Re: BUG #1609: Bug in interval datatype for 64 Bit timestamps

Oliver Siegmar <oliver@siegmar.net> writes:
> It removed the bug, but also added a new one (hopefully only one ;-))

I don't think it's a new bug, seeing that I didn't change the code
for the ISO case.

I see the imprecise result only in the non-integer-datetime case; is
it acting differently for you?

If it is only the float case, some imprecision is to be expected.

            regards, tom lane