Re: VACUUM FULL changes the order of rows in a table?

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Douglas Trainor <trainor(at)uic(dot)edu>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: VACUUM FULL changes the order of rows in a table?
Date: 2002-07-24 07:42:45
Message-ID: 9569.1027496565@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Douglas Trainor <trainor(at)uic(dot)edu> writes:
> WARNING: Not sure if what I am about to describe is a bug.

As Scott comments, it isn't. Under *no* circumstances does Postgres
guarantee anything about the physical order of rows in a table.
This is entirely in accordance with the SQL spec --- row ordering is
not considered significant at all in the computational model. It's
worth noting that ORDER BY is treated by the spec as a mere output
decoration issue; you cannot ORDER the result of a sub-select, only
the final results being sent to the client. Fundamentally SQL does
not think row order is a valid consideration at all.

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message pgsql-bugs 2002-07-24 16:02:45 Bug #718: request for improvement of /? to show /d+ /l+ commands
Previous Message Bruce Momjian 2002-07-23 23:49:53 Re: pg_ctl -w option does not behave as documented