Horton

Re: psql's FETCH_COUNT (cursor) is not being respected for CTEs

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Daniel Verite <daniel@manitou-mail.org>, Jakub Wartak <jakub.wartak@enterprisedb.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2023-01-04T16:36:51Z
Lists: pgsql-hackers 
Robert Haas <robertmhaas@gmail.com> writes:
> On Wed, Jan 4, 2023 at 10:22 AM Daniel Verite <daniel@manitou-mail.org> wrote:
>> A solution would be for psql to use PQsetSingleRowMode() to retrieve
>> results row-by-row, as opposed to using a cursor, and then allocate
>> memory for only FETCH_COUNT rows at a time.

> Is there any reason that someone hasn't, like, already done this?

As you well know, psql's FETCH_COUNT mechanism is far older than
single-row mode.  I don't think anyone's tried to transpose it
onto that.  I agree that it seems like a good idea to try.
There will be more per-row overhead, but the increase in flexibility
is likely to justify that.

			regards, tom lane

Commits

  1. Further review for re-implementation of psql's FETCH_COUNT feature.

  2. Re-implement psql's FETCH_COUNT feature atop libpq's chunked mode.

  3. Support retrieval of results in chunks with libpq.

  4. Attempt to fix newly added Memoize regression test