| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Emmanuel Charpentier <charpent(at)bacbuc(dot)dyndns(dot)org> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Aggregates with non-commutative transition functions |
| Date: | 2003-02-14 00:10:58 |
| Message-ID: | 2214.1045181458@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Emmanuel Charpentier <charpent(at)bacbuc(dot)dyndns(dot)org> writes:
> In other words, may I guarantee that :
> select recnum, glue(linenum)as authors from (select recnum, linenum, author
> from cit_authors where <some conditions on recnum> order by recnum,
> linenum) as foo;
> will indeed give me the authors in the original order ?
The query as given is illegal; you'd need to add "GROUP BY recnum" to
the outer query to make it legal. And once you do that, I don't think
you can rely on the ordering of the rows.
This problem has been discussed before, and I think we came up with some
workaround for it, but I don't have time to go trolling the archives for
the solution at the moment.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Greg Stark | 2003-02-14 01:04:07 | Re: index scan with index cond on first column doesn't recognize sort order of second column |
| Previous Message | Tom Lane | 2003-02-14 00:01:30 | Re: set returning functions in v7.3 |