Re: How to create unique index on multiple columns where the combination doesn't matter?

["David G. Johnston" <david.g.johnston@xxxxxxxxx>]

Maybe try combining them into a single array then performing array comparisons...

On Wednesday, March 22, 2017, Glen Huang <hey.hgl@xxxxxxxxx> wrote:

Thanks.

Didn't realize it could be implemented with a exclusion constraint. The comparing between any two row definitely sounds like the right direction. But I'm still having a hard time figuring out how i should write the `exclude_element WITH operator` part, which I think, should detect if specified columns consist of the same items, regardless the order? could `exclude_element` contains multiple columns? (from the syntax it looks like it's impossible) And is there such an operator to compare multiple columns?

On 23 Mar 2017, at 1:04 AM, David G. Johnston <david.g.johnston@xxxxxxxxx> wrote:

On Wed, Mar 22, 2017 at 9:54 AM, Glen Huang <hey.hgl@xxxxxxxxx> wrote:

Hello,

If I have a table like

CREATE TABLE relationship (
  obj1 INTEGER NOT NULL REFERENCES object,
  obj2 INTEGER NOT NULL REFERENCES object,
  obj3 INTEGER NOT NULL REFERENCES object,
  ...
)

And I want to constrain that if 1,2,3 is already in the table, rows like 1,3,2 or 2,1,3 shouldn't be allowed.

Is there a general solution to this problem?

Sorry if the question is too basic, but I couldn't find the answer in the doc, at least not in the chapter on unique index.

The most direct option to consider is a exclusion constraint.

https://www.postgresql.org/docs/current/static/ddl-constraints.html (bottom of page)

David J.