Re: R: feature proposal ...

Andrew Dunstan <andrew@dunslane.net> writes:
> So why not do what everyone is agreed on now?

I wasn't agreed on it ;-)

The primary objection I've got is that I think this will be a very
considerable increment of work for exactly zero increment in
functionality, compared to being able to copy from a view.  (If you're
not seeing why, consider that COPY is a utility statement not an
optimizable statement; you'd have to change that classification, with
resultant impacts all across the system.)  There are other places
where the effort could be more usefully spent.
        regards, tom lane