Re: Inconsistent results from seqscan and gist indexscan

Richard Guo <guofenglinux@gmail.com> writes:
> On Fri, Nov 26, 2021 at 5:23 PM Julien Rouhaud <rjuju123@gmail.com> wrote:
>> On Fri, Nov 26, 2021 at 2:10 PM Richard Guo <guofenglinux@gmail.com>
>> wrote:
>>> Seems point_inside() does not handle NaN properly.

>> This is unfortunately a known issue, which was reported twice ([1] and
>> [2]) already.  There's a patch proposed for it:
>> https://commitfest.postgresql.org/32/2710/ (adding Horiguchi-san in
>> Cc).

> Ah, I missed the previous threads. Good to know there is a patch fixing
> it.

Note that that patch seems pretty well stalled; if you'd like to
see it move forward, please pitch in and help review.

(Maybe we should scale back the patch's ambitions, and just try
to get the seqscan/indexscan inconsistency fixed.)

            regards, tom lane