Re: Support NOT VALID / VALIDATE constraint options for named NOT NULL constraints - Mailing list pgsql-hackers

On 2025-Mar-31, jian he wrote:

> hi.
> in notnull-notvalid.patch
> 
> + if (coninfo->contype == 'c')
> + keyword = "CHECK CONSTRAINT";
> + else
> + keyword = "INVALID NOT NULL CONSTRAINT";
> we have a new TocEntry->desc kind.

Yeah, I wasn't sure that this change made much actual sense.  I think it
may be better to stick with just CONSTRAINT.  There probably isn't
sufficient reason to have a different ->desc value.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/