Re: Inconsistent Behavior of GROUP BY ROLLUP in v17 vs master

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

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

To: Richard Guo <guofenglinux@gmail.com>

Cc: David Rowley <dgrowleyml@gmail.com>, 邱宇航 <iamqyh@gmail.com>, Bruce Momjian <bruce@momjian.us>, PostgreSQL-development <pgsql-hackers@postgresql.org>

Date: 2025-10-17T16:40:41Z

Lists: pgsql-hackers

Commits

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

Fix test case from 40c242830
- ee49f2cf447a 18.1 landed
Fix pushdown of degenerate HAVING clauses
- 40c2428307b8 18.1 landed
- 18d261409348 19 (unreleased) landed
Allow pushdown of HAVING clauses with grouping sets
- 67a54b9e83d3 18.0 cited
Mark expressions nullable by grouping sets
- f5050f795aea 18.0 cited

I wrote:
> I started to look at this again, and now I'm thinking that there is
> indeed an issue related to "Query 1".

Oh, scratch that, I see my mistake: I was thinking of "1" as a
constant, but actually we must be interpreting it per SQL92 rules
as a reference to output column 1.  So that's why adding or dropping
the "id" output column changes the results of the grouping.

-ENOCAFFEINE ... sorry for the noise.  I'll go back to studying
the HAVING issue.

			regards, tom lane