ShapeType

Define shapes and types and use them for data/structure validation.

---

API

Type

Types allow you to define what type of data you expect a value to be.

Type Definitions:

• Type.bool() for a boolean value
• Type.number() for a number
• Type.string() for a string
• Type.value(val1[, val2, val3]) for an exact match of one or more values.
• Type.datetime() for a Date object
• Type.null() for a null value
• Type.undefined() for an undefined value
• Type.object() for an object literal
  • Note: You'll often want to use a Shape instead of Type.object() (see below).
• Type.array() for an array.
  • Note: You'll often want to use arrayOf() instead of Type.array()(see below).
• Type.custom(isType[, name]) allows you to define a custom Type.
  • isType should be a method that accepts a value and returns a boolean.
  • name is an optional value that will be stored as Type.value internally.
  • Example: const EmailType = Type.custom(val => isEmail(val), 'EMAIL');

Type Methods

• .compare(val): returns true/false whether val matches the Type.
  • Example:
    • EmailType.compare('[e-mail]') ~ true
    • EmailType.compare('test=test') ~ false
• .validate(val): returns an validation object assessing whether val matches the Type.
  • Example:
    • EmailType.validate('[e-mail]') ~ { invalidTypeFields: [] }
    • EmailType.validate('test=test') ~ { invalidTypeFields: [ 'test=test' ] }
• .or(): allows you to chain a list of Types together
  • Example: const assignedTo = Type.number().or(Type.null())

---

arrayOf(Type|Shape)

Indicates that there is an array of the provided value.

When validate is called on a value defined by arrayOf(), an empty array will be returned if all values pass their tests. An array of validation objects with an additional index key will be returned if any of the array's values do not pass their test.

---

optional()

Wraps a Type, Shape, or ArrayContainer to indicate that a given field isn't required in a shape definition. Tests will still pass if the value is omitted.

const Event = defineShape({
  id: Type.number(),
  name: Type.string(),
  createdBy: optional(User),
  scheduledAt: optional(Type.datetime()),
  guests: arrayOf(User),

---

defineShape({})

Returns a Shape when provided with an object of key/Type pairs.

const User = defineShape({
  id: Type.number(),
  name: Type.string(),
  assignedTo: Type.number().or(Type.null()),
})

You can also nest Shapes within other Shapes.

const Event = defineShape({
  id: Type.number(),
  name: Type.string(),
  createdBy: User,
  guests: arrayOf(User),
})

Shape Methods

• .compare(obj) returns true/false whether obj matches the pattern defined by the Shape.
  • Example:
    • User.compare({ id: 2, name: 'Test', assignedTo: 12 }) ~ true
    • User.compare({ id: 2, name: 'Test', assignedTo: 'Joe' }) ~ false
• .partialCompare(obj) same as .compare(obj), but only tests for the keys present in obj.
  • Example:
    • User.partialCompare({ id: 2 }) ~ true
    • User.partialCompare({ id: 'Toby' }) ~ false
• .validate(obj) returns a validation object reflecting the test results of the individual values defined by the shape.
  • The validation object consists of three arrays:
    • missingFields: fields that are defined in the Shape but missing from the object
    • extraFields: fields that are not defined in the Shape but are included in the object
    • invalidTypeFields: fields that don't match the Type defined by the Shape
  • Example:
    • User.validate({ id: 2, name: 'Test', assignedTo: 12 }) ~ { missingFields: [], extraFields: [], invalidTypeFields: [] }
    • User.validate({ id: 2, name: 'Test', assignedTo: 'Joe' }) ~ { missingFields: [], extraFields: [], invalidTypeFields: [ 'assignedTo' ] }
    • User.validate({ name: 'Test', assignedTo: 12 }) ~ { missingFields: [ 'id ], extraFields: [], invalidTypeFields: [] }
    • User.validate({ id: 2, name: 'Test', assignedTo: 12, isDog: true }) ~ { missingFields: [], extraFields: [ 'isDog' ], invalidTypeFields: [] }
• .partialValidate(obj) same as .validate(obj), but only tests for the keys present in obj.
  • Example:
    • User.partialValidate({ id: 2 }) ~ { missingFields: [], extraFields: [], invalidTypeFields: [] }
    • User.partialCompare({ id: 'Toby' }) ~ { missingFields: [], extraFields: [], invalidTypeFields: [ 'id' ] }

---

extendShape(existingShape, obj)

Allows you to build off of an existing shape. A new Shape instance will be returned.

const Rectangle = defineShape({
  length: Type.number(),
  width: Type.number(),
});
 
const Box = extendShape(Rectangle, { height: Type.number() });

---

Additional Examples

Usage in testing

You can use Shapes with a testing library such as Jest for type assertions.

it ('should return the user', () => {
  expect(UserShape.compare(user)).toEqual(true);	
});

Usage in field validation

const missingFields = UserShape.validate(user));
if (missingFields.length > 0){
  onError(missingFields);
} else {
  onSave(user);
}