2010YOUY01 commented on code in PR #26925:
URL: https://github.com/apache/datafusion/pull/26925#discussion_r4422685002
##########
datafusion/physical-plan/src/aggregates/group_values/multi_group_by/clustered.rs:
##########
@@ -396,35 +403,40 @@ mod tests {
}
for key_type in key_types {
for nullable in [false, true] {
- let rows = 1025;
- let first: ArrayRef =
- Arc::new(Int32Array::from_iter((0..rows).map(|row| {
- let group = (row / 7) as i32;
- let key = group / 4;
- // Swap neighboring key values while keeping each run
contiguous.
- (!nullable || group >= 4).then_some(if unsorted {
- key ^ 1
- } else {
- key
- })
- })));
- let second = clustered_key_array(&key_type, rows, nullable)?;
- for descending in [false, true] {
- let mut input = vec![Arc::clone(&first),