Horton

Re: Add proper planner support for ORDER BY / DISTINCT aggregates

David Rowley <dgrowleyml@gmail.com>

View thread View on postgr.es
From: David Rowley <dgrowleyml@gmail.com>
To: PostgreSQL Developers <pgsql-hackers@lists.postgresql.org>
Date: 2021-07-12T12:04:22Z
Lists: pgsql-hackers

Attachments

On Sun, 13 Jun 2021 at 03:07, David Rowley <dgrowleyml@gmail.com> wrote:
>
> Please find attached my WIP patch.  It's WIP due to what I mentioned
> in the above paragraph and also because I've not bothered to add JIT
> support for the new expression evaluation steps.

I've split this patch into two parts.

0001 Adds planner support for ORDER BY aggregates.

0002 is a WIP patch for DISTINCT support.  This still lacks JIT
support and I'm still not certain of the best where to store the
previous value or tuple to determine if the current one is distinct
from it.

The 0001 patch is fairly simple and does not require much in the way
of changes in the planner aside from standard_qp_callback().
Surprisingly the executor does not need a great deal of work here
either.  It's just mostly about skipping the normal agg(.. ORDER BY)
code when the Aggref is presorted.

David

Commits

  1. Don't presort ORDER BY/DISTINCT Aggrefs with volatile functions

  2. Add enable_presorted_aggregate GUC

  3. Remove pessimistic cost penalization from Incremental Sort

  4. Fix hypothetical problem passing the wrong GROUP BY pathkeys

  5. Remove unused fields from ExprEvalStep

  6. Improve performance of ORDER BY / DISTINCT aggregates

  7. Refactor function parse_subscription_options.