Murphy
Murphy
And also, if this algorithm(space-saving) is fundamentally inaccurate, we can consider other algorithms. But before that we can focus on the performance optimization to make it practical. If you want...
> Hi @murphyatwork I would like to proceed with the solution using the approach you suggested. Please find my atttached proposal and assign me the issue. > > ## Proposal:...
According to my previous profiling, I can still see the bottleneck is counter `maintain_ordering` ``` - 44.99% 1.72% pip_exec_com starrocks_be [.] starrocks::AggregateFunctionBatchHelper::upd▒ - 43.26% starrocks::AggregateFunctionBatchHelper::update_batch_single_state ▒ - 35.39% starrocks::ApproxTopKState::_maintain_ordering ▒...
> Hi @murphyatwork Are we looking to optimize this by writing a new algorithm other than Space Saving or we are going to ignore considering the worst case is less...
> @Dshadowzh Absolutely, here is a minimal example that includes nested objects, as well as arrays of objects: > > ```sql > SELECT > CAST(json_parse( > '{"a": 5, "b_obj": {"bb":...
```sql MySQL> with common as ( select parse_json('{"a": 1}') -> 'a' as c1, '1' as c2 ) select c1 = c2, cast(c1 as string) = c2 from common; +---------+-----------------------------------+ |...
@Mergifyio rebase
@Mergifyio rebase
@cursor review
@Mergifyio rebase