Type system

Type expressions appear in schema property definitions under the type key. Types are inferred for resource files, so explicit type annotations are only needed in schemas. This page covers every type constructor and how they compose.

Primitive types ​

properties:
  name:
    type: string
  birth_date:
    type: date
  active:
    type: boolean

Nullable types ​

Append ? to any type to make it nullable. The field accepts either the declared type or null. Omitted nullable fields default to null.

type: string?
type: date?
type: list[string]?

List types ​

list[T] is a list where every element has type T.

type: list[string]
type: list[number]
type: list[Task]

Dict types ​

dict[K, V] is a homogeneous mapping from keys of type K to values of type V.

type: dict[string, number]

Fixed-key object types ​

Inline a mapping to define a nested object type:

type:
  street:
    type: string
  city:
    type: string
  zip:
    type: number

Union types ​

Use | to combine types. Accepts any member.

type: string | number
type: string | null
type: list[string] | null

Parentheses group when needed:

type: (string | number)?    # string, number, or null
type: list['a' | 'b' | 'c']

A YAML list also works:

type: [string, number]

Enum types ​

A union of string or number literals:

type: 'todo' | 'in_progress' | 'done'
type: 1 | 2 | 3

List syntax also works:

type: ['todo', 'in_progress', 'done']

String literals must be quoted to distinguish them from type names.

Literal types ​

A type with exactly one valid value:

type: 1           # must be the number 1
type: "draft"     # must be the string "draft"

Schema references ​

Use a schema name as a type to declare a link field. The field expects a fref to a resource conforming to that schema.

type: Person           # link to a Person resource
type: Person?          # optional link
type: list[Task]       # list of links