Robert Haas <robertmhaas@gmail.com> writes:
> On Tue, Aug 2, 2016 at 5:42 PM, David G. Johnston
> <david.g.johnston@gmail.com> wrote:
> The fact that pg_dump is emitting COMMENT ON DATABASE at all is
> fundamentally wrong given the existing division-of-labor decisions,
> namely that pg_dump is responsible for objects within a database
> not for database-level properties.
> I think a while back somebody had the idea of making COMMENT ON
> CURRENT_DATABASE or COMMENT ON CURRENT DATABASE work, which seems like
> an elegant solution to me. Of course, I just work here.
I'm fairly annoyed at David for having selectively quoted from private
email in a public forum, but that was one of the points I touched on
in material that he cut. The point I tried to make to him is that
possibly COMMENT ON CURRENT DATABASE is a portion of a holistic solution,
but it's only a portion. We need to rethink exactly what pg_dump is
supposed to do with database-level properties. And if it does need
COMMENT ON CURRENT DATABASE, it likely also needs GRANT ... ON CURRENT
DATABASE, ALTER CURRENT DATABASE OWNER TO, ALTER CURRENT DATABASE SET,
and maybe some other things.
regards, tom lane