LoopBack VS typetta

Backend is usually an API you design, can simply use Rails for this, basic Express.js or some framework like https://loopback.io/ if you want javascript otherwise you want 'system design' https://github.com/donnemartin/system-design-primer for all the load balancers/dbms if that's what you mean.

LoopBack – Highly Extensible Node.js and TypeScript framework

We deliberately decided not to use "kitchen-sink included" frameworks such as Laravel, Django, Loopback or Nest.js to keep things simple.

Loopback

I'd say check out https://loopback.io

This interface defines the basic CRUD capabilities that any entity should provide. This contract is based on frameworks like Loopback, Django RF, RoR and Spring. The Service.kt looks something like:

Have you ever tried Typetta? It's a Typescript ORM based on GraphQL model definition. ;-)

Have you ever heard of Typetta? We proudly built a Typescript ORM for both SQL and noSQL databases that gives you the ability to build a completely auto-generated GraphQL server to access your data in less than 5 minutes.

Give a try to Typetta!

The GitHub repository is the following: https://github.com/twinlogix/typetta The documentation: https://twinlogix.github.io/typetta/

I think you can't implement every security policy for every type you can request in every resolver. It requires a lot of code duplication and increase complexity. It's simply not maintainable. I ended up thinking that security policies should be applied by the layer that is in charge to load the graph, but this is just my opinion. I discussed all this things in a Medium article (https://blog.devgenius.io/how-to-secure-your-graphql-api-part-1-eb41f66ec07e) where I also proposed a solution based on a library that me and my team built to simplify the data access in a GraphQL context (it's name is Typetta and you can find more details here https://twinlogix.github.io/typetta/).

I used Mongoose in the past, but I found it rough in terms of typing, that is important in a TS context. I also tried the official driver but, you know, it's only a driver. I (and my team) ended up building something new, a sort of a type-safe open-source ORM we named Typetta (https://twinlogix.github.io/typetta/). We are using it in some production projects, with a very good feedback, but my questions is still there: how do others face the problem of managing a data layer on top of MongoDB?