Re: Assigning fixed OIDs to system catalogs and indexes

Поиск
Список
Период
Сортировка
От Bruce Momjian
Тема Re: Assigning fixed OIDs to system catalogs and indexes
Дата
Msg-id 200504181724.j3IHOLC00828@candle.pha.pa.us
обсуждение исходный текст
Ответ на Re: Assigning fixed OIDs to system catalogs and indexes  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: Assigning fixed OIDs to system catalogs and indexes  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
Tom Lane wrote:
> Christopher Kings-Lynne <chriskl@familyhealth.com.au> writes:
> >> So some changing-oid operations like vacuum full, reindex, etc will not
> >> affect these system catalogs?
> 
> > Is it possible to cluster system tables?
> 
> No, and yes.  CLUSTER, REINDEX, and similar things change the
> relfilenode, not the relation OID.
> 
> If you are looking for things that will fundamentally break the system,
> try renaming one of the indexes of a core system catalog ;-).  We could
> avoid that dependency if we went by OIDs instead.

Right now you can use a temporary table in place of a system
table using the temporary schema.  Would that still work?test=> BEGIN;BEGINtest=> CREATE temporary TABLE pg_class(x
INT);CREATETABLEtest=> INSERT INTO pg_class VALUES (1);INSERT 0 1test=> SELECT * FROM pg_class; x--- 1(1 row)
 

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
359-1001+  If your life is a hard drive,     |  13 Roberts Road +  Christ can be your backup.        |  Newtown Square,
Pennsylvania19073
 


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: Problem with PITR recovery
Следующее
От: Tom Lane
Дата:
Сообщение: Re: Problem with PITR recovery