【Python】一键爬取快手所有关注账号的全部视频

FirmTao

PS：更新了一下，之前防盗链忘记改了，现在需要添加自己的

最开始只是为了下载快|手关注的一个账号的全部视频，写完代码后又想下载全部关注的视频，
但又不想一个一个的手动操作，所有一想直接一键搞定，走起

以下为抓包过程：

先进入网站，网址 快手官网，不知道的自己百度

1、进入后登录自己的账号，然后点击进入主页   
  
   

进入后左侧可以看到自己的关注人数，右侧是具体的关注账号，我们要获取的就是右侧的数据

   

2、右键打开检测，切换到 network 窗口，再选中第三行工具栏的 Fetch/XHR ，然后刷新一下当前页面，可以看到数据包

  

3、依次点击 graphql 数据包，找到需要的数据，这个列表就是我们需要的关注数据  visionProfileUserList

  

4、这个关注的数据获取到后，之后就是去找个人所有视频的数据啦啦啦，随便打开一个人的主页，可以看到全部视频



5、跟第 2 步一样，刷新出数据包，然后依次点击 graphql 数据包，找到需要的数据

  

6、OK，数据全部找到，接下来就是写代码了，话不多说直接上代码

      代码可能写的不好，本人刚接触，大佬请指正批评，让我进步

     提示：有四个参数，你只需要填两个 一个是Cookie   Cookie填的是登录后的，另一个是User-Agent

  

   User-Agent 在Cookie的下面！！！

   
7、这是运行效果图

      


8、最后请各位兄弟姐妹们有什么给什么，帮忙顶一顶，出问题可以评论区留言。多说一句，python是3.9版本

[Asm] 纯文本查看 复制代码

# coding:utf-8
import json
import os
import pprint
import re
import string
import sys
import time
import ctypes
import os
import platform
from zhon.hanzi import punctuation
import requests
import urllib3

urllib3.disable_warnings()


# 请求网页
def req_data(url, id, pcursor, ck, ua):
    # 请求头
    headers = {
        'content-type': 'application/json',
        'Cookie': ck,
        'Host': 'www.kuaishou.com',
        'Origin': 'https://www.kuaishou.com',
        'Referer': 'https://www.kuaishou.com/profile/' + id,
        'User-Agent': ua
    }
    # 请求参数
    data = {
        'operationName': 'visionProfilePhotoList',
        'query': "query 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      type\n      author {\n  "
                 "      id\n        name\n        following\n        headerUrl\n        headerUrls {\n          cdn\n "
                 "         url\n          __typename\n        }\n        __typename\n      }\n      tags {\n        "
                 "type\n        name\n        __typename\n      }\n      photo {\n        id\n        duration\n      "
                 "  caption\n        likeCount\n        realLikeCount\n        coverUrl\n        coverUrls {\n        "
                 "  cdn\n          url\n          __typename\n        }\n        photoUrls {\n          cdn\n         "
                 " url\n          __typename\n        }\n        photoUrl\n        liked\n        timestamp\n        "
                 "expTag\n        animatedCoverUrl\n        stereoType\n        videoRatio\n        "
                 "profileUserTopPhoto\n        __typename\n      }\n      canAddComment\n      currentPcursor\n      "
                 "llsid\n      status\n      __typename\n    }\n    hostName\n    pcursor\n    __typename\n  }\n}\n",
        'variables': {'userId': id, 'pcursor': pcursor, 'page': 'profile'}
    }
    data = json.dumps(data)
    data_json = requests.post(url=url, headers=headers, data=data, timeout=6.05).json()
    # pprint.pprint(data_json)
    return data_json


# 清洗文件名
def rep_char(chars):
    eg_punctuation = string.punctuation
    ch_punctuation = punctuation
    # print("所有标点符号：", eg_punctuation, ch_punctuation)
    for item1 in eg_punctuation:
        chars = chars.replace(item1, '')
    for item2 in ch_punctuation:
        chars = chars.replace(item2, '')
    chars = chars.replace(' ', '').replace('\n', '').replace('\xa0', '').replace('\r', '')
    return chars


# 磁盘内存检查
def get_free_space():
    folder = os.path.abspath(sys.path[0])
    if platform.system() == 'Windows':
        free_bytes = ctypes.c_ulonglong(0)
        ctypes.windll.kernel32.GetDiskFreeSpaceExW(ctypes.c_wchar_p(folder), None, None, ctypes.pointer(free_bytes))
        return free_bytes.value / 1024 / 1024 / 1024
    else:
        st = os.statvfs(folder)
        return st.f_bavail * st.f_frsize / 1024 / 1024


