Re: [PATCH] Integer overflow in timestamp[tz]_part() and date/time boundaries check

On 3/16/16, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> So I fixed that up and committed it, with a very short set of new
> regression test cases. I seriously doubt that the other ones add
> enough value to be worth trying to make them work in both float- and
> int-timestamp cases; though if you want to submit a new patch to
> add more test cases we could consider it. My feeling about it is that
> that kind of testing does nothing for errors of omission (ie functions
> that fail to range-check their results), which is the most likely sort
> of bug here, and so I doubt it's worth adding them to regression tests
> that many people will run many times a day for a long time to come.
>
> regards, tom lane

Thank you very much! If you decide such kind of tests is not
necessary, it is enough for me.

Is there any reason to leave JULIAN_MINDAY and JULIAN_MAXDAY which are
not used now?
Also why JULIAN_MAXMONTH is set to "6" whereas
{DATE|TIMESTAMP}_END_JULIAN use "1" as month?

--
Best regards,
Vitaly Burovoy