universal-scheme-validator
v0.0.1
Published
Validation schemas for class-based data models
Maintainers
Readme
universal-scheme-validator
Robust and easy-to-use validation schemas for class-based data models in TypeScript.
Installation
Install the package using npm:
npm install universal-scheme-validatorBasic Usage
Define your classes or data types, create a validation schema using the ValidationSchema interface, and evaluate your instances using the runSchemaValidation function.
import { ValidationSchema, runSchemaValidation } from 'universal-scheme-validator';
class User {
name!: string;
email!: string;
age?: number;
}
// 1. Define the validation schema
const userSchema: ValidationSchema<User> = {
name: {
isRequired: true,
minLength: 3,
maxLength: 50,
},
email: {
isRequired: true,
isEmail: true,
},
age: {
isNumber: { isZero: false },
isMin: 18,
},
};
// 2. Create an instance and test validation
try {
const user = new User();
user.name = 'Ro'; // Minimum length is 3
// 3. Execute validation
// Throws an exception if there are errors by default
runSchemaValidation(user, userSchema);
} catch (error) {
console.log(error.errors);
// Expected output:
// {
// "name": ["The value must be at least 3 characters long"],
// "email": ["The property 'email' is required"]
// }
}Validation Options
The runSchemaValidation function accepts a third parameter for configuration options:
strict(boolean): Iftrue, throws an error when the object contains properties that are not defined in the schema.output("exception" | "record"): Determines the format in which errors are returned."exception"(default): Throws aValidationErrorexception."record": Returns an object where the keys are the properties and the values are arrays of the found errors.
Options example:
// Returns errors without throwing an exception and validates in strict mode
const errores = runSchemaValidation(user, userSchema, {
output: 'record',
strict: true,
});Validation of Nested Objects and Arrays
You can validate complex structures, arrays, and nested objects using the nestedSchema rule.
const schemaComplejo: ValidationSchema = {
direccion: {
isObject: true,
nestedSchema: {
calle: { isRequired: true, isNotEmpty: true },
codigoPostal: { isNumber: true },
},
},
etiquetas: {
isArray: { type: 'string' },
isRequired: true,
},
};It is also possible to directly pass an array of instances to runSchemaValidation(arrayOfInstances, schema).
Available Rules (ValidationRule)
The following validation rules are available through the ValidationRule interface and can be used on any property:
Text and Strings
isString: boolean: Verifies that the value is a string.minLength: number: Minimum required length of a string.maxLength: number: Maximum required length of a string.isEmail: boolean: Verifies if the string is a valid email.isUrl: boolean: Verifies if the string is a valid URL.isPath: PathValidationOptions: Verifies if the string is a valid path. Allows configuring extra options (e.g., no spaces, not empty).isUUID: boolean | { version?: number, provider?: "standard" | "oracle" }: Validates that it's a valid UUID identifier.isJSON: boolean | { expectedStructure?: ExpectedStructure }: Verifies if the string is a valid JSON.isNotAlpha: boolean | object: Validates that the text does not contain purely alphabetical characters, allowing numbers, accents, or symbols.
Numbers
isNumber: boolean | { isZero?: boolean, type?: NumberSchema }: Verifies that the value is numeric.isMin: number: Minimum expected numeric value.isMax: number: Maximum expected numeric value.
Data Types and Structure
isType: PrimitiveType: Verifies that the type matches a primitive ("string" | "number" | "boolean").isBoolean: boolean: Validates that the input value is boolean.isArray: boolean | { type: PrimitiveType, strict?: boolean }: Validates that the structure is an array.isObject: boolean | { allowEmpty?: boolean, allowArrays?: boolean }: Verifies that the value is an object.
Required and Equality
isRequired: boolean: Strict property, throws an error if the value or property is null, undefined, or non-existent in the instance.isNotEmpty: boolean: The value cannot be empty (works with strings, arrays, objects).isEqualTo: any: Checks if the input value is strictly equal to the provided value.containsValue: any[]: Checks if the tested value includes any of the elements from the options array.
Classes and Nested Schemas
isInstance: class: Verifies if the provided value is an instance of a specified class.nestedSchema: An optional object with additional validation schema for embedded/nested structures.
Errors and Exceptions
By default, if a validation fails, the library will throw an instance of ValidationError which can be accessed from @catch.
import { ValidationError } from 'universal-scheme-validator';
try {
// ...
} catch (error) {
if (error instanceof ValidationError) {
console.log(error.errors);
}
}