Re: Correction of RowMark Removal During Sel-Join Elimination

Andrei Lepikhov <lepihov@gmail.com>

View thread View on postgr.es

From: Andrei Lepikhov <lepihov@gmail.com>

To: Greg Sabino Mullane <htamfids@gmail.com>

Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Alexander Korotkov <aekorotkov@gmail.com>

Date: 2025-08-12T09:40:15Z

Lists: pgsql-hackers

Attachments

v1-0001-Force-RowMark-Sanity-Checking.patch (text/plain) patch v1-0001

On 11/8/2025 20:15, Greg Sabino Mullane wrote:
> I'm not convinced this is an improvement from someone just coming in to 
> this part of the code, especially given (for example) the comment right 
> above it:
> 
>   * Determine if the inner table can duplicate outer rows.  We must
>   * bypass the unique rel cache here since we're possibly using aThanks for your feedback.
I made some minor adjustments to the comments to make the code more 
consistent. Sure, rrel and krel don't seem like the best solution - I 
guess natives could find less wordy synonyms than just dumb 
'keeping_rel' and 'removing_rel'. But it is simpler to track which 
relation and RowMark should be removed.

-- 
regards, Andrei Lepikhov

Commits

Improve RowMark handling during Self-Join Elimination
- 8951941fb8a4 18.0 landed
- 5f6f951f88e5 19 (unreleased) landed
Refactor variable names in remove_self_joins_one_group()
- d713cf9b65ae 19 (unreleased) landed
- 8202b0e83d09 18.0 landed
Reduce "Var IS [NOT] NULL" quals during constant folding
- e2debb64380e 19 (unreleased) cited
Reimplement parsing and storage of default expressions and constraint
- eabc714a916b 7.1.1 cited