Re: Adjusting a mode function for the deprecation of anyarray - Mailing list pgadmin-support

From Aditya Toshniwal
Subject Re: Adjusting a mode function for the deprecation of anyarray
Date
Msg-id CAM9w-_m_uPgn3uUkhWcZGhptcEdp7on4P6A2nMg_iNVuB=X9rQ@mail.gmail.com
Whole thread Raw
In response to Adjusting a mode function for the deprecation of anyarray  (Wells Oliver <wells.oliver@gmail.com>)
List pgadmin-support
Hi Wells,

This is a pgAdmin mailing list. You will get better support on PostgreSQL mailing list - pgsql-admin@postgresql.org.

On Sun, Jan 15, 2023 at 2:45 AM Wells Oliver <wells.oliver@gmail.com> wrote:
We have had and used this basic dumb useful aggregate in PG 13 and earlier:

CREATE OR REPLACE FUNCTION _final_mode(anyarray)
  RETURNS anyelement AS
$BODY$
    SELECT a
    FROM unnest($1) a
    GROUP BY 1
    ORDER BY COUNT(1) DESC, 1
    LIMIT 1;
$BODY$
LANGUAGE sql IMMUTABLE;

CREATE AGGREGATE mode(anyelement) (
  SFUNC=array_append,
  STYPE=anyarray,
  FINALFUNC=_final_mode,
  INITCOND='{}'
);

I can't seem to figure out how to adjust the mode(anyelement) bit now for PG 14 and the loss of anyarray. Changing STYPE to anyelement yields "function array_append(anyelement, anyelement) does not exist" and changing it to anycompatiblearray yields:

ERROR:  cannot determine transition data type
DETAIL:  A result of type anycompatiblearray requires at least one input of type anycompatible, anycompatiblearray, anycompatiblenonarray, anycompatiblerange, or anycompatiblemultirange.

What's the obvious thing I am missing here? Thank you.

--


--
Thanks,
Aditya Toshniwal
pgAdmin Hacker | Software Architect | edbpostgres.com
"Don't Complain about Heat, Plant a TREE"

pgadmin-support by date:

Previous
From: Wells Oliver
Date:
Subject: Adjusting a mode function for the deprecation of anyarray
Next
From: Akshay Joshi
Date:
Subject: pgAdmin 4 v6.19 Released