Dandandan commented on code in PR #3530:
URL: https://github.com/apache/arrow-datafusion/pull/3530#discussion_r975397476
##########
datafusion/core/src/physical_plan/sorts/sort.rs:
##########
@@ -816,14 +831,14 @@ fn sort_batch(
batch: RecordBatch,
schema: SchemaRef,
expr: &[PhysicalSortExpr],
+ fetch: Option<usize>,
) -> ArrowResult<BatchWithSortArray> {
- // TODO: pushup the limit expression to sort
let sort_columns = expr
.iter()
.map(|e| e.evaluate_to_sort_column(&batch))
.collect::<Result<Vec<SortColumn>>>()?;
- let indices = lexsort_to_indices(&sort_columns, None)?;
+ let indices = lexsort_to_indices(&sort_columns, fetch)?;
Review Comment:
I think the current change already buffers `num_input_batches * limit` by
the way, as it is applied before adding them to the buffer. As far as I can see
adding the second to `lexsort_to_indices` will reduce mainly the output of the
individual sorts to `fetch` - which is of course beneficial too as that reduces
time to sort and limit the input again to `take` and input to
`SortPreservingMergeStream`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
