On Oct 14, 2004, at 6:29 AM, Giovanni Bajo wrote:
[sorry, sent to gcc@ by mistake]
Benjamin Kosnik wrote:
I'm more in favor of compile-time construction via metaprogramming
(only as complicated as is absolutely necessary, of course...). In
the long run, this seems like the more robust solution. Also, one
without build dependencies. It seems like either way you'll have some
complexity, it's just where you choose to put it.
I'm not sure. We could commit the generated <tuple> to CVS much like
we do with
configure for instance, so that there is no build dependence at all.
The only
cons I see with the generation program is that hacking could get much
harder,
depending on how the generation program is written.
I assume you're aware of the tricks that the Boost people play,
exploiting the fact that the C preprocessor is Turing complete?
I'm not endorsing those tricks; I find them appalling, and I'd like
to see the preprocessor die. Just thought we ought to be aware
of all options, if we aren't already.