On Mon, 2021-04-19 at 08:00 +0000, PG Doc comments form wrote:
> The 'round' mathematical function documentation doesn't specify its rounding
> strategy. If this would be documented then users can rely on its rounding
> behavior. For double precision the rounding probably depends on your
> platform. For numeric the function seems to round to nearest with ties going
> away from zero (sometimes referred to as "round half up). 6.5 -> 7.0 and
> -6.5 -> -7.0.
Here is a proposed patch.
Yours,
Laurenz Albe
From a0c40fd431820f6d1ce96f5bc3f41bd7665028be Mon Sep 17 00:00:00 2001
From: Laurenz Albe <laurenz.a...@cybertec.at>
Date: Thu, 22 Apr 2021 18:53:14 +0200
Subject: [PATCH] Document tie breaking behavior of round()
---
doc/src/sgml/func.sgml | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 67d802b02c..fcc09d5231 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -1568,7 +1568,9 @@ repeat('Pg', 4) <returnvalue>PgPgPgPg</returnvalue>
<returnvalue>double precision</returnvalue>
</para>
<para>
- Rounds to nearest integer
+ Rounds to nearest integer. For <type>numeric</type>, ties are
+ broken by rounding away from 0. For <type>double precision</type>,
+ the tie breaking behavior is architecture dependent.
</para>
<para>
<literal>round(42.4)</literal>
@@ -1583,7 +1585,7 @@ repeat('Pg', 4) <returnvalue>PgPgPgPg</returnvalue>
</para>
<para>
Rounds <parameter>v</parameter> to <parameter>s</parameter> decimal
- places
+ places. Ties are broken by rounding away from 0.
</para>
<para>
<literal>round(42.4382, 2)</literal>
--
2.26.3
