Convert Figma logo to code with AI

typestack logoclass-validator

Decorator-based property validation for classes.

10,945
789
10,945
286

Top Related Projects

String validation

22,814

Dead simple Object schema validation

33,503

TypeScript-first schema validation with static type inference

13,713

The fastest JSON schema Validator. Supports JSON Schema draft-04/06/07/2019-09/2020-12 and JSON Type Definition (RFC8927)

20,860

The most powerful data validation library for JS

A simple and composable way to validate data in JavaScript (and TypeScript).

Quick Overview

Class-validator is a TypeScript/JavaScript library for object validation. It uses decorators and non-decorators based approaches to validate objects against defined constraints, making it easy to ensure data integrity in applications.

Pros

  • Seamless integration with TypeScript, leveraging decorators for clean and readable validation rules
  • Extensive set of built-in validation decorators covering common use cases
  • Supports custom validation functions and asynchronous validation
  • Works well with popular frameworks like Express.js and TypeORM

Cons

  • Requires TypeScript for full decorator support, which may not be suitable for all projects
  • Learning curve for developers unfamiliar with decorator-based validation
  • Some users report issues with circular dependencies in complex object structures
  • Documentation could be more comprehensive, especially for advanced use cases

Code Examples

  1. Basic property validation:
import { IsEmail, MinLength } from 'class-validator';

class User {
  @IsEmail()
  email: string;

  @MinLength(6)
  password: string;
}
  1. Custom validation function:
import { ValidatorConstraint, ValidatorConstraintInterface, Validate } from 'class-validator';

@ValidatorConstraint({ name: 'customText', async: false })
class CustomTextValidator implements ValidatorConstraintInterface {
  validate(text: string) {
    return text.includes('hello');
  }
}

class Post {
  @Validate(CustomTextValidator, {
    message: 'Text must include the word "hello"'
  })
  text: string;
}
  1. Asynchronous validation:
import { ValidateIf, ValidationArguments, ValidatorConstraint, ValidatorConstraintInterface, Validate } from 'class-validator';

@ValidatorConstraint({ async: true })
class IsUserAlreadyExist implements ValidatorConstraintInterface {
  async validate(username: string) {
    const user = await fetchUserByUsername(username);
    return !user;
  }
}

class User {
  @Validate(IsUserAlreadyExist, {
    message: 'User $value already exists'
  })
  username: string;
}

Getting Started

  1. Install the package:
npm install class-validator
  1. Enable decorators in your tsconfig.json:
{
  "experimentalDecorators": true
}
  1. Use in your code:
import { validate } from 'class-validator';
import { User } from './user.model';

const user = new User();
user.email = 'invalid-email';
user.password = '123';

validate(user).then(errors => {
  if (errors.length > 0) {
    console.log('Validation failed:', errors);
  } else {
    console.log('Validation successful');
  }
});

Competitor Comparisons

String validation

Pros of validator.js

  • Lightweight and simple to use, with no dependencies
  • Supports both server-side and client-side validation
  • Extensive set of built-in validation rules

Cons of validator.js

  • Lacks TypeScript support and type safety
  • No decorators or class-based validation approach
  • Limited customization options for complex validation scenarios

Code Comparison

class-validator:

import { IsEmail, MinLength } from 'class-validator';

class User {
  @IsEmail()
  email: string;

  @MinLength(6)
  password: string;
}

validator.js:

const validator = require('validator');

const email = 'test@example.com';
const password = 'password123';

const isValid = validator.isEmail(email) && validator.isLength(password, { min: 6 });

Summary

class-validator is a TypeScript-focused library that uses decorators for class-based validation, offering strong type safety and integration with ORMs. validator.js is a simpler, more lightweight option that works well for both client and server-side validation but lacks TypeScript support and advanced features.

Choose class-validator for TypeScript projects with complex domain models, and validator.js for simpler validation needs or when working in plain JavaScript environments.

22,814

Dead simple Object schema validation

Pros of yup

  • More flexible schema definition with a chainable API
  • Better support for custom validation logic
  • Easier to use with form libraries like Formik

Cons of yup

  • Slightly more verbose syntax for simple validations
  • Less integrated with TypeScript decorators
  • Smaller community and ecosystem compared to class-validator

Code Comparison

class-validator:

import { IsString, MinLength, IsEmail } from 'class-validator';

class User {
  @IsString()
  @MinLength(2)
  name: string;

  @IsEmail()
  email: string;
}

yup:

import * as Yup from 'yup';

const userSchema = Yup.object().shape({
  name: Yup.string().min(2).required(),
  email: Yup.string().email().required(),
});

Both libraries offer robust validation capabilities, but they differ in their approach. class-validator uses decorators and is more tightly integrated with TypeScript, while yup provides a more flexible, chainable API that can be easier to customize. class-validator is often preferred in larger, class-based projects, especially those using TypeORM or NestJS. yup is popular in React applications and works well with form libraries like Formik. The choice between them often depends on the specific project requirements and the developer's preference for declarative vs. imperative validation styles.

