#!/usr/bin/env python
# -*- coding:utf-8 -*-
# @FileName  :test.py
# @Time      :2024/4/27
# @Author    :coding
# @email     :
from multiprocessing import Pool

from DrissionPage import ChromiumPage
from DataRecorder import Recorder
import time
import random
import os
import datetime
from util import mongo_manager

douyin_videos = mongo_manager('douyin_videos', db='public_data')


def countdown(n):
    for i in range(n, 0, -1):
        print(f'\r倒计时{i}秒', end='')  # \r让光标回到行首 ，end=''--结束符为空，即不换行
        time.sleep(1)  # 让程序等待1秒
    else:
        print('\r倒计时结束')


def sign_in():
    sign_in_page = ChromiumPage()
    sign_in_page.get('https://www.douyin.com/?recommend=1')
    print('请扫码登录')
    countdown(30)


def delete_file(file_path):
    """用于删除初始化excel文件"""
    # 检查文件是否存在
    if os.path.exists(file_path):
        # 删除文件
        os.remove(file_path)
        print(f"已删除初始化excel文件：{file_path}")
    else:
        print(f"文件不存在：{file_path} ")


def run(keyword='纯人声朗诵'):
    page = ChromiumPage()
    # page
    url = f'https://www.douyin.com/search/{keyword}?type=video'
    # page.clear_cache()
    # page.refresh()
    page.listen.start('www.douyin.com/aweme/v1/web/search/item', method='GET')  # 开始监听，指定获取包含该文本的数据包
    page.get(url)
    # 新建一个excel表格，用来保存数据
    init_file_path = f'数据/抖音搜索-{keyword}.xlsx'
    os.makedirs(os.path.dirname(init_file_path), exist_ok=True)
    r = Recorder(path=init_file_path, cache_size=100)
    lens = 0
    # 循环提取视频
    while lens <= 300:
        # 生成一个随机时间
        random_time = random.uniform(2, 3)
        # 下滑页面，随机等待
        page.scroll.to_bottom()
        time.sleep(random_time)
        # 下滑页面，随机等待
        page.scroll.to_bottom()
        time.sleep(random_time)

        res = page.listen.wait(timeout=5)  # 等待并获取一个数据包
        try:
            # json_data和浏览器的开发者工区中response一样
            json_data = res.response.body

        except Exception as e:
            print('没有监听到更多视频，结束运行。', e)
            time.sleep(10)
            break
        try:
            videos = json_data['data']
            # 提取视频信息  # videos是所有视频列表
        except Exception as e:
            print('没有监听到更多视频，结束运行。', e)
            time.sleep(10)
            res = page.listen.wait(timeout=5)  # 等待并获取一个数据包
            # json_data和浏览器的开发者工区中response一样
            json_data = res.response.body

            videos = json_data['data']
        for v in videos:
            # print(v)
            time.sleep(10)
            # 视频id
            aweme_id = v['aweme_info']['aweme_id']
            # 视频标题
            title = v['aweme_info']['desc'].strip()
            # 发布时间
            create_time = v['aweme_info']['create_time']
            # 作者名称
            nickname = v['aweme_info']['author']['nickname']
            # 作者id
            uid = v['aweme_info']['author']['uid']
            # 认证信息
            enterprise_verify_reason = v['aweme_info']['author']['enterprise_verify_reason']
            # 作者主页地址
            sec_uid = v['aweme_info']['author']['sec_uid']
            # 作者粉丝数
            follower_count = v['aweme_info']['author']['follower_count']
            # 点赞
            digg_count = v['aweme_info']['statistics']['digg_count']
            # 评论
            comment_count = v['aweme_info']['statistics']['comment_count']
            # 收藏
            collect_count = v['aweme_info']['statistics']['collect_count']
            # 转发
            share_count = v['aweme_info']['statistics']['share_count']

            video_url = "https://www.douyin.com/video/" + aweme_id

            sec_uid = "https://www.douyin.com/user/" + sec_uid
            duration = v['aweme_info']['video']['duration']
            # 将时间戳转换为datetime对象
            create_time = datetime.datetime.fromtimestamp(create_time)
            # 将datetime对象转换为字符串
            create_time = create_time.strftime('%Y-%m-%d %H:%M:%S')

            print(aweme_id, title, digg_count, comment_count, collect_count, share_count)
            info = {'关键词': keyword, '视频标题': title, '发布时间': create_time, '点赞': digg_count,
                    '评论': comment_count, '收藏': collect_count, '转发': share_count, '作者名称': nickname,
                    '作者粉丝数': follower_count, '视频ID': aweme_id, '视频链接': video_url,
                    '作者主页地址': sec_uid, '时长(秒)': int(duration) / 1000,
                    '作者ID': uid, '作者认证信息': enterprise_verify_reason,
                    'data_type':'舱内'}
            r.add_data(info)
            try:
                info["_id"] = info['视频ID']
                douyin_videos.insertOne(info)
                lens += 1
            except Exception as e:
                print('插入失败', e)

        time.sleep(3)
        # 如果还有视频，继续循环。如果没有视频，结束循环。
        has_more = json_data['has_more']
        if has_more != 1:
            break

    # 爬虫结束，避免丢数据，强制保存文件
    r.record()

    print('**' * 10)

    print(
        f'抖音关键词【{keyword}】总计获取到【{lens}】条视频')
    # 文件重命名
    final_file_path = f'数据/抖音搜索-{keyword}-{lens}条视频.xlsx'
    r.record(final_file_path)
    print(f'成功将文件另存为：{final_file_path}')

    # 删除初始文件
    delete_file(init_file_path)

    # 关闭浏览器
    page.close()


def main():
    pool = Pool(processes=4)
    pool.map(run, keywords)
    pool.close()
    pool.join()


if __name__ == '__main__':
    # # 1、登录，第2次运行可以注释掉这一步骤
    sign_in()

    keywords =['狗 车内视频','猫 车内视频']
    for keyword in keywords:
        run(keyword=keyword)
        print(keyword, '采集完毕,休息100秒')
        time.sleep(100)
        # break