TAI zone?

[Guy Harris]

On Jul 1, 2011, at 5:39 AM, Clive D.W. Feather wrote:

> I'm afraid that's wrong. TAI *does* have a concept of years, months, days,
> hours, and minutes. It makes perfect sense to say TZ=TAI and you'll get a
> meaningful answer (which is currently 34 seconds ahead of TZ=UTC).

...but there's no way to have a TAI zone that could convert a POSIX time_t to TAI with localtime() (whether with the Olson code and database with anything else), given that POSIX says "seconds since the Epoch" really means "non-leap seconds since the Epoch".

Then again, as POSIX explicitly admits, there's no way to convert a POSIX time_t to UTC with localtime() 100% of the time.