Date: Wed, 2 Jun 2021 21:39:20 -0500
From: Justin Pryzby <pryzby@telsasoft.com>
To: Tomas Vondra <tomas.vondra@enterprisedb.com>
Cc: Dean Rasheed <dean.a.rasheed@gmail.com>, pgsql-hackers@postgresql.org
Subject: Re: PoC/WIP: Extended statistics on expressions (\d in old client)
Message-ID: <20210603023920.GD14099@telsasoft.com>
References: <2f2a3b3b-93e4-723f-144e-bc21f10d6b61@enterprisedb.com>
 <20210118045331.GF8560@telsasoft.com>
 <CAEZATCVnQccV_Z7AuvbvquaOYxUj+cVFk48pPGASYHFjXWgW5g@mail.gmail.com>
 <3f20da4d-89dc-0912-98d2-b6f0ef06c87f@enterprisedb.com>
 <CAEZATCUuj5FPjtOU2=7Z2pTC8LEvrf79+Go7R8yQ+eE0ezfgfQ@mail.gmail.com>
 <f865af90-c19c-769e-8c65-19692e245479@enterprisedb.com>
 <20210122022958.GC27167@telsasoft.com>
 <acf27932-8159-8872-8fca-902c7a35b351@enterprisedb.com>
 <20210122040101.GF27167@telsasoft.com>
 <04b100f9-201b-0bc8-c613-0f918451b599@enterprisedb.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <04b100f9-201b-0bc8-c613-0f918451b599@enterprisedb.com>
User-Agent: Mutt/1.9.4 (2018-02-28)
Archived-At: 
 <https://www.postgresql.org/message-id/20210603023920.GD14099%40telsasoft.com>
Precedence: bulk

On Fri, Jan 22, 2021 at 02:09:04PM +0100, Tomas Vondra wrote:
> On 1/22/21 5:01 AM, Justin Pryzby wrote:
> > On Fri, Jan 22, 2021 at 04:49:51AM +0100, Tomas Vondra wrote:
> > > > > | Statistics objects:
> > > > > |     "public"."s2" (ndistinct, dependencies, mcv) ON  FROM t
> > > 
> > > Umm, for me that prints:
> > 
> > >      "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?
> > 
> > Oh.  It's because I was using /usr/bin/psql and not ./src/bin/psql.
> > I think it's considered ok if old client's \d commands don't work on new
> > server, but it's not clear to me if it's ok if they misbehave.  It's almost
> > better it made an ERROR.
> > 
> 
> Well, how would the server know to throw an error? We can't quite patch the
> old psql (if we could, we could just tweak the query).

To refresh: stats objects on a v14 server which include expressions are shown
by pre-v14 psql client with the expressions elided (because the attnums don't
correspond to anything in pg_attribute).

I'm mentioning it again since, even though I knew about this earlier in the
year, it caused some confusion for me again just now while testing our
application.  I had the v14 server installed but the psql symlink still pointed
to the v13 client.

There may not be anything we can do about it.
And it may not be a significant issue outside the beta period: more typically,
the client version would match the server version.

-- 
Justin