Re: Probably bug in implementation of numerical constant parsing

[Duncan Cross <duncan.cross@...>]

On Mon, Sep 17, 2012 at 7:03 PM, Coda Highland <chighland@gmail.com> wrote:
> But anyway, Joao's point is that if we didn't have number-string
> coercion, you would have to have written "tostring(64)..s", which
> wouldn't have this problem

Don't forget that the __concat metamethod allows for the possibility
of odd constructions using the concatenation operator anyway. It's
unusual, sure, but just because usage is unusual is no excuse for
related core semantics/implementation to be anything other than
rock-solid.

-Duncan