Horton

Re: BUG #18764: server closed the connection unexpectedly

David Rowley <dgrowleyml@gmail.com>

View thread
From: David Rowley <dgrowleyml@gmail.com>
To: Richard Guo <guofenglinux@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, dqetool@126.com, pgsql-bugs@lists.postgresql.org
Date: 2025-01-10T01:35:52Z
Lists: pgsql-bugs

Commits

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

  1. Fix UNION planner datatype issue

  2. Allow planner to use Merge Append to efficiently implement UNION

On Thu, 9 Jan 2025 at 20:38, Richard Guo <guofenglinux@gmail.com> wrote:
>
> On Thu, Jan 9, 2025 at 12:44 PM David Rowley <dgrowleyml@gmail.com> wrote:
> > Can't you just look up the setop's type from op->colTypes instead of
> > looking the type up via the sortop? i.e. the attached?
>
> Yeah, right.  Using SetOperationStmt.colTypes is more convenient here.

I pushed a revised patch. I removed the check for the sortop. I first
also thought I'd overlooked that, but it turned out that's fine as it
was as standard_qp_callback() handles checking for un-sortable
SortGroupClauses.

Jason, you can expect a fix for this in 17.3. In the meantime, you
could add an explicit cast to numeric in the subquery, e.g:

SELECT c2 AS ca2, c2 AS ca3
FROM t0
UNION
SELECT DISTINCT ca8 AS ca5, ca7::numeric AS ca6
FROM (
SELECT c1 AS ca7, c2 AS ca8 FROM t0
) AS ta1 JOIN (
SELECT c1 AS ca10, c1 AS ca11
FROM t0
) AS ta2 ON TRUE;

Thanks for the report.

David