From: Tom Lane <tgl@sss.pgh.pa.us>
To: "Regina Obe" <lr@pcorp.us>
cc: pgsql-hackers@lists.postgresql.org, Jeff Davis <pgsql@j-davis.com>
Subject: Re: Order changes in PG16 since ICU introduction
In-reply-to: <000b01d97465$c34bbd60$49e33820$@pcorp.us>
References: <000b01d97465$c34bbd60$49e33820$@pcorp.us>
Comments: In-reply-to "Regina Obe" <lr@pcorp.us>
	message dated "Fri, 21 Apr 2023 11:27:19 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <3353946.1682092131.1@sss.pgh.pa.us>
Date: Fri, 21 Apr 2023 11:48:51 -0400
Message-ID: <3353947.1682092131@sss.pgh.pa.us>
Archived-At: <https://www.postgresql.org/message-id/3353947.1682092131%40sss.pgh.pa.us>
Precedence: bulk

"Regina Obe" <lr@pcorp.us> writes:
> A couple of days ago, our PostGIS PG16 bots started failing with order
> changes in text.
> We have our tests set to locale=c

> It seems since April 20th, our tests that rely on sorting characters
> changed.
> As noted in this ticket:

> https://trac.osgeo.org/postgis/ticket/5375

> I'm assuming it's result of icu change:
> https://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=fcb21b3ac
> dcb9a60313325618fd7080aa36f1626

> I suspect all our bots are compiling with icu enabled. But I haven't
> confirmed.

If they actually are using locale C, I would say this is a bug.
That should designate memcmp sorting and nothing else.

			regards, tom lane