| From: | Kyotaro HORIGUCHI <horiguchi(dot)kyotaro(at)lab(dot)ntt(dot)co(dot)jp> |
|---|---|
| To: | pgsql-hackers(at)postgresql(dot)org |
| Cc: | tgl(at)sss(dot)pgh(dot)pa(dot)us, robertmhaas(at)gmail(dot)com |
| Subject: | Re: Get more from indices. |
| Date: | 2013-11-22 07:59:10 |
| Message-ID: | 20131122.165910.65987817.horiguchi.kyotaro@lab.ntt.co.jp |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hello. I found a bug(?) in thsi patch as I considered on another
patch.
In truncate_useless_pathkeys, if query_pathkeys is longer than
pathkeys made from index columns old patch picked up the latter
as IndexPath's pathkeys. But the former is more suitable
according to the context here.
the attched pathkey_and_uniqueindx_v4_20131122.patch is changed
as follows.
- Rebased to current master.
- truncate_useless_pathkeys returns root->query_pathkeys when
the index is fully-ordered and query_pathkeys contains the
pathkeys made from index columns.
regards,
--
Kyotaro Horiguchi
NTT Open Source Software Center
| Attachment | Content-Type | Size |
|---|---|---|
| pathkey_and_uniqueindx_v4_20131122.patch | text/x-patch | 8.5 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kyotaro HORIGUCHI | 2013-11-22 07:59:27 | Re: Using indices for UNION. |
| Previous Message | Haribabu kommi | 2013-11-22 06:42:15 | Re: Heavily modified big table bloat even in auto vacuum is running |