On 1/22/21 10:00 AM, Dean Rasheed wrote:
> On Fri, 22 Jan 2021 at 04:46, Justin Pryzby <pryzby@telsasoft.com> wrote:
>>
>> I think you'll maybe have to do something better - this seems a bit too weird:
>>
>> | postgres=# CREATE STATISTICS s2 ON (i+1) ,i FROM t;
>> | postgres=# \d t
>> | ...
>> | "public"."s2" (ndistinct, dependencies, mcv) ON i FROM t
>>
>
> I guess that's not surprising, given that old psql knows nothing about
> expressions in stats.
>
> In general, I think connecting old versions of psql to newer servers
> is not supported. You're lucky if \d works at all. So it shouldn't be
> this patch's responsibility to make that output nicer.
>
Yeah. It's not clear to me what exactly could we do with this, without
"backpatching" the old psql or making the ruleutils.c consider version
of the psql. Neither of these seems possible/acceptable.
I'm sure this is not the only place showing "incomplete" information in
old psql on new server.
regards
--
Tomas Vondra
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
