Revu: Customer Insight & Analysis Platform

Revu is a product review intelligence platform that scrapes product reviews (currently works for Amazon. Revu then analyzes sentiment and topics, and generates concise summaries and insight reports. It pairs a FastAPI backend (with Celery workers), a React + Vite frontend, Redis for queueing, and MongoDB for persistence.

This is a quick developer and operator guide: setup, API examples, project layout, testing, deployment notes, and contribution guidelines.

Demo

revu-demo-1762035753892.mp4

Project structure

Revu/
├── backend/
│   ├── app/
│   │   ├── main.py                # FastAPI app
│   │   ├── worker.py              # Celery worker setup
│   │   ├── routes/                # API route handlers
│   │   ├── services/              # Scraper, summarizer, sentiment, topic extractor
│   │   └── models/                # Pydantic schemas
│   ├── Dockerfile
│   └── requirements.txt
├── frontend/
│   ├── src/                       # React + Vite app
│   └── package.json
├── docker-compose.yml
├── README.md
├── LICENSE
├── CONTRIBUTING.md
├── CODE_OF_CONDUCT.md
└── tests/                         # Small end-to-end and unit tests

Getting Started

Option 1: Docker (Recommended)

The fastest way to get everything running with consistent parity across environments.

1. Copy environment template and set secrets:

cp .env.example .env
# Edit .env with your configuration (see Environment Variables table below)

2. Build and start all services:

docker-compose up --build -d

3. Verify services are running:

• Frontend: http://localhost:3000
• Backend API (Swagger docs): http://localhost:8000/docs
• Redis: localhost:6379

Note: The scraper uses a single Apify actor for reviews. Ensure your Apify account has access to the configured actor and that APIFY_API_TOKEN is set. If you see an error like "Apify scraping failed: You must rent a paid Actor...", either rent/enable the actor in Apify or change the reviews actor via environment variables described below.

Option 2: Local Development (Native)

If you prefer running services outside Docker:

Backend setup:

cd backend
python -m venv ../.venv
..\.venv\Scripts\activate   # Windows (use `source ../.venv/bin/activate` on Linux/Mac)
pip install -r requirements.txt
python -m uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

Start Celery worker (separate terminal):

cd backend
..\.venv\Scripts\activate
celery -A app.worker.celery_app worker -l info --pool=solo

Frontend setup:

cd frontend
npm ci
npm run dev

Prerequisites for local dev:

• Python 3.11+
• Node.js 18+ and npm
• Redis running (use Docker: docker run -d -p 6379:6379 redis:7-alpine or install locally)
• MongoDB (Atlas URI or local instance)

Environment Variables

Configure these in .env (for Docker) or backend/.env (for local backend):

Variable	Description	Default / Example
MONGO_URI	MongoDB connection string	mongodb+srv://<username>:<password>@cluster.mongodb.net/revu
MONGO_DB	Database name	revu
JWT_SECRET	Secret for JWT token signing	change-me-to-secure-random-string
JWT_EXPIRES_MIN	JWT token expiration (minutes)	60
REDIS_URL	Redis connection URL	redis://localhost:6379/0
SCRAPER_MAX_REVIEWS	Max reviews per scrape (hard capped at 300)	300
APIFY_API_TOKEN	Apify API token	your token
APIFY_REVIEWS_ACTOR	Optional: reviews actor id	axesso_data/amazon-reviews-scraper
APIFY_REVIEWS_EXTRA_INPUT	Optional: JSON to merge into actor run input (actor-specific flags)	{ "includeProductInfo": true }
SUMMARY_BACKEND	Summarizer engine: gemini or textrank	gemini
GEMINI_API_KEY	Google Gemini API key (required if using gemini backend)	Your API key
GEMINI_MODEL	Optional: specific Gemini model name	gemini-1.5-flash

Scraper behavior and limits

• Default per-user daily limit: 3 products/day (enforced via Redis)
• Each scrape is capped at 300 reviews to avoid blocking
• The scraper uses a single Apify actor to fetch reviews (configurable via APIFY_REVIEWS_ACTOR)
• Respect robots and site terms: this tool is provided for research and controlled usage

Testing

Run backend tests:

cd backend
# Activate virtual environment first
python -m pytest tests/
python -m pytest tests/test_account_routes.py  # profile & password endpoints

Frontend tests/build:

cd frontend
npm ci
npm run build
npm run test

Contributing

See our contribution guidelines for how to write bug reports and feature requests, and read more about our PR workflow. Be sure to run tests and include clear descriptions in PRs.

Code of Conduct

Our community follows the Contributor Covenant as the basis for our Code of Conduct. Please read and adhere to it.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Maintainers

• Primary maintainer: harshitaphadtare
• Contact: harshita.codewiz@gmail.com