Python爬虫实战之快手高清无水印视频爬取：以后再也不用去视频水印了.

1.这里我拿这个博主视频为例：www.kuaishou.com/profile/3xy…

2.爬取步骤：

1，快手的视频是通过post请求得到的，我们需要知道他是通过怎样的post请求去获取的：

2，通过检查可以发现post请求过来的数据在graphql这个网址里面（注意请求过来的有两个graphql，只有一个是有数据）：

3，与其他博主的主页进行对比发现只有 variables 的userID和Referer是不同的，也就是网址中最后一段标识；

3,分析完成直接代码展示：

import requests
import json
from pprint import pprint

def main(lian_id):
    url = 'https://www.kuaishou.com/graphql'
    headers = {
        'Cookie': 'kpf=PC_WEB; clientid=3; didv=1674819691332; did=web_a5bf1c2bbfdbff8caf853821bd3de0f4; ktrace-context=1|MS43NjQ1ODM2OTgyODY2OTgyLjIyODE3MTE5LjE2NzU1ODcxMzQxNzcuMTUzNDU0|MS43NjQ1ODM2OTgyODY2OTgyLjU1NTcyOTg2LjE2NzU1ODcxMzQxNzcuMTUzNDU1|0|graphql-server|webservice|false|NA; clientid=3; kpn=KUAISHOU_VISION',
        # 'Origin':'https://www.kuaishou.com',
        'Referer': f'https://www.kuaishou.com/profile/{lian_id}',
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.51 Safari/537.36'
    }
    date = {
        'operationName': "visionProfilePhotoList",
        'query': "fragment photoContent on PhotoEntity {\n  id\n  duration\n  caption\n  originCaption\n  likeCount\n  viewCount\n  realLikeCount\n  coverUrl\n  photoUrl\n  photoH265Url\n  manifest\n  manifestH265\n  videoResource\n  coverUrls {\n    url\n    __typename\n  }\n  timestamp\n  expTag\n  animatedCoverUrl\n  distance\n  videoRatio\n  liked\n  stereoType\n  profileUserTopPhoto\n  musicBlocked\n  __typename\n}\n\nfragment feedContent on Feed {\n  type\n  author {\n    id\n    name\n    headerUrl\n    following\n    headerUrls {\n      url\n      __typename\n    }\n    __typename\n  }\n  photo {\n    ...photoContent\n    __typename\n  }\n  canAddComment\n  llsid\n  status\n  currentPcursor\n  tags {\n    type\n    name\n    __typename\n  }\n  __typename\n}\n\nquery visionProfilePhotoList($pcursor: String, $userId: String, $page: String, $webPageArea: String) {\n  visionProfilePhotoList(pcursor: $pcursor, userId: $userId, page: $page, webPageArea: $webPageArea) {\n    result\n    llsid\n    webPageArea\n    feeds {\n      ...feedContent\n      __typename\n    }\n    hostName\n    pcursor\n    __typename\n  }\n}\n",
        'variables': {'userId': f"{lian_id}", 'pcursor': "", 'page': "profile"}
    }
    html = requests.post(url=url, headers=headers, json=date)
    # pprint(html.json())
    feeds = html.json()['data']['visionProfilePhotoList']['feeds']
    s = 1
    for i in feeds:
        s += 1
        tite = i['photo']['caption']
        url = i['photo']['manifest']['adaptationSet'][0]['representation'][0]['backupUrl'][0]
        print(tite, url)
        data = requests.get(url=url, headers=headers)
        f = open(f'快手高清视频\{s}{tite}.mp4', 'wb')
        f.write(data.content)
        f.close()


if __name__ == '__main__':
    while True:
        lian_jie = input("请输入博主主页链接: ")
        lian_id = lian_jie.split('/')[-1]
        main(lian_id)

4，爬取效果展示：

5，有什么问题都可以在评论区中评论，或者私信博主，制作不易麻烦给博主留下一个小赞，谢谢！