xtime/libtai/"right"

[Antoine Leca]

Markus Kuhn wrote:
> 
> Bradley White wrote on 1998-10-10 18:53 UTC:
> > People seem to dislike struct xtime because of the
> > ability to represent bogus leap seconds.
> 
> Just for the record: ISO 8601 style display UTC as well as struct tm
> also has the ability to represent bogus leap seconds such as
> 01:02:60.123.

But they do not give it any sense (except on input for mktime()).

Also just for the record:
2000-06-30 23:59:60,456Z is a wrong timestamp these days, but it
might very well be a correct one in 16 months from now (assuming
the Earth's rotation variation does not change too much until
that time).

Same for the equivalent TIME_UTC value using struct xtime.

But as noted by Bradley and David before, this is not a problem when
you are only talking about inode times.


Antoine