Re: pg_upgrade test chatter

Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> On 2021-Oct-19, Tom Lane wrote:
>> We could dodge that, with modern versions of psql, by issuing
>> two -c switches.

> Isn't it easier to pass client_min_messages via PGOPTIONS?

> PGOPTIONS="-c client_min_messages=warning" psql -c "drop database if exists foo"

Yeah, my original thought had been to hack this at the test level.
However, I felt like it'd be worth adding this code because we could
apply it elsewhere in pg_regress.c to save several psql sessions
(and hence backend starts) per regression DB creation.  That's not a
huge win, but it'd add up.

			regards, tom lane