Telegram Video Compressor (Cloudflare Worker Edition)

This is a Cloudflare Worker that compresses videos sent to a Telegram bot.

DEV: mohammadham

Architecture

This project uses a hybrid architecture to handle video processing, which is not natively supported in the Cloudflare Workers environment.

1. Cloudflare Worker (video-compressor-worker):

   • Acts as the main entry point for the Telegram bot.
   • Handles all interactions with the Telegram API.
   • Generates presigned URLs for direct uploads to Cloudflare R2.
   • Triggers the video processing service.
   • Notifies the user when the compressed video is ready.

2. Video Processing Service (video-processor):

   • A separate service (in this case, a Python script) that can run ffmpeg.
   • Downloads the original video from R2.
   • Compresses the video using ffmpeg.
   • Uploads the compressed video back to R2.
   • Notifies the Cloudflare Worker when processing is complete.

3. Cloudflare R2:

   • Used for storing the original and compressed videos.

Deployment

1. Cloudflare Worker

Manual Deployment:

1. Clone this repository.
2. Navigate to the video-compressor-worker directory.
3. Install dependencies: npm install
4. Set up your wrangler.toml file with your Cloudflare account details.
5. Create a .dev.vars file and add the following secrets:
   • BOT_TOKEN: Your Telegram bot token.
   • R2_ACCESS_KEY_ID: Your R2 access key ID.
   • R2_SECRET_ACCESS_KEY: Your R2 secret access key.
   • R2_ACCOUNT_ID: Your Cloudflare account ID.
   • PROCESSOR_URL: The URL of your video processing service.
6. Deploy the worker: npm run deploy

2. Video Processing Service

1. Navigate to the video-processor directory.
2. Install dependencies: pip install -r requirements.txt
3. Create a .env file and add the following variables:
   • R2_ACCOUNT_ID: Your Cloudflare account ID.
   • R2_ACCESS_KEY_ID: Your R2 access key ID.
   • R2_SECRET_ACCESS_KEY: Your R2 secret access key.
   • R2_BUCKET_NAME: The name of your R2 bucket.
   • BOT_TOKEN: Your Telegram bot token.
4. Run the service. You will need a way to expose this service to the internet, for example, by running it on a VPS or as a serverless function on a platform like AWS Lambda or Google Cloud Functions.

Original Python Project

The original Python project can be found in the python-version directory. this project It's a port of the original TG-videoCompress project, rewritten in TypeScript to run on the Cloudflare serverless platform.