33,503

TypeScript-first schema validation with static type inference

Pros of Zod

  • Runtime type checking without decorators, making it more flexible and easier to use in non-class-based architectures
  • Better TypeScript integration, providing strong type inference out of the box
  • More comprehensive validation capabilities, including complex object shapes and arrays

Cons of Zod

  • Steeper learning curve due to its more complex API
  • Potentially larger bundle size, which may impact performance in client-side applications
  • Less seamless integration with existing class-based architectures

Code Comparison

class-validator:

import { IsString, IsInt, Min, Max } from 'class-validator';

class User {
  @IsString()
  name: string;

  @IsInt()
  @Min(0)
  @Max(100)
  age: number;
}

Zod:

import { z } from 'zod';

const UserSchema = z.object({
  name: z.string(),
  age: z.number().int().min(0).max(100),
});

type User = z.infer<typeof UserSchema>;

Both libraries offer robust validation capabilities, but Zod provides a more functional approach with better TypeScript integration. class-validator is more suitable for class-based architectures and may be easier to adopt for developers familiar with decorators. The choice between the two depends on project requirements, architecture, and developer preferences.

13,713

The fastest JSON schema Validator. Supports JSON Schema draft-04/06/07/2019-09/2020-12 and JSON Type Definition (RFC8927)

Pros of Ajv

  • Faster performance, especially for large datasets
  • More comprehensive JSON Schema support
  • Supports custom keywords for extended validation

Cons of Ajv

  • Steeper learning curve, especially for complex schemas
  • Less intuitive for object-oriented programming paradigms
  • Requires separate schema definition, which can be verbose

Code Comparison

class-validator example:

class User {
  @IsEmail()
  email: string;

  @MinLength(6)
  password: string;
}

Ajv example:

const schema = {
  type: 'object',
  properties: {
    email: { type: 'string', format: 'email' },
    password: { type: 'string', minLength: 6 }
  },
  required: ['email', 'password']
};

Summary

class-validator is more intuitive for TypeScript users and integrates well with class-based architectures. It offers decorators for easy validation definition within classes. Ajv, on the other hand, provides better performance and more extensive JSON Schema support. It's more suitable for complex validation scenarios and large-scale applications where performance is critical. The choice between the two depends on the specific project requirements, team expertise, and performance needs.

20,860

The most powerful data validation library for JS

Pros of Joi

  • More flexible schema definition with a chainable API
  • Extensive built-in validation rules and customization options
  • Better performance for large-scale validation tasks

Cons of Joi

  • Steeper learning curve due to its extensive API
  • Less seamless integration with TypeScript compared to class-validator
  • Requires more boilerplate code for schema definition

Code Comparison

class-validator:

import { IsString, IsInt, Min, Max } from 'class-validator';

class User {
  @IsString()
  name: string;

  @IsInt()
  @Min(18)
  @Max(100)
  age: number;
}

Joi:

const Joi = require('joi');

const userSchema = Joi.object({
  name: Joi.string().required(),
  age: Joi.number().integer().min(18).max(100).required()
});

Key Differences

  • class-validator uses decorators for validation rules, while Joi uses a chainable API
  • class-validator is more tightly integrated with TypeScript, providing better type inference
  • Joi offers more flexibility in schema definition and validation rules
  • class-validator is easier to use with existing class structures, while Joi requires separate schema definitions

Both libraries have their strengths and are suitable for different use cases. class-validator excels in TypeScript projects and when working with existing class structures, while Joi offers more flexibility and extensive validation options for complex schemas.

A simple and composable way to validate data in JavaScript (and TypeScript).

Pros of Superstruct

  • More flexible and can be used with plain JavaScript objects
  • Supports custom types and coercion
  • Lighter weight and has fewer dependencies

Cons of Superstruct

  • Less integrated with TypeScript decorators
  • May require more manual setup for complex validation scenarios
  • Doesn't have built-in support for some specific validations (e.g., IsEmail)

Code Comparison

class-validator:

import { IsString, MinLength, IsEmail } from "class-validator";

class User {
  @IsString()
  @MinLength(2)
  name: string;

  @IsEmail()
  email: string;
}

Superstruct:

import { object, string, size, email } from "superstruct";

const User = object({
  name: size(string(), 2, Infinity),
  email: email(),
});

Summary

Both class-validator and Superstruct are powerful validation libraries for JavaScript/TypeScript. class-validator is more tightly integrated with TypeScript and uses decorators, making it a good choice for projects heavily relying on TypeScript features. Superstruct, on the other hand, offers more flexibility and can be used with plain JavaScript objects, making it suitable for a wider range of projects. The choice between the two depends on the specific needs of your project and your preferred coding style.

Convert Figma logo designs to code with AI

Visual Copilot

Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.

Try Visual Copilot

README

class-validator

Build Status codecov npm version install size

Allows use of decorator and non-decorator based validation. Internally uses validator.js to perform validation. Class-validator works on both browser and node.js platforms.

Table of Contents

Installation

npm install class-validator --save

Note: Please use at least npm@6 when using class-validator. From npm@6 the dependency tree is flattened, which is required by class-validator to function properly.

Usage

Create your class and put some validation decorators on the properties you want to validate:

import {
  validate,
  validateOrReject,
  Contains,
  IsInt,
  Length,
  IsEmail,
  IsFQDN,
  IsDate,
  Min,
  Max,
} from 'class-validator';

export class Post {
  @Length(10, 20)
  title: string;

  @Contains('hello')
  text: string;

  @IsInt()
  @Min(0)
  @Max(10)
  rating: number;

  @IsEmail()
  email: string;

  @IsFQDN()
  site: string;

  @IsDate()
  createDate: Date;
}

let post = new Post();
post.title = 'Hello'; // should not pass
post.text = 'this is a great post about hell world'; // should not pass
post.rating = 11; // should not pass
post.email = 'google.com'; // should not pass
post.site = 'googlecom'; // should not pass

validate(post).then(errors => {
  // errors is an array of validation errors
  if (errors.length > 0) {
    console.log('validation failed. errors: ', errors);
  } else {
    console.log('validation succeed');
  }
});

validateOrReject(post).catch(errors => {
  console.log('Promise rejected (validation failed). Errors: ', errors);
});
// or
async function validateOrRejectExample(input) {
  try {
    await validateOrReject(input);
  } catch (errors) {
    console.log('Caught promise rejection (validation failed). Errors: ', errors);
  }
}

Passing options

The validate function optionally expects a ValidatorOptions object as a second parameter:

export interface ValidatorOptions {
  skipMissingProperties?: boolean;
  whitelist?: boolean;
  forbidNonWhitelisted?: boolean;
  groups?: string[];
  dismissDefaultMessages?: boolean;
  validationError?: {
    target?: boolean;
    value?: boolean;
  };

  forbidUnknownValues?: boolean;
  stopAtFirstError?: boolean;
}

IMPORTANT The forbidUnknownValues value is set to true by default and it is highly advised to keep the default. Setting it to false will result unknown objects passing the validation!

Validation errors

The validate method returns an array of ValidationError objects. Each ValidationError is:

{
    target: Object; // Object that was validated.
    property: string; // Object's property that haven't pass validation.
    value: any; // Value that haven't pass a validation.
    constraints?: { // Constraints that failed validation with error messages.
        [type: string]: string;
    };
    children?: ValidationError[]; // Contains all nested validation errors of the property
}

In our case, when we validated a Post object, we have such an array of ValidationError objects:

[{
    target: /* post object */,
    property: "title",
    value: "Hello",
    constraints: {
        length: "$property must be longer than or equal to 10 characters"
    }
}, {
    target: /* post object */,
    property: "text",
    value: "this is a great post about hell world",
    constraints: {
        contains: "text must contain a hello string"
    }
},
// and other errors
]

If you don't want a target to be exposed in validation errors, there is a special option when you use validator:

validator.validate(post, { validationError: { target: false } });

This is especially useful when you send errors back over http, and you most probably don't want to expose the whole target object.

Validation messages

You can specify validation message in the decorator options and that message will be returned in the ValidationError returned by the validate method (in the case that validation for this field fails).

import { MinLength, MaxLength } from 'class-validator';

export class Post {
  @MinLength(10, {
    message: 'Title is too short',
  })
  @MaxLength(50, {
    message: 'Title is too long',
  })
  title: string;
}

There are few special tokens you can use in your messages:

  • $value - the value that is being validated
  • $property - name of the object's property being validated
  • $target - name of the object's class being validated
  • $constraint1, $constraint2, ... $constraintN - constraints defined by specific validation type

