From: PG Bug reporting form <noreply@postgresql.org>
To: pgsql-bugs@lists.postgresql.org
Cc: imchifan@163.com
Subject: BUG #19883: Partition pruning errors on SubPlan inside = ANY array
Date: Wed, 11 Nov 2026 14:04:19 +0000
Message-ID: <19883-5fe20b9d0ae341dc@postgresql.org> (raw)
The following bug has been logged on the website:
Bug reference: 19883
Logged by: Qifan Liu
Email address: imchifan@163.com
PostgreSQL version: 18.6
Operating system: Linux/amd64
Description:
On PostgreSQL 18.6 on Linux/amd64, a SELECT against a range-partitioned
table fails with `ERROR: SubPlan found with no parent plan` when an `= ANY`
array element contains an uncorrelated SubPlan whose expression evaluates to
NULL.
The predicate is valid SQL and should return no rows. `NULL::integer IN
(SELECT v ...)` evaluates to NULL, its integer cast remains NULL, and
comparison with the resulting one-element NULL array is not true for either
row. Startup partition pruning should not attempt to evaluate the SubPlan in
a context without a parent PlanState.
How to repeat:
```sql
DROP TABLE IF EXISTS items CASCADE;
DROP TABLE IF EXISTS lookup_values;
CREATE TABLE items (id integer) PARTITION BY RANGE (id);
CREATE TABLE items_low
PARTITION OF items FOR VALUES FROM (MINVALUE) TO (10);
CREATE TABLE items_high
PARTITION OF items FOR VALUES FROM (10) TO (MAXVALUE);
INSERT INTO items VALUES (1), (11);
CREATE TABLE lookup_values (v integer);
INSERT INTO lookup_values VALUES (1);
SELECT id
FROM items
WHERE id = ANY
(ARRAY[(NULL::integer IN
(SELECT v FROM lookup_values))::integer])
ORDER BY id;
DROP TABLE items CASCADE;
DROP TABLE lookup_values;
```
Save this as `reproducer.sql` and run:
```
psql -X -f reproducer.sql
```
The client reports:
```
ERROR: SubPlan found with no parent plan
```
The corresponding server log records the failing statement:
```
ERROR: SubPlan found with no parent plan
STATEMENT: SELECT id
FROM items
WHERE id = ANY
(ARRAY[(NULL::integer IN
(SELECT v FROM lookup_values))::integer])
ORDER BY id;
```
Instead, the SELECT should complete normally and return zero rows. The issue
was reproduced on PostgreSQL 18.6; no other versions were tested. This
occurs when startup partition pruning processes an `= ANY` expression whose
array contains an uncorrelated SubPlan that evaluates to NULL.
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients from your mail client; the link fills in
the headers below: reply via email
Some clients, webmail especially, drop In-Reply-To and References from
a link, and the reply then starts a new thread: use the mbox below.
To: pgsql-bugs@postgresql.org
Cc: noreply@postgresql.org, imchifan@163.com
Subject: Re: BUG #19883: Partition pruning errors on SubPlan inside = ANY array
In-Reply-To: <19883-5fe20b9d0ae341dc@postgresql.org>
References: <19883-5fe20b9d0ae341dc@postgresql.org>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox