Custom validation decorators in NestJS - Step-by-Step Execution

Concept Flow - Custom validation decorators

Define validation logic function

↓

Create decorator using ValidatorConstraint

↓

Apply decorator to DTO property

↓

NestJS runs validation on DTO

↓

Custom validator checks value

↓

Allow request

This flow shows how you define a custom validator, create a decorator, apply it to data, and how NestJS uses it to accept or reject input.

Execution Sample

NestJS

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

@ValidatorConstraint({ async: false })
class IsEvenConstraint implements ValidatorConstraintInterface {
  validate(value: number) { return value % 2 === 0; }
}

export function IsEven(validationOptions?: ValidationOptions) {
  return function (object: Object, propertyName: string) {
    registerDecorator({
      target: object.constructor,
      propertyName: propertyName,
      options: validationOptions,
      constraints: [],
      validator: IsEvenConstraint,
    });
  };
}

Defines a custom decorator @IsEven that checks if a number is even.

Execution Table

Step	Action	Input Value	Validation Result	Effect
1	Create IsEvenConstraint class	-	-	Validator logic ready
2	Define @IsEven decorator	-	-	Decorator function ready
3	Apply @IsEven to DTO property 'age'	-	-	Property marked for validation
4	Validate DTO with age=4	4	true	Passes validation
5	Validate DTO with age=5	5	false	Fails validation, error returned
6	Validate DTO with age=0	0	true	Passes validation
7	Validate DTO with age=-2	-2	true	Passes validation
8	Validate DTO with age=3.5	3.5	false	Fails validation, error returned
9	Validation ends	-	-	Request accepted or rejected based on result

💡 Validation stops after checking the input value with the custom validator logic.

Variable Tracker

Variable	Start	After Step 4	After Step 5	After Step 6	After Step 7	After Step 8	Final
value	-	4	5	0	-2	3.5	-
validationResult	-	true	false	true	true	false	-

Key Moments - 3 Insights

Why does the validation fail for 3.5 even though it's a number?

The custom validator checks if the number is even using modulo (% 2 === 0). Since 3.5 % 2 is not zero, it fails. See execution_table row 8.

What happens if we forget to register the decorator with registerDecorator?

The validation will not run because NestJS won't know about the custom validator. The decorator must call registerDecorator as shown in execution_table step 2.

Can the validator handle negative even numbers?

Yes, negative even numbers like -2 pass because modulo works the same. See execution_table step 7.

Visual Quiz - 3 Questions

Look at the execution_table at step 5. What is the validation result for input value 5?

A true

B false

C undefined

D error

💡 Hint

Check the 'Validation Result' column at step 5 in the execution_table.

At which step does the validation pass for the input value 0?

A Step 4

B Step 6

C Step 7

D Step 8

💡 Hint

Look for input value 0 in the 'Input Value' column and see its validation result.

If the validate method returned true for all numbers, how would the validation results change?

A All inputs would pass validation

B Only even numbers would pass

C Only odd numbers would pass

D Validation would always fail

💡 Hint

Consider what happens if validate always returns true regardless of input.

Concept Snapshot

Custom validation decorators in NestJS:
- Create a class implementing ValidatorConstraintInterface
- Define validate() with your logic
- Use @ValidatorConstraint() on the class
- Create a decorator function calling registerDecorator
- Apply decorator to DTO properties
- NestJS runs your validator during validation pipeline

Full Transcript

This example shows how to create a custom validation decorator in NestJS using class-validator. First, you define a class with a validate method that checks if a number is even. Then, you decorate this class with @ValidatorConstraint. Next, you create a decorator function that registers this validator with registerDecorator. When you apply this decorator to a DTO property, NestJS will run your custom validation logic during request validation. The execution table traces validation for different input values, showing which pass or fail. Key moments clarify common confusions like handling decimals or negative numbers. The visual quiz tests understanding of validation results at specific steps. This approach helps you add your own rules beyond built-in validators in NestJS.

Practice

1. What is the main purpose of creating a custom validation decorator in NestJS?

easy

A. To create new routes in the application

B. To automatically generate database schemas

C. To handle HTTP requests and responses

D. To define your own rules for validating data inputs

Solution

1. Step 1: Understand validation decorators

   Validation decorators in NestJS are used to check if data meets certain rules before processing.

2. Step 2: Purpose of custom validation decorators

   Custom validation decorators let you create your own specific rules that are not covered by built-in validators.

