Re: [patch 2/2] Assert leftover cleanups in TRY_CATCH

[Jan Kratochvil <jan.kratochvil@redhat.com>]

On Mon, 06 May 2013 20:50:47 +0200, Tom Tromey wrote:
> >>>>> "Jan" == Jan Kratochvil <jan.kratochvil@redhat.com> writes:
> Jan> C++ exceptions solve it all, everyone knows it, it is simple,
> Jan> effective and at least in comparison with the existing GDB system
> Jan> it is foolproof.
> 
> I know, and I agree that it would yield a better gdb, but I don't think
> it is going to happen.

I can't seriously reply these questions anymore.  GCC already requires C++ so
why GDB cannot?  I do not remember any valid reason against C++ from all the
GDB discussions around it.

GDB still is barely usable for real C++ application debugging, debugging
multiple virtual class inheritance does not work, one has to use printfs
instead.  Inferior breakpoint with conditional to stop only after thousands of
iterations is so unusably slow it is faster to rebuild the inferior with the
conditional put into inferior's source. etc. etc.

And with all this work ahead continuously wasting engineering time on
reimplementing C++-in-C...

Although I was waiting for -Wc++-compat by Matt Rice at least as a first stap
but that probably won't happen by Matt Rice so it needs a reassignment.


Jan