Re: cataloguing NOT NULL constraints

On 2024-May-13, Robert Haas wrote:

> It seems to me that the practical thing to do about this problem is
> just decide not to solve it. I mean, it's currently the case that if
> you establish a PRIMARY KEY when you create a table, the columns of
> that key are marked NOT NULL and remain NOT NULL even if the primary
> key is later dropped. So, if that didn't change, we would be no less
> compliant with the SQL standard (or your reading of it) than we are
> now.
[...]
> So I don't really think it's a great idea to change this behavior, but
> even if it is, is it such a good idea that we want to sink the whole
> patch set repeatedly over it, as has already happened twice now? I
> feel that if we did what Tom suggested a year ago in
> https://www.postgresql.org/message-id/3801207.1681057430@sss.pgh.pa.us
> -- "I'm inclined to think that this idea of suppressing the implied
> NOT NULL from PRIMARY KEY is a nonstarter and we should just go ahead
> and make such a constraint" [...]

Hmm, I hadn't interpreted Tom's message the way you suggest, and you may
be right that it might be a good way forward. I'll keep this in mind
for next time.

--
Álvaro Herrera PostgreSQL Developer — https://www.EnterpriseDB.com/
"No es bueno caminar con un hombre muerto"