youtube更新视频自动采集，自动更新到国内腾讯平台第二版（每24小时执行一次）

整个自动化流程已经完全走通，代码内增加了多种异常处理

第一版链接（很粗糙）：https://www.jianshu.com/p/d837e70eb33a
提醒：有些模块是在我项目文件夹下自己写的，pip安装不了；

from pytube import YouTube
import requests
from fake_useragent import UserAgent
import re
from selenium import webdriver
import ssl
from lxml import etree
import time
from selenium.webdriver.common.keys import Keys
from Google import GoogleTranslate
import redis

opt = webdriver.ChromeOptions()
opt.add_argument('--headless')
#更换头部
cookie = [{'name': 'pt2gguin', 'value': 'o0654921690'}, {'name': 'RK', 'value': 'mSglPLOFQ8'}, {'name': 'ptcz', 'value': '5eb1aeb628b6ab67aa306285a78434959b385f8e39415bda4ff4db3ea7763d75'}, {'name': 'pgv_pvid', 'value': '116706325'}, {'name': 'pgv_pvi', 'value': '1972684800'}, {'name': 'pgv_si', 'value': 's3116448768'}, {'name': 'pgv_info', 'value': 'ssid'}, {'name': 'uin', 'value': 'o0654921690'}, {'name': 'skey', 'value': '@7pMFazBu1'}, {'name': 'ptisp', 'value': 'ctc'}, {'name': 'ts_refer', 'value': 'www.baidu.com/link'}, {'name': 'ts_uid', 'value': '3500005680'}, {'name': 'userid', 'value': '5661917'}, {'name': 'fname', 'value': '%E5%BC%BA%E8%BF%AB%E7%97%87%E7%9A%84%E5%B0%8F%E4%B9%90%E8%B6%A3'}, {'name': 'fimgurl', 'value': 'http%3A%2F%2Finews.gtimg.com%2Fnewsapp_ls%2F0%2F1807099981_200200%2F0'}, {'name': 'omtoken', 'value': '3799d6a2c9'}, {'name': 'omtoken_expire', 'value': '1531158354'}, {'name': 'alertclicked', 'value': '%7C%7C'}, {'name': 'rmod', 'value': '1'}]

ua = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.62 Safari/537.36'
opt.add_argument('user-agent="%s"' % ua)
# opt.add_argument('cookie=%s'% cookie)
driver = webdriver.Chrome(chrome_options=opt)
driver.maximize_window()
redis_cli = redis.Redis(host='127.0.0.1',port=6379)
class QIE:
    def __init__(self):
        self.ua = UserAgent()
        # 上传异常计数，超过阈值放弃当前上传的视频
        self.upload_error_handling_num = 0

    # 拿到目标视频的链接
    def get_video_urls(self):
        while True:

            # index_page = 'https://www.youtube.com/channel/UC6XKMuLChaxW4Isw0BTZI8A/videos'
            index_page = 'https://www.youtube.com/channel/UC6XKMuLChaxW4Isw0BTZI8A/videos?view=0&flow=grid&sort=dd'
            headers = {
                'User-Agent':self.ua.random
            }
            proxies = {
                'https':'127.0.0.1:1087'
            }
            response = requests.get(index_page,headers=headers,proxies=proxies).text
            list_urls = re.findall(r'"(/watch\?v=.+?)"',response)
            if len(list_urls)>0:
                list_urls = set(list_urls)
                list_urls = list(list_urls)
                print(list_urls)
                # 取前3个视频
                if len(list_urls)>3:
                    list_urls = list_urls[:3]
                return list_urls
            else:
                print('匹配失败')
                print(response)
                time.sleep(2)

    # 下载所有需要的视频
    def download_vides(self):
        result = self.get_video_urls()
        if result != 'no':
            # 文件名列表
            self.names = []
            # 下载计数
            num = 0
            for url in result:
                redis_result = redis_cli.get(url)
                if redis_result == None:
                    redis_cli.set(url,1)
                    print('{}已经写入redis'.format(url))
                    print('一共{}个视频'.format(len(result)))
                    ssl._create_default_https_context = ssl._create_unverified_context
                    url = 'https://www.youtube.com' + url
                    yt = YouTube(url=url,proxies={'https':'127.0.0.1:1087'})
                    data = yt.streams.first()
                    file_name = data.default_filename
                    print('视频开始下载......')
                    # 调用google翻译生成新的文件名
                    go = GoogleTranslate(file_name)
                    file_name = go.get_translated()
                    print(file_name)
                    data.download(filename=file_name)
                    num += 1
                    print('第%d个视频下载完成，文件名是:%s' % (num,file_name))
                    self.names.append(file_name + '.mp4')
                    print(self.names)
            if num > 0:
                return 'ok'
            else:
                return 'no'
        else:
            return 'no'

