Re: explicitly defining table length + _ARG proposal

[Roberto Ierusalimschy <roberto@...>]

> On Tue, 22 Jun 2010 07:31:12 +0300, David Manura <dm.lua@math2.org> wrote:
> 
> >To followup [1], can someone remind me what is the rationale for Lua
> >5.1 tables not storing an explicit table length, which thereby
> >prevents trailing nils from being regularly counted in the table
> >length?  I'm not suggesting storing length in a regular table key,
> >like the Lua 5.0 "n" field which affects things like pairs iteration,
> >but rather storing it in some hidden location accessible only via the
> ># operator.
> 
> The best summary was here (replace "definition" with "implementation"):
> http://lua-users.org/lists/lua-l/2010-06/msg00100.html
> 
> Main issue is that automatically maintaining the hidden "n" has a
> runtime cost that is not acceptable for all users.

The semantics also seem quite problematic. How the implementation will
know whether "t[#t] = nil" is shrinking the list or simply changing
its last value to nil? One should be able to do both operations.

If you want an explicit table length, use it, storing it in a regular
table key.

-- Roberto