On Mon, Jun 20, 2016 at 3:29 PM, David G. Johnston <david.g.johnston@gmail.com> wrote: > The entire theory here looks whacked - and seems to fall into the "GUCs > controlling results" bucket of undesirable things.
As far as I can see, this entire email is totally wrong and off-base, because the whole thing seems to be written on the presumption that single_copy is a GUC, when it's actually a structure member. If there was some confusion about that, you could have spent 5 seconds running "git grep" before writing this email, or you could have tried "SET single_copy" and discovered, hey, there's no such GUC.
Furthermore, I think that describing something that you obviously haven't taken any time to understand as "whacked" is not very nice. For that matter, I think that describing something you *have* taken time to understand as "whacked" is not very nice.
Point taken.
I don't think my entire post depends solely upon this being a GUC though.
I've burned too many brain cells on this already, though, to dive much deeper.
Internal or external I do think the number and type of flags described here, for the purposes described, seems undesirable from an architectural standpoint. I do not and cannot offer up more than that generally due to knowledge and resource constraints. I tried to frame things up relative to my understanding of existing, non-parallel, idioms, both to understand it better myself and to throw out another POV from a fresh perspective. I'll admit its one with some drawbacks but its offered in good faith.
Please do with it as you will and accept my apology for the poor choice of colloquialism.