Horton

Re: [18] Policy on IMMUTABLE functions and Unicode updates

Jeff Davis <pgsql@j-davis.com>

View thread
From: Jeff Davis <pgsql@j-davis.com>
To: Jeremy Schneider <schneider@ardentperf.com>, Laurenz Albe <laurenz.albe@cybertec.at>
Cc: Daniel Verite <daniel@manitou-mail.org>, Noah Misch <noah@leadboat.com>, Peter Eisentraut <peter@eisentraut.org>, Robert Haas <robertmhaas@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@postgresql.org
Date: 2024-07-24T17:35:38Z
Lists: pgsql-hackers 
On Tue, 2024-07-23 at 06:31 -0600, Jeremy Schneider wrote:
> Other RDBMS are very careful not to corrupt databases, afaik
> including function based indexes, by changing Unicode. I’m not aware
> of any other RDBMS that updates Unicode versions in place; instead
> they support multiple Unicode versions and do not drop the old ones.

I'm curious about the details of what other RDBMSs do.

Let's simplify and say that there's one database-wide collation at
version 1, and the application doesn't use any COLLATE clause or other
specifications for queries or DDL.

Then, version 2 of that collation becomes available. When a query comes
into the database, which version of the collation does it use, 1 or 2?
If it uses the latest available (version 2), then all the old indexes
are effectively useless.

So I suppose there's some kind of migration process where you
rebuild/fix objects to use the new collation, and when that's done then
you change the default so that queries use version 2. How does all that
work?

Regards,
	Jeff Davis