Defining Types: Structs and Enums

While MON is flexible, sometimes you need to enforce a specific structure for your data. The optional MON type system helps you do just that, preventing typos and ensuring data integrity and schema validation. Unlike JSON, which relies on external JSON Schema definitions, or TOML with its more limited type capabilities, MON provides built-in mechanisms for defining robust configuration schemas. The two main ways to define types are Enums and Structs.

Enums (#enum): A Fixed Set of Choices

An Enum is a type that can only be one of several predefined values. It’s perfect for things like status, categories, or modes.

Syntax:

{
    // Defines a type named `Status` that can only be one of these three values.
    Status: #enum {
        Active,
        Inactive,
        Pending,
    },
}

To use an enum value, you reference it with a dollar sign $ followed by the type name and the variant name.

Example Usage:

{
    Status: #enum { Active, Inactive, Pending },

    // Using the enum value
    current_status: $Status.Active,
}

Structs (#struct): Defining an Object’s Schema

A Struct defines the “shape” of an object. It specifies what keys are allowed, what type their values should be, and can provide default values for optional keys.

Syntax:

{
    // Defines a schema for a User object
    User: #struct {
        // `id` is a required field of type Number
        id(Number),

        // `username` is a required field of type String
        username(String),

        // `email` is optional and defaults to null if not provided
        email(String) = null,

        // `is_active` is optional and defaults to `true`
        is_active(Boolean) = true,
    },
}

Built-in Types: You can use String, Number, Boolean, Null, Array, Object, and Any (which allows any value).

Challenge 4: Define a Product Schema

Create a file product.mon. Inside it, define a #struct named Product. The Product struct should have:

1. A required id of type String.
2. A required name of type String.
3. A required price of type Number.
4. An optional in_stock field of type Boolean that defaults to false.

---

Answer to Challenge 4

{
    Product: #struct {
        id(String),
        name(String),
        price(Number),
        in_stock(Boolean) = false,
    },
}

Fantastic! You’ve defined your first custom types. But how do you actually use them to validate your data? That’s next!