joi VS class-validator

Package Feature Merit Demerit Link Zod TypeScript-first schema validation with static type inference Method chaining and easy to get ts type Newer package zod.dev joi This is almost same to zod Have many plugins Doesn't support static type inference https://github.com/sideway/joi Yup This is almost same to zod Have many plugins Missing some feature, function, deep object, promise... ... https://github.com/jquense/yup class-validator Validate class properties Have many features Code isn't clean https://github.com/typestack/class-validator ajv Validate by json scheme Enable to share schem to other languages Not supported typescript https://ajv.js.org/

Using Joi, a data validator for JavaScript.

See how I've used joi for validating the request body. You can install and use different libraries like this inside the helpers/routes. Joi can be installed with the following command:

3. joi

⭐ Recommended: Doing this manually is a ton of work, and if you're dealing with unknown objects often it would be better to lean onto a validation library to simplify it. My personal favorite is superstruct, although lots of people like joi or zod. They all do the same thing

The class-validator / class-transform looks really nice for formatting/validation but there is also joi. suretype is attractive since theoretically typeconv could generate jsonSchema for json-schema-faker and Open API docs.

Since Typescript is just a language and not a library of some sorts, it doesn't add runtime functionality. If this is what you're after, there's so many libraries that help converting unknown's into known types - joi, yup, superstruct (my choice), purify-ts Codec, io-ts

What does this library provide that is not provided by Yup, Zod, Superstruct (a personal favorite), Joi, Nope, computed-types, typanion, and all the other validation libraries?

True, I remember there's Yup and joi for JavaScript

You can learn more from the documentation here.

This is the request body when creating a new task. The IsString decorator provided by class-validator rejects all types other than strings, including undefined or null. The task name value object does not consider the possibility of values other than strings.

In this class, we are using the famous class-validator to decorate our request parameters. This way it is more clear how we are going to use our request object and everything.

We add validation using https://github.com/typestack/class-validator.

I will make sure we get this documented.

Also OP can check out class-transformer + class-validator if he want to properly instantiate objects and validate them with precise and verbose output of potential format errors. OP can even go the god tier way by setting up a custom RxJS operator that does that for him easily.

Start by installing the class-transformer and class-validator packages for dto validation (see more in the docs), sharp for image optimization, and the S3 client packages.

The description of the request body model in NestJS is similar to the description of Ditsmod models, but NestJS uses decorators provided by the utility class-validator. In NestJS, it is customary to name the request body model with the ending *Dto (this is an abbreviation of Data transfer object):

You could use Zod or Superstruct to validate the types of values, or use typestack/class-validator if it makes sense.

Also, title of before article was I made 1,000x faster TypeScript validator library, but today's benchmark is showing that typescript-json is maximum 6,500x times faster. It's just because I added a new library class-validator in the benchmark and it is much slower than previous zod. Of course, I've continuously tuned the performance, but it just improved the performance about 2x times only.