In today's blog post you will learn how to validate objects in .NET using various techniques and libraries.
We'll explore what is the best way to validate objects in .NET in 2024.
Object validation is a crucial aspect of any .NET application.
Ensuring that your data is valid before processing, can prevent numerous issues, from simple bugs to critical security vulnerabilities.
There are 2 popular approaches to validating objects in .NET:
- DataAnnotations
- FluentValidation
I will show why I consider FluentValidation
the best approach to validate object in .NET in 2024.
On my website: antondevtips.com I share .NET and Architecture best practices.
Subscribe to become a better developer.
Validation with DataAnnotations
DataAnnotations have been part of .NET for years and remain a straightforward way to enforce validation rules on your models by using attributes.
Here's a simple example:
public class User
{
[Required(ErrorMessage = "Username is required.")]
[StringLength(100, ErrorMessage = "Username must be less than 100 characters.")]
public string Username { get; set; }
[Required(ErrorMessage = "Email is required.")]
[EmailAddress(ErrorMessage = "Invalid email address.")]
public string Email { get; set; }
[Range(18, 99, ErrorMessage = "Age must be between 18 and 99.")]
public int Age { get; set; }
}
While DataAnnotations might seem like a concise way to define validation and have entity and validation in one place, this approach has its drawbacks:
- Tight Coupling: Data annotations tightly couple the entity classes with the validation logic, making it challenging to switch to a different validation framework without modifying the entity classes.
- Limited Flexibility: Data annotations provide limited flexibility compared to FluentValidation configurations. Complex validation scenarios may be difficult or impossible to express using data annotations alone.
- Code Clutter: Embedding validation logic within entity classes can clutter the codebase, especially as the application grows larger. This violates the Single Responsibility Principle and makes the code harder to maintain.
- Limited Reusability: While data annotations can be reused across multiple entities, they are not flexible when it comes to validation scenarios where validation depends on other properties of an object.
As your validation needs to grow more complex, DataAnnotations can become cumbersome and less maintainable.
Validation with FluentValidation Library
FluentValidation offers a more flexible and powerful approach for validation.
With FluentValidation, you define your validation logic in a separate class, keeping your models clean and your validation logic expressive:
public class UserValidator : AbstractValidator<User>
{
public UserValidator()
{
RuleFor(user => user.Username)
.NotEmpty()
.Length(1, 100);
RuleFor(user => user.Email)
.NotEmpty()
.EmailAddress();
RuleFor(user => user.Age)
.InclusiveBetween(18, 99);
}
}
FluentValidation
has built-in default error messages, but you can customize them if you need to:
public class UserValidator : AbstractValidator<User>
{
public UserValidator()
{
RuleFor(user => user.Username)
.NotEmpty().WithMessage("Username is required.")
.Length(1, 100).WithMessage("Username must be less than 100 characters.");
RuleFor(user => user.Email)
.NotEmpty().WithMessage("Email is required.")
.EmailAddress().WithMessage("Invalid email address.");
RuleFor(user => user.Age)
.InclusiveBetween(18, 99).WithMessage("Age must be between 18 and 99.");
}
}
With FluentValidation you can inject a dependency into a Validator constructor and, for example, use localization provider to get a localized error message.
Here's how to call a validator:
var validator = new UserValidator();
var result = validator.Validate(user);
if (!result.IsValid)
{
foreach (var failure in result.Errors)
{
Console.WriteLine($"Property {failure.PropertyName} failed validation. Error: {failure.ErrorMessage}");
}
}
You can also use asynchronous method to validate an object:
var validator = new UserValidator();
var result = await validator.ValidateAsync(user);
FluentValidation
offers the following advantages:
- Flexibility: FluentValidation provides more flexibility compared to data annotations. It allows for more complex configurations and validation rules that might not be achievable with data annotations alone.
- Separation of Concerns: FluentValidation promotes better separation of concerns by keeping validation logic separate from domain classes. This improves code readability and maintainability, as it separates validation concerns from domain logic.
- Explicitness: FluentValidation configurations are explicit and self-documenting. Developers can easily understand the validation rules by reading the FluentValidation code, which helps in understanding the validation logic without inspecting the domain classes directly.
- Reuse and Composition: FluentValidation configurations can be reused across multiple entities or projects. Validation classes can be composed, allowing for easier management and maintenance of complex validation logic.
That's why I consider FluentValidation
the best approach to validate object in .NET.
Let's explore some more complex examples on how to validate objects with FluentValidation
.
