On Thu, Feb 14, 2019 at 4:57 PM Michael Paquier <michael@paquier.xyz> wrote:
>
> On Wed, Feb 13, 2019 at 12:08:50PM +0100, Chris Travers wrote:
> > If the point is visibility in \dx it seems to me we want to fix the \dx
> > query.
>
> Yes, I got to think a bit more about that case, and there are cases
> where this actually works properly as this depends on the objects
> defined in the extension. Fixing \dx to not show up extensions
> defined in temp schemas of other sessions is definitely a must in my
> opinion, and I would rather drop the rest of the proposal for now. A
> similar treatment is needed for \dx+.
I'd vote for accepting the extension creation in temporary schemas and
fixing \dx and \dx+. However the error raised by creating extensions
in temporary schema still looks strange to me. Since we don't search
functions and operators defined in temporary schemas (which is stated
by the doc) unless we use qualified function name we cannot create
extensions in temporary schema whose functions refer theirs other
functions. I'd like to fix it or to find a workaround but cannot come
up with a good idea yet.
Regards,
--
Masahiko Sawada
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center