Hi,
I'd also like to mention d-pointers as an alternative to smart pointers. I
learned about d-pointers are with the Qt toolkit and they are used
extensively in the kde desktop environment.
d-pointers are used to manage shared data. They decouple the data from the
methods of the object. The data resides on the shared part and the methods
reside on a wrapper object that contain a pointer to the data, thus the
d-pointer name. The d-pointer pattern is also known as 'pimpl' or 'cheshire
cat'.
I like d-pointers because they completely abstract the fact that you are
using a pointer, and make you think that you are using the object itself.
You can pass the object by value, and use it as if it were a native type.
Creating a d-pointer object is quite tedious, but you only do it once, and
in most cases the effort pays off.
Maybe d-pointers aren't as versatile as smart pointers, in fact they are
just a restricted and more friendly form of smart ptr. In some cases I think
they make your life easier, and specially for 'shared resource management'.
Have a look at http://doc.trolltech.com/qq/qq02-data-sharing-with-class.html
to find out more.
Ignacio Castaño
cas...@ya...
_______________________________________________________________
Copa del Mundo de la FIFA 2002
El único lugar de Internet con vídeos de los 64 partidos.
¡Apúntante ya! en http://fifaworldcup.yahoo.com/fc/es/
|