Example of usage:

import { MinLength, MaxLength } from 'class-validator';

export class Post {
  @MinLength(10, {
    // here, $constraint1 will be replaced with "10", and $value with actual supplied value
    message: 'Title is too short. Minimal length is $constraint1 characters, but actual is $value',
  })
  @MaxLength(50, {
    // here, $constraint1 will be replaced with "50", and $value with actual supplied value
    message: 'Title is too long. Maximal length is $constraint1 characters, but actual is $value',
  })
  title: string;
}

Also you can provide a function, that returns a message. This allows you to create more granular messages:

import { MinLength, MaxLength, ValidationArguments } from 'class-validator';

export class Post {
  @MinLength(10, {
    message: (args: ValidationArguments) => {
      if (args.value.length === 1) {
        return 'Too short, minimum length is 1 character';
      } else {
        return 'Too short, minimum length is ' + args.constraints[0] + ' characters';
      }
    },
  })
  title: string;
}

Message function accepts ValidationArguments which contains the following information:

  • value - the value that is being validated
  • constraints - array of constraints defined by specific validation type
  • targetName - name of the object's class being validated
  • object - object that is being validated
  • property - name of the object's property being validated

Validating arrays

If your field is an array and you want to perform validation of each item in the array you must specify a special each: true decorator option:

import { MinLength, MaxLength } from 'class-validator';

export class Post {
  @MaxLength(20, {
    each: true,
  })
  tags: string[];
}

This will validate each item in post.tags array.

Validating sets

If your field is a set and you want to perform validation of each item in the set you must specify a special each: true decorator option:

import { MinLength, MaxLength } from 'class-validator';

export class Post {
  @MaxLength(20, {
    each: true,
  })
  tags: Set<string>;
}

This will validate each item in post.tags set.

Validating maps

If your field is a map and you want to perform validation of each item in the map you must specify a special each: true decorator option:

import { MinLength, MaxLength } from 'class-validator';

export class Post {
  @MaxLength(20, {
    each: true,
  })
  tags: Map<string, string>;
}

This will validate each item in post.tags map.

Validating nested objects

If your object contains nested objects and you want the validator to perform their validation too, then you need to use the @ValidateNested() decorator:

import { ValidateNested } from 'class-validator';

export class Post {
  @ValidateNested()
  user: User;
}

Please note that nested object must be an instance of a class, otherwise @ValidateNested won't know what class is target of validation. Check also Validating plain objects.

It also works with multi-dimensional array, like :

import { ValidateNested } from 'class-validator';

export class Plan2D {
  @ValidateNested()
  matrix: Point[][];
}

Validating promises

If your object contains property with Promise-returned value that should be validated, then you need to use the @ValidatePromise() decorator:

import { ValidatePromise, Min } from 'class-validator';

export class Post {
  @Min(0)
  @ValidatePromise()
  userId: Promise<number>;
}

It also works great with @ValidateNested decorator:

import { ValidateNested, ValidatePromise } from 'class-validator';

export class Post {
  @ValidateNested()
  @ValidatePromise()
  user: Promise<User>;
}

Inheriting Validation decorators

When you define a subclass which extends from another one, the subclass will automatically inherit the parent's decorators. If a property is redefined in the descendant, class decorators will be applied on it from both its own class and the base class.

import { validate } from 'class-validator';

class BaseContent {
  @IsEmail()
  email: string;

  @IsString()
  password: string;
}

class User extends BaseContent {
  @MinLength(10)
  @MaxLength(20)
  name: string;

  @Contains('hello')
  welcome: string;

  @MinLength(20)
  password: string;
}

let user = new User();

user.email = 'invalid email'; // inherited property
user.password = 'too short'; // password wil be validated not only against IsString, but against MinLength as well
user.name = 'not valid';
user.welcome = 'helo';

validate(user).then(errors => {
  // ...
}); // it will return errors for email, password, name and welcome properties

Conditional validation

The conditional validation decorator (@ValidateIf) can be used to ignore the validators on a property when the provided condition function returns false. The condition function takes the object being validated and must return a boolean.

import { ValidateIf, IsNotEmpty } from 'class-validator';

export class Post {
  otherProperty: string;

  @ValidateIf(o => o.otherProperty === 'value')
  @IsNotEmpty()
  example: string;
}

In the example above, the validation rules applied to example won't be run unless the object's otherProperty is "value".

Note that when the condition is false all validation decorators are ignored, including isDefined.

