Question

I’m working on my first real MVC application and I’m trying to follow general OOP best practices. I’m refactoring some simple business logic that I had in a controller into my domain model. I’ve been doing some reading lately and it seems pretty clear that I should put the logic somewhere in a domain model entity class in order to avoid the “anemic domain model” anti-pattern.

The application will allow people to purchase leases for parking spaces. Rates are determined by the length of the spot and whether or not the customer is a member of the business park.

So I have entity classes in my domain model that look like this (simplified):

public class Customer
{
    int ID { get; set; }
    string Name { get; set; }
    bool IsMember { get; set; }
}

public class ParkingSpace
{
    int ID { get; set; }
    int Length { get; set; }
}

public class ParkingSpaceLease
{
    int ID { get; set; }
    DateTime OpenDate { get; set; }
    DateTime CloseDate { get; set; }
    Customer Customer { get; set; }
    ParkingSpace ParkingSpace { get; set; } 
}

Edit: Just to clarify the LeaseQuote is not an entity class as it is just used to display the cost breakdown to perspective customers and is not persisted anywhere.

public class LeaseQuote
{
    int SubTotal { get; set; }
    int Discount { get; set; }
    int Total { get; set; }
}

Now as a feature of the application I need to be able to generate quotes for different customer and parking space combinations. The quotes will normally be accessed outside the context of actually creating a lease such as when a customer calls up to inquire about a price.

So what is the best way to go about this? Does it make sense to instantiate a new ParkingSpaceLease object inside the controller just to call a GetQuote method on it?

var lease = new ParkingSpaceLease();
var quote = lease.GetQuote(length: 168, isMember: true);
return Json(quote);

Or should the LeaseQuote class have the method?

var leaseQuote = new LeaseQuote();
var quote = leaseQuote.GetQuote(length: 168, isMember: true);
return Json(quote);

It feels strange putting the logic in the actual ParkingSpaceLease class. I guess it feels kind of “heavy” to create a new lease object when I know that I’m not going to actually do anything with it other than access the GetQuote method which seems kind of like a separate service.

So where should the GetQuote method go and why should it go there?

Answer 1

It almost sounds like your LeaseQuote isn’t an entity and more of a business level class. I mean, you’re not storing it in the database anywhere, are you? And it’s not a part of another data object.

When I see this

Now as a feature of the application I need to be able to generate quotes for different customer and parking space combinations. The quotes will normally be accessed outside the context of actually creating a lease such as when a customer calls up to inquire about a price.

I think of a method signature like this

public LeaseQuote GetQuote(Customer customer, ParkingSpace parkingSpace, int length)

But with that in mind, I’d probably also want to store information about the cost of the parking space within the ParkingSpace entity and (if applicable) the customer’s discount in the Customer entity.

Where would this stuff go? In a model class (business model, not LINQ or Entity model) that accesses your entities and serves as a provider for your controller.

Now I know that’s not using your models exactly as written. And it could just be personal bias. But when I think about data models and data entities, they should not have any addon methods outside of what’s coming back from the database. They should just represent the data unaltered as it appears in the database. If you’re acting on the data, that belongs in a tier above the data entities.

Update:

What I am curious about from your example is why one would want to pass the full Entity objects (Customer and Parking Space) versus just the properties needed to perform the calculation?

It depends on your standard of code. Exposing the entity itself could be dangerous if the consuming code manipulates the entity. I prefer passing the entity mainly because that’s what I’m used to. But I’m also careful not to manipulate the entity on the way in. That, and I think the method signature reflects what the GetQuote method is focused on; it’s related to a customer and a parking space.

I could also make the case that if more fields go into the Entity later that can effect the GetQuote method, then the method signature doesn’t have to change. In this case, only the implementation for GetQuote has to change.

Short answer: Preference.