Vetrina

Vetrina is a full-stack e-commerce management platform built with Angular, Spring Boot, MySQL, and a Python AI service. It brings together product browsing, carts, orders, customer management, analytics, reviews, store administration, and an AI assistant for data-aware business questions.

Features

• Role-based authentication for ADMIN, CORPORATE, and USER accounts
• Product catalog with product detail pages, categories, and admin product workflows
• Cart and order management
• Customer, review, shipment, and profile management
• Corporate dashboard and analytics views with Chart.js visualizations
• Admin panels for users, stores, categories, settings, audit logs, and reports
• AI assistant powered by FastAPI and LangGraph for guided analytics and SQL-backed answers
• MySQL database bootstrap with Docker Compose

Tech Stack

Layer	Technology
Frontend	Angular 21, TypeScript, SCSS, RxJS, Chart.js
Backend	Java 17, Spring Boot, Spring Security, Spring Data JPA, JWT
Database	MySQL 8
AI Service	Python, FastAPI, LangGraph, Groq, pandas, Plotly
Tooling	Maven, npm, Docker Compose

Project Structure

.
+-- frontend/      # Angular application
+-- backend/       # Spring Boot REST API
+-- ai-service/    # FastAPI AI assistant service
+-- database/      # MySQL Docker Compose and initialization SQL
+-- docs/          # Project documentation and technical report
`-- init_db.sql    # Root-level database initialization script

Prerequisites

• Node.js and npm
• Java 17
• Maven, or the included Maven wrapper
• Docker and Docker Compose
• Python 3.10 or newer

Getting Started

1. Clone the repository

git clone <repository-url>
cd ECommerceProject

2. Start the database

cd database
docker compose up -d

The database runs on localhost:3307 and creates the ecommerce_db schema. Default local credentials are:

username: root
password: root

3. Start the backend

cd ../backend
./mvnw spring-boot:run

On Windows PowerShell:

.\mvnw.cmd spring-boot:run

The API runs at http://localhost:8080.

4. Start the AI service

The AI assistant is optional for the core store flows, but required for /api/ai chat features.

cd ../ai-service
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

On Windows PowerShell:

cd ..\ai-service
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

If your AI provider requires secrets, add them in ai-service/.env.

5. Start the frontend

cd ../frontend
npm install
npm run start

Open http://localhost:4200.

The Angular dev server uses frontend/proxy.conf.json to forward:

• /api to http://localhost:8080
• /api/ai to http://localhost:8000

Common Commands

Frontend

cd frontend
npm run start
npm run build
npm run test

Backend

cd backend
./mvnw test
./mvnw spring-boot:run

Database

cd database
docker compose up -d
docker compose down

Configuration

Backend configuration lives in backend/src/main/resources/application.properties.

Important local defaults:

spring.datasource.url=jdbc:mysql://localhost:3307/ecommerce_db
spring.datasource.username=root
spring.datasource.password=root
app.internal.secret=super-secret-internal-key-12345

For production or shared deployments, replace local database credentials, JWT-related secrets, Stripe keys, and internal service secrets with environment-specific secure values.

API Documentation

The backend includes SpringDoc OpenAPI support. After starting the backend, check:

http://localhost:8080/swagger-ui.html

Depending on the SpringDoc route mapping, the UI may also be available at:

http://localhost:8080/swagger-ui/index.html