piaoquan_crawler/shipinhao/shipinhao_search/shipinhao_search.py

# -*- coding: utf-8 -*-
# @Author: wangkun
# @Time: 2023/4/25
import difflib
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 selenium.common import NoSuchElementException
from selenium.webdriver.common.by import By
sys.path.append(os.getcwd())
from common.feishu import Feishu
from common.publish import Publish
from common.common import Common
from common.public import get_config_from_mysql


class ShipinhaoSearch:
    i = 0

    @classmethod
    def start_wechat(cls, log_type, crawler, sheetid, env):
        # try:
        Common.logger(log_type, crawler).info('启动微信')
        if env == "dev":
            chromedriverExecutable = "/Users/wangkun/Downloads/chromedriver/chromedriver_v86/chromedriver"
        else:
            chromedriverExecutable = '/Users/piaoquan/Downloads/chromedriver'
        caps = {
            "platformName": "Android",  # 手机操作系统 Android / iOS
            "deviceName": "Android",  # 连接的设备名（模拟器或真机），安卓可以随便写
            "platforVersion": "11",  # 手机对应的系统版本（Android 11）
            "appPackage": "com.tencent.mm",  # 被测APP的包名，乐活圈 Android
            "appActivity": ".ui.LauncherUI",  # 启动的Activity名
            "autoGrantPermissions": True,  # 让 appium 自动授权 base 权限，
            # 如果 noReset 为 True，则该条不生效（该参数为 Android 独有），对应的值为 True 或 False
            "unicodekeyboard": True,  # 使用自带输入法，输入中文时填True
            "resetkeyboard": True,  # 执行完程序恢复原来输入法
            "noReset": True,  # 不重置APP
            "recreateChromeDriverSessions": True,  # 切换到非 chrome-Driver 会 kill 掉 session，就不需要手动 kill 了
            "printPageSourceOnFailure": True,  # 找不到元素时，appium log 会完整记录当前页面的 pagesource
            "newCommandTimeout": 6000,  # 初始等待时间
            "automationName": "UiAutomator2",  # 使用引擎，默认为 Appium，
            # 其中 Appium、UiAutomator2、Selendroid、Espresso 用于 Android，XCUITest 用于 iOS
            "showChromedriverLog": True,
            "chromeOptions": {"androidProcess": "com.tencent.mm:appbrand0"},
            # "chromeOptions": {"androidProcess": "com.tencent.mm:tools"},
            'enableWebviewDetailsCollection': True,
            'setWebContentsDebuggingEnabled': True,
            'chromedriverExecutable': chromedriverExecutable,
        }
        driver = webdriver.Remote("http://localhost:4723/wd/hub", caps)
        driver.implicitly_wait(10)

        time.sleep(5)
        # 根据词搜索视频
        cls.search_video(log_type=log_type,
                         crawler=crawler,
                         driver=driver,
                         sheetid=sheetid,
                         env=env)

        # except Exception as e:
        #     Common.logger(log_type, crawler).error('start_wechat异常，重启APP:{}\n', e)

    # 查找元素
    @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_webview(cls, log_type, crawler, driver: WebDriver):
        while True:
            Common.logger(log_type, crawler).info('切换到webview')
            webview = driver.contexts
            driver.switch_to.context(webview[1])

    @classmethod
    def search_video(cls, log_type, crawler, driver: WebDriver, sheetid, env):
        word_list = get_config_from_mysql(log_type, crawler, env, "search_word", action="")
        for word in word_list:
            driver.implicitly_wait(10)
            Common.logger(log_type, crawler).info('点击搜索按钮')
            driver.find_element(By.ID, 'com.tencent.mm:id/j5t').click()
            time.sleep(0.5)
            Common.logger(log_type, crawler).info(f'输入视频标题:{word}')
            driver.find_element(By.ID, 'com.tencent.mm:id/cd7').clear().send_keys(word)
            Common.logger(log_type, crawler).info('点击搜索')
            driver.press_keycode(AndroidKey.ENTER)
            driver.find_elements(By.ID, 'com.tencent.mm:id/oi4')[0].click()

            cls.check_to_webview(log_type, crawler, driver)

            video_list = cls.search_elements(driver, '//div[@class="unit"]/*[2]')
            Common.logger(log_type, crawler).info('点击"视频号"分类')
            video_list[0].click()
            time.sleep(5)

            index = 0
            while True:
                if index == 30:
                    Common.logger(log_type, crawler).info(f'"{word}"已抓取视频数:{index}\n')
                    break
                try:
                    if cls.search_elements(driver, '//*[@class="double-rich double-rich_vertical"]') is None:
                        Common.logger(log_type, crawler).info('窗口已销毁\n')
                        return

