Pushed this. The mechanism of creating a new slot from the source, then later advancing the LSN of the new slot using the updated values from the source slot, seems quite clever. I reworded the comment that explained how it is supposed to work; please double-check to ensure I got it right.
I renamed the opr_sanity-induced wrappers. I see no reason to get very precise about what's what ... these functions are all identical. Thanks everyone! -- Álvaro Herrera https://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
