Re: Schema variables - new implementation for Postgres 15

On 2024-Jan-30, Dmitry Dolgov wrote:

> Yep, in this constellation the implementation holds much better (in
> terms of memory) in my create/let/drop testing.
>
> I've marked the CF item as ready for committer, but a note for anyone
> who would like to pick up it from here -- we're talking about first 5
> patches here, up to the memory cleaning after DROP VARIABLE. It doesn't
> mean the rest is somehow not worth it, but I believe it's a good first
> step.

Hmm, I think patch 16 is essential, because the point of variable shadowing
is a critical aspect of how the whole thing works. So I would say that
a first step would be those first five patches plus 16.

I want to note that when we discussed this patch series at the dev
meeting in FOSDEM, a sort-of conclusion was reached that we didn't want
schema variables at all because of the fact that creating a variable
would potentially change the meaning of queries by shadowing table
columns. But this turns out to be incorrect: it's _variables_ that are
shadowed by table columns, not the other way around.

--
Álvaro Herrera PostgreSQL Developer — https://www.EnterpriseDB.com/
"No hay ausente sin culpa ni presente sin disculpa" (Prov. francés)