| 1 |
On 5/26/05, Caleb Tennis <caleb@g.o> wrote: |
| 2 |
|
| 3 |
> On Thursday 26 May 2005 02:30 am, Duncan wrote: |
| 4 |
> > So the KDE problem... Is that what's causing all those virtual function |
| 5 |
> > but destructor isn't virtual type warnings whenever I compile a KDE ebuild |
| 6 |
> > with gcc4? |
| 7 |
> |
| 8 |
> No, that's just shoddy C++ coding that also needs to be fixed. |
| 9 |
|
| 10 |
Sorry to cut in like this, but I get tired of comments like this. |
| 11 |
While this specific case might very well represent "shoddy" coding, |
| 12 |
there are perfectly good reasons to have non-virtual destructors in |
| 13 |
classes designed to be base classes. |
| 14 |
|
| 15 |
One common case is when COM-style reference counting is used. In such |
| 16 |
a scenario leaf classes always implement a release() member function |
| 17 |
which deletes 'this' when the reference count goes to zero. With this |
| 18 |
configuration there is no need for a virtual destructor since |
| 19 |
destructor chaining always happens from a leaf class. |
| 20 |
|
| 21 |
I hope one day GCC's warning for this will be smart enough to at least |
| 22 |
not complain about non-public non-virtual destructors in classes that |
| 23 |
otherwise contain virtual member functions. Then the warning will be |
| 24 |
useful and perhaps indicate "shoddy C++ coding". :-) |
| 25 |
|
| 26 |
// Andreas |
| 27 |
|
| 28 |
-- |
| 29 |
gentoo-dev@g.o mailing list |
Archives