| From: | Mark Dilger <mark(dot)dilger(at)enterprisedb(dot)com> |
|---|---|
| To: | Zhihong Yu <zyu(at)yugabyte(dot)com> |
| Cc: | Tomas Vondra <tomas(dot)vondra(at)enterprisedb(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>, Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: Use extended statistics to estimate (Var op Var) clauses |
| Date: | 2021-08-28 17:40:08 |
| Message-ID: | E2517511-7474-4E8B-88C4-12CCB2BC0591@enterprisedb.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> On Aug 28, 2021, at 10:18 AM, Zhihong Yu <zyu(at)yugabyte(dot)com> wrote:
>
> I wonder if the queries with (a=a) or (a<a) clauses are worth this additional complexity to address.
> Has anyone seen such clause in production queries ?
You might expect clauses like WHERE FALSE to be unusual, but that phrase gets added quite a lot by query generators. Somebody could add "WHERE a != a" in a misguided attempt to achieve the same thing.
I wouldn't be terribly surprised if query generators might generate queries with a variable number of tables joined together with comparisons between the joined tables, and in the degenerate case of only one table end up with a table column compared against itself.
You could argue that those people need to fix their queries/generators to not do this sort of thing, but the end user affected by such queries may have little ability to fix it.
—
Mark Dilger
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Justin Pryzby | 2021-08-28 18:16:18 | extended stats objects are the only thing written like "%s"."%s" |
| Previous Message | Zhihong Yu | 2021-08-28 17:18:06 | Re: Use extended statistics to estimate (Var op Var) clauses |