Realio- Go MicroService Application

Realio is a Micro-service architecture built with Go and Python, while adhering to clean architecture principles, provides a scalable and modular way to handle core functionalities like property listings, user management, search, and recommendations. This architecture allows users to view, filter, and get personalised recommendations for properties they may wish to buy. Below is an in-depth look at how to structure and implement these services:

1. Microservices Overview

This application can be broken down into multiple Micro-services, each handling a specific function. Here’s a high-level overview of key services:

Core Microservices

1. User Service (Go): Manages user accounts, authentication, and profiles.
2. Property Service (Go): Manages property listings, including CRUD (Create, Read, Update, Delete) operations.
3. Search Service (Go): Provides search functionality with filters for property type, location, price range, etc.
4. Recommendation Service (Go): This service uses machine learning to provide property recommendations based on user preferences and past interactions.
5. Booking and Scheduling Service (Go): Manages property viewing appointments and scheduling.
6. Messaging Service (Go): Facilitates communication between buyers, sellers, and agents.
7. Notification Service (Go): This service sends alerts and updates to users via email, SMS, and in-app notifications.

Each service is designed to operate independently, allowing for modular deployment and scaling based on user demand.

Detailed Service Architecture

1. User Service (Go):

• Responsibilities: Manages user registration, authentication, and user profile data.
• Database: PostgreSQL for structured user data.
• Endpoints:
  • POST /register: Register a new user.
  • POST /verify_user: Verify new users OTP.
  • POST /resend_otp: Resend User OTP.
  • POST /check_user_email: Check if Users’ Emails Already Exist.
  • POST /logout: Logout User.
  • POST /login: Authenticate a user.
  • GET /profile/{user_id}: Retrieve user profile details.
  • GET /refresh_token: Refreshes access token when user token is expired.
• Storage: Cloudinary or Amazon S3 for User Image Storage.
• Implement Role-Based Access Control (RBAC) for users, agents, and admins.

2. Property Listings (Go):

• Responsibilities: Manages property listings, CRUD operations, and image storage.
  • Database: PostgreSQL for structured user data and Kafka as a Messaging Broker.
• Endpoints:
  • GET /properties: Retrieve a list of properties.
  • POST /properties: Add a new property.
  • PUT /properties/{property_id}: Edit property Added.
  • DELETE /properties/{property_id}: Delete property Added.
  • GET /properties/{property_id}: View property details.

3. Search Service (Go)

• Responsibilities: Provides search and filtering functionality.
• Database: Elasticsearch for search indexing.
• Event Management: Data Sync at Intervals, When properties are created, updated, or deleted in the property management service, it can publish events containing relevant property details. These events can be sent through a message broker (Kafka) to which the search service subscribes. The search service then indexes the properties in Elasticsearch based on these events.
• Endpoints:
  • GET /search: Retrieve properties based on search criteria.

4. Booking and Scheduling Service (Go)

• Responsibilities: Manages property view scheduling and booking confirmations.
• Database: PostgreSQL to handle structured booking data.
• Endpoints:
  • POST /bookings: Create a booking.
  • PUT /bookings/{booking_id}: Modify or reschedule a booking.
  • DELETE /bookings/{booking_id}: Cancel a booking.
  • GET /bookings: Retrieve a list of bookings.
  • GET /bookings/{booking_id}: Edit property details.

5. Messaging Service (Go)

• Responsibilities:
  • Manages conversations and messages between users.
  • Supports both real-time communication via Socket.IO and persistent message storage in MongoDB.
  • Exposes RESTful API endpoints for sending, retrieving, and deleting messages in conversations
• Database: MongoDB for storing messages and conversations.
• Realtime Connection: SocketIO for real-time
• Endpoints:
  • POST /messages: Send a message.
  • GET /messages/{conversation_id}: Retrieve messages in a conversation.
  • DELETE /messages/{conversation_id}: Delete messages in a conversation.
• Socket IO:
  • join_room: Users join a conversation room.
  • send_message: Sends a message to the room.
  • receive_message: Listens for new messages in the room.
• Database Design: Using MongoDB is suitable here due to its flexibility with nested data, allowing efficient storage and retrieval of messages and conversations.

6. Notification Service (Go)

• Responsibilities: Sends notifications for bookings, messages, or updates.
• Database: Redis for temporary storage and PostgreSQL for persistent notification storage.
• Endpoints:
  • POST /notifications: Send a notification.
  • GET /notifications/{user_id}: Fetch a list of user notifications.
  • PUT /notification/{notification_id}: Mark Notification as Read.
  • PUT /notifications: Mark All Notifications as Read.

