Thread

Domains on ranges should create a multirange

Paul A Jungwirth <pj@illuminatedcomputing.com> — 2025-12-03T18:59:28Z

Hi Hackers,

While reviewing
https://www.postgresql.org/message-id/CACJufxGoAmN_0iJ%3DhjTG0vGpOSOyy-vYyfE%2B-q0AWxrq2_p5XQ%40mail.gmail.com,
I noticed that when you create a domain on a rangetype, you don't get
a corresponding multirange. For instance:

```
[v19devel:5432][481380] mr=# create domain int4range_d as int4range
check (value <> '[4,5)');
CREATE DOMAIN
[v19devel:5432][481380] mr=# select oid, typname, typtype, typbasetype
from pg_type where oid >= 20000;
  oid  |   typname    | typtype | typbasetype
-------+--------------+---------+-------------
 24694 | _int4range_d | b       |           0
 24695 | int4range_d  | d       |        3904
(2 rows)
```

This means there is no way to get a multirange over a domain'ed rangetype.

On the other hand you *can* get a multirange over a rangetype whose
subtype has a domain:

```
[v19devel:5432][481380] mr=# select oid, typname, typtype, typbasetype
from pg_type where oid > 24695;
  oid  |      typname       | typtype | typbasetype
-------+--------------------+---------+-------------
 24697 | _integer_d         | b       |           0
 24698 | integer_d          | d       |          23
 24700 | _int4_d_range      | b       |           0
 24701 | int4_d_multirange  | m       |           0
 24702 | _int4_d_multirange | b       |           0
 24703 | int4_d_range       | r       |           0
(6 rows)
```

This feels like a bug. Are there any backwards compatibility concerns,
if I write a patch to create a multirange when a domain over a range
is created?

Yours,

-- 
Paul              ~{:-)
pj@illuminatedcomputing.com