Re: [boost] [intrusive] rtti_base class proposition

Subject: Re: [boost] [intrusive] rtti_base class proposition
From: Ireneusz Szpilewski (irek_at_[hidden])
Date: 2009-10-22 00:08:00

Eugene Wee wrote:
> Hi,
>
> On Thu, Oct 22, 2009 at 5:26 AM, Ireneusz Szpilewski
> <irek_at_[hidden]> wrote:
>   
>>>> And how exactly are you supposed to make that change at runtime?
>>>>
>>>>         
>> What do you suggest?
>>
>> heart as a member of Me?
>>
>> Before:
>>
>> class Me
>> {
>>        MyHeart heart;
>> };
>>
>> After:
>>
>> class Me
>> {
>>   SomebodyElsesHeart heart;
>> }
>>
>> It can't be done at runtime either.
>>     
>
> If MyHeart and SomebodyElsesHeart publicly inherited from say, a Heart
> (abstract) base class, and if Me's heart member variable was a (smart)
> pointer to Heart, then you could indeed change the actual (sub)type of
> the Heart object that a Me object owns at runtime. This is a common
> technique, methinks.
>
> Regards,
> Eugene
> _______________________________________________
> Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

So, what do you suggest?
Heart* as a member of Me?
Let's try:

struct Me
{
     Me()
    {
      heart = new(MyHeart);
   }

   ~Me()
   {
      delete heart;
   }
 
private:
Heart* heart;
};

What if MyHeart has SingMySong() method while Heart does not?
How can I access it?

Ireneusz Szpilewski