README.md

metadata

title: Odyssey Backend
emoji: 🎥
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 5.38.2
app_file: app.py
pinned: false
short_description: Video upload backend for Odyssey adventure game

🎥 Odyssey Backend

Backend service for the Odyssey interactive video adventure game.

Features

• Video Upload: Upload generated adventure videos to HuggingFace dataset
• UUID Generation: Each video gets a unique identifier
• Public URLs: Serve videos via HuggingFace CDN
• Gradio API: Simple API for programmatic access

Architecture

• Frontend: odyssey - Interactive game interface
• Backend: This Space - Video upload service
• Storage: odyssey-videos - Dataset of generated videos

API Usage

JavaScript/TypeScript

import { Client } from "@gradio/client";

const client = await Client.connect("Fraser/odyssey-backend");

// Upload video
const result = await client.predict("/upload_video", {
    video_file: blob // or file handle
});

// Parse result
const data = JSON.parse(result.data[0]);
console.log(data.uuid, data.url, data.share);

Python

from gradio_client import Client

client = Client("Fraser/odyssey-backend")

# Upload video
result = client.predict("/upload_video", video_file="path/to/video.mp4")

Deployment

Environment Variables

Set these in Space Settings → Repository secrets:

• HF_TOKEN: HuggingFace token with write permissions to Fraser/odyssey-videos dataset

Local Development

pip install -r requirements.txt
export HF_TOKEN=your_token_here
python app.py
# Server runs at http://localhost:7860

Dataset Structure

All videos are stored in Fraser/odyssey-videos:

{uuid}.mp4
{uuid}.mp4
...

Each video can be accessed via:

• Direct URL: https://huggingface.co/datasets/Fraser/odyssey-videos/resolve/main/{uuid}.mp4
• Frontend URL: https://fraser-odyssey.static.hf.space/?id={uuid}

Tech Stack

• Gradio: API framework and web interface
• HuggingFace Hub: Dataset storage backend
• Python: Core server logic