Re: Lua next version

[Cosmin Apreutesei <cosmin.apreutesei@...>]

> But a true table count would be one such tool that
> would improve the possibilities for such implementations.

That was exactly what I was thinking when I wondered if the current
infrastructure could support an efficient implementation of a library
of common abstractions based on tables (as a C or lua lib)... after
all, they are made for this purpose exactly (considering the "kernel
language" quality of lua).

IMHO (I'm a novice -- hope I'm not just making noise here), either
more information about tables is exposed to the user (like those
counts), or, for a truly blastingly efficient library (efficient to
the standards that lua users have been spoiled to expect) would have
to be implemented in the VM... something which I wouldn't mind at all
having.