💳 Credit System – Credit Analysis and Granting Platform

A modern platform for credit analysis and approval, featuring traditional and social login, a microservices architecture, asynchronous communication via RabbitMQ, PDF generation, AWS S3 integration, and much more.

Ideal for simulating a real-world credit system with a robust, scalable, and secure architecture.

---

🗂 Project Overview

This project is composed of five microservices:

• user-service: Handles user registration, authentication, social login, proposal submission, and access to proposal history.

• proposal-service: Receives submitted proposals, stores them with user information and a status of "PENDING", then forwards them for analysis.

• credit-analysis-service: Analyzes proposals based on the user's CPF and a randomly generated score. Returns the analyzed proposal with the status updated to "APPROVED" or "REJECTED".

• notification-service: Sends emails regarding proposal status, user validation links, and links to download receipts.

• documentation-service: Generates PDF receipts for "APPROVED" proposals, uploads them to AWS S3, and sends the download link to the notification-service.

• Tests: Includes 244 unit and integration tests, with 100% code coverage.

✨ Features

• 🔐 User registration and authentication (traditional and social login with Google)
  On the first Google login, only the user's email and a default password are stored in the database.
  A temporary validation link is sent via email, allowing the user to complete missing data (name, CPF, password, and income) to become a valid user.
• 📧 Email notifications for validation, proposal status, and document download (via JavaMail)
• 📄 PDF receipt generation for approved proposals
• ☁️ Integration with AWS S3 for uploading and retrieving receipts
• 🔁 Asynchronous communication between services via RabbitMQ
• 🌐 Synchronous REST communication between services via WebClient + OAuth2 (Client Credentials)
• 🧠 Automated credit analysis with random score logic
• 🧾 Account deletion request flow: users must submit a reason; administrators receive a notification and decide whether to approve the deletion
• ⛔ Security filters to block access from unvalidated users
• 🔄 Temporary links with expiration (10 minutes) for email validation.
• 🔄 Temporary links with expiration (1 hour) for receipt download.
• ⏱️ Rate limiting: only one proposal submission every 24 hours
• ♻️ PDF regeneration and re-sending if the link has expired
• 💥 Circuit Breaker and fallback mechanism for service resilience (using Resilience4j)
• 📑 API documentation with OpenAPI and Swagger UI
• 📊 Application monitoring with Prometheus and Grafana
• 🐳 Dockerized services and deployment via Docker Hub

🧑‍💻 Technologies

🔧 Backend & Security

• Spring Boot, Spring Security, OAuth2 (Resource & Auth Server), JWT
• Jakarta Bean Validation, MapStruct, Thymeleaf

📡 Communication

• RabbitMQ (Async), WebFlux + WebClient (Sync)

📦 Cloud & Storage

• AWS S3, JavaMailSender

📊 Monitoring & Documentation

• SpringDoc OpenAPI, Spring Boot Actuator, Micrometer, Prometheus, Grafana

🧪 Testing

• JUnit, Mockito, TestContainers, WireMock, REST Assured

🛠️ Resilience & Patterns

• Resilience4j (Circuit Breaker, Fallback), Design Patterns (Factory, Strategy, Builder, etc.)

🧱 Design Patterns

The project applies well-known design patterns to improve maintainability, testability, and scalability:

• Factory
• Strategy
• Singleton
• Builder

🧪 Development Practices

• TDD (Test-Driven Development)
• BDD (Behavior-Driven Development)

🚀 Getting Started

To run this system locally, you need Docker and MySQL Workbench. No manual dependency installation is required — all services run via Docker Compose.

---

🐳 Step 1: Run Docker Compose

Download and use this docker-compose.yml file:Docker file

⚠️ Before running, set the following environment variables in the file:

🔐 Social Login (Google)

• GOOGLE_CLIENT
• GOOGLE_SECRET

☁️ AWS S3 Access (requires an IAM user with S3FullAccess)

• ACCESS_KEY
• SECRET_KEY

📧 Email Configuration (for notifications)

• EMAIL
• EMAIL_PASSWORD → This must be an "App Password", not your regular email password.

Then, navigate to the folder containing the file and run:

docker-compose up 

🛠️ Step 2: Manual Setup (one-time only)

Open MySQL Workbench and create two separate connections:

• Connection 1: localhost:3307 → User DB
• Connection 2: localhost:3308 → Proposal DB
  (Username and password for both: root)

Now, insert the following values manually:

🔐 User DB (Roles table)

Id	Name
1	ADMIN
2	USER

🔑 Proposal DB (Client table)

ClientId	ClientSecret	Scope
userserviceclient	$2a$10$KJz93hA5eQ9re6RIeXZwl.eqFb5au6//25IRdw19f/T9bsa5WpFR2	ADMIN

---

📬 Try it Out

Once everything is set up, open the API documentation:

🔗 Swagger UI

📞 Info

Murillo Marques

A LinkedIn post demonstrates the main flow of this API — feel free to check it out: LinkedIn Post

Email