Question

I’m working on an EntityFramework application and I have run across a problem when using a struct instead of a class with Database.SqlQuery.

The struct looks like this:

internal struct Column
{
    public int object_id { get; set; }
    public string name { get; set; }
}

and the call to SqlQuery looks like this:

using (var db = new DbContext("ConnectionString"))
{
    var columns = db.Database.SqlQuery<Column>(query);
    DoSomething(columns.ToList()); // Exception thrown here
}

I get an ArgumentException with the message “Expression of type ‘MyService.Column’ cannot be used for return type ‘System.Object'” when Column is a struct. I changed it to a class and it works fine.

I’m happy enough to use a class here, but I’m wondering why a struct would fail when used in this way.

Answer 1

EntityFramework provides an object oriented representation of your data. Because of that, the result needs to be an object, as it has to maintain state. If you use a struct, the data will be copied every time you pass the column between methods and that makes no sense, as it represent the same value from the database.