ahmed-dinar / digital-library

Simple Digital Library

Live Demo

A live demo is available at https://lib.dinar.sh.

Backend and frontend is hosted on Vercel using Github Actions. Database (PostgreSQL) is hosted on Neon. As these are in free tier, demo might be a little bit slow.

Technology Stack

• Backend

  • NestJS
    • Nest has built-in support for TypeScript and a modular architecture, making it easy to organize and scale large applications.
    • Nest provides built-in support for dependency injection (DI), validation, microservices, monolith, rate limiter, documentation, etc., which can help maintain code out of the box to production quality.
  • Prisma (ORM)
    • Prisma ensures type-safety by generating typings in TypeScript and includes built-in migrations.
  • Jest (Test)

• Frontend

  • Next.js
    • React framework supports server-side rendering (SSR), filter-based routing and server components. It simplifies the development process, optimizes performance and enhances SEO.
  • Ant Design
  • Tailwind CSS
    • An easy to use CSS utility helps to design website using predefined building blocks.

• Database

  • PostgreSQL 15
    • Considering the requirements of the digital library, book records with fixed fields (title, summary, author etc), a SQL database is likely the most suitable choice.
    • SQL provide transaction support, strong consistency with ACID property which is required for book record integrity.
    • SQL can perform complex Queries which is required to retrieve data from book, author, genre etc entries.

• DevOps

  • Github Actions (CI/CD)
  • Docker (local run)

Running the dockerized app

Note

Please ensure that you have docker and docker compose installed in your machine. Docker compose version should be >= 2.2

To run the app, three Docker containers are required: one for the backend, one for the frontend, and another for the database.

# Clone the code to your local machine
$ git clone https://github.com/ahmed-dinar/digital-library.git

# Go to the app code directory
$ cd digital-library

# Running backend, frontend and database
$ docker-compose up -d

# To see the logs
$ docker-compose logs -f

Using the app

• Frontend is running on port http://localhost:8080.
• Backend is running on port http://localhost:3000.

Seeding the database

To seed the database with some demo book records:

• Go to http://localhost:8080/backdoor
• Click the appropriate button to seed or clear the seed, as shown in the attached screenshot below.

Performance Consideration

• Caching

  • Implement caching mechanisms to store frequently accessed book records to reduce database load and response time.
  • Tools like Redis or memcached can be used for caching.

• Load Balancing

  • Add multiple instance of server to balance the growing request.
  • Add a load balancer to distribute incoming traffic evenly across multiple instances.
  • This helps in scaling horizontally and handling increased user traffic effectively.

• Rate limiting

  • Add a rate limiter to prevent misuse and denial-of-service (DoS) attacks.
  • Currently we are allowing all users operation without authentication, we can limit request by user IP address within a specified timeframe.

• Containerize Scaling

  • We can utilize Docker and Kubernetes to achieve horizontal scaling.
  • By adding and removing containers based on the traffic, our app can handle traffic loads efficiently.

• Database Performance

  • We could reduce and limit some column length, E.g. book title, name etc to a meaningful limit.
  • We could add indexing on frequent used columns, E.g. book title, book summary (text search), author and genre name (search)
  • As the request grow, we could have database replicas. E.g. read and write replica
  • Partition the database into multiple shards as it grows to distribute data across multiple servers and improve query performance.
  • Utilize database caching to cache frequently accessed book entries, as reads are more common than write operations, to reduce query latency.
  • If we need to serve media files (book images, audio etc), we could use a object store (E.g. AWS S3).
  • Database backup can help to recovery of any data loss.

• Frontend Optimization

  • We could implement client-side caching and pagination to reduce the amount of data fetched from the server.
  • Use CDN to host frontend static assets (CSS, JavaScript) for faster loading times.
  • Utilize lazy loading for images and components for faster loading.

• Monitoring and Logging

  • Add logging to both server and frontend, we can track any errors and perform troubleshooting in production.
  • We could use tools like Sentry (error tracker), Hotjar, analytics (Google analytics etc) in frontend to track real time monitoring.
  • We could use GCP, AWS or other logs service in backend to monitor incident real time.

Further Improvements

• Development improvements

  • Add more logs in book add/remove/update services to track who/when/what is modifying.
  • Add more comments on code where necessary.
  • Add API documentation tool (E.g swagger) to maintain auto document backend APIs.
  • Generate and share a postman API collection.
  • Improve UX in modifying author and genres, in search book option by multiple fields, more sort functionalities.

• Frontend testing

  • We should add frontend unit and e2e testing to make the UI UX quality better and consistent.
  • Framework like Playwright, Cypress can be used to test UI.

• Authentication and Authorization

  • We could introduce authentication before perform any modification (create, update, delete).
  • We could have authorization to limit perform modification only books that are owned by a user.

• Backend error mapping

  • We could add more meaningful error from backend. E.g. adding error codes to error type can help to debug production errors.

• Internalization

  • We could add support for multiple languages to provide users more better experience.
  • Next.js has built-in i18n supports we can utilize that.

• User specific features

  • Features like add a book to favourite, track reading progress, add to user collections can help user to find the app more useful.
  • Suggested books based on user interaction and reading history can improve UX better.
  • A socializing feature like an user can add other user as a friend and see/share their interests.

• Additional features

  • Information validation based on user votes (genres attached to book, author etc) can help the reliability of contents.
  • Like, comment, rating on books can improve interest of users.
  • Introduce audio book supports to listen to the book.
  • A pdf/ebook reading experience built int in the app.
  • We could have a home page section with top/popular/new books.

Current Database Design

• We have four tables. Book, Author, Genre and Edition.
• Books are unique across the system by their title.
• Authors are unique across the system by their name.
• Genres are unique across the system by their name.
• A Book can be associated to multiple Authors and Genres and vice versa.
• A new edition of book should have have a unique ISBN associated to it.

Development

Note

Please ensure that you have node >= 18.17.0 or higher installed.

Please find the development guides in server and client directory readme.

License

License under MIT please find the file here.