| From: | Alexander Pyhalov <a(dot)pyhalov(at)postgrespro(dot)ru> |
|---|---|
| To: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Inconsistency between try_mergejoin_path and create_mergejoin_plan |
| Date: | 2024-06-17 14:51:30 |
| Message-ID: | c59ec04a2fef94d9ffc35a9b17dfc081@postgrespro.ru |
| Views: | Whole Thread |Raw Message |Download mbox |Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi.
There's the following inconsistency between try_mergejoin_path() and
create_mergejoin_plan().
When clause operator has no commutator, we can end up with mergejoin
path.
Later create_mergejoin_plan() will call get_switched_clauses(). This
function can error out with
ERROR: could not find commutator for operator XXX
The similar behavior seems to be present also for hash join.
Attaching a test case (in patch) and a possible fix.
--
Best regards,
Alexander Pyhalov,
Postgres Professional
| Attachment | Content-Type | Size |
|---|---|---|
| 0001-Merge-and-hash-joins-need-more-checks-with-custom-op.patch | text/x-diff | 5.8 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Melanie Plageman | 2024-06-17 14:53:27 | Re: Show WAL write and fsync stats in pg_stat_io |
| Previous Message | Jacob Champion | 2024-06-17 14:11:06 | Re: Direct SSL connection and ALPN loose ends |