0

I run across Orika recently.

And I couldn't find a good explanation of why I should use it. If I have, say, a User domain object, why not use that? Why do I need to create a UserDTO with more or less the same members.

Of course there are times when I need to hide some fields. But that doesn't explain the need to have dozens of libraries.

Can someone explain to me why I should not re-use domain objects from one architectural boundary to another? Saying boundary to include layers or micro-service interfaces or anything similar.

Amanuel Nega
  • 1,899
  • 1
  • 24
  • 42

1 Answers1

1

It all depends! Good designpatterns for big systems are often overkill for smaller ones. Is the data you are getting really the same as your logical intuitive domain objects or are there extra data there.

Do you find yourself in the situation described in the answer to this post, then DTO it up. DTO's exist to limit the amount of expensive network operations by transmitting more data in each request. Say you have 'User' and 'AddressDetail' domain objects, and that you could get the data for both these objects in a single call(and the data is useful in the same area of the application) then you'd use a dto and send all the data at once.

It can be hard predicting how your system will grow(especially if you are working against a living API which someone else controls), and data transfer object on some level provides a clear separation of responsibility which is often a good thing.

I'd say reuse domain objects with caution in large systems.

Adam
  • 2,845
  • 2
  • 32
  • 46