Re: Collation versioning - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: Collation versioning
Date
Msg-id 4f60612c-a7b5-092d-1532-21ff7a106bd5@2ndquadrant.com
Whole thread Raw
In response to Re: Collation versioning  (Christoph Berg <myon@debian.org>)
Responses Re: Collation versioning  (Christoph Berg <myon@debian.org>)
Re: Collation versioning  (Thomas Munro <thomas.munro@enterprisedb.com>)
List pgsql-hackers
On 12/09/2018 13:25, Christoph Berg wrote:
> Re: Peter Eisentraut 2018-09-12 <0447ec7b-cdb6-7252-7943-88a4664e7bb7@2ndquadrant.com>
>>> Naive idea: make that catalog shared? Collations are system-wide after
>>> all.
>>
>> By the same argument, extensions should be shared, but they are not.
> 
> But extensions put a lot of visible stuff into a database, whereas a
> collation is just a line in some table that doesn't get into the way.

How about C functions?  They are just a system catalog representation of
something that exists on the OS.

Anyway, we also want to support application-specific collation
definitions, so that users can CREATE COLLATION
"my_specific_requirements" and use that that in their application, so
global collations wouldn't be appropriate for that.

Moreover, the fix for a collation version mismatch is, in the simplest
case, to go around and REINDEX everything.  Making the collation or
collation version global doesn't fix that.  It would actually make it
harder because you couldn't run ALTER COLLATION REFRESH VERSION until
after you have rebuilt all affected objects *in all databases*.

-- 
Peter Eisentraut              http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services


pgsql-hackers by date:

Previous
From: Tatsuo Ishii
Date:
Subject: Re: Unused argument from execute_sql_string()
Next
From: Paul Guo
Date:
Subject: Re: [Patch] Create a new session in postmaster by calling setsid()