|
+
|
|
|
+ @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="expose--adapt-parent"]')
|
|
|
+ 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)
|
|
|
+
|
|
|
+
|
|
|
+
|
|
|
+ @classmethod
|
|
|
+ def get_video_url(cls, log_type, crawler, driver: WebDriver, video_title_element):
|
|
|
+ for i in range(3):
|
|
|
+ cls.search_elements(driver, '//wx-view[@class="expose--adapt-parent"]')
|
|
|
+ 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[@class="video-player--video"]')
|
|
|
+ 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
|
|
|
+ cls.check_to_applet(log_type=log_type, crawler=crawler, env=env, driver=driver)
|
|
|
+ time.sleep(1)
|
|
|
+ page = 0
|
|
|
+ element_name = cls.search_elements(driver, "//wx-view[@class='box_user_nick_txt']")
|
|
|
+ driver.execute_script("arguments[0].scrollIntoView({block:'center',inline:'center'});",
|
|
|
+ element_name[0])
|
|
|
+ time.sleep(3)
|
|
|
+ count = len(element_name)
|
|
|
+ for i in range(count-1):
|
|
|
+ element_name[i].click()
|
|
|
+ user_name = element_name[i].text
|
|
|
+ time.sleep(10)
|
|
|
+ while True:
|
|
|
+ if cls.search_elements(driver, '//wx-view[@class="expose--adapt-parent"]') 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 = []
|
|
|
+ driver.press_keycode(AndroidKey.BACK)
|
|
|
+ continue
|
|
|
+ page_source = driver.page_source
|
|
|
+ soup = BeautifulSoup(page_source, 'html.parser')
|
|
|
+ soup.prettify()
|
|
|
+ video_url_elements = soup.findAll("wx-view", class_="expose--adapt-parent")
|
|
|
+ video_url_elements = list(set(video_url_elements).difference(set(cls.element_list)))
|
|
|
+ cls.element_list = list(set(video_url_elements) | set(cls.element_list))
|
|
|
+ Common.logger(log_type, crawler).info(f"正在抓取第{page + 1}页，共:{len(video_url_elements)}条视频")
|
|
|
+ Common.logging(log_type, crawler, env, f"正在抓取第{page + 1}页，共:{len(video_url_elements)}条视频")
|
|
|
+ if len(video_url_elements) == 0:
|
|
|
+ for i in range(5):
|
|
|
+ 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_url_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_="album--album-cover-title").text
|
|
|
+ # 播放量字符串
|
|
|
+ play_str = video_element.find("wx-view", class_="album--album-cover-views").text
|
|
|
+
|
|
|
+ # 封面 URL
|
|
|
+ cover_url = video_element.find("wx-image", class_="album--album-cover-bg")["src"]
|
|
|
+ play_cnt = int(play_str.replace("+", "").replace("次播放", ""))
|
|
|
+ # 发布时间
|
|
|
+ play_time = video_element.find("wx-view", class_="album--album-time").text
|
|
|
+ date_three_days_ago_string = (date.today() + timedelta(days=-3)).strftime("%Y.%m.%d")
|
|
|
+ rule = play_time > date_three_days_ago_string
|
|
|
+ if rule == False:
|
|
|
+ Common.logger(log_type, crawler).info(f"发布时间小于3天，发布时间：{rule}")
|
|
|
+ Common.logging(log_type, crawler, env, f"发布时间小于3天，发布时间：{rule}")
|
|
|
+ break
|
|
|
+ out_video_id = md5(video_title.encode('utf8')).hexdigest()
|
|
|
+ out_video_id = out_video_id + "user"
|
|
|
+ out_user_id = md5(user_name.encode('utf8')).hexdigest()
|
|
|
+ out_user_id = out_user_id + "user"
|
|
|
+ video_dict = {
|
|
|
+ "video_title": video_title,
|
|
|
+ "video_id": out_video_id,
|
|
|
+ "duration_str": '',
|
|
|
+ "duration": 0,
|
|
|
+ "play_str": play_str,
|
|
|
+ "play_cnt": play_cnt,
|
|
|
+ "like_str": "",
|
|
|
+ "like_cnt": 0,
|
|
|
+ "comment_cnt": 0,
|
|
|
+ "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": cover_url,
|
|
|
+ "cover_url": cover_url,
|
|
|
+ "session": f"xiaoniangao-{int(time.time())}"
|
|
|
+ }
|
|
|
+
|
|
|
+ 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")
|
|
|
+ Common.logging(log_type, crawler, env, '无效视频')
|
|
|
+ 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")
|
|
|
+ Common.logging(log_type, crawler, env, "不满足抓取规则")
|
|
|
+ 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')
|
|
|
+ Common.logging(log_type, crawler, env, '已中过滤词')
|
|
|
+ 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')
|
|
|
+ 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_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"]
|
|
|
+ Common.logger(log_type, crawler).info(f"video_dict:{video_dict}")
|
|
|
+ Common.logging(log_type, crawler, env, f"video_dict:{video_dict}")
|
|
|
+ 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")
|
|
|
+ page += 1
|
|
|
+ driver.press_keycode(AndroidKey.BACK)
|
|
|
+ driver.execute_script("arguments[0].scrollIntoView({block:'center',inline:'center'});",
|
|
|
+ element_name[i])
|
|
|
+ Common.logger(log_type, crawler).info("f已抓取完用户{user_name}，休眠 5 秒")
|
|
|
+ Common.logging(log_type, crawler, env, "f已抓取完用户{user_name}，休眠 5 秒")
|
|
|
+ time.sleep(5)
|
|
|
+
|
|
|
+
|
|
|
+
|
|
|
+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}}
|
|
|
+ XiaoNianGaoPlusRecommend.start_wechat("recommend", "xiaoniangao", "dev", rule_dict1, 6267141)
|