3. Final Answer:

   To define your own rules for validating data inputs -> Option D

4. Quick Check:

   Custom validation decorators = custom data rules [OK]

Hint: Custom decorators create your own data validation rules [OK]

Common Mistakes:

• Confusing validation decorators with routing or database features
• Thinking custom decorators generate schemas automatically
• Assuming they handle HTTP requests directly

2. Which of the following is the correct way to start defining a custom validator class in NestJS?

easy

A. Use @Injectable decorator on the class

B. Use @Controller decorator on the class

C. Use @ValidatorConstraint decorator on the class

D. Use @Module decorator on the class

Solution

1. Step 1: Identify the decorator for custom validators

   In NestJS, custom validation classes must be marked with @ValidatorConstraint to tell the system they are validators.

2. Step 2: Differentiate from other decorators

   @Controller is for routing, @Injectable for services, and @Module for modules, not for validation.

3. Final Answer:

   Use @ValidatorConstraint decorator on the class -> Option C

4. Quick Check:

   @ValidatorConstraint marks custom validator classes [OK]

Hint: Custom validators use @ValidatorConstraint decorator [OK]

Common Mistakes:

• Using @Controller or @Module instead of @ValidatorConstraint
• Forgetting to decorate the validator class
• Confusing service decorators with validation decorators

3. Given this custom validator code snippet, what will be the output when validating the value "abc123"?

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

@ValidatorConstraint({ name: 'isAlphaOnly', async: false })
export class IsAlphaOnly implements ValidatorConstraintInterface {
  validate(text: string) {
    return /^[A-Za-z]+$/.test(text);
  }
  defaultMessage() {
    return 'Text must contain only letters';
  }
}

medium

A. Validation fails because 'abc123' contains numbers

B. Validation passes because 'abc123' contains letters

C. Validation passes because regex allows numbers

D. Validation throws an error due to syntax

Solution

1. Step 1: Analyze the regex in validate method

   The regex /^[A-Za-z]+$/ matches only strings with letters from start to end, no numbers allowed.

2. Step 2: Check the test string 'abc123'

   This string contains numbers '123', so it does not match the regex and validate returns false.

3. Final Answer:

   Validation fails because 'abc123' contains numbers -> Option A

4. Quick Check:

   Regex /^[A-Za-z]+$/ rejects numbers [OK]

Hint: Check regex carefully to see what characters it allows [OK]

Common Mistakes:

• Assuming partial matches pass validation
• Ignoring numbers in the test string
• Thinking regex allows digits when it does not

4. Identify the error in this custom validation decorator code:

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

@ValidatorConstraint({ async: true })
export class IsEvenNumber implements ValidatorConstraintInterface {
  validate(value: number) {
    return value % 2 === 0;
  }
}

export function IsEven(validationOptions?: ValidationOptions) {
  return function (object: Object, propertyName: string) {
    registerDecorator({
      name: 'isEven',
      target: object.constructor,
      propertyName: propertyName,
      options: validationOptions,
      validator: IsEvenNumber,
    });
  };
}

medium

A. The validate method should return a Promise because async is true

B. The registerDecorator call is missing the validate method

C. The decorator function should not accept validationOptions

D. The class should not have @ValidatorConstraint decorator

Solution

1. Step 1: Check async flag and validate method

   The @ValidatorConstraint decorator has async: true, so validate must return a Promise<boolean>.

2. Step 2: Validate method returns boolean, not Promise

   Currently, validate returns a boolean directly, which conflicts with async: true.

3. Final Answer:

   The validate method should return a Promise because async is true -> Option A

4. Quick Check:

   async: true requires Promise return [OK]

Hint: If async: true, validate must return Promise<boolean> [OK]

Common Mistakes:

• Returning boolean instead of Promise when async is true
• Omitting @ValidatorConstraint decorator
• Misusing registerDecorator parameters

5. You want to create a custom validation decorator that checks if a string property contains at least one uppercase letter and one digit. Which approach correctly combines the concepts to implement this in NestJS?

hard

A. Use built-in decorators @IsUppercase and @IsNumber together on the property

B. Create a class with @ValidatorConstraint, implement validate() using regex, then use registerDecorator in a function to apply it

C. Create a service with @Injectable and call it inside the DTO without decorators

D. Use only registerDecorator with a simple function without a class or @ValidatorConstraint

Solution

1. Step 1: Understand custom validation structure

   Custom validators require a class decorated with @ValidatorConstraint implementing validate() method.

2. Step 2: Use registerDecorator to link the class to a decorator function

   This function applies the validator to the property in the DTO.

3. Step 3: Implement validate() with regex checking uppercase and digit presence

   This ensures the string meets both conditions in one check.

4. Final Answer:

   Create a class with @ValidatorConstraint, implement validate() using regex, then use registerDecorator in a function to apply it -> Option B

5. Quick Check:

   Custom validator class + registerDecorator function = correct approach [OK]

Hint: Combine @ValidatorConstraint class with registerDecorator function [OK]

Common Mistakes:

• Skipping the validator class and using only registerDecorator
• Trying to use services instead of decorators for validation
• Assuming built-in decorators cover combined conditions