    # 连接youtube和腾讯
    def context(self):
        result = self.download_vides()
        if result == 'ok':
            for file_name in self.names:
                self.set_cookie()
                self.upload(file_name)


    # 上传所有本地的视频，每日最多3条
    def set_cookie(self):
        # 请求目标地址
        url = 'https://om.qq.com/article/articlePublish#/!/view:article?typeName=multivideos'
        driver.get(url)
        driver.implicitly_wait(18)
        # 给目标地址添加cookie
        driver.delete_all_cookies()
        for c in cookie:
            new = dict(c,**{
            "domain": ".qq.com",
            "expires": "",
            'path': '/',
            'httpOnly': False,
            'HostOnly': False,
            'Secure': False,
        })
            driver.add_cookie(new)
        # 再次请求目标地址
        driver.get(url)
        driver.implicitly_wait(18)
        
    # 文件导入，上传
    def upload(self,file_name):

        # 将本地文件上传到页面的input标签
        path1 = '/Users/admin/Documents/qiehao2/'
        # path1 = '/home/ubuntu/Documents/tencent/'

        # file_name = "化妆变身迪士尼公主.mp4"
        # 标题名需要小于30个字符
        # file_name = file_name.split(' ')[0]
        file_name2 = path1+file_name
        driver.find_element_by_xpath('//input').send_keys(file_name2)
        path = '进入视频上传主页.png'
        driver.save_screenshot(path)
        print('进入视频上传主页截图保存成功')
        # 尝试cookie登陆，如果失败模拟登陆
        try:
            status = driver.find_element_by_xpath('//span[@class="text-title"]').get_attribute('textContent').strip()
            print(status)
        except:
            print('进入失败，可能cookie已经过期，重新模拟登陆')
            driver.delete_all_cookies()
            url = 'https://om.qq.com/article/articlePublish#/!/view:article?typeName=multivideos'
            driver.get(url)
            driver.implicitly_wait(18)
            driver.find_element_by_xpath('//input[@name="email"]').send_keys('xxxx你的账号')
            driver.find_element_by_xpath('//input[@name="password"]').send_keys('oooo你的密码')
            driver.find_element_by_xpath('//button[@class="btnLogin btn btn-primary"]').click()
            time.sleep(30)
            driver.get(url)
            driver.implicitly_wait(18)
            self.upload(file_name)
        else:
            # 等待上传完成
            num = 0
            while True:
                time.sleep(60)
                num += 1
                status = driver.find_element_by_xpath('//span[@class="text-title"]').get_attribute('textContent').strip()
                path = '载入进度.png'
                driver.save_screenshot(path)
                print('已经上传了第%d分钟'%num)
                if status == '共1个视频，已上传1个':
                    print('ok')
                    break
            path = 'success.png'
            driver.save_screenshot(path)
            print('文件导入成功')
            #滚动到目标位置 选择分类
            # 上传错误处理
            self.upload_error_handling(file_name)
            
