Re: Question about Postgresql time fields(possible bug)

In article <1292(dot)1136913298(at)sss(dot)pgh(dot)pa(dot)us>,
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:

> Harald Fuchs <hf0923x(at)protecting(dot)net> writes:
>> Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
>>> A leap second will show as 24:00:00. It is a valid time.

>> Shouldn't such a leap second be represented as '... 23:59:60'?

> People who didn't like 24:00:00 would complain about that, too ;-)

Well, Richard T. Snodgrass says in "Developing Time-Oriented Database
Applications in SQL" (pg. 81) the following:

> Most days have 24 hours. The day in April that daylight saving time
> kicks in has only 23 hours; the day in October that daylight saving
> time ends contains 25 hours. Similarly, minutes can have 62 seconds
> (though up to 1999 only one leap second has ever been added to any
> particular minute), as mentioned in this standard [44, p. 25].

where ref [44] is

> ISO, Database Language SQL. ISO/IEC 9075: 1992. ANSI X3.135-1992

To me this sounds like 23:59:60, doesn't it?