Re: PoC/WIP: Extended statistics on expressions

On 1/22/21 3:29 AM, Justin Pryzby wrote:
> This already needs to be rebased on 55dc86eca.
>
> And needs to update rules.out.
>

Whooops. A fixed version attached.

> And doesn't address this one:
>
> On Sun, Jan 17, 2021 at 10:53:31PM -0600, Justin Pryzby wrote:
>> | postgres=# CREATE TABLE t(i int);
>> | postgres=# CREATE STATISTICS s2 ON (i+1) ,(i+1+0) FROM t;
>> | postgres=# \d t
>> | Table "public.t"
>> | Column | Type | Collation | Nullable | Default
>> | --------+---------+-----------+----------+---------
>> | i | integer | | |
>> | Indexes:
>> | "t_i_idx" btree (i)
>> | Statistics objects:
>> | "public"."s2" (ndistinct, dependencies, mcv) ON FROM t
>>
>> on ... what ?
>

Umm, for me that prints:

test=# CREATE TABLE t(i int);
CREATE TABLE
test=# CREATE STATISTICS s2 ON (i+1) ,(i+1+0) FROM t;
CREATE STATISTICS
test=# \d t
Table "public.t"
Column | Type | Collation | Nullable | Default
--------+---------+-----------+----------+---------
i | integer | | |
Statistics objects:
"public"."s2" ON ((i + 1)), (((i + 1) + 0)) FROM t

which I think is OK. But maybe there's something else to trigger the
problem?

regards

--
Tomas Vondra
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company