Thread

Re: BUG #19340: Wrong result from CORR() function

Dean Rasheed <dean.a.rasheed@gmail.com> — 2025-12-02T23:48:28Z

On Tue, 2 Dec 2025 at 23:24, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Dean Rasheed <dean.a.rasheed@gmail.com> writes:
> > I played around with having just 2 extra array elements, constX and
> > constY equal to the common value if all the values are the same, and
> > NaN otherwise.
>
> Hmm.
>
> > Doing it that way does lead to one difference though: all-NaN inputs
> > leads to a NaN result, whereas your patch produces NULL for that case.
>
> Yeah, I did it as I did precisely because I wanted all-NaN-input to be
> seen as a constant.  But you could make an argument that NaN is not
> really a fixed value but has more kinship to the "we don't know what
> the value is" interpretation of SQL NULL.

I think that's more consistent with the general policy that most
(all?) math functions have, where if any input is NaN, the result is
NaN.

Regards,
Dean