Message-ID: <bf166909-8a8d-2d5f-ed5e-4a4d5121bde7@enterprisedb.com>
Date: Mon, 12 Jun 2023 22:38:59 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.11.0
Content-Language: en-US
To: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>,
 Peter Eisentraut <peter.eisentraut@enterprisedb.com>
From: Tomas Vondra <tomas.vondra@enterprisedb.com>
Subject: Shouldn't construct_array_builtin and deconstruct_array_builtin agree
 on types?
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Archived-At: <https://www.postgresql.org/message-id/bf166909-8a8d-2d5f-ed5e-4a4d5121bde7%40enterprisedb.com>
Precedence: bulk

Hi,

I noticed that these two function, introduced in d746021de18b, disagree
on what types they support. For construct_array_builtin, we support
these types:

 - CHAROID:
 - CSTRINGOID:
 - FLOAT4OID:
 - INT2OID:
 - INT4OID:
 - INT8OID:
 - NAMEOID:
 - OIDOID:
 - REGTYPEOID:
 - TEXTOID:
 - TIDOID:

while deconstruct_array_builtin supports just a subset:

 - CHAROID:
 - CSTRINGOID:
 - FLOAT8OID:
 - INT2OID:
 - OIDOID:
 - TEXTOID:
 - TIDOID:

I ran into this for INT4OID. Sure, I can just lookup the stuff and use
the regualr deconstruct_array, but the asymmetry seems a bit strange. Is
that intentional?

regards

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