From: BRIDGES Dick (Dick.Bridges_at_[hidden])
Date: 2005-08-26 11:43:06
> -----Original Message-----
> From: boost-bounces_at_[hidden]
> On Behalf Of Andreas Huber
> Sent: Thursday, August 25, 2005 9:14 AM
> To: boost_at_[hidden]
> Subject: Re: [boost] [statechart] Problems if I make dtorS virtual?
> BRIDGES Dick wrote:
> > I followed the instructions to build the examples. Changed to the
> > example directory and ran <<bjam "sTOOLS=gcc">>. I got lazy and
> > ran the PingPong* progs. Guess I should have mentioned that. %>]
> Oh, sorry. The fact that you saw the warnings led me to believe that
> examples weren't built with bjam.
Again my exposition was lacking. I built my code and the examples with
bjam and make in unsuccessful attempts to get myself out from between
rock (warnings,warnings everywhere) and that hard place (convincing QA
setting the -Wno-non-virtual-dtor flag for *ALL* of our code that
the library is acceptable).
> > Any chance that, in the absence of some future evidence to the
> > contrary, those dtorS can be made virtual in the library?
> I'm at least reluctant to do that. Given the abundance of platforms
> support C++ we can never be sure that there is not one where
> will suffer significantly when the dtors are made virtual. From the 3
> compilers I use GCC is the only one that a) has such a non-virtual
> warning *and* b) doesn't have a pragma to turn off warnings.
> the latter will soon be taken care of...
I understand. Balancing performance and safety in this dynamic
of promises and deadlines must be very difficult. I envy your ability
admire what you've accomplished.
Just a thought - might it be worthwhile to introduce a policy to allow
user to choose between maximum performance and the comforting
of virtual dtorS to go with those virtual functions? :-)
> Andreas Huber
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk