Defining a schema

Goal: check that a value conforms to a predefined set of rules.

Instead of using if/else for every parameter we can define a structure that represents our desired data. That's what a schema is all about!

Effect schema is a set of APIs inside effect that allows to decode and encode data based on a schema.

Schema became part of effect core since version 3.10 (course update commit).

If you are using an older version of effect, you need to install a separate @effect/schema package.

pnpm add @effect/schema

The main change is that all the modules previously exported from @effect/schema are now exported from effect directly:

import {
  Arbitrary,
  AST, 
  SchemaAST, 
  FastCheck,
  JSONSchema,
  ParseResult,
  Pretty,
  Schema
} from "@effect/schema"
} from "effect"

The link to Effect playgrounds in this course are still using the old @effect/schema package.

Under the hood schema will check that the data conforms with every rule we defined, or report a detailed error of what went wrong.

In practice schema looks like a more powerful typescript interface.

Instead of defining a plain typescript interface:

interface Pokemon {
  id: number;
  order: number;
  name: string;
  height: number;
  weight: number;
}

We create a value that represents the same schema using Schema:

• Schema.Struct: schema for an object with properties
• Schema.Number: schema for a number
• Schema.String: schema for a string

import { Schema } from "effect";

const Pokemon = Schema.Struct({
  id: Schema.Number,
  order: Schema.Number,
  name: Schema.String,
  height: Schema.Number,
  weight: Schema.Number,
});

This looks strikingly similar to a normal interface, doesn't it?