| From: | "Daniel Verite" <daniel(at)manitou-mail(dot)org> |
|---|---|
| To: | "Nishant Sharma" <nishant(dot)sharma(at)enterprisedb(dot)com> |
| Cc: | "Robert Haas" <robertmhaas(at)gmail(dot)com>,pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Problem in 'ORDER BY' of a column using a created collation? |
| Date: | 2025-10-02 11:11:13 |
| Message-ID: | 3feee389-e1d5-4319-bbb2-680b3df71b10@manitou-mail.org |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Nishant Sharma wrote:
> Testing sort order for '1' & 'a' using ICU library with collation =
> 'ja-u-kr-latn-digit'
>
> With Method ucol_strcollUTF8():
> SORT ORDER ASC : '1', 'a'
>
> With Method ucol_nextSortKeyPart() (i.e transform and memcmp):
> SORT ORDER ASC : 'a', '1'
Yes, ucol_strcoll() with kr-latn-digit has been flawed for a
long time [1] and still is in the development branch of ICU.
There's a recent ticket about this [2] with a reproducer similar
to yours.
[1]
https://www.postgresql.org/message-id/flat/153201618542.1404.3611626898935613264%40wrigleys.postgresql.org
[2] https://unicode-org.atlassian.net/browse/ICU-23016
Best regards,
--
Daniel Vérité
https://postgresql.verite.pro/
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ajin Cherian | 2025-10-02 11:23:34 | Re: Improve pg_sync_replication_slots() to wait for primary to advance |
| Previous Message | Hayato Kuroda (Fujitsu) | 2025-10-02 10:40:24 | RE: POC: enable logical decoding when wal_level = 'replica' without a server restart |