                    Common.logger(log_type, crawler).info('获取视频列表\n')
                    video_elements = cls.search_elements(driver, '//wx-view[@class="double-rich double-rich_vertical"]')
                    if video_elements is None:
                        Common.logger(log_type, crawler).warning(f'video_elements:{video_elements}')
                        return

                    video_element_temp = video_elements[index:]
                    if len(video_element_temp) == 0:
                        Common.logger(log_type, crawler).info('到底啦~~~~~~~~~~~~~\n')
                        return

                    for i, video_element in enumerate(video_element_temp):
                        if video_element is None:
                            Common.logger(log_type, crawler).info('到底啦~\n')
                            return
                        cls.i += 1
                        cls.search_elements(driver, '//wx-view[@"double-rich double-rich_vertical"]')

                        Common.logger(log_type, crawler).info(f'拖动"视频"列表第{cls.i}个至屏幕中间')
                        time.sleep(3)
                        driver.execute_script("arguments[0].scrollIntoView({block:'center',inline:'center'})",
                                              video_element)
                        Common.logger(log_type, crawler).info("点击进入视频详情")
                        video_element.click()
                        time.sleep(3)
                        cls.get_video_info(log_type=log_type,
                                           crawler=crawler,
                                           driver=driver,
                                           sheetid=sheetid)
                    Common.logger(log_type, crawler).info('已抓取完一组视频，休眠10秒\n')
                    time.sleep(10)
                    index = index + len(video_element_temp)
                except Exception as e:
                    Common.logger(log_type, crawler).info(f"get_videoList:{e}\n")
                    cls.i = 0

    @classmethod
    def get_video_info(cls, log_type, crawler, driver: WebDriver, sheetid):
        # webview = driver.contexts
        Common.logger(log_type, crawler).info('切回NATIVE_APP\n')
        driver.switch_to.context('NATIVE_APP')

        # 点击播放器，获取视频时长
        # Common.logger(log_type, crawler).info('暂停播放')
        pause_btn = driver.find_element(By.ID, 'com.tencent.mm:id/eh4')
        pause_btn.click()
        start_time = driver.find_element(By.ID, 'com.tencent.mm:id/l59').get_attribute('name')
        start_time = int(start_time.split(':')[0]) * 60 + int(start_time.split(':')[-1])
        try:
            end_time = driver.find_element(By.ID, 'com.tencent.mm:id/l7i').get_attribute('name')
        except NoSuchElementException:
            end_time = driver.find_element(By.ID, 'com.tencent.mm:id/g73').get_attribute('name')
        end_time = int(end_time.split(':')[0]) * 60 + int(end_time.split(':')[-1])
        duration = start_time + end_time

        # 点赞
        like_id = driver.find_element(By.ID, 'com.tencent.mm:id/k04')
        like_cnt = like_id.get_attribute('name')
        if like_cnt == "" or like_cnt == "喜欢":
            like_cnt = 0
        elif '万' in like_cnt:
            like_cnt = float(like_cnt.split('万')[0]) * 10000
        elif '万+' in like_cnt:
            like_cnt = float(like_cnt.split('万+')[0]) * 10000
        else:
            like_cnt = float(like_cnt)

        # 分享
        share_id = driver.find_element(By.ID, 'com.tencent.mm:id/jhv')
        share_cnt = share_id.get_attribute('name')
        if share_cnt == "" or share_cnt == "转发":
            share_cnt = 0
        elif '万' in share_cnt:
            share_cnt = float(share_cnt.split('万')[0]) * 10000
        elif '万+' in share_cnt:
            share_cnt = float(share_cnt.split('万+')[0]) * 10000
        else:
            share_cnt = float(share_cnt)

        # 收藏
        favorite_id = driver.find_element(By.ID, 'com.tencent.mm:id/fnp')
        favorite_cnt = favorite_id.get_attribute('name')
        if favorite_cnt == "" or favorite_cnt == "收藏":
            favorite_cnt = 0
        elif '万' in favorite_cnt:
            favorite_cnt = float(favorite_cnt.split('万')[0]) * 10000
        elif '万+' in favorite_cnt:
            favorite_cnt = float(favorite_cnt.split('万+')[0]) * 10000
        else:
            favorite_cnt = float(favorite_cnt)

        # 评论
        comment_id = driver.find_element(By.ID, 'com.tencent.mm:id/bje')
        comment_cnt = comment_id.get_attribute('name')
        if comment_cnt == "" or comment_cnt == "评论":
            comment_cnt = 0
        elif '万' in comment_cnt:
            comment_cnt = float(comment_cnt.split('万')[0]) * 10000
        elif '万+' in comment_cnt:
            comment_cnt = float(comment_cnt.split('万+')[0]) * 10000
        else:
            comment_cnt = float(comment_cnt)

        # 把视频信息写入飞书feeds文档
        values = [[time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(int(time.time()))),
                   "视频号搜索",
                   "标题",
                   duration,
                   like_cnt,
                   share_cnt,
                   favorite_cnt,
                   comment_cnt]]
        time.sleep(1)
        Feishu.update_values(log_type, crawler, sheetid, 'F2:Z2', values)
        Common.logger(log_type, crawler).info('视频信息写入飞书文档成功\n')


if __name__ == '__main__':
    ShipinhaoSearch.start_wechat(log_type="search",
                                 crawler="shipinhao",
                                 sheetid="xYWCzf",
                                 env="dev")

    pass