Re: Bypassing cursors in postgres_fdw to enable parallel plans

From: Jelte Fennema-Nio <postgres(at)jeltef(dot)nl>
To: Rafia Sabih <rafia(dot)pghackers(at)gmail(dot)com>
Cc: Robert Haas <robertmhaas(at)gmail(dot)com>, Yilin Zhang <jiezhilove(at)126(dot)com>, KENAN YILMAZ <kenan(dot)yilmaz(at)localus(dot)com(dot)tr>, Andy Fan <zhihuifan1213(at)163(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Bypassing cursors in postgres_fdw to enable parallel plans
Date: 2026-09-02 19:14:04
Message-ID: CAGECzQSTwxPsLrT9OUO_i8BUv958UeaTL9yHodb=DthbzZgQHA@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Wed, 2 Sept 2026 at 13:49, Rafia Sabih <rafia(dot)pghackers(at)gmail(dot)com> wrote:
> Please find the reworked patchset in the attachment.
> Looking forward to your feedback on this one.

I tried this patch out with the "postgres compatible" endpoint I built
for MotherDuck[1]. DuckDB (currently) doesn't support cursors, so that
is blocking us from supporting postgres_fdw. With this patchset (and a
few small bugfixes on my side) I was able to get it to work with
motherduck its postgres endpoint.

[1]: https://motherduck.com/blog/motherduck-now-speaks-postgres/

P.S. I did not review any of the code.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2026-09-02 19:15:49 Re: GRAPH_TABLE pattern WHERE clause is not coerced to boolean
Previous Message Nathan Bossart 2026-09-02 18:57:51 auto-vectorize varbit bitwise operators