Re: Show 'Rows Removed by Filter' with two fractional digits in EXPLAIN

From: Andrei Lepikhov <lepihov(at)gmail(dot)com>
To: Ilia Evdokimov <ilya(dot)evdokimov(at)tantorlabs(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>
Cc: PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Show 'Rows Removed by Filter' with two fractional digits in EXPLAIN
Date: 2026-11-10 10:33:54
Message-ID: de65de5d-ad84-4f83-a30b-3095035c3a98@gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 10/11/2026 08:12, Ilia Evdokimov wrote:
> On 11/10/26 01:59, Andrei Lepikhov wrote:
>
>> Here's the next version of the patch. It passed 'make check-world' on the
>> current master (c074737f687).
>
> One small nit: could we also use a precision of 2 in the else branch of `nloops
>> 0`, so that it prints 0.00 rather than 0? Looking at the changes in [0], its
> parser will accept both integer and fractional values, so this suggestion is
> mainly for consistency.

I doubt that. What you want is the never-executed, non-text format case. In this
case, I see the following:

"Actual Rows": 0,

To ensure, I checked it on the current master with the query:

explain (analyze, costs off, timing off, summary off, buffers off, format json)
select * from
generate_series(1,0) g, lateral
(select * from pg_class c
where c.relpages + g > 0
offset 0
) s;

So, as I understand it, adding floating-point to the 'else' section causes
inconsistency with the 'Actual Rows' format. Am I wrong?

--
regards, Andrei Lepikhov,
pgEdge

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Dave Cramer 2026-11-10 10:39:14 Re: Proposal to allow setting cursor options on Portals
Previous Message Aya Iwata (Fujitsu) 2026-11-10 10:32:45 Re: BGWORKER_INTERRUPTIBLE workers are never restarted after DROP/ALTER/CREATE DATABASE