README.qtcopy: -no-stl or no -no-stl?

Lubos Lunak l.lunak at
Wed Jan 21 09:24:53 GMT 2004

On Tuesday 20 of January 2004 20:34, Bo Thorsen wrote:
> On Tuesday 20 January 2004 14:51, Dirk Mueller wrote:
> > On Tuesday 20 January 2004 10:30, Bo Thorsen wrote:
> > > cripple peoples Qt, when they might download Qt apps that need
> > > exceptions. And yes, there are examples of such apps.
> >
> > Might be a stupid question, but why do you expect that Qt would do
> > anything with exceptions?
> >
> > it is still fine to compile Qt without exceptions support. If you want
> > exceptions fine - use them. But don't expect that Qt makes use of them.
> >
> > what do you think what will happen when exceptions cross signal/slot
> > borders, even when Qt is compiled with exception support? Qt does not
> > handle any exceptions at all, so letting them pass through Qt is
> > probably producing very unwanted things.
> >
> > And if you don't pass the exceptions through Qt - you can compile Qt
> > without exceptions support. Thats why we recommend it.
> Just throw an exception in a slot, or something that is called from a
> slot. Qt without exception support crashes. Qt with exception support
> handles this nicely.
> Or throw an exception in any of the template methods Qt has for
> overriding, with the same results.
> Or subclass a Qt class, and override something that's called by a
> non-overridden method...
> There are so many ways to crash Qt if you use exceptions but have a Qt
> without it. And they are all fixed by not using that old workaround.

 I think what Dirk was trying to say was that just removing -fno-exceptions is 
not going to magically make Qt exceptions aware. Qt code won't generally 
expect an exception to be thrown, and may leak resources or misbehave. Just 
e.g. have a look at QObject::activate_signal() and start counting things that 
may go wrong. Before fixing at least the parts where exceptions are most 
likely there's no point in compiling Qt with exceptions enabled.

Lubos Lunak
KDE developer
SuSE CR, s.r.o.  e-mail: l.lunak at , l.lunak at
Drahobejlova 27  tel: +420 2 9654 2373
190 00 Praha 9   fax: +420 2 9654 2374
Czech Republic

More information about the kde-core-devel mailing list