环境搭建
本代码使用了os、time、requests库，仅只有requests库需要下载
pip install requests抓包方法
抓包
代码编写
本代码保护四部分
第一部分为请求头、cookies的信息(自行抓取，无代码展示)
第二部分为根据关键词获取视频信息的函数
def get_video_url(keyword, pcursor):
# 定义请求头
json_data = {
'operationName': 'visionSearchPhoto',
'variables': {
'keyword': keyword,
'page': 'search',
'pcursor': str(pcursor)
},
'query': 'fragment photoContent on PhotoEntity {\n __typename\n id\n duration\n caption\n originCaption\n likeCount\n viewCount\n commentCount\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}\n\nfragment recoPhotoFragment on recoPhotoEntity {\n __typename\n id\n duration\n caption\n originCaption\n likeCount\n viewCount\n commentCount\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}\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 ...recoPhotoFragment\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 visionSearchPhoto($keyword: String, $pcursor: String, $searchSessionId: String, $page: String, $webPageArea: String) {\n visionSearchPhoto(keyword: $keyword, pcursor: $pcursor, searchSessionId: $searchSessionId, page: $page, webPageArea: $webPageArea) {\n result\n llsid\n webPageArea\n feeds {\n ...feedContent\n __typename\n }\n searchSessionId\n pcursor\n aladdinBanner {\n imgUrl\n link\n __typename\n }\n __typename\n }\n}\n',
}
# 进行异常抛出操作
try:
# 发送post请求
response = requests.post('https://www.kuaishou.com/graphql', cookies=cookies, headers=headers, json=json_data).json()
feeds = response["data"]["visionSearchPhoto"]["feeds"]
# 如果有内容则执行
if feeds:
# 循环取各个视频的信息
for feed in feeds:
# 获取视频名称
name = feed["author"]["name"]
# 获取视频链接
video_url = feed["photo"]["manifest"]["adaptationSet"][0]["representation"][0]["url"]
# 调用下载视频函数进行下载视频
dow_status = dowload_video(name=name, video_url=video_url, dir_name=keyword)
# 如果下载视频函数返回了true则返回true
if dow_status:
return True
# 如果未返回任何信息则暂停1秒
time.sleep(1)
# 否则返回True
else:
return True
# 如果上述代码运行异常则返回true
except:
return True第三部分为下载函数
def dowload_video(video_url, name, dir_name):
try:
# 发送get请求获取视频内容
content = requests.get(url=video_url, headers=headers, cookies=cookies).content
# 首先判断video目录是否存在 其次判断video/dir_name目录是否存在 如果不存在则创建
try:
try:
os.mkdir("video/")
except:
pass
os.mkdir(f"video/{dir_name}/")
except:
pass
# 使用open读入文件
with open(f"video/{dir_name}/{name}.mp4", "wb") as f:
# 写入文件内容
f.write(content)
# 输出提示
print(f"{name}下载完毕")
except:
# 如果发送错误则返回true
return True第四部分为程序主入口
if __name__ == '__main__':
keyword = input("请输入要爬取视频的类型:")
page = int(input("请输入要爬取的页数:"))
for i in range(0, page):
status = get_video_url(keyword=keyword, pcursor=i+1)
if status:
print("也许已下载完相应页数的视频，也许被风控了，将自动退出程序")
break
time.sleep(1)
评论