Re: NEXT VALUE FOR <sequence>

>>>>> "Tom" == Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:

>>> The spec clearly says one value per row, not one per statement;
>>> so command ID is very definitely not the right thing.

>> I think (command ID, estate->es_processed) would work.

Tom> Not terribly well, eg each new transaction starts over at
Tom> command ID 1. You could fix that particular objection by also
Tom> tracking virtual xid. But the bigger issue is that using
Tom> es_processed for this seems like an utter hack. It's not meant
Tom> to be anything but statistical, and it's not maintained anyway
Tom> for non-canSetTag queries (ie, DO ALSO rule commands). That
Tom> reflects the fact that what it's meant to do is count the number
Tom> of rows returned to the executor's caller, which isn't
Tom> necessarily the definition we'd need here.

Maybe it would make sense to do something with a SubPlan, rather than
trying to hide everything inside a function?

--
Andrew (irc:RhodiumToad)