Re: [gtkmm] Re: [glade--] Could raw C++ pointers be eliminated?
"Murray Cumming" <[email protected]> Thu, 21 Oct 2004 09:50:04 +0200 (CEST)
| Newsgroups | gmane.comp.gnome.glademm,gmane.comp.gnome.gtkmm |
|---|---|
| Message-ID | <[email protected]> |
> I hold what you may view as a radical position. I believe that > FRAMEWORKS should be designed so that applications programmers should > not have to think about or deal with raw pointers at all, unless > collection classes are being written, or some other class that deals > with things with dynamic lifetimes. Pointers are useful, and every useful language and framework uses them. Java doesn't call them pointers (it calls them references, I think), but it uses their equivalent far more than we do. If you want to use smartpointers then use smartpointers. Maybe you would like all objects to act like smartpointers, but - I prefer the syntax to show where we deviate from normal C++ memory management. So, "smartpointer<TheClass> theInstance" clearly shows that the instance will have the lifetime dictated by smartpointer, while "TheClass theInstance" will act like a regular C++ object. - There are practical C++ problems with this when dealing with class hierarchies. And if you don't like the code that glademm generates then don't use glademm and write your own, or use libglademm. I am unlikely to discuss this lots unless I am very bored. Murray Cumming [email protected] www.murrayc.com www.openismus.com