On 2/12/18 23:00, Michael Paquier wrote:
> Hm.  Wouldn't it be enough to just spread the use of
> TestLib::check_pg_config and use SKIP on the tests where things cannot
> be run then?  I see no need to invent an extra facility if all the
> control is already in pg_config.h.  For slapd, you can actually just
> check if it can be executed in the PATH defined at the top of
> 001_auth.pl in $slapd, and skip the tests if the thing cannot be run.
> We have enough flexibility in perl and TAP to allow that to work
> reliably.

Maybe that would work.

We still need a way to configure whether we want to run tests that open
TCP/IP listen sockets.

-- 
Peter Eisentraut              http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Reply via email to