feat(optimizer): improve pull_up_correlated_predicate_agg_rule to optimize TPCH Q17 #15383
Conversation
I am not sure if it could be better in stream because the streaming OverWindow operator is more expensive |
It reminds me why our agg executor could emit messages when the barrier arrives, however, window executor can't do it in the same way? |
There was a problem hiding this comment.
generally LGTM, maybe add the case #15026 (comment) in the planner test?
| let mut agg_null_bitset = FixedBitSet::with_capacity(agg.base.schema().len()); | ||
| for pos in null_agg_pos { | ||
| agg_null_bitset.insert(pos); | ||
| } |
There was a problem hiding this comment.
Are you looking for FixedBitSet::from_iter?
There was a problem hiding this comment.
I need to ensure the capacity of this bitset is agg.base.schema().len().
I hereby agree to the terms of the RisingWave Labs, Inc. Contributor License Agreement.
What's changed and what's your intention?
countaggcall forpull_up_correlated_predicate_agg_ruleif it can be proved that the row generated by empty input finally being able to be eliminated.Windowoperator, but it is out of the scope of this PR.Checklist
./risedev check(or alias,./risedev c)Documentation
Release note
If this PR includes changes that directly affect users or other significant modifications relevant to the community, kindly draft a release note to provide a concise summary of these changes. Please prioritize highlighting the impact these changes will have on users.