Horton

Re: Replace IN VALUES with ANY in WHERE clauses during optimization

Laurenz Albe <laurenz.albe@cybertec.at>

View thread
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: Alena Rybakina <a.rybakina@postgrespro.ru>, Ivan Kush <ivan.kush@tantorlabs.com>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2024-10-03T20:19:56Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Stabilize regression test from c0962a113.

  2. Convert 'x IN (VALUES ...)' to 'x = ANY ...' then appropriate

  3. Extract make_SAOP_expr() function from match_orclause_to_indexcol()

On Thu, 2024-10-03 at 23:10 +0300, Alena Rybakina wrote:
> On 03.10.2024 22:52, Ivan Kush wrote:
> > 
> > Hello, hackers! I with my friends propose the patch to replace IN 
> > VALUES to ANY in WHERE clauses.
> > 
> > # Intro
> > 
> > The `VALUES` in the `IN VALUES` construct is replaced with with an 
> > array of values when `VALUES` contains 1 column. In the end it will be 
> > replaced with ANY by the existing function makeA_Expr 
> > (src/backend/nodes/makefuncs.c)
> > 
> > This improves performance, especially if the values are small.
> 
> Anlrey Lepikhov and I recently described this in an article [0] here and 
> the implementation already exists, but for now it was posted a binary 
> application for testing. The acceleration is significant I agree.
> 
> [0] https://danolivo.substack.com/p/7456653e-9716-4e91-ad09-83737784c665

I believe that the speed improvement is significant, but who writes a
query like

  ... WHERE col IN (VALUES (1), (2), (3))

when they could write the much shorter

  ... WHERE col IN (1, 2, 3)

which is already converted to "= ANY"?

I wonder if it is worth the extra planning time to detect and improve
such queries.

Yours,
Laurenz Albe