Objection.js VS Knex

Objection.js is a SQL-friendly ORM for Node.js that supports various relational databases, including PostgreSQL, MySQL, and SQLite. It provides a flexible and expressive query builder. Objection.js is known for its expressive syntax, allowing developers to build complex queries easily. It supports eager loading, transactions, and migrations.

I don't think there's a best per say, but we did recently use Objection on our project. Did the job well, only issue is there's no constructor for the DB Models but it's just something you work around (https://vincit.github.io/objection.js/)

If you aren't using TypeScript I'd very much recommend Objection.js, I've used it multiple times and no complaints so far. You can pass raw SQL queries to it as well so I'm sure it would be a good fit for your project!

I'd also like to point out objection js is no longer actively maintained. I'm going to switch my work's codebase from it eventually because of it.

You should look into Objection.js. IMHO using that will make your life much easier as it seems you are trying to reimplement it's features in this sample code.

Next, let’s set up the database for our newly created Nest application. I’ll be using PostgreSQL, but you can use any of the databases Knex supports. To interact with our database, we’ll be using Objection.js, which is an ORM for Node.js built on top Knex. For this tutorial, we’ll be using Nest Objection, a Nest module for Objection.

Knex.js

AdonisJS core team has created/maintains Lucid. It is a SQL query builder, and an Active Record ORM built on top of Knex.

Now, why not use an ORM? I've seen performance issues too many times with ORMs. I prefer writing my own SQL to avoid surprises. After all, I know the database schema and writing code for a specific purpose very often leads to better performance than generic code. ORMs have to support all kinds of database schemas. I only have to support mine. Having successfully used Knex.js in NodeJS (a popular query builder) in the past, I know writing SQL queries myself is not hard and provides very good performance.

Given the dynamic nature of the schema, we employ Knex, a query builder, for database access.

https://www.npmjs.com/package/knex - 779 issues

# related: https://github.com/knex/knex/issues/2354

This article will be focused on a security issue that I found in Knex and how to mitigate it, but I'll also talk briefly about the social aspects of this problem.

SQL is an old, irregular language to work with, but it is better known than HCL and SQL already has it's own Pulumi/CDK in the form of every ORM with introspection (like Javascript's Prisma, Python's Django, Go's XO etc) and QueryBuilder (LINQ, Knex, etc) in whatever programming language you prefer. You probably already know it.