On 2020-04-23 00:27, Tom Lane wrote:
This should not only fix the regression instability but offer a pretty significant speedup for real-world use of the triggers view, so I propose squeezing it in even though we're past feature freeze.
I think that's fine.
However ... what about the back branches? As noted, we are seeing this now in the back branches, at least v12. Are we willing to make a change in the information_schema in back branches to make the instability go away? (And if not, how else could we fix it?)
That would be okay with me. -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
