Studio Schema

🧩 Plain JavaScript objects with runtime type guarantees. For node and browsers.

Usage

Defining a schema:

const { schema, object, string, integer, opt } = require('@studio/schema');

/**
 * @typedef {ReturnType<typeof person>} Person
 */
const person = schema(
  object({
    name: string, // mandatory
    age: opt(integer) // optional
  })
);

Schema Validation

The schema is a function that can be used to validate a given object. It throws if non-optional properties are missing, a value has the wrong type, or undeclared properties are present.

person({ name: 123 }); // throws
person({ name: 'Test', customer: true }); // throws
person({ name: 'Test', age: true }); // throws
person({ name: 'Test', age: 7.5 }); // throws

person({ name: 'Test' }); // ok
person({ name: 'Test', age: 7 }); // ok

Validators

Validators are functions that verify values and objects. A validator function returns false if a value doesn't meet the expectation. Studio Schema comes with a set of pre-defined validators for primitive types, enums, objects, arrays and maps. Validators can be nested and reused:

const { schema, literal, string, object, array } = require('@studio/schema');

const status = literal('LOADING', 'LOADED', 'SAVING', 'SAVED');
const person = object({
  first_name: string,
  last_name: string,
  tags: array(string)
});

/**
 * @typedef {import('@studio/schema').Infer<typeof personModel>} PersonModel
 */
const personModel = schema(object({ status, person }));

Readers

A reader validates a given object and returns a Proxy that makes the object read-only and verifies that only defined properties are accessed.

const person = person.read({ name: 'Test', age: 7 });

const name = person.name; // ok
const age = person.age; // ok
const customer = person.customer; // throws
person.name = 'Changed'; // throws

Writers

A writer accepts an empty or partial object and returns a Proxy that validates any accessed, assigned or deleted properties. To verify that no non-optional properties are missing, use mySchema.verify(writer).

const alice = person.write({ name: 'Alice' });

alice.customer = true; // throws
alice.name = 'Changed'; // ok
alice.age = 7; // ok

Errors

All schema validation errors have a code property with the value E_SCHEMA.

Types

Validators and schemas creates TypeScript types from the given structure. Resolve the types like this:

/**
 * @typedef {import('@studio/schema').Infer<typeof mySchema>} MySchema
 */
const mySchema = schema(someValidator);

/**
 * @typedef {import('@studio/schema').Infer<typeof myLiteral>} MyLiteral
 */
const myLiteral = literal('foo', 'bar');

API

This module exports the schema function which carries the entire API, so you can require it in two ways:

const schema = require('@studio/schema');

const person = schema({ name: schema.string });

With destructuring:

const { schema, string } = require('@studio/schema');

const person = schema({ name: string });

• schema(validator[, options]): Returns a new schema with the given validator. validator The validator to use for the schema. These options are supported:
  • error_code: The code property to define on errors. Defaults to E_SCHEMA.
• defined: Is a validator that accepts any value other than undefined.
• boolean: Is a validator that accepts true and false.
• number: Is a validator that accepts finite number values.
• number.min(min): Is a validator that accepts finite number values >= min.
• number.max(max): Is a validator that accepts finite number values <= max.
• number.range(min, max): Is a validator that accepts finite number values >= min and <= max.
• integer: Is a validator that accepts finite integer values.
• integer.min(min): Is a validator that accepts finite integer values >= min.
• integer.max(max): Is a validator that accepts finite integer values <= max.
• integer.range(min, max): Is a validator that accepts finite integer values

  = min and <= max.

• string: Is a validator that accepts string values.
• string.regexp(re): Is a validator that accepts string values matching the given regular expression.
• string.length.{min,max,range}: Verifies the string length with an integer validator.
• literal(value_1, value_2, ...): Returns a validator that matches against a list of primitive values. Can be used to define constants or enumerations.
• all(validator_1, validator_2, ...): Returns a validator where all of the given validators have to match.
• one(validator_1, validator_2, ...): Returns a validator where one of the given validators has to match.
• opt(validator): Returns an optional validator.
• object(properties): Returns an object validator. The given object maps object keys to validators.
• object.any: Is a validator that accepts arbitrary object values.
• array(itemValidator): Returns an array. Each element in the array has to match the given validator..
• array.any: Is a validator that accepts arbitrary array values.
• map(keyValidator, valueValidator): Returns a map validator for key-value pairs where keyValidator and valueValidator are the validators for the object key and value pairs.
• validator(test[, toString]): Creates a custom validator for the given test function. The optional toString argument can be a function that returns a string, or a string defining the validator name used in error messages. Defaults to <custom validator>.
• E_SCHEMA: The code property exposed on schema validation errors.

Note that all validator functions are also exposed on schema.

Schema API

The schema created by schema(validator) is a function that throws a TypeError if the given value does not match the validator, and returns the value otherwise. For object and array, the schema also allows to create proxy objects that validate reading, assigning and deleting properties:

• reader = mySchema.read(data[, options]): Creates a schema compliant reader for the given data. If the given data does not match the schema, an exception is thrown. The returned reader throws on any property modification or on an attempt to read an undefined property. These options are supported:
  • error_code: The code property to define on errors. Defaults to E_SCHEMA.
• writer = mySchema.write([data[, options]]): Creates a writer with optional initial data. If the given data does not match the schema, an exception is thrown. The returned writer throws on undefined property modification, if an assigned value is invalid, or on an attempt to read an undefined property. These options are supported:
  • error_code: The code property to define on errors. Defaults to E_SCHEMA.
• data = mySchema.verify(writer): Checks if any properties are missing in the given writer and returns the unwrapped data. Throws if the given object is not a schema writer.
• data = reader_or_writer.toJSON(): returns the original object.

Note that schema readers and writers can be safely used with JSON.stringify.

License

MIT

Made with ❤️ on 🌍