Re: C# versus C++ Constructor Problem

BTW, you may be interested in the following references:
http://blogs.msdn.com/abhinaba/archive/2006/02/28/540357.aspx
http://blogs.msdn.com/scottwil/archive/2005/01/14/353177.aspx
http://www.artima.com/cppsource/nevercall.html

As well as Scott Meyer's Effective C++; he discusses calling virtual
methods in c'tors (and why you shouldn't)

On Thu, 22 Jun 2006 11:09:02 -0400, David Thompson
<[email protected]> wrote:

>I'm converting some C++ code to C# and have found something that I'd like
some clarification on.
>If I define the following:
>
>public class Class1
>{
>    public Class1()
>    {
>         initialize();
>    }
>
>    public virtual void initialize()
>    {
>         // Do something
>    }
>
>}
>
>public class Class2 : Class1
>{
>     private SomeOtherClass soc;
>     public Class2() { soc = new SomeOtherClass(); }
>
>     public override void initialize()
>     {
>           base.initialize();
>           soc.SomeFunction();   // error here.
>     }
>}
>
>Now looking at this, in C++ the callstack when creating a new Class2 is:
>
>Class2()
>Class1()
>Class1::initialize()
>
>but in C# the stack becomes:
>Class2()
>Class1()
>Class2.initialize()
>
>Now since Class2's constructor hasn't completed yet, then soc doesn't
exist and an exception is
>thrown because soc doesn't exist yet.
>
>Is there any way to get the C++ behavior in C# or do I have to look at re-
writing the code? I think I
>understand why it happens (dealing with the type information on the
class), but it doesn't seem
>correct. Anybody got some advice?

===================================
This list is hosted by DevelopMentor®  http://www.develop.com

View archives and manage your subscription(s) at http://discuss.develop.com