# 保存数据
def save(url, page, ck, ua, selfid):
    except_lit = []
    count = 0
    idlist = get_all_ids(url, page, ck, ua, selfid)
    for id in idlist:
        count = count + 1
        print(f'第{count}位关注：{id} 全部视频下载中...')
        num = 0
        # 循环下载视频，直到 page == 'no_more'
        while page != 'no_more':
            time.sleep(1)
            data = req_data(url, id, page, ck, ua)
            # 获取翻页的参数
            next_page_Pcursor = data['data']['visionProfilePhotoList']['pcursor']
            page = next_page_Pcursor
            print(next_page_Pcursor)
            data_list = data['data']['visionProfilePhotoList']['feeds']
            for item in data_list:
                num = num + 1
                video_name = item['photo']['caption']
                video_url = item['photo']['photoUrl']
                author = item['author']['name']
                author = rep_char(author)
                video_name = rep_char(video_name)
                path = './video1'
                if not os.path.exists(path + '/' + author + '/'):
                    os.makedirs(path + '/' + author + '/')
                filepath = path + '/' + author + '/' + str(num) + '.' + video_name + '.mp4'
                if os.path.exists(filepath):
                    print(f'{num}、 {video_name} >>> 已存在!!!')
                    # time.sleep(1)
                    continue
                # 请求二进制视频数据
                try:
                    video_content = requests.get(url=video_url, timeout=(3, 7)).content
                except:
                    strss = f'{author}_{num}video_name：{video_url}'
                    except_lit.append(strss)
                    continue
                with open(filepath, mode='wb') as f:
                    f.write(video_content)
                print(f'{num}、 {video_name} >>> 下载完成!!!')
                # 判断剩余容量是否充足
                free_space = get_free_space()
                if free_space <= 1:
                    break
        # pcursor = page 这个变量的值必须为空，不用动他，它是换页的参数
        page = ''
        print(f'第{count}位关注：{id} 全部视频下载完成！！！')
    with open('yc_info.txt', 'a') as f:
        f.write(except_lit)
        print('异常保存成功')
    print(except_lit)


# 获取全部关注页面数据
def req_follow_data(url, pcursor, ck, ua, selfid):
    # 请求头
    headers = {
        'content-type': 'application/json',
        'Cookie': ck,
        'Host': 'www.kuaishou.com',
        'Origin': 'https://www.kuaishou.com',
        'Referer': 'https://www.kuaishou.com/profile/' + selfid,
        'User-Agent': ua
    }
    # 请求参数
    data = {
        'operationName': 'visionProfileUserList',
        'query': 'query visionProfileUserList($pcursor: String, $ftype: Int) {\n  visionProfileUserList(pcursor: '
                 '$pcursor, ftype: $ftype) {\n    result\n    fols {\n      user_name\n      headurl\n      '
                 'user_text\n      isFollowing\n      user_id\n      __typename\n    }\n    hostName\n    pcursor\n   '
                 ' __typename\n  }\n}\n',
        'variables': {'ftype': 1, 'pcursor': pcursor}
    }
    data = json.dumps(data)
    follow_json = requests.post(url=url, headers=headers, data=data).json()
    # pprint.pprint(follow_json)
    return follow_json


# 获取全部关注的id
def get_all_ids(url, page, ck, ua, selfid):
    id_list = []
    num = sign = 0
    # 循环保存id，直到 Pcursor == 'no_more'
    while page != 'no_more':
        time.sleep(1)
        follow_data = req_follow_data(url, page, ck, ua, selfid)
        # 获取翻页的参数
        next_pcursor = follow_data['data']['visionProfileUserList']['pcursor']
        page = next_pcursor
        sign = sign + 1
        print(f'第{sign}页:{next_pcursor}')
        fols_list = follow_data['data']['visionProfileUserList']['fols']
        for item in fols_list:
            num = num + 1
            user_name = item['user_name']
            user_id = item['user_id']
            id_list.append(user_id)
            print(f'{num}、 {user_name}：{user_id} >>> ID获取成功！！！')
    print(id_list)
    print(id_list)
    return id_list


if __name__ == '__main__':
    link = 'https://www.kuaishou.com/graphql'
    # pcursor这个变量的值开始必须为空，不用动他，它是换页的参数
    # selfid 是自己账号网址的最后面那一串 例如 https://www.kuaishou.com/profile/3xkfgnn9hkacbwc selfid 就是 3xkfgnn9hkacbwc
    selfid = ''
    pcursor = ''
    # ck =''  引号中间填登录后的 Cookie 值
    ck =''
    # ua = '' 引号中间填 User-Agent
    ua = ''
    save(link, pcursor, ck, ua, selfid)

dmq336

感谢楼主分享。

Hiaappy

收藏了，很好的软件

NNYY

感谢楼主分享！

SDU123

感谢楼主分享

lvcaolhx

楼主分享下，怎样模拟登陆拿到cookies?

miaomiaofu

谢谢分享，

rewin

麻烦问一下，下载下来的带快手水印吗？

FirmTao

lvcaolhx 发表于 2021-12-23 12:48
楼主分享下，怎样模拟登陆拿到cookies?

cookie是登录自己的账号后直接F12找到的啊，上面有图，至于你说的模拟登录要用到selenium，我这里不需要模拟登录啊

FirmTao

rewin 发表于 2021-12-23 13:35
麻烦问一下，下载下来的带快手水印吗？

没有水印哦，反正我下载的没有水印

ychiou

有成品吗？