Re: Remaining dependency on setlocale() - Mailing list pgsql-hackers

From Jeff Davis
Subject Re: Remaining dependency on setlocale()
Date
Msg-id cf976b924de917c762b53e0d7bcb5ded2d0b1aa3.camel@j-davis.com
Whole thread Raw
In response to Re: Remaining dependency on setlocale()  (Jeff Davis <pgsql@j-davis.com>)
List pgsql-hackers
On Tue, 2026-01-06 at 14:20 -0800, Jeff Davis wrote:
> Looks good to me.

I closed the entry for this commitfest, though there are a few loose
ends:

  * Not sure what to do about strerror & gettext. One idea is to wrap
both in a uselocale() everywhere. For windows, we use the special
setlocale() mode. For NetBSD, I don't know if we can or should do
anything -- maybe just prevent changing lc_messages?

  * pg_strcasecmp() still used many places in the backend. We should
consider replacing with an ASCII variant or one that uses the database
collation.

  * A number of places still rely on isalpha(), etc., for simple
parsing (e.g. ltree).

Regards,
    Jeff Davis




pgsql-hackers by date:

Previous
From: Masahiko Sawada
Date:
Subject: Re: RepOrigin vs. replorigin
Next
From: Masahiko Sawada
Date:
Subject: Re: pg_upgrade: optimize replication slot caught-up check