| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Clifford Wolf <clifford(dot)wolf(at)linbit(dot)com> |
| Cc: | pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: Bug in query planer ? |
| Date: | 2006-02-01 17:19:55 |
| Message-ID: | 1839.1138814395@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
Clifford Wolf <clifford(dot)wolf(at)linbit(dot)com> writes:
> However, it would be great to have get_restriction_variable() and
> get_attstatsslot() extended so they can pass the most common values
> from the statistics cache thru expressions, as described in my earlier
> mail.
This would only be helpful if the most-common-values list describes
practically all of the column population, which isn't typically the case.
In any case I'm not sure why you're resistant to maintaining an index
on an expression that you are frequently querying by --- that index
could have more direct use than just cueing ANALYZE what to figure
statistics on.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2006-02-01 17:32:39 | Re: Huge number of disk writes after migration to 8.1 |
| Previous Message | Patrick Rotsaert | 2006-02-01 16:57:50 | Re: BUG #2225: Backend crash -- BIG table |