Python爬取抖音视频（2）

最近看到之前写的抖音爬虫失效了，其实方法还能用，只是多了一个参数而已，直接从页面获取就可以了。
之前看到一个获取视频列表不用加密参数_signature，的链接。
下载的视频还是去水印版本的。

• 不用_signature参数来获取视频列表。
• 增加了dytk参数。
• 主要使用协程来加速下载

协程的教程：链接

1. 获取用户的url链接
2. 模拟请求，直接获取dytk参数、id参数
3. 构建获取视频url地址参数，https://www.amemv.com这个域名下的请求不需要_signature
4. 然后就是下载视频了。

# -*- coding: utf-8 -*-
# File  : douyin.py
# Author: HuWenBo
# Date  : 2018/10/26
# 使用协程爬取抖音个人下的所有视频

import os
import requests
import re
import sys
import asyncio
import aiohttp

headers = {
    'user-agent': 'Mozilla/5.0 (iPhone; CPU iPhone OS 11_0 like Mac OS X) AppleWebKit/604.1.38 (KHTML, like Gecko) '
                  'Version/11.0 Mobile/15A372 Safari/604.1'
}

VIDEO_URLS, PAGE = [], 1


def get_info(url):
    """
    :param url: 用户的链接
    :return:返回name，dytk，user_id 参数
    """
    name = None
    dytk = None
    user_id = None
    try:
        response = requests.get(url, headers=headers)
        user_id = response.url.split('/')[5].split('?')[0]
        name = re.search(r'class="nickname">(.*?)<', response.text)[1]
        dytk = re.search(r"dytk: '(.*?)'", response.text)[1]
    except (TypeError, IndexError):
        sys.stdout.write('Waring：输入的链接错误')
    except requests.exceptions:
        sys.stdout.write('Waring：链接错误')
    finally:
        return name, user_id, dytk


def make_dir(name):
    """
    建立文件夹
    :param name: 用户名称
    :return:
    """
    if not os.path.isdir(name):
        os.mkdir(name)
    else:
        pass


def get_all_video(user_id, max_cursor, dytk):
    """
    获取视频的地址
    :param user_id:
    :param max_cursor:
    :param dytk:
    :return:
    """
    url = "https://www.amemv.com/aweme/v1/aweme/post/?"
    params = {'user_id': user_id,
              'count': 21,
              'max_cursor': max_cursor,
              'dytk': dytk}
    try:
        response = requests.get(url=url, params=params, headers=headers)
        if response.status_code == 200:
            datas = response.json()
            for data in datas['aweme_list']:
                name = data.get('share_info').get('share_desc')
                url = data.get('video').get('play_addr').get('url_list')[0].replace('playwm', 'play')
                VIDEO_URLS.append([name, url])
            if datas['has_more'] == 1 and datas.get('max_cursor') != 0:
                global PAGE
                print(f'收集第{PAGE}页视频')
                PAGE += 1
                return get_all_video(user_id, datas.get('max_cursor'), dytk)
            else:
                print('收集完成')
                return VIDEO_URLS
        else:
            print('状态码：', response.status_code)
            return None
    except Exception as e:
        print('Waring：', e)
        return


async def download_video(index, name, video_name, url):
    """
    下载视频
    :param index:   视频id
    :param name:    用户名称
    :param video_name:  视频名称
    :param url:     下载url
    :return:
    """
    print(f'正在下载第{index}个视频：{video_name}')
    video_path = '{}/{}.mp4'.format(name, video_name)
    if not os.path.isfile(video_path):
        try:
            async with aiohttp.ClientSession() as session:
                async with session.get(url=url, headers=headers, ssl=False) as response:
                    with open(video_path, 'wb') as f:
                        while True:
                            chunk = await response.content.read(1024)
                            f.write(chunk)
                            if not chunk:
                                break
                        print(f'下载完成第{index}个视频：{video_name}')
        except Exception as e:
            print('waring：download faild', video_name, e)
            return
    else:
        print('文件已存在')


def main():
    url = 'http://v.douyin.com/dEorkn/'
    name, user_id, dytk = get_info(url)
    if not (name, user_id, dytk):
        return
    make_dir(name)
    get_all_video(user_id, 0, dytk)
    print(f'{name}：总共有{len(VIDEO_URLS)}个视频')
    tasks = []
    for index, item in enumerate(VIDEO_URLS, 1):
        video_name = item[0]
        url = item[1]
        tasks.append(asyncio.ensure_future(download_video(index, name, video_name, url)))
    loop = asyncio.get_event_loop()
    loop.run_until_complete(asyncio.wait(tasks))
    loop.run_until_complete(asyncio.sleep(0))
    loop.close()
    print(f'{name}视频下载完成！')


if __name__ == '__main__':
    main()