Re: roundoff problem in time datatype

Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> Where are we on this?

We haven't decided what to do.

I think my preference is to allow '24:00:00' (but not anything larger)
as a valid input value of the time datatypes. This for two reasons:
* existing dump files may contain such values
* it's consistent with allowing, eg, '12:13:60', which we
allow even though it's certainly not a valid leap second.

The alternative is to try to catch all places where 23:59:59.something
could get rounded up to 24:00:00, but that looks messy, and it would
introduce a gotcha into calculations on time values.

regards, tom lane