This project provides a web interface to query and understand Toronto Municipal Code bylaws. It uses a Flutter frontend, a Flask (Python) backend, MongoDB Atlas for data storage and vector search, and the Google Gemini API for generating answers based on retrieved bylaw context.

• Natural Language Queries
  Ask questions about Toronto bylaws in plain English.

• Retrieval-Augmented Generation (RAG)
  Finds relevant bylaw sections using MongoDB Atlas Vector Search before generating an answer.

• AI-Powered Answers
  Uses the Google Gemini API to synthesize information from retrieved bylaws and answer user questions.

• Source Linking
  Provides links to the original PDF source documents for the relevant bylaw chapters.

• Web Interface
  Clean and responsive UI built with Flutter.

• Frontend: Flutter (Dart)
• Backend: Flask (Python)
• Database: MongoDB Atlas (Vector Search)
• AI Model: Google Gemini API

pymongo, google-generativeai, Flask, python-dotenv, sentence-transformers, PyMuPDF (fitz), beautifulsoup4

http, flutter_markdown, url_launcher, google_fonts, flutter_animate, loading_animation_widget

GDSC_Hacks_2025/
├── flask_api/ # Backend Flask application
│ ├── templates/ # (If any Flask templates exist)
│ ├── static/ # (If any static files exist)
│ ├── venv/ # Python virtual environment
│ ├── .env # Environment variables (DO NOT COMMIT)
│ ├── requirements.txt # Python dependencies
│ ├── app.py # Main Flask app
│ ├── create_database.py
│ ├── embed_vectors.py
│ ├── parse_html.py
│ ├── query_database.py
│ ├── python_to_gemini.py
│ ├── chunk_text.py
│ ├── lawmcode.htm # Source HTML file (manual download)
│ └── ...
│
├── flutter_frontend/ # Frontend Flutter app
│ ├── lib/
│ │ └── main.dart
│ ├── assets/
│ ├── web/
│ ├── pubspec.yaml
│ └── ...
│
└── README.md # This file\

• Git
• Python 3.8+ and pip
• Flutter SDK
• A MongoDB Atlas Account (Free M0 Tier)
• A Google Gemini API Key (get it from Google AI Studio)

cd flask_api

# Create virtual environment
python -m venv venv

# Activate virtual environment
# Windows:
.\venv\Scripts\activate
# macOS/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt
🧾 Download Bylaw Source HTML
Go to: https://www.toronto.ca/legdocs/municode/toronto-code-chapters.shtml

Save the HTML content as lawmcode.htm in the flask_api folder.

🔐 Create a .env file

DATABASE_LOGIN=username:password GEMINI_API_KEY=your_gemini_api_key 🏗 Populate the Database

python create_database.py ⚠️ This script downloads many PDFs, embeds content, and uploads to MongoDB. It may take some time.

💻 2. Frontend Setup (flutter_frontend)

cd ../flutter_frontend flutter pub get 🔧 Configure Backend URL Edit lib/main.dart:

final uri = Uri.parse('http://127.0.0.1:5000/api/query'); // For local dev For deployment, change this to your hosted backend URL.

🔑 Environment Variables Set these for the backend via .env (locally) or platform secrets (in production):

DATABASE_LOGIN: MongoDB Atlas login (username:password)

GEMINI_API_KEY: Google Gemini API key

▶️ Running Locally Start the Flask Backend

cd flask_api source venv/bin/activate # if not already activated flask run --host=0.0.0.0 # Runs on port 5000 Start the Flutter Frontend

cd ../flutter_frontend flutter run -d chrome Ensure the API URL in main.dart points to your local backend.

🚀 Deployment Frontend (Flutter Web)

flutter build web --release Host build/web on:

Firebase Hosting

GitHub Pages

Netlify

Vercel

Render

Backend (Flask) Deploy flask_api to:

Render

PythonAnywhere

Fly.io

Google Cloud Run

AWS (free tier)

Use Gunicorn (included in requirements.txt) and set debug=False.

🔒 Deployment Security Tips CORS: Only allow frontend domain.

Secrets: Use environment variables. Never commit .env.

DB Access: Restrict MongoDB Atlas to your backend IPs.

HTTPS: Always enable HTTPS in production.

🤝 Contributing (Add guidelines if you'd like others to contribute)

📝 License (Add your license here, e.g., MIT License)