A Telegram bot that provides humorous predictions with a modular architecture.

• Telegram Bot API token (get it from BotFather)

1. Build the Docker image:

   docker build -t vorkybot .
   

2. Run the container:

   docker run -e API_KEY=your_telegram_bot_token -d --name vorkybot vorkybot
   

1. Set your API_KEY as an environment variable:

   export API_KEY=your_telegram_bot_token
   

2. Start the service:

   docker-compose up -d
   

1. Add a new stack in Portainer
2. Upload or paste the docker-compose.yml file
3. Set environment variables:
   • API_KEY: your_telegram_bot_token
4. Deploy the stack

• /future command: Get a random humorous prediction (limited to once per 24 hours per user)

You can provide your own predictions file instead of using the built-in ones:

1. Create a text file with one prediction per line

2. Mount this file when running Docker:

   # Using docker run
   docker run -e API_KEY=your_telegram_bot_token -v /path/to/your/predictions.txt:/app/data/predictions.txt:ro -d --name vorkybot vorkybot

   With docker-compose, uncomment and edit the volume mounting line:

   volumes:
     - ./data:/app/data
     - /path/to/your/predictions.txt:/app/data/predictions.txt:ro

3. The bot will automatically use your external predictions file if it exists

The bot follows a modular architecture with separation of concerns:

• Models: Data structures and business logic

  • Prediction: Handles prediction selection from available replies

• Services: Core functionality providers

  • CooldownManager: Manages user cooldown periods for predictions

• Bot: Telegram integration

  • VorkyBot: Main bot class that handles Telegram integration and command processing

This architecture makes the code more maintainable, testable, and follows best practices like dependency injection and single responsibility principle.