Re: Faster methods for getting SPI results (460% improvement)

On 4/6/17 9:21 PM, Andres Freund wrote:
>> Personally I'm way more excited about what a SPI feature like this
>> could do for plpgsql than about what it can do for plpython. If the
>> latter is what floats your boat, that's fine; but I want a feature
>> that we can build on for other uses, not a hack that we know we need
>> to redesign next month.

Yeah, I thought about plpgsql and I can't see any way to make that work
through an SPI callback (perhaps just due to my ignorance on things C).
I suspect what plpgsql actually wants is a way to tell SPI to start the
executor up, a function that pulls individual tuples out of the
executor, and then a function to shut the executor down.

> Dislike of the proposed implementation, alternative proposals, and the
> refutation of the "absolutely no way to do more without breaking plpy"
> argument leads to me to conclude that this should be returned with
> feedback.

Agreed.
--
Jim Nasby, Chief Data Architect, Austin TX
OpenSCG http://OpenSCG.com