Am Thu, Jun 13, 2024 at 09:49:46AM +0200 schrieb Laurenz Albe: > > > Note that with DROP COLLATION you can only remove collations > > > that belong to the encoding of your current database. > > src/backend/catalog/namespace.c: > > /* > * get_collation_oid - find a collation by possibly qualified name > * > * Note that this will only find collations that work with the current > * database's encoding. > */ > Oid > get_collation_oid(List *collname, bool missing_ok) Thanks. Are collations per-database or per-cluster objects ? I am asking because I seem to not be enabled to 1) use a collation that's intended for an encoding different from the database encoding -> makes sense 2) remove a collation that's intended for an encoding different from the database encoding -> so far so good, ignore them, but 3) update collation version information in pg_collations for collations intended for an encoding different from the database encoding (ALTER COLLATION ... REFRESH VERSION fails) which in effect would mean that -- upon change of collation versions in the underlying operating system (ICU update, libc update) -- one would have to live with outdated version information in pg_collations short of dump/sed/restore or some such ? I'm pretty sure I am overlooking something. Best, Karsten -- GPG 40BE 5B0E C98E 1713 AFA6 5BC0 3BEA AC80 7D4F C89B
