Horton

Re: Strange behavior with polygon and NaN

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Cc: bruce@momjian.us, sbjesse@gmail.com, pgsql-hackers@lists.postgresql.org
Date: 2020-08-26T12:18:49Z
Lists: pgsql-hackers 
Kyotaro Horiguchi <horikyota.ntt@gmail.com> writes:
> At Tue, 25 Aug 2020 19:03:50 -0400, Bruce Momjian <bruce@momjian.us> wrote in 
>> I can confirm that this two-month old email report still produces
>> different results with indexes on/off in git master, which I don't think
>> is ever correct behavior.

> I agree to that the behavior is broken. 

Yeah, but ... what is "non broken" in this case?  I'm not convinced
that having point_inside() return zero for any case involving NaN
is going to lead to noticeably saner behavior than today.

			regards, tom lane

Commits

  1. In geo_ops.c, represent infinite slope as Infinity, not DBL_MAX.

  2. Fix FPeq() and friends to get the right answers for infinities.

  3. Extend the geometric regression test cases a little.

  4. Remove underflow error in float division with infinite divisor.