From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Andreas Kretschmer <akretschmer(at)spamfence(dot)net> |
Cc: | pgsql-sql(at)postgresql(dot)org |
Subject: | Re: unique constraint definition within create table |
Date: | 2015-12-02 14:46:09 |
Message-ID: | 14838.1449067569@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
Andreas Kretschmer <akretschmer(at)spamfence(dot)net> writes:
> i'm trying to create a table with 2 int-columns and a constraint that a
> pair of (x,y) cannot be as (y,x) inserted:
> test=# create table foo(u1 int,u2 int, unique (least(u1,u2),greatest(u1,u2)));
> ERROR: syntax error at or near "("
> LINE 1: create table foo(u1 int,u2 int, unique (least(u1,u2),greates...
> I know, i can solve that in this way:
> test=*# create table foo(u1 int,u2 int);
> CREATE TABLE
> test=*# create unique index idx_foo on foo(least(u1,u2),greatest(u1,u2));
> CREATE INDEX
> But is there a way to define the unique constraint within the create table - command?
No. Per SQL standard, the argument of a UNIQUE (or PRIMARY KEY)
constraint can only be a list of column names.
We would consider relaxing that, except it would break the
information_schema views for such constraints.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Adrian Klaver | 2015-12-02 14:59:58 | Re: unique constraint definition within create table |
Previous Message | Andreas Kretschmer | 2015-12-02 06:36:21 | unique constraint definition within create table |