On Mon, Dec 8, 2025 at 9:51 PM Ron Johnson <[email protected]> wrote:

> On Mon, Dec 8, 2025 at 9:40 PM Igor Korot <[email protected]> wrote:
>
>> Hi, ALL,
>> Consider the following scenario:
>>
>> CREATE TABLE test(a INT, b VARCHAR(256), c INT, d VARCHAR(256), /*
>> more fields follows*/);
>> CREATE UNIQUE INDEX test_x( b, c, d );
>>
>> Now I try to do:
>>
>> INSERT INTO test VALUES( 0, 'abc', 12345, (SELECT foo FROM bar),
>> /*more data follow*/);
>>
>> My problem is:
>>
>> The SELECT can either return data or NULL.
>> Everything is good when the data is returned, but the insert fails
>> when the NULL is returned, because the field "d" is a part of UNIQUE
>> INDEX.
>>
>> However,, I'd like to still insert the record and I'd like to do
>> something like:
>>
>> INSERT INTO test VALUES( 0, 'abc', 12345, IF( (SELECT foo FROM bar) ==
>> NULL, "postgres", <select_result>), /*more data follow*/);
>>
>> What would be the best way to achieve this?
>>
>
> https://www.postgresql.org/docs/15/sql-createindex.html section on NULLS
> DISTINCT says
> "Specifies whether for a unique index, null values should be considered
> distinct (not equal). *The default* is that they are *distinct*, so that
> a *unique index could contain multiple null values in a column*."
>
> That seems to mean multiple rows can have NULL in column "d".
>

It does work:

dba=# create table foo (a int, b int, c int, d int);
CREATE TABLE
dba=# create unique index i_foo_u1 on foo (a, b, d);
CREATE INDEX
dba=#
dba=# insert into foo values (1, 1, 1, 1);
INSERT 0 1
dba=# insert into foo values (2, 2, 2, null);
INSERT 0 1
dba=# insert into foo values (3, 3, 3, null);
INSERT 0 1
dba=# insert into foo values (4, 4, 4, null);
INSERT 0 1

-- 
Death to <Redacted>, and butter sauce.
Don't boil me, I'm still alive.
<Redacted> lobster!

Reply via email to