On 2020-Jan-14, Tom Lane wrote: > I can't get terribly excited about persuading that test to cover this > trivial little bit of logic, but if you are, I won't stand in the way.
Hmm, that's a good point actually: the patch changed several places to inject the FOREIGN keyword, so in order to cover them all it would need several additional regexps, not just one. I'm not sure that 002_pg_dump.pl is prepared to do that without unsightly contortions. Anyway, other than that minor omission the patch seemed good to me, so I don't oppose Tomas pushing the version I posted yesterday. Or I can, if he prefers that. -- Álvaro Herrera https://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
