Horton

Re: Removing unneeded self joins

Alexander Korotkov <aekorotkov@gmail.com>

View thread
From: Alexander Korotkov <aekorotkov@gmail.com>
To: Andrey Lepikhov <a.lepikhov@postgrespro.ru>
Cc: "Gregory Stark (as CFM)" <stark.cfm@gmail.com>, Michał Kłeczek <michal@kleczek.org>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2023-10-04T00:12:02Z
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 →

  1. Remove GUC_NOT_IN_SAMPLE from enable_self_join_elimination

  2. Put enable_self_join_elimination into postgresql.conf.sample

  3. Get rid of ojrelid local variable in remove_rel_from_query()

  4. Implement Self-Join Elimination

  5. Revert: Remove useless self-joins

  6. Replace lateral references to removed rels in subqueries

  7. Replace relids in lateral subquery parse tree during SJE

  8. Forbid SJE with result relation

  9. Fix misuse of RelOptInfo.unique_for_rels cache by SJE

  10. Replace the relid in some missing fields during SJE

  11. Revert 56-bit relfilenode change and follow-up commits.

  12. Stabilize timetz test across DST transitions.

  13. Speed up finding EquivalenceClasses for a given set of rels

  14. Fix mark-and-restore-skipping test case to not be a self-join.

Hi!

I think this is a neat optimization.

On Tue, Sep 12, 2023 at 4:58 PM Andrey Lepikhov
<a.lepikhov@postgrespro.ru> wrote:
> On 5/7/2023 21:28, Andrey Lepikhov wrote:
> > During the significant code revision in v.41 I lost some replacement
> > operations. Here is the fix and extra tests to check this in the future.
> > Also, Tom added the JoinDomain structure five months ago, and I added
> > code to replace relids for that place too.
> > One more thing, I found out that we didn't replace SJs, defined by
> > baserestrictinfos if no one self-join clause have existed for the join.
> > Now, it is fixed, and the test has been added.
> > To understand changes readily, see the delta file in the attachment.
> Here is new patch in attachment. Rebased on current master and some
> minor gaffes are fixed.

I went through the thread and I think the patch gets better shape.  A
couple of notes from my side.
1) Why replace_relid() makes a copy of lids only on insert/replace of
a member, but performs deletion in-place?
2) It would be nice to skip the insertion of IS NOT NULL checks when
they are not necessary.  [1] points that infrastructure from [2] might
be useful.  The patchset from [2] seems committed mow.  However, I
can't see it is directly helpful in this matter.  Could we just skip
adding IS NOT NULL clause for the columns, that have
pg_attribute.attnotnull set?

Links
1. https://www.postgresql.org/message-id/2375492.jE0xQCEvom%40aivenronan
2.  https://www.postgresql.org/message-id/flat/830269.1656693747%40sss.pgh.pa.us


------
Regards,
Alexander Korotkov