Darren New wrote:
I'm having a hard time imagining how the "hold on to a pointer to the
frame" approach will work in all cases if these are first-class
values, and I have a hard time figuring out how ugly this will be
given the lack of GC and the potential oddness of
constructor/destructor semantics in C++.
Yeah, that is not clear to me either. I think the only way it'd make
sense would be if you reified your stack, which seems like a most un-C++
thing to do.
But I'm not interested enough in C++ to work through the proposal
and figure it out. I'm just noticing that the baroqueness of C++
would seem to complicate every new feature in an NxM kind of way.
Oh yeah, definitely, although the C++0X committee is pretty clever when
it comes to mitigating the NxM effect, they never get it down much below
Nx1. ;-)
--Chris
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
