Tom,
> Actually, the definition of the single-parameter variants of age() is
> age(current_date, $1)
> not
> age(now(), $1)
>
> I'm not sure this is wrong, but perhaps it should be better documented.
Hmmm ... well, if that's the definition, then we probably don't need to change
i
Josh Berkus <[EMAIL PROTECTED]> writes:
> Description: The age() built-in function sometimes truncates hours, minutes
> and seconds, and sometimes it doesn't, depending on the parameters. This
> seems inconsistent and confusing.
> gforge=> select age('2004-01-01'::TIMESTAMP);
> age
> --
Folks,
Pardon me if this is redundant, but again the pgsql-bugs archives are not
working.
Summary: Various versions of AGE() give inconsistent results
Versions Tested: 7.4.2, 7.4.3, 8.0b4
Severity: Annoyance
Description: The age() built-in function sometimes truncates hours, minutes
and seco