7. Recommendation Service (Python)

• Responsibilities: The recommendation service remains in Python, leveraging machine learning libraries like Scikit-Learn and Reinforcement Learning for personalised recommendations. This service connects seamlessly with other Go-based services, serving as an independent Micro-service for generating property suggestions based on user history and interactions.
• Database: Redis for temporary storage and PostgreSQL for persistent notification storage.

• Endpoints:

Scalability Considerations

• Load Balancer: Use NGINX or AWS ELB to balance the load between instances. **Optional
• Database Scaling: Use Read Replicas for scaling reads. Sharding for large datasets. **Optional
• Caching: Use Redis to cache frequently accessed data like property listings.
• Search: Use Elasticsearch for handling complex search queries.
• Microservices: Break down large components into microservices (e.g., separate services for User, Property, Booking).

Security Considerations

• Data Encryption: Use HTTPS for secure communication. Encrypt sensitive data like passwords.
• Access Control: Implement proper authorization for different roles (buyers, sellers, agents, admins).
• Input Validation: Sanitize inputs to prevent SQL injection and other common vulnerabilities.
• Rate Limiting: Prevent abuse by implementing rate limits on critical endpoints.

Monitoring & Maintenance

• Use Prometheus and Grafana for monitoring and alerting.
• Use ELK Stack (Elasticsearch, Logstash) for logging and visualization.
• Automate deployments with CI/CD tools like GitHub Actions.
• Regular backups for databases using tools like AWS Backup or scheduled cron jobs.

Third-Party Integrations

• Crash Analytics: Sentry
• Geolocation & Maps: Google Maps API.
• Analytics: Google Analytics.
• Notifications: Firebase for push notifications.

DATABASE TABLE:

User Table:

Field	Type	Description
id	UUID	Primary key
name	String	User's full name
email	String	User's email (unique)
password	String	Hashed password
role	Enum	buyer, seller, agent, admin
phone	String	Contact number
created_at	Timestamp	Timestamp of registration
updated_at	Timestamp	Timestamp of last update

User Session Table Schema

Column	Data Type	Description
session_id	UUID	Unique identifier for each session.
user_id	UUID	Foreign key linking to the user table (identifies the user).
token	VARCHAR (255)	The session token, which can be a JWT or another token format.
created_at	TIMESTAMP	Timestamp of when the session was created.
expires_at	TIMESTAMP	Timestamp of when the session expires.
last_activity	TIMESTAMP	Tracks the last activity time for session timeout checks.
ip_address	VARCHAR (45)	The IP address from which the session was initiated.
user_agent	VARCHAR (255)	The user agent (browser or device info) for the session.
is_active	BOOLEAN	Indicates whether the session is currently active.
revoked_at	TIMESTAMP	Timestamp for when the session was revoked, if applicable.
device_info	JSON or VARCHAR	(Optional) Stores additional device details if needed.

Property Table:

Field	Type	Description
id	UUID	Primary key
title	String	Property title
description	Text	Detailed description
price	Decimal	Price of the property
type	Enum	House, Apartment, Land, etc.(Optional)
address	String	Address details
zip_code	String	Zip code(Optional)
owner_id	UUID	Reference to the user (seller)
images	Array	List of image URLs(Optional)
no_of_bed_rooms	Int	Number of Bed Rooms
no_of_bath_rooms	Int	Number of Bath Rooms
no_of_toilets	Int	Number of Toilets
geo_location	JSON	Latitude & longitude(Optional)
status	Enum	Available, Sold, Rented, etc.
created_at	Timestamp	Timestamp of listing
updatedAt	Timestamp	Timestamp of last update

Booking Table:

Field	Type	Description
id	UUID	Primary key
property_id	UUID	Reference to Property table
user_id	UUID	Reference to User table (buyer)
booking_date	Date	Date of the booking
status	Enum	Pending, Confirmed, Cancelled
created_at	Timestamp	Timestamp of booking
updated_at	Timestamp	Timestamp of last update

Message Table:

Field	Type	Description
id	UUID	Primary key
conversation_id	UUID	Reference to Conversation table
sender_id	UUID	Reference to Send on User table
content	String	Message Content
read	Bool	Confirm if message has been read by User
created_at	Timestamp	Timestamp of booking
updated_at	Timestamp	Timestamp of last update
receiver_id	UUID	Reference to Receiver on User table

Conversation Table:

Field	Type	Description
id	UUID	Primary key
sender_id	UUID	Reference to Send on User table
last_conversion	String	Message Content
read	Bool	Confirm if message has been read by User
created_at	Timestamp	Timestamp of booking
updated_at	Timestamp	Timestamp of last update
receiver_id	UUID	Reference to Receiver on User table