RE: Planner picks n² query plan when available

2024-11-25 Thread Toto guyoyg
e57f1194d9f6543bc7948c16411b), or is that expected behavior ? (My understanding is that this wouldn't solve the performance problem of the original query given the const array limitation but this may improve other cases.) Thanks again for your time! De

Planner picks n² query plan when available

2024-11-21 Thread Toto guyoyg
Offending O(n²) query: ```sql SELECT id FROM indexed_table WHERE indexed_value = ANY (ARRAY[1,2,...]) ``` I'm not posting this on the `pgsql-performance` mailing list because this is about fixing the issue, not working around it. I'm not posting this on the `pgsql-bugs` mailing list because docu