Validating Nested Objects with FluentValidation
Let's explore an example on how to validate nested objects:
public class Product
{
public string Name { get; set; }
public decimal Price { get; set; }
public Manufacturer Manufacturer { get; set; }
}
public class Manufacturer
{
public string Name { get; set; }
public string Country { get; set; }
}
using FluentValidation;
public class ManufacturerValidator : AbstractValidator<Manufacturer>
{
public ManufacturerValidator()
{
RuleFor(manufacturer => manufacturer.Name)
.NotEmpty();
RuleFor(manufacturer => manufacturer.Country)
.NotEmpty();
}
}
public class ProductValidator : AbstractValidator<Product>
{
public ProductValidator()
{
RuleFor(product => product.Name)
.NotEmpty();
RuleFor(product => product.Price)
.GreaterThan(0);
RuleFor(product => product.Manufacturer)
.NotNull()
.SetValidator(new ManufacturerValidator());
}
}
As you can see in the example above, you can call the SetValidator
method to add ManufacturerValidator
to a Product
model.
Here's how to call a validator:
var validator = new ProductValidator();
var result = validator.Validate(product);
if (!result.IsValid)
{
foreach (var failure in result.Errors)
{
Console.WriteLine($"Property {failure.PropertyName} failed validation. Error: {failure.ErrorMessage}");
}
}
Asynchronous Validation with Fluent Validation
To perform asynchronous validation with FluentValidation
, you can use the MustAsync
method for custom asynchronous rules.
Here's an example of how to validate a Product
model using async validation:
public class ProductValidator : AbstractValidator<Product>
{
public ProductValidator()
{
RuleFor(product => product.Name)
.NotEmpty()
.MustAsync(BeUniqueName).WithMessage("Product name must be unique.");
RuleFor(product => product.Price)
.GreaterThan(0)
RuleFor(product => product.Manufacturer)
.NotNull()
.SetValidator(new ManufacturerValidator());
RuleFor(product => product.Manufacturer.Name)
.MustAsync(ManufacturerExists).WithMessage("Manufacturer does not exist.");
}
private async Task<bool> BeUniqueName(string name, CancellationToken cancellationToken)
{
// Simulate an async check to a database or external service
await Task.Delay(100, cancellationToken); // Simulated delay
// Check if the name is unique (this would be a database or service call in a real application)
return name != "ExistingProductName";
}
private async Task<bool> ManufacturerExists(string manufacturerName, CancellationToken cancellationToken)
{
// Simulate an async check to a database or external service
await Task.Delay(100, cancellationToken); // Simulated delay
// Check if the manufacturer exists (this would be a database or service call in a real application)
return manufacturerName != "NonExistentManufacturer";
}
}
Generally, I don't recommend checking for database uniqueness in the validators, but some people do it in validators.
I find these business checks to fit in the other places. But you should be aware of this feature.
When using asynchronous validation in a validator class - you must call the ValidateAsync
method, otherwise an exception will be thrown:
var validator = new ProductValidator();
var result = await validator.ValidateAsync(product);
Conditional Validation with Fluent Validation
FluentValidation
provides a way to apply conditional validation rules using the When
and Unless
methods.
These methods allow you to specify conditions under which certain validation rules should be applied or ignored.
The When
method applies a rule only if the specified condition is true, for example:
public class ProductValidator : AbstractValidator<Product>
{
public ProductValidator()
{
RuleFor(product => product.Name)
.NotEmpty();
// Validate the Price when IsDiscounted is false
RuleFor(product => product.Price)
.GreaterThan(0)
.When(product => !product.IsDiscounted);
// Validate the DiscountPercentage when IsDiscounted is true
RuleFor(product => product.DiscountPercentage)
.InclusiveBetween(0, 100)
.When(product => product.IsDiscounted);
}
}
The Unless
method works in the opposite way to When. It applies a rule unless the specified condition is true. For example:
// Validate Price unless IsDiscounted is true
RuleFor(product => product.Price)
.GreaterThan(0)
.Unless(product => product.IsDiscounted);
Dependent Property Validation with FluentValidation
In FluentValidation you can easily create a validation rule that checks whether an EndDate
is greater than StartDate
:
RuleFor(x => x.EndDate)
.GreaterThan(x => x.StartDate)
.WithMessage("EndDate must be after StartDate.");
This example and the ones with product discounts showcase validation rules that depend on the other fields.
I will feel sorry for those who will try to implement custom DataAnnotation
attributes that rely on other properties of an object.
Creating Custom Validators in FluentValidation
In FluentValidation
you can easily create custom validators.
Let's create a custom BeAlphanumeric
validator for a Product
model:
using FluentValidation;
public static class CustomValidators
{
public static IRuleBuilderOptions<T, string> BeAlphanumeric<T>(this IRuleBuilder<T, string> ruleBuilder)
{
return ruleBuilder.Must(value => value.All(char.IsLetterOrDigit))
.WithMessage("'{PropertyName}' must contain only alphanumeric characters.");
}
}
public class ProductValidator : AbstractValidator<Product>
{
public ProductValidator()
{
RuleFor(product => product.Name)
.NotEmpty().WithMessage("Product name is required.")
.BeAlphanumeric();
// Other validation rules...
}
}
By creating an extension method to IRuleBuilder
you can use your custom validators as if it was a standard validation method like NotEmpty
.
On my website: antondevtips.com I share .NET and Architecture best practices.
Subscribe to become a better developer.
Top comments (0)