Content-Type: text/plain; charset="iso-8859-15"
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: Change initdb default to the builtin collation provider
From: "Daniel Verite" <daniel@manitou-mail.org>
To: "Robert Haas" <robertmhaas@gmail.com>
Cc: "Jeff Davis" <pgsql@j-davis.com>,pgsql-hackers@postgresql.org
In-Reply-To: 
 <CA+TgmoYPfiz6_QjkiKykwziFDZFzPVf_x77rChYBu=OGVOmbYw@mail.gmail.com>
Date: Wed, 11 Mar 2026 17:28:05 +0100
Message-Id: <ec3d1357-d8c4-47d1-887b-ec2f0b8a7e67@manitou-mail.org>
Archived-At: 
 <https://www.postgresql.org/message-id/ec3d1357-d8c4-47d1-887b-ec2f0b8a7e67%40manitou-mail.org>
Precedence: bulk

	Robert Haas wrote:

> To be honest, I'd probably be ready to support making the default
> encoding UTF8 regardless of the environment, and you have to use -E
> if you want anything else. I think there are still people using
> other encodings, but I believe it to be a small minority at this
> point.

It would be interesting to have the point of view of Asian users about
this. Recently, the suggestion to retire GB18030 in favor of UTF-8 was
met with the objection that GB18030 was likely preferred by users from
China [1].  Another example against UTF-8 that I found notable, is
Tatsuo Ishii mentioning that Japanese users tend use --no-locale
rather than UTF-8 locales [2].

Also, it's not obvious how initdb could choose an UTF-8 locale
regardless of the environment.
For instance, let's say it finds LC_ALL=3D"fr_FR.iso885915@euro", what
would it do? Maybe look at the UTF-8 locales on the system.  Here's a
subset of what it would find on my system:

C.utf8
en_AG
en_AG.utf8
en_AU.utf8
en_BW.utf8
en_CA.utf8
en_DK.utf8
en_GB.utf8
en_HK.utf8
en_IE.utf8
...
tr_TR.utf8


=46rom that kind of list, which locale should it pick and why?

Personally I think that ignoring the environment's LC_* for the
collations would be fine if we went for builtin/C.UTF-8 by default, as
$subject suggests.  But the level of enthusiasm for that from the
community seems much lower than it would need to be for that kind of
change to be acceptable.


[1]
https://www.postgresql.org/message-id/45b4b689-0e78-4d30-a5f9-1a39d01ab2b7%=
40ww-it.cn
[2]
https://www.postgresql.org/message-id/20230608.104535.2171011311090815110.t=
-ishii%40sranhm.sra.co.jp


Best regards,
--=20
Daniel V=E9rit=E9=20
https://postgresql.verite.pro/