用Python下载知乎视频，非常实用！

 

请点击此处输入图片描述

Python下载知乎视频。

# -*- coding: utf-8 -*-

"""

下载知乎视频：

依赖：

pip install requests

mac 安装 ffmpeg:

brew install ffmpeg

"""

import re

import uuid

import subprocess

import requests

# 下边 cookie 请打开知乎打开浏览器开发者工具随便找一个请求复制 cookie，千万不要泄露出去

HEADERS = {

'cookie': '' # TODO

}

QUALITY = 'ld' # 支持是 'ld' 'sd' 'hd' 分别是低清、中清、高清

def get_video_ids_from_url(url):

"""

回答或者文章的 url

"""

html = requests.get(url, headers=HEADERS).text

video_ids = re.findall(r'data-lens-id="(\d+)"', html)

if video_ids:

return set([int(video_id) for video_id in video_ids])

return []

def yield_video_m3u8_url_from_video_ids(video_ids):

for video_id in video_ids:

api_video_url = 'https://lens.zhihu.com/api/videos/{}'.format(int(video_id))

r = requests.get(api_video_url, headers=HEADERS)

playlist = r.json()['playlist']

m3u8_url = playlist[QUALITY]['play_url']

yield m3u8_url

def download(url):

video_ids = get_video_ids_from_url(url)

m3u8_list = list(yield_video_m3u8_url_from_video_ids(video_ids))

filename = '{}.mp4'.format(uuid.uuid4())

for idx, m3u8_url in enumerate(m3u8_list):

print('download {}'.format(m3u8_url))

subprocess.call(['ffmpeg', '-i', m3u8_url, filename.format(str(idx))])

if __name__ == '__main__':

# 贴上你需要下载的 回答或者文章的链接

url = '--------'

download(url)