Feedback-ID: ie3de48e3:Fastmail
Date: Wed, 25 Feb 2026 17:12:56 +0100
From: =?utf-8?Q?=C3=81lvaro?= Herrera <alvherre@kurilemu.de>
To: VASUKI M <vasukianand0119@gmail.com>
Cc: pgsql-translators@lists.postgresql.org,
	Daniel Gustafsson <daniel@yesql.se>
Subject: Re: [PATCH] Fix incorrect Spanish translation and remove obsolete
 FIXME comments
Message-ID: <202602251555.pzsyfbfto36y@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: 
 <CAE2r8H4VNdJvS0jg+f9eL6KYAS=wosWD6pF0FVbfDLPEXi2-xw@mail.gmail.com>
Archived-At: 
 <https://www.postgresql.org/message-id/202602251555.pzsyfbfto36y%40alvherre.pgsql>
Precedence: bulk

Hello,

Many thanks for the report!

On 2026-Feb-20, VASUKI M wrote:

> While reviewing the Spanish translation file (src/backend/po/es.po),
> I found an incorrect translation in the message:
> 
>   "physical replication slot \"%s\" exists, but \"wal_level\" < \"replica\""
> 
> The current translation refers to a logical replication slot
> ("replicación lógica"), which is incorrect in this context.
> This patch corrects the translation to use "replicación física",
> so that the message accurately reflects the original meaning.

Your patch applies only to branch 17, but the translation is actually
incorrect in all branches.  (More precisely, the message fix applies;
the FIXME removals were all reported as merge failures).  Moreover, in
branches 14 through 16, the translation said "replica" instead of
"logical", which is what the original has.  So I fixed that as well.

(Another point is that in this repository the file is actually named
es/postgres.po, so what I actually did was "patch es/postgres.po <
/tmp/yourpatch" rather than any git commands.)

I used this command to search for other potentially affected
translations,
 msggrep --no-wrap -K -e 'replication slot.*exists.*but' es/postgres.po
which found none.


I didn't commit the FIXME removals, because it'd be a pain to merge
them, and also even if they no longer reference a relevant code
location, they are pointing out an issue that continues to exist --
namely, that the original messages shouldn't have the string "logical"
or "replica" as the value being compared to wal_level, but instead they
should do something lik this:
            ereport(FATAL,
                    (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
                     errmsg("logical replication slot \"%s\" exists, but \"%s\" < \"%s\"",
                            NameStr(cp.slotdata.name), "wal_level", "replica"),
                     errhint("Change \"%s\" to be \"%s\" or higher.",
                             "wal_level", "replica")));

However, I never got around to fixing the code on the grounds there were
only two affected messages; I think this ultimately proved a judgement
mistake, because when the required level was changed from logical to
replica, we all had to update the translations.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/
"Ed is the standard text editor."
      http://groups.google.com/group/alt.religion.emacs/msg/8d94ddab6a9b0ad3