Mime-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset=UTF-8; format=Flowed
Date: Wed, 15 Nov 2023 04:27:49 -0600
Message-Id: <CWZBBRR6YA8D.8EHMDRGLCKCD@neon.tech>
To: "pgsql-hackers" <pgsql-hackers@postgresql.org>
From: "Tristan Partin" <tristan@neon.tech>
Subject: On non-Windows, hard depend on uselocale(3)
Archived-At: 
 <https://www.postgresql.org/message-id/CWZBBRR6YA8D.8EHMDRGLCKCD%40neon.tech>
Precedence: bulk

I have been working on adding using thread-safe locale APIs within=20
Postgres where appropriate[0]. The patch that I originally submitted=20
crashed during initdb (whoops!), so I worked on fixing the crash, which=20
led me to having to touch some code in chklocale.c, which became=20
a frustrating experience because chklocale.c is compiled in 3 different=20
configurations.

> pgport_variants =3D {
>   '_srv': internal_lib_args + {
>     'dependencies': [backend_port_code],
>   },
>   '': default_lib_args + {
>     'dependencies': [frontend_port_code],
>   },
>   '_shlib': default_lib_args + {
>     'pic': true,
>     'dependencies': [frontend_port_code],
>   },
> }

This means that some APIs I added or changed in pg_locale.c, can't be=20
used without conditional compilation depending on what variant is being=20
compiled. Additionally, I also have conditional compilation based on=20
HAVE_USELOCALE and WIN32.

I would like to propose removing HAVE_USELOCALE, and just have WIN32,=20
which means that Postgres would require uselocale(3) on anything that=20
isn't WIN32.

[0]: https://www.postgresql.org/message-id/CWMW5OZBWJ10.1YFLQWSUE5RE9@neon.=
tech

--=20
Tristan Partin
Neon (https://neon.tech)