Review-Summary

A FastAPI-based backend that collects and summarizes items. Reviews are stored in a PostgreSQL database, and summaries are generated using a local Ollama model mistral.

---

🚀 Features

• Submit and store detailed reviews
• Generate natural language summaries using LLM
• FastAPI for RESTful endpoints
• PostgreSQL as the backend database
• Docker support for consistent setup

---

🛠️ Prerequisites

Before you begin, make sure you have the following installed:

• Docker

• Ollama installed locally

• Pull the model you'll be using:

  ollama pull mistral

• Python 3.11+ (optional, for local development)

---

⚙️ Environment Variables

Create a .env file in the root directory of your project:

POSTGRES_USER=youruser
POSTGRES_PASSWORD=yourpassword
POSTGRES_DB=yourdb
DB_HOST=yourdb
DB_PORT=yourport

This is used by both the FastAPI app and the PostgreSQL database.

---

🐳 Running the Dockerized App

1. Build and Start Containers

Run this from the root of your project:

docker-compose up --build

This will:

• Build the FastAPI container (web)
• Start PostgreSQL using the official postgres:16 image
• Mount and initialize the database from docker-entrypoint-initdb.d/
• Run your FastAPI app at http://localhost:8000
• Arbitrarily set port as 15432 to avoid conflict with 5432- you may need to set a different port in yml if you're already using 15432

2. Start Ollama (Locally)

Ollama must be installed and running on your host machine, not inside Docker.

Start it separately:

ollama serve

Ensure you’ve pulled the required model (e.g. mistral):

ollama pull mistral

---

🔎 API Access

Once running:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

---

📁 Project Structure

.
├── app/
│   ├── main.py                  # FastAPI routes
│   ├── db.py                    # DB connection and insert functions
│   └── summarize.py             # Ollama summarizer logic
├── docker-entrypoint-initdb.d/
│   ├── 01-schema.sql     # SQL setup
│   └── 02-seed.sql       # Sample data         
├── requirements.txt
├── Dockerfile
├── docker-compose.yml
├── .env
└── README.md

---

🧪 Local Development (Optional)

If you're not using Docker, you can run the app locally like this:

python -m venv venv
source venv/bin/activate    # or 'venv\Scripts\activate' on Windows

pip install -r requirements.txt
uvicorn app.main:app --reload

---

🔐 Security To-Do

• Add authentication (JWT, OAuth)
• Use HTTPS via reverse proxy like Nginx
• Salt + hash passwords securely
• Disable direct DB port access in production

---

👥 Contributors

• Data Science – James Kim
• Backend – Austin Myers
• Cybersecurity – Xavier Gutter
• Design/UX – Chavis Ferguson

---

📝 License

This project is private and proprietary. All rights reserved.

---

📬 Feedback

Feel free to open issues or pull requests for improvements or bugs.