Re: BUG #3431: age() gets the days wrong

"Pelle Johansson" <pelle(at)morth(dot)org> writes:
> The age() function seem to work by first counting months until less than a
> month remains to to the second argument, then counting days left. This
> doesn't give the correct result, as shown by this example:

age() isn't intended to give the same result as minus, so I'm not
convinced that this behavior is wrong. I'm more worried about what
applications we might break by changing it --- age() has always worked
the way it does now.

regards, tom lane