| From: | Vladimir Sitnikov <sitnikov(dot)vladimir(at)gmail(dot)com> |
|---|---|
| To: | Ildar Musin <i(dot)musin(at)postgrespro(dot)ru> |
| Cc: | Oleg Bartunov <obartunov(at)gmail(dot)com>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org>, Robert Haas <robertmhaas(at)gmail(dot)com>, Alexander Korotkov <a(dot)korotkov(at)postgrespro(dot)ru>, Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com> |
| Subject: | Re: Index Onlys Scan for expressions |
| Date: | 2016-09-08 18:58:17 |
| Message-ID: | CAB=Je-FroJAi-v8PeuDX3xW0JC8Y6RrD=bfg3uzqeOjd6zHuuw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Ildar> Could you please try the patch and tell if it works for you?
I've tested patch6 against recent head. The patch applies with no problems.
The previous case (filter on top of i-o-s) is fixed. Great work.
Here are the test cases and results:
https://gist.github.com/vlsi/008e18e18b609fcaaec53d9cc210b7e2
However, it looks there are issues when accessing non-indexed columns.
The error is "ERROR: variable not found in subplan target list"
The case is 02_case2_fails.sql (see the gist link above)
The essence of the case is "create index on substr(vc, 1, 128)"
and assume that majority of the rows have length(vc)<128.
Under that conditions, it would be nice to do index-only-scan
and filter (like in my previous case), but detect "long" rows
and do additional recheck for them.
Vladimir
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Heikki Linnakangas | 2016-09-08 18:59:41 | Re: Tuplesort merge pre-reading |
| Previous Message | Tom Lane | 2016-09-08 18:57:40 | Re: Re: GiST optimizing memmoves in gistplacetopage for fixed-size updates [PoC] |