lazy
Creates a lazy schema.
const Schema = lazy<TGetter>(getter);
Generics
TGetterextends (input: unknown) => BaseSchema
Parameters
getterTGetter
Explanation
The getter function is called lazily to retrieve the schema. This is necessary to be able to access the input through the first argument of the getter function and to avoid a circular dependency for recursive schemas.
Returns
SchemaLazySchema<TGetter>
Examples
The following examples show how lazy can be used.
Binary tree schema
Recursive schema to validate a binary tree.
Due to a TypeScript limitation, the input and output types of recursive schemas cannot be inferred automatically. Therefore, you must explicitly specify these types using
BaseSchema.
type BinaryTree = {
element: string;
left: BinaryTree | null;
right: BinaryTree | null;
};
const BinaryTreeSchema: BaseSchema<BinaryTree> = object({
element: string(),
left: nullable(lazy(() => BinaryTreeSchema)),
right: nullable(lazy(() => BinaryTreeSchema)),
});
Lazy union schema
Schema to validate a discriminated union of objects.
In most cases,
unionandvariantare the better choices for creating such a schema. I recommend usinglazyonly in special cases.
const LazyUnionSchema = lazy((input) => {
if (input && typeof input === 'object' && 'type' in input) {
switch (input.type) {
case 'email':
return object({
type: literal('email'),
email: string([email()]),
});
case 'url':
return object({
type: literal('url'),
url: string([url()]),
});
case 'date':
return object({
type: literal('date'),
date: string([isoDate()]),
});
}
}
return never();
});
Related
The following APIs can be combined with lazy.
Schemas
any,array,bigint,blob,boolean,date,enum_,instance,intersect,literal,map,nan,never,nonNullable,nonNullish,nonOptional,null_,nullable,nullish,number,object,optional,picklist,record,set,special,string,symbol,tuple,undefined_,union,unknown,variant,void_