Doug Rogers wrote:
Sorry about that - it sounds unkind that way. I meant 'in the case where #t has no sensible semantics, I would rather have ipairs() stop at the first nil.'Roberto Ierusalimschy wrote:Yes, I understand that completely. I still think it makes more sense for ipairs() to depend on something guaranteed to stop at the first nil rather than on #t - something that has no sensible semantics.I disagree. In my point of view, there is no sensible semantics for #t when the table has holes.
Doug