Embula Restaurant Chatbot Backend (Python FastAPI)

This is the Python FastAPI version of the Embula Restaurant Chatbot backend, powered by Mistral AI. It handles:

• Menu queries (filter by vegan/vegetarian/non-veg)
• Table availability & reservations
• Restaurant info (location, contact, opening hours)
• Safe SQL access (restricted tables are blocked)
• Frontend integration ready

---

Features

• FastAPI: High-performance, easy-to-learn, production-ready framework
• Mistral AI Integration: Uses mistral-small-latest for intelligent natural language understanding
• Context Awareness: Maintains conversation history per session (in-memory)
• System Personality: Friendly restaurant assistant
• Profanity & SQL Injection Filter: Sanitizes user input using better-profanity and a custom sanitizer
• Database-Safe Queries: Only reads from allowed tables: food_item, food_item_ingredients, reservations, restaurant_tables, discounts
• Frontend Integration Ready: API endpoints can be called via JS, React, Next.js, or any frontend

---

Project Structure

Project Structure

embula-chatbot-python/
├── src/
│   ├── config/         # Configuration settings
│   ├── models/         # Pydantic models
│   ├── routers/        # API endpoints
│   ├── services/       # Business logic
│   ├── utils/          # Helpers (Memory store)
│   └── main.py         # Application entry point
├── .env                # Environment variables
├── requirements.txt    # Python dependencies
└── README.md           # Documentation

Setup Instructions

1.Prerequisites

• Python 3.9+
• MySQL 8+ (if using a real database)
• Mistral AI API Key
• pip package manager

2. Create Virtual Environment (Recommended):

   python -m venv venv
   # Windows
   .\venv\Scripts\activate
   # Mac/Linux
   source venv/bin/activate

3. Install Dependencies:

   pip install -r requirements.txt

4. Configure Environment:

   • Open .env file.
   • Add your Mistral API Key: MISTRAL_API_KEY=your_actual_key_here

5. Run the Server:

   # Run from the root directory
   python src/main.py
   # OR using uvicorn directly
   uvicorn src.main:app --reload