Whitelisting

Even if your object is an instance of a validation class it can contain additional properties that are not defined. If you do not want to have such properties on your object, pass special flag to validate method:

import { validate } from 'class-validator';
// ...
validate(post, { whitelist: true });

This will strip all properties that don't have any decorators. If no other decorator is suitable for your property, you can use @Allow decorator:

import {validate, Allow, Min} from "class-validator";

export class Post {

    @Allow()
    title: string;

    @Min(0)
    views: number;

    nonWhitelistedProperty: number;
}

let post = new Post();
post.title = 'Hello world!';
post.views = 420;

post.nonWhitelistedProperty = 69;
(post as any).anotherNonWhitelistedProperty = "something";

validate(post).then(errors => {
  // post.nonWhitelistedProperty is not defined
  // (post as any).anotherNonWhitelistedProperty is not defined
  ...
});

If you would rather to have an error thrown when any non-whitelisted properties are present, pass another flag to validate method:

import { validate } from 'class-validator';
// ...
validate(post, { whitelist: true, forbidNonWhitelisted: true });

Passing context to decorators

It's possible to pass a custom object to decorators which will be accessible on the ValidationError instance of the property if validation failed.

import { validate } from 'class-validator';

class MyClass {
  @MinLength(32, {
    message: 'EIC code must be at least 32 characters',
    context: {
      errorCode: 1003,
      developerNote: 'The validated string must contain 32 or more characters.',
    },
  })
  eicCode: string;
}

const model = new MyClass();

validate(model).then(errors => {
  //errors[0].contexts['minLength'].errorCode === 1003
});

Skipping missing properties

Sometimes you may want to skip validation of the properties that do not exist in the validating object. This is usually desirable when you want to update some parts of the object, and want to validate only updated parts, but skip everything else, e.g. skip missing properties. In such situations you will need to pass a special flag to validate method:

import { validate } from 'class-validator';
// ...
validate(post, { skipMissingProperties: true });

When skipping missing properties, sometimes you want not to skip all missing properties, some of them maybe required for you, even if skipMissingProperties is set to true. For such cases you should use @IsDefined() decorator. @IsDefined() is the only decorator that ignores skipMissingProperties option.

Validation groups

In different situations you may want to use different validation schemas of the same object. In such cases you can use validation groups.

IMPORTANT Calling a validation with a group combination that would not result in a validation (eg: non existent group name) will result in a unknown value error. When validating with groups the provided group combination should match at least one decorator.

import { validate, Min, Length } from 'class-validator';

export class User {
  @Min(12, {
    groups: ['registration'],
  })
  age: number;

  @Length(2, 20, {
    groups: ['registration', 'admin'],
  })
  name: string;
}

let user = new User();
user.age = 10;
user.name = 'Alex';

validate(user, {
  groups: ['registration'],
}); // this will not pass validation

validate(user, {
  groups: ['admin'],
}); // this will pass validation

validate(user, {
  groups: ['registration', 'admin'],
}); // this will not pass validation

validate(user, {
  groups: undefined, // the default
}); // this will not pass validation since all properties get validated regardless of their groups

validate(user, {
  groups: [],
}); // this will not pass validation, (equivalent to 'groups: undefined', see above)

There is also a special flag always: true in validation options that you can use. This flag says that this validation must be applied always no matter which group is used.

Custom validation classes

If you have custom validation logic you can create a Constraint class:

  1. First create a file, lets say CustomTextLength.ts, and define a new class:

    import { ValidatorConstraint, ValidatorConstraintInterface, ValidationArguments } from 'class-validator';
    
    @ValidatorConstraint({ name: 'customText', async: false })
    export class CustomTextLength implements ValidatorConstraintInterface {
      validate(text: string, args: ValidationArguments) {
        return text.length > 1 && text.length < 10; // for async validations you must return a Promise<boolean> here
      }
    
      defaultMessage(args: ValidationArguments) {
        // here you can provide default error message if validation failed
        return 'Text ($value) is too short or too long!';
      }
    }
    

    We marked our class with @ValidatorConstraint decorator. You can also supply a validation constraint name - this name will be used as "error type" in ValidationError. If you will not supply a constraint name - it will be auto-generated.

    Our class must implement ValidatorConstraintInterface interface and its validate method, which defines validation logic. If validation succeeds, method returns true, otherwise false. Custom validator can be asynchronous, if you want to perform validation after some asynchronous operations, simply return a promise with boolean inside in validate method.

    Also we defined optional method defaultMessage which defines a default error message, in the case that the decorator's implementation doesn't set an error message.

  2. Then you can use your new validation constraint in your class:

    import { Validate } from 'class-validator';
    import { CustomTextLength } from './CustomTextLength';
    
    export class Post {
      @Validate(CustomTextLength, {
        message: 'Title is too short or long!',
      })
      title: string;
    }
    

    Here we set our newly created CustomTextLength validation constraint for Post.title.

