Skip to main content

支持AI总结的社交媒体聚合解析器

Project description

ParseHub

支持AI总结的社交媒体聚合解析器

视频总结会调用 whisper-1 模型

基于该项目开发的 Tg Bot:
@ParsehubBot | https://github.com/z-mio/parse_hub_bot

支持的平台:
Twitter 视频|图文
Instagram 视频|图文
微博 视频|图文
贴吧 视频|图文
小红书 视频|图文
Youtube 视频|音乐
Facebook 视频
Bilibili 视频|动态
抖音|TikTok 视频|图文
......

安装

pip install parsehub

安装yt-dlp: wiki


[!IMPORTANT]

注意

Linux用户在导入skia-python包时可能会遇到以下报错

libGL.so.1: cannot open shared object file: No such file or directory

Windows用户在缺少Microsoft Visual C++ Runtime时可能会遇到以下报错

ImportError: DLL load failed while importing skia: The specified module could not be found.

解决方法

ubuntu用户

apt install libgl1-mesa-glx

ArchLinux用户

pacman -S libgl

centos用户

yum install mesa-libGL -y

Windows用户

下载链接Microsoft Visual C++ 2015 Redistributable Update 3 RC

使用

from parsehub import ParseHub
from parsehub.config import ParseConfig, DownloadConfig
import asyncio


async def main():
    ph = ParseHub(config=ParseConfig())
    result = await ph.parse('https://twitter.com/aobuta_anime/status/1827284717848424696')
    print(result)

    # dr = await result.download()
    # print(dr.media)
    # sr = await dr.summary()
    # await dr.delete()

    sr = await result.summary(download_config=DownloadConfig())
    print(sr.content)


if __name__ == '__main__':
    asyncio.run(main())

环境变量

名称 描述 默认值
PROVIDER 模型提供商, 暂只支持openai openai
API_KEY API Key
BASE_URL API 地址 https://api.openai.com/v1
MODEL AI总结使用的模型 gpt-4o-mini
PROMPT AI总结提示词 You are a useful assistant to summarize the main points of articles and video captions. Summarize 3 to 8 points in "Simplified Chinese" and summarize them all at the end.

参考项目

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

parsehub-1.0.9.tar.gz (57.0 kB view details)

Uploaded Source

Built Distribution

parsehub-1.0.9-py3-none-any.whl (80.0 kB view details)

Uploaded Python 3

File details

Details for the file parsehub-1.0.9.tar.gz.

File metadata

  • Download URL: parsehub-1.0.9.tar.gz
  • Upload date:
  • Size: 57.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.4.16

File hashes

Hashes for parsehub-1.0.9.tar.gz
Algorithm Hash digest
SHA256 7a73a25fc9e106d58808748646708d317604ff2bbe9eacf539b4a6ed9e217851
MD5 85fb79b23a314107e91b615e56c53310
BLAKE2b-256 63c0a75aea650932699934d14ea1522be4c0aab67507716b9ad3d956e6f4c64e

See more details on using hashes here.

File details

Details for the file parsehub-1.0.9-py3-none-any.whl.

File metadata

File hashes

Hashes for parsehub-1.0.9-py3-none-any.whl
Algorithm Hash digest
SHA256 a46da1d7005020c0bf2292ce42bcc1e0b89650a4c66fcd51b087a4953d1a2228
MD5 ac304f385791ef4102b61e144f2b9f9d
BLAKE2b-256 a41d54b50c6eb6e6d0d602f95a34c416efd34bc266a4d1bd8f02f6265a698437

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page