I'm just starting on a class to handle client connections to a TCP server. Here is the code I've written thus far:
Imports System.Net.Sockets
Imports System.Net
Public Class Client
Private _Socket As Socket
Public Property Socket As Socket
Get
Return _Socket
End Get
Set(ByVal value As Socket)
_Socket = value
End Set
End Property
Public Enum State
RequestHeader ''#Waiting for, or in the process of receiving, the request header
ResponseHeader ''#Sending the response header
Stream ''#Setup is complete, sending regular stream
End Enum
Public Sub New()
End Sub
Public Sub New(ByRef Socket As Socket)
Me._Socket = Socket
End Sub
End Class
So, on my overloaded constructor, I am accepting a reference to an instance of a System.Net.Sockets.Socket
, yes?
Now, on my Socket
property, when setting the value, it is required to be ByVal
. It is my understanding that the instance in memory is copied, and this new instance is passed to value
, and my code sets _Socket
to reference this instance in memory. Yes?
If this is true, then I can't see why I would want to use properties for anything but native types. I'd imagine there can be quite a performance hit if copying class instances with lots of members. Also, for this code in particular, I'd imagine a copied socket instance wouldn't really work, but I haven't tested it yet.
Anyway, if you could either confirm my understanding, or explain the flaws in my foggy logic, I would greatly appreciate it.