Re: pairs(t, skey) and ipairs(t, skey)

[steve donovan <steve.j.donovan@...>]

On Wed, Oct 2, 2013 at 3:27 PM, Andrew Starks <andrew.starks@trms.com> wrote:
> I too have wanted a sequence. I'm surprised there isn't a library for one,

I'm sure one of our clever people would do one in an afternoon, but I
for one will not be rushing to download it - just don't see the
compelling need.  If I have a list of objects, then 'false' is a
perfectly good non-hole with the right semantics.

If we were all paranoid about checking data _internal_ to a program,
then productivity would dive out of sight - _external_ to a system,
yes. Paranoia is an appropriate attitude when dealing with the rest of
the world.

There is something to be said for not being too picky about data - if
I provide a function for working with the 'array' part it will assume
#t is valid and only access 1..#t.  Any other element coming along for
the ride isn't a problem and not worth complaining about.

steve d.