On Mon, Nov 15, 2021 at 5:23 PM Tatsuro Yamada < tatsuro.yamada...@nttcom.co.jp> wrote:
> > > I'm not confident that if I would use this, so let's wait to see if > someone > > else wants to give a +1. > > Okay, but you agree that there are DBAs and users who want to see the > list of constraints, I think. > My opinion is this doesn't exist because there isn't any demand for it. > For example, it will be easier to understand how many foreign key > constraints are in the DB. That isn't a very compelling metric. Metrics also don't seem to be the primary motivation for the psql \d commands. I envision them mostly useful when writing a query and wanting a query refresher as to what is valid/available. In that context looking at constraints in the context of a single table makes sense. Looking at all constraints is considerably less so. Especially since constraints mostly impact insert queries and those only affect a single table. If the only motivation for this is "feature completion" - since we have so many other \d commands already implemented - I say we should pass. David J.