  3. And use validator as usual:

    import { validate } from 'class-validator';
    
    validate(post).then(errors => {
      // ...
    });
    

You can also pass constraints to your validator, like this:

import { Validate } from 'class-validator';
import { CustomTextLength } from './CustomTextLength';

export class Post {
  @Validate(CustomTextLength, [3, 20], {
    message: 'Wrong post title',
  })
  title: string;
}

And use them from validationArguments object:

import { ValidationArguments, ValidatorConstraint, ValidatorConstraintInterface } from 'class-validator';

@ValidatorConstraint()
export class CustomTextLength implements ValidatorConstraintInterface {
  validate(text: string, validationArguments: ValidationArguments) {
    return text.length > validationArguments.constraints[0] && text.length < validationArguments.constraints[1];
  }
}

Custom validation decorators

You can also create a custom decorators. Its the most elegant way of using a custom validations. Lets create a decorator called @IsLongerThan:

  1. Create a decorator itself:

    import { registerDecorator, ValidationOptions, ValidationArguments } from 'class-validator';
    
    export function IsLongerThan(property: string, validationOptions?: ValidationOptions) {
      return function (object: Object, propertyName: string) {
        registerDecorator({
          name: 'isLongerThan',
          target: object.constructor,
          propertyName: propertyName,
          constraints: [property],
          options: validationOptions,
          validator: {
            validate(value: any, args: ValidationArguments) {
              const [relatedPropertyName] = args.constraints;
              const relatedValue = (args.object as any)[relatedPropertyName];
              return typeof value === 'string' && typeof relatedValue === 'string' && value.length > relatedValue.length; // you can return a Promise<boolean> here as well, if you want to make async validation
            },
          },
        });
      };
    }
    
  2. Put it to use:

    import { IsLongerThan } from './IsLongerThan';
    
    export class Post {
      title: string;
    
      @IsLongerThan('title', {
        /* you can also use additional validation options, like "groups" in your custom validation decorators. "each" is not supported */
        message: 'Text must be longer than the title',
      })
      text: string;
    }
    

In your custom decorators you can also use ValidationConstraint. Lets create another custom validation decorator called IsUserAlreadyExist:

  1. Create a ValidationConstraint and decorator:

    import {
      registerDecorator,
      ValidationOptions,
      ValidatorConstraint,
      ValidatorConstraintInterface,
      ValidationArguments,
    } from 'class-validator';
    
    @ValidatorConstraint({ async: true })
    export class IsUserAlreadyExistConstraint implements ValidatorConstraintInterface {
      validate(userName: any, args: ValidationArguments) {
        return UserRepository.findOneByName(userName).then(user => {
          if (user) return false;
          return true;
        });
      }
    }
    
    export function IsUserAlreadyExist(validationOptions?: ValidationOptions) {
      return function (object: Object, propertyName: string) {
        registerDecorator({
          target: object.constructor,
          propertyName: propertyName,
          options: validationOptions,
          constraints: [],
          validator: IsUserAlreadyExistConstraint,
        });
      };
    }
    

    note that we marked our constraint that it will by async by adding { async: true } in validation options.

  2. And put it to use:

    import { IsUserAlreadyExist } from './IsUserAlreadyExist';
    
    export class User {
      @IsUserAlreadyExist({
        message: 'User $value already exists. Choose another name.',
      })
      name: string;
    }
    

Using service container

Validator supports service container in the case if want to inject dependencies into your custom validator constraint classes. Here is example how to integrate it with typedi:

import { Container } from 'typedi';
import { useContainer, Validator } from 'class-validator';

// do this somewhere in the global application level:
useContainer(Container);
let validator = Container.get(Validator);

// now everywhere you can inject Validator class which will go from the container
// also you can inject classes using constructor injection into your custom ValidatorConstraint-s

Synchronous validation

If you want to perform a simple non async validation you can use validateSync method instead of regular validate method. It has the same arguments as validate method. But note, this method ignores all async validations you have.

