Subj : Re: Can you mimic atexit() for class functions? To : borland.public.cpp.borlandcpp From : "Benjamin Pratt" Date : Thu Oct 09 2003 05:00 pm Ed, I'm not trying to start a war. Please read on. > I did not speak of the order of destruction of globals versus statics. I > did not discuss statics at all. > I told him that if he was uncomfortable with atexit then he could have an > instance of a class destruct at the end of the program declaring a global > instance of the class. We know that. Jeff simply offered a warning that if Gary should use the destructor of a global to trigger his callback code, he should be aware that the destructor may or may not be called before other global or static objects are destroyed. If that destructor needs to interact with other global/static objects at program exit, they may or may not still exist. > > And no, I did not look to see that others had jumpted in to kick me in the > shorts, arguing against me on my positions on things I actually did not > mention, and no, I do not appreciate that. > Jeff did not 'kick you in the shorts'. He simply was giving extra information, and since you are the athority around here, he asked you to clarify his advice if he was wrong. > And no, we were not discussing the order of globals and statics being > destructed. Right, Jeff was. But your reply to Jeff was off target. Apperently, you didn't notice that he had added to the discussion by introducing the order of global destruction. That's OK. As a TeamB member we know that you are busy responding to posts and may miss something. That is why I responded to your post to indicate that Jeff's point was indeed valid. Again, as the athority, I asked you to refute it if we were wrong. > > Now why are you in this thread, beating me up Well, I didn't until first put on the defense by your post. However, I appologise for that. .