ORM

An ORM is an Object-relational mapping and is one of the most important concepts when dealing with communication between a server and a database. An ORM provides a mapping between objects in memory (Defined classes such a User or Comment) and Relational tables in a database. This allows you to create a Data Transfer Object that knows how to write objects stored in memory to a database, and read the results from an SQL or another query language, back into memory. In this book, we will talk about three different ORMs: two relational and one for a NoSQL database. TypeORM is one of the most mature and popular ORMs for Node.js and thus has a very wide and flushed out feature set. It is also one of the packages that Nest provides its own packages for: @nestjs/typeorm. It is incredibly powerful and has support for many databases like MySQL, PostgreSQL, MariaDB, SQLite, MS SQL Server, Oracle, and WebSQL. Along with TypeORM, Sequelize is also another ORM for relational data.

If TypeORM is one of the most popular ORMs, then Sequelize is THE most popular in the Node.js world. It is written in plain JavaScript but has TypeScript bindings through the sequelize-typescript and @types/sequelize packages. Sequelize boasts strong transaction support, relations, read replication and many more features. The last ORM covered in this book is one that deals with a non-relational, or NoSQL, database. The package mongoose handles object relations between MongoDB and JavaScript. The actual mapping between the two is much closer than with relational databases, as MongoDB stores its data in JSON format, which stands for JavaScript Object Notation. Mongoose is also one of the packages that has a @nestjs/mongoose package and provides the ability to query the database through query chaining.