    # 封装上传错误处理
    def upload_error_handling(self,file_name):
        try:
            command = driver.find_element_by_xpath('//span[contains(./text(),"请选择分类")]')
        except:
            # 不超过阈值重试
            self.upload_error_handling_num +=1
            if self.upload_error_handling_num < 4:
                print('文件上传出现异常，准备重试，重试第{}次'.format(self.upload_error_handling_num))
                self.upload(file_name)
            else:
                # 超过阈值初始化计数器，放弃本条视频上传；
                self.upload_error_handling_num = 0
                print('上传异常次数过多，可能本条视频已经传过了，放弃。准备上传下一条......')
                pass
        else:
            driver.execute_script("arguments[0].scrollIntoView();", command)
            command.click()
            time.sleep(1)
            # 滚动到目标位置,输入内容并且回车
            command = driver.find_element_by_xpath('//div[@class="chosen-search"]/input')
            driver.execute_script("arguments[0].scrollIntoView();", command)
            command.send_keys('少儿节目')
            command.send_keys(Keys.ENTER)
            # 将页面滚动到底部
            js = "var q=document.documentElement.scrollTop=10000"
            driver.execute_script(js)
            time.sleep(10)
            # 选择标签，默认选择9个标签
            self.get_tag()
            # 选择封面图
            driver.find_element_by_xpath('//button[contains(./text(),"设置封面")]').click()
            # 滚动到目标位置并且点击
            # 如果封面图加载超时过长那么就放弃本次上传
            num = 0
            while True:
                time.sleep(60)
                try:
                    command = driver.find_element_by_xpath(
                        '//div[@class="upload-cover-block"]/span/img[contains(./@src,"1.jpg")]')
                    print('图片加载好了')
                    driver.execute_script("arguments[0].scrollIntoView();", command)
                    command.click()
                    break
                except:
                    num += 1
                    print('图片还没加载好，继续等待')
                    # 放弃上传
                    if num > 10:
                        print('放弃本次上传')
                        break
            # 如果加载超时，放弃执行接下来语句
            try:
                driver.find_element_by_xpath('//button[contains(./text(),"确定")]').click()
                path = '封面设置成功.png'
                driver.save_screenshot(path)
                # 发布
                time.sleep(3)
                num = 0
                while True:
                    try:
                        driver.find_element_by_xpath('//button[@action="publish"]').click()
                        path = '上床成功.png'
                        driver.save_screenshot(path)
                        print('上传成功')
                        break
                    except:
                        print('设置封面图窗口关闭中')
                        time.sleep(5)
                        num += 1
                        if num > 3:
                            print('上传按钮等待超时')
                            break
            except:
                # 放弃上传
                pass

    # 填写标签
    def get_tag(self):
        single = driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[1]/a').get_attribute('textContent').strip()
        print(single)
        if single != '暂无':
            # 滚动到标签位置
            command = driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[1]/a')
            driver.execute_script("arguments[0].scrollIntoView();", command)
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[1]/a').click()
            except:
                print('没有推荐的标签')
                driver.find_element_by_xpath('//div/input[contains(./@id,"videoTags")]').send_keys('玩具')
                driver.find_element_by_xpath('//div/input[contains(./@id,"videoTags")]').send_keys(Keys.ENTER)
                # driver.find_element_by_xpath('//label[contains(./text(),"标签")]').click()
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[2]/a').click()
            except:
                print('没有第二个推荐的标签')
                driver.find_element_by_xpath('//div/input[contains(./@id,"videoTags")]').send_keys('欧美流行')
                driver.find_element_by_xpath('//div/input[contains(./@id,"videoTags")]').send_keys(Keys.ENTER)
                # driver.find_element_by_xpath('//label[contains(./text(),"标签")]').click()
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[3]/a').click()
            except:
                print('没有第三个推荐的biaoqian ')

            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[4]/a').click()
            except:
                print('没有第四个推荐的标签')
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[5]/a').click()
            except:
                print('没有第五个推荐的标签')
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[6]/a').click()
            except:
                print('没有第六个推荐的标签')
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[7]/a').click()
            except:
                print('没有第七个推荐的标签')
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[8]/a').click()
            except:
                print('没有第八个推荐的标签')
            try:
                driver.find_element_by_xpath('//li[contains(./text(),"推荐标签")]/following::li[9]/a').click()
            except:
                print('没有第九个推荐的标签')
        else:
            # 滚动到目标位置，手动输入标签
            command = driver.find_element_by_xpath('//div/input[contains(./@id,"videoTags")]')
            driver.execute_script("arguments[0].scrollIntoView();", command)
            command.send_keys('玩具')
            command.send_keys(Keys.ENTER)
            command.send_keys('欧美流行')
            command.send_keys(Keys.ENTER)

    # 每24小时更新一次
    def main(self):
        while True:
            self.context()
            time.sleep(24*60*60)







if __name__ == '__main__':
    qie = QIE()
    qie.main()