爬取抖音某个博主下面的全部视频

import json
from time import time

import requests
from requests import urllib3
from jsonpath import jsonpath

urllib3.disable_warnings()

headers = {
    ‘Accept-Encoding‘: ‘gzip‘,
    ‘X-SS-REQ-TICKET‘: ‘1581660428182‘,
    ‘sdk-version‘: ‘1‘,
    ‘Cookie‘: ‘install_id=103352043068; ttreq=1$d29781e6817415b2336bf64a341c5c52a774276d; d_ticket=f618ea679f55d5c2f8b80a9a6d094563e59df; odin_tt=a84c4ba69760c5508dc4deb0fc8d455602a0feaf37eb88ffc1ff6160c74134380e27fde9e3145770266377b503270b6a; msh=LwK4GHwRJ3HW0pYhRKoude--bdE; sid_guard=a0ff9b655c3fc7fd489a4f29c564b8b0%7C1581657235%7C5184000%7CTue%2C+14-Apr-2020+05%3A13%3A55+GMT; uid_tt=0849a342334709951bf7d36eca9b71ac; sid_tt=a0ff9b655c3fc7fd489a4f29c564b8b0; sessionid=a0ff9b655c3fc7fd489a4f29c564b8b0‘,
    ‘x-tt-token‘: ‘00a0ff9b655c3fc7fd489a4f29c564b8b0348cef63112f9c2de76300a6428a0d14d46d0638e964f7d6e4a44d229e0eebe25f‘,
    ‘X-Gorgon‘: ‘040120d9400141032e52e896e6e108d5d8030c421707c7b1da70‘,
    ‘X-Khronos‘: ‘1581660428‘,
    ‘Host‘: ‘aweme.snssdk.com‘,
    ‘Connection‘: ‘Keep-Alive‘,
    ‘User-Agent‘: ‘okhttp/3.10.0.1‘
}

url = ‘https://aweme.snssdk.com/aweme/v1/aweme/post/?source=0&max_cursor=0&sec_user_id=MS4wLjABAAAAQgBRnIwhA5hR4wi7-TjXzE2-8ir5uwBpbT2a-gukccc&count=20&os_api=23&device_type=MI%205s&device_platform=android&ssmix=a&iid=103352043068&manifest_version_code=972&dpi=270&uuid=500000000189568&version_code=972&app_name=aweme&cdid=955be6dc-f79f-4480-ab01-0409eff3f75b&version_name=9.7.2&ts=1581660430&openudid=862bef236d1d353f&device_id=70748665911&resolution=810*1440&os_version=6.0.1&language=zh&device_brand=Xiaomi&app_type=normal&ac=wifi&update_version_code=9708&aid=1128&channel=tengxun_new&_rticket=1581660428228‘
resp = requests.get(url, headers=headers, stream=True, verify=False)
result = json.loads(resp.text)
video_url_list = []
video_url_list1 = jsonpath(result, ‘$..play_addr_lowbr‘)
desc_name_list = jsonpath(result, ‘$..desc‘)
print(desc_name_list)
for item in video_url_list1:
    video_url_list.append(item[‘url_list‘][0])
print(video_url_list)

for i, url1 in enumerate(video_url_list):
    resp1 = requests.get(url1)
    with open(‘{}.mp4‘.format(desc_name_list[i]), ‘wb‘)as f:
        for content in resp1.iter_content(1024):
            if content:
                f.write(content)
#因为只是爬了一个博主一个版面的视频，往下拉，还会有类似的请求就没有爬取，都一样，只是url里面的max_cursor变成了时间戳*1000的值，count变成10，每次下拉的cookie也可能会变（我没有试验过）
#比如我爬取的这个博主有62个视频，抓包发现第一次url里面count是20，后面五次count都是10，获取70个但博主只有62个，从第二次请求开始count由20变为10
#要大量爬取的话要多准备点抖音账号登录获取cookie，可以采用redis+flask的方法搭建cookie池，详情可以参见崔庆才大佬的博客，抖音cookie过期比较快，貌似十分钟左右就过期了，可能还要用到代理