From: Tom Lane <tgl@sss.pgh.pa.us>
To: Erik Wienhold <ewie@ewie.name>
cc: Alexey Murz Korepov <murznn@gmail.com>, pgsql-admin@lists.postgresql.org
Subject: Re: Ways to change a database collation with removing duplicates
In-reply-to: <1378811615.328761.1678562412852@office.mailbox.org>
References: 
 <CAL5pyKvXL69bRrONgkutz=xeCUQz6QO7GN_JndqBD0Mq5OJJ-Q@mail.gmail.com>
 <1378811615.328761.1678562412852@office.mailbox.org>
Comments: In-reply-to Erik Wienhold <ewie@ewie.name>
	message dated "Sat, 11 Mar 2023 20:20:12 +0100"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <1253566.1678568003.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Sat, 11 Mar 2023 15:53:23 -0500
Message-ID: <1253567.1678568003@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/1253567.1678568003%40sss.pgh.pa.us>
Precedence: bulk

Erik Wienhold <ewie@ewie.name> writes:
> On 11/03/2023 19:23 CET Alexey Murz Korepov <murznn@gmail.com> wrote:
>> Could anyone suggest to me the ways to change a database collation with
>> removing all the duplicates, caused by this change?

> Collations can only affect uniqueness if they are nondeterministic or if=
 you
> have functional indexes, e.g. using lower(text) for a case-insensitive u=
nique
> index.  Otherwise the collations only affect text ordering.

Yeah.  I suspect that what actually happened here was a previous change in
the host system's sort ordering (cf [1]), leading to text indexes becoming
functionally corrupt and unable to enforce uniqueness correctly, after
which you accumulated some unintentional duplicates.  If you try
reindex'ing on the source database you'll probably find that it fails with
the same errors.  I don't know of any automatic tools for fixing up such
duplications, and wouldn't trust one hugely anyway --- you'll probably
need manual curation of the fixes.

			regards, tom lane

[1] https://wiki.postgresql.org/wiki/Locale_data_changes