Re: [Proposal] Arbitrary queries in postgres_fdw

From: rtorre(at)carto(dot)com
To: Rafa de la Torre <rtorre(at)carto(dot)com>
Cc: David Fetter <david(at)fetter(dot)org>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: [Proposal] Arbitrary queries in postgres_fdw
Date: 2019-11-05 10:09:34
Message-ID: CA+Fz15khMu6XSL1rEkY6KXYLgwVL8E+tfE_0qaQoRz1M=mJSQQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sun, Oct 27, 2019 at 7:07 PM David Fetter <david(at)fetter(dot)org> wrote:
> There's a SQL MED standard feature for CREATE ROUTINE MAPPING that
> does something similar to this. Might it be possible to incorporate
> it into the previous patch that implemented that feature?

Supporting CREATE ROUTINE MAPPING goes a level beyond
postgres_fdw. It'd require adding new DDL syntax elements to the
parser, catalog and callbacks for the FDW's to support them.

For the record, there's a very interesting thread on this topic (that
you participated in):
https://www.postgresql.org/message-id/flat/CADkLM%3DdK0dmkzLhaLPpnjN2wBF5GRpvzOr%3DeW0EWdCnG-OHnpQ%40mail.gmail.com

I know they have different semantics and may turn more limiting, but
for certain use cases, the `extensions` parameter of postgres_fdw may
come in handy (shipping function calls to the foreign end from
extensions present in both local and foreign).

For my use case, which is retrieving catalog info before any CREATE
FOREIGN TABLE, CREATE ROUTINE MAPPING is not really a good fit.

Thank you for pointing out anyway.
-Rafa

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Kyotaro Horiguchi 2019-11-05 10:27:48 Re: Refactor parse analysis of EXECUTE command
Previous Message k.jamison@fujitsu.com 2019-11-05 09:58:22 RE: [Patch] Optimize dropping of relation buffers using dlist