On Sat, Apr 4, 2009 at 4:16 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Yeah. I did add a regression test for the specific case of '30 days'
> vs '1 month', which we know is a pain point for this particular data
> type. Generating values at random doesn't seem like it's really likely
> to teach us much though.
Yeah, I was going to suggest random values but came to the same conclusion.
I think each datatype would have to provide some list of "interesting"
values. But given that list it would be possible to do the same
regression tests for every data type.
I suppose it would require a plpgsql function using PERFORM to
implement though. ick.
--
greg