Re: Tying an object's ownership to datdba

John Naylor <john.naylor@enterprisedb.com>

From: John Naylor <john.naylor@enterprisedb.com>
To: Noah Misch <noah@leadboat.com>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2021-03-25T13:55:57Z
Lists: pgsql-hackers
On Thu, Mar 25, 2021 at 12:07 AM Noah Misch <noah@leadboat.com> wrote:
>
> > In the refactoring patch, there is a lingering comment reference to
roles_has_privs_of(). Aside from that, it looks good to me. A possible
thing to consider is an assert that is_admin is not null where we expect
that.
>
> Thanks.  The next version will contain the comment fix and
> "Assert(OidIsValid(admin_of) == PointerIsValid(is_admin));".
>
> > The database owner role patch looks good as well.
>
> Do you plan to put the CF entry into Ready for Committer, or should the
> patches wait for another review?

I've marked it Ready for Committer.

--
John Naylor
EDB: http://www.enterprisedb.com

Commits

  1. Add "pg_database_owner" default role.

  2. Merge similar algorithms into roles_is_member_of().