Interesting problem needs GURUS of managed/unmanaged C++ compliler

Hi, all

I am running into a very interesting problem, which has been bothering me for a while.

I have a base class

class B
{
public:
virtual bool fun() =0;
}

and two derived classes

class A : public B
{
public:
virtual bool fun()
{
return false;
}

}

and

class C : public B
{
public:
virtual bool fun()
{
return false;
}

}

One thing that is very interesting is that if the following line appears anywhere in my managed code, whether it is executed or not,

A a;

the function C:fun's return value will become true.

What I am trying to understand is the difference it makes with or without "A a". Any hint or suggestion will be really appreciated!

Thanks.

Could you post some (complete) code that reproduces the error?

- petter

Actually I could not reproduce the problem in another context.

I am using Excel + COM + Managed C++ + Unmanaged C++. I cannot post a small piece of the code to reproduce the error.

What I am trying to understand is the difference it makes with or without "A a".

Well, the difference is that with 'A a;' you're creating an instance of A using its default constructor. So, it really depends on what that constructor is doing.

- petter

Hi, Peter,

Thanks for your help.

I cannot agree with you. Since "A a" only declares a local variable in some function, the classs will not be instantiated until "A a;" is executed at running time when its default constructor will be called.

In my case, no matter where I put " A a", C:fun() will fail. I suspect it relates the RTI table of the base classs B, but I am not sure of how.