Question

I ran into an issue using a struct today that caught me off guard, and I was hoping someone could shed some light on it for me.

I have a struct defined like this:

public struct PaymentDetail {     public Decimal Amount{get;set;}     public string CheckNumber{get;set;}     public PaymentType PaymentType{get;set;} } 

I have a class that contains this info

public class Transaction {     public PaymentDetail Payment{get;} } 

I have a presentation model which in which I want to set the underlying properties like this

public class ViewModel {     public Decimal Amount     {         get{return _Transaction.PaymentDetail.Amount;}         set         {              //This is the offending line of code              _Transaction.PaymentDetail.Amount = value;              RaisePropertyChanged('Amount');         }     } } 

What’s wierd is I can make this work if I change the Payment property to a public field like this:

public class Transaction {     public PaymentDetail Payment; } 

there’s obviosuly something I don’t understand about structs that are causing this. Is this a bad idea? Is there a better way? What am I doing wrong?

Answer 1

First – don’t have mutable structs (i.e. a struct where you can change the values after construction, via setters etc). That is the primary cause of confusion here.

The point is; when you call a property (like Payment) you get a copy of the value (in your local stack area). For a class, that is a copy of the reference (no problem). For a struct it is a copy of the struct itself. Any changes to that value will be discarded, so the compiler has stopped you from losing data.

When it is a public field, you are mutating the original value directly, which is why it doesn’t mind. But mutating structs really isn’t a good idea.

Make PaymentDetail a class; that is the correct solution here…

In .NET, structs aren’t ‘objects without behaviour’ – they are ‘value-types’. Things like ‘a currency/value pair’, ‘a time range’, etc might make valid structs – but not PaymentDetail.