I have a Spring MVC application which returns Json and Xml based on what is requested per client call. I am using Jackson and Xstream to let Spring do the de-serialization of my java object into json or xml output.
My java object contains a bunch of attributes, at least 30. I would like to know if there is a way I can let Spring control which fields of my java object will be present in the json or xml based on a header or parameter attribute. So the client application will be able to identify itself and the backend will return only the fields necessary or "visible" for that specific client app. Of course I could go to the nasty approach of hard coding, but I would not like to do that as the number of client applications can increase or decrease and having a deployment anytime it happens with code changes is out of context.
Is there a way to instruct spring/jackson/xstream to control the output based on some providaded value?
I did a quick implementation and my current solution works like this: I have an xml with a list of client IDs (I use these ids to identify my client app) and for each ID I have a list of attributes that the client app needs from the java object. I created a interceptor and between the controller and the view, my interceptor gets the header information with the client ID, get the list of attributes and using the BeanWrapper (http://docs.spring.io/spring/docs/2.0.x/reference/validation.html) to create a new object with only the attributes required by the client with data, all the others remain null (I instruct Jackson and Xtream) to ignore null attributes. This approach works fine but I was wondering if there is another/better way to do this.
Thank you TL