Manual validation

There are several method exist in the Validator that allows to perform non-decorator based validation:

import { isEmpty, isBoolean } from 'class-validator';

isEmpty(value);
isBoolean(value);

Validation decorators

DecoratorDescription
Common validation decorators
@IsDefined(value: any)Checks if value is defined (!== undefined, !== null). This is the only decorator that ignores skipMissingProperties option.
@IsOptional()Checks if given value is empty (=== null, === undefined) and if so, ignores all the validators on the property.
@Equals(comparison: any)Checks if value equals ("===") comparison.
@NotEquals(comparison: any)Checks if value not equal ("!==") comparison.
@IsEmpty()Checks if given value is empty (=== '', === null, === undefined).
@IsNotEmpty()Checks if given value is not empty (!== '', !== null, !== undefined).
@IsIn(values: any[])Checks if value is in an array of allowed values.
@IsNotIn(values: any[])Checks if value is not in an array of disallowed values.
Type validation decorators
@IsBoolean()Checks if a value is a boolean.
@IsDate()Checks if the value is a date.
@IsString()Checks if the value is a string.
@IsNumber(options: IsNumberOptions)Checks if the value is a number.
@IsInt()Checks if the value is an integer number.
@IsArray()Checks if the value is an array
@IsEnum(entity: object)Checks if the value is a valid enum
Number validation decorators
@IsDivisibleBy(num: number)Checks if the value is a number that's divisible by another.
@IsPositive()Checks if the value is a positive number greater than zero.
@IsNegative()Checks if the value is a negative number smaller than zero.
@Min(min: number)Checks if the given number is greater than or equal to given number.
@Max(max: number)Checks if the given number is less than or equal to given number.
Date validation decorators
@MinDate(date: Date | (() => Date))Checks if the value is a date that's after the specified date.
@MaxDate(date: Date | (() => Date))Checks if the value is a date that's before the specified date.
String-type validation decorators
@IsBooleanString()Checks if a string is a boolean (e.g. is "true" or "false" or "1", "0").
@IsDateString()Alias for @IsISO8601().
@IsNumberString(options?: IsNumericOptions)Checks if a string is a number.
String validation decorators
@Contains(seed: string)Checks if the string contains the seed.
@NotContains(seed: string)Checks if the string not contains the seed.
@IsAlpha()Checks if the string contains only letters (a-zA-Z).
@IsAlphanumeric()Checks if the string contains only letters and numbers.
@IsDecimal(options?: IsDecimalOptions)Checks if the string is a valid decimal value. Default IsDecimalOptions are force_decimal=False, decimal_digits: '1,', locale: 'en-US'
@IsAscii()Checks if the string contains ASCII chars only.
@IsBase32()Checks if a string is base32 encoded.
@IsBase58()Checks if a string is base58 encoded.
@IsBase64(options?: IsBase64Options)Checks if a string is base64 encoded.
@IsIBAN()Checks if a string is a IBAN (International Bank Account Number).
@IsBIC()Checks if a string is a BIC (Bank Identification Code) or SWIFT code.
@IsByteLength(min: number, max?: number)Checks if the string's length (in bytes) falls in a range.
@IsCreditCard()Checks if the string is a credit card.
@IsCurrency(options?: IsCurrencyOptions)Checks if the string is a valid currency amount.
@IsISO4217CurrencyCode()Checks if the string is an ISO 4217 currency code.
@IsEthereumAddress()Checks if the string is an Ethereum address using basic regex. Does not validate address checksums.
@IsBtcAddress()Checks if the string is a valid BTC address.
@IsDataURI()Checks if the string is a data uri format.
@IsEmail(options?: IsEmailOptions)Checks if the string is an email.
@IsFQDN(options?: IsFQDNOptions)Checks if the string is a fully qualified domain name (e.g. domain.com).
@IsFullWidth()Checks if the string contains any full-width chars.
@IsHalfWidth()Checks if the string contains any half-width chars.
@IsVariableWidth()Checks if the string contains a mixture of full and half-width chars.
@IsHexColor()Checks if the string is a hexadecimal color.
@IsHSL()Checks if the string is an HSL color based on CSS Colors Level 4 specification.
@IsRgbColor(options?: IsRgbOptions)Checks if the string is a rgb or rgba color.
@IsIdentityCard(locale?: string)Checks if the string is a valid identity card code.
@IsPassportNumber(countryCode?: string)Checks if the string is a valid passport number relative to a specific country code.
@IsPostalCode(locale?: string)Checks if the string is a postal code.
@IsHexadecimal()Checks if the string is a hexadecimal number.
@IsOctal()Checks if the string is a octal number.
@IsMACAddress(options?: IsMACAddressOptions)Checks if the string is a MAC Address.
@IsIP(version?: "4"|"6")Checks if the string is an IP (version 4 or 6).
@IsPort()Checks if the string is a valid port number.
@IsISBN(version?: "10"|"13")Checks if the string is an ISBN (version 10 or 13).
@IsEAN()Checks if the string is an if the string is an EAN (European Article Number).
@IsISIN()Checks if the string is an ISIN (stock/security identifier).
@IsISO8601(options?: IsISO8601Options)Checks if the string is a valid ISO 8601 date format. Use the option strict = true for additional checks for a valid date.
@IsJSON()Checks if the string is valid JSON.
@IsJWT()Checks if the string is valid JWT.
@IsObject()Checks if the object is valid Object (null, functions, arrays will return false).
@IsNotEmptyObject()Checks if the object is not empty.
@IsLowercase()Checks if the string is lowercase.
@IsLatLong()Checks if the string is a valid latitude-longitude coordinate in the format lat, long.
@IsLatitude()Checks if the string or number is a valid latitude coordinate.
@IsLongitude()Checks if the string or number is a valid longitude coordinate.
@IsMobilePhone(locale: string)Checks if the string is a mobile phone number.
@IsISO31661Alpha2()Checks if the string is a valid ISO 3166-1 alpha-2 officially assigned country code.
@IsISO31661Alpha3()Checks if the string is a valid ISO 3166-1 alpha-3 officially assigned country code.
@IsLocale()Checks if the string is a locale.
@IsPhoneNumber(region: string)Checks if the string is a valid phone number using libphonenumber-js.
@IsMongoId()Checks if the string is a valid hex-encoded representation of a MongoDB ObjectId.
@IsMultibyte()Checks if the string contains one or more multibyte chars.
@IsNumberString(options?: IsNumericOptions)Checks if the string is numeric.
@IsSurrogatePair()Checks if the string contains any surrogate pairs chars.
@IsTaxId()Checks if the string is a valid tax ID. Default locale is en-US.
@IsUrl(options?: IsURLOptions)Checks if the string is a URL.
@IsMagnetURI()Checks if the string is a magnet uri format.
@IsUUID(version?: UUIDVersion)Checks if the string is a UUID (version 3, 4, 5 or all ).
@IsFirebasePushId()Checks if the string is a Firebase Push ID
@IsUppercase()Checks if the string is uppercase.
@Length(min: number, max?: number)Checks if the string's length falls in a range.
@MinLength(min: number)Checks if the string's length is not less than given number.
@MaxLength(max: number)Checks if the string's length is not more than given number.
@Matches(pattern: RegExp, modifiers?: string)Checks if string matches the pattern. Either matches('foo', /foo/i) or matches('foo', 'foo', 'i').
@IsMilitaryTime()Checks if the string is a valid representation of military time in the format HH:MM.
@IsTimeZone()Checks if the string represents a valid IANA time-zone.
@IsHash(algorithm: string)Checks if the string is a hash The following types are supported:md4, md5, sha1, sha256, sha384, sha512, ripemd128, ripemd160, tiger128, tiger160, tiger192, crc32, crc32b.
@IsMimeType()Checks if the string matches to a valid MIME type format
@IsSemVer()Checks if the string is a Semantic Versioning Specification (SemVer).
@IsISSN(options?: IsISSNOptions)Checks if the string is a ISSN.
@IsISRC()Checks if the string is a ISRC.
@IsRFC3339()Checks if the string is a valid RFC 3339 date.
@IsStrongPassword(options?: IsStrongPasswordOptions)Checks if the string is a strong password.
Array validation decorators
@ArrayContains(values: any[])Checks if array contains all values from the given array of values.
@ArrayNotContains(values: any[])Checks if array does not contain any of the given values.
@ArrayNotEmpty()Checks if given array is not empty.
@ArrayMinSize(min: number)Checks if the array's length is greater than or equal to the specified number.
@ArrayMaxSize(max: number)Checks if the array's length is less or equal to the specified number.
@ArrayUnique(identifier?: (o) => any)Checks if all array's values are unique. Comparison for objects is reference-based. Optional function can be speciefied which return value will be used for the comparsion.
Object validation decorators
@IsInstance(value: any)Checks if the property is an instance of the passed value.
Other decorators
@Allow()Prevent stripping off the property when no other constraint is specified for it.

Defining validation schema without decorators

Schema-based validation without decorators is no longer supported by class-validator. This feature was broken in version 0.12 and it will not be fixed. If you are interested in schema-based validation, you can find several such frameworks in the zod readme's comparison section.

Validating plain objects

Due to nature of the decorators, the validated object has to be instantiated using new Class() syntax. If you have your class defined using class-validator decorators and you want to validate plain JS object (literal object or returned by JSON.parse), you need to transform it to the class instance via using class-transformer).

Samples

Take a look on samples in ./sample for more examples of usages.

Extensions

There are several extensions that simplify class-validator integration with other modules or add additional validations:

Release notes

See information about breaking changes and release notes here.

Contributing

For information about how to contribute to this project, see TypeStack's general contribution guide.

NPM DownloadsLast 30 Days