Fetchy API - Vercel & Supabase Backend

Video download API for Fetchy iOS app using yt-dlp, migrated to Vercel and Supabase.

Architecture

• Vercel: Handles API requests and yt-dlp execution (Node.js).
• Supabase Database: Stores job status and logs.
• Supabase Storage: Stores downloaded video files.

Setup

1. Supabase Setup:

   • Create a new project on Supabase.
   • Create a table named jobs with the following columns:
     • id: int8 (Identity)
     • status: text
     • progress: float8
     • message: text
     • title: text
     • file_path: text
     • log: text
     • url: text
     • created_at: timestamptz (default: now())
   • Create a storage bucket named downloads (set it to public or private as needed).

2. Environment Variables: Copy .env.example to .env and fill in your Supabase credentials.

3. Install Dependencies:

npm install

Deployment to Vercel

1. Install Vercel CLI: npm i -g vercel
2. Deploy: vercel
3. Add Environment Variables in Vercel Dashboard:
   • SUPABASE_URL
   • SUPABASE_SERVICE_ROLE_KEY
   • SUPABASE_BUCKET_NAME (optional, defaults to downloads)

API Endpoints

POST /api/download

Start a download job.

Request:

{
  "url": "https://youtube.com/watch?v=...",
  "quality": "1080p"
}

GET /api/status/:jobId

Get job status and progress.

GET /api/download/:jobId

Redirects to a signed URL for the completed file.

GET /api/log/:jobId

Get raw yt-dlp log output.