On 8/1/21 8:51 AM, Tom Lane wrote:
> Adrian Klaver <adrian.klaver@aklaver.com> writes:
>
> Now it's possible that the wording didn't change exactly at the 2/3
> boundary, but I doubt it. So it looks to me like the alleged plpython3u
> language is actually invoking python 2.something, which is unlikely
> to work well at all.
In that vein the OP could run:
DO $$
import sys
plpy.notice(sys.version)
$$ LANGUAGE plpython3u;
in the database to see what plpython3u is actually pointing at?
>
> regards, tom lane
>
--
Adrian Klaver
adrian.klaver@aklaver.com