Skip to main content

schemaFactory

npm Package Size Security Scorecard CI

Use schemaFactory to build custom schemas when built-in helpers are not enough.

Use this only for advanced custom validators.
Most app code should use string, number, object, array, or, union, and other standard schema builders.

import {schemaFactory} from '@livon/schema';

const slug = schemaFactory({
  name: 'Slug',
  type: 'string',
  ast: () => ({type: 'string', name: 'Slug'}),
  validate: (input) => {
    if (typeof input !== 'string') {
      throw {message: 'Expected string'};
    }
    return input.trim().toLowerCase();
  },
});

const value = slug.parse('hello-world');

Parameters​

  • name (string, required): schema name.
  • type (string, required): schema type id used in AST/errors.
  • ast ((ctx) => AstNode, required): AST builder.
  • validate ((input, ctx) => TValue, required): runtime validator.
  • doc (SchemaDoc, optional): metadata attached to AST/doc output.
  • chain (Record<string, (value, ctx) => (...args) => nextValue>, optional): fluent chain methods.

Chain API​

  • Chain methods are defined inside chain.
  • Each entry receives (value: TValue, ctx: SchemaContext) and returns a function for method args.
  • Each chain call returns SchemaWithChain<TNext, TChain>, so chains stay immutable.

Minimal chain definition:

const text = schemaFactory({
  name: 'Text',
  type: 'string',
  ast: () => ({type: 'string', name: 'Text'}),
  validate: (input) => String(input),
  chain: {
    maxLen: (value: string) => (max: number): string => value.slice(0, max),
  },
});

const parsed = text.maxLen(10).parse('hello world');

When to use schemaFactory:

  • building custom schema primitives
  • adding custom chain methods
  • keeping AST + runtime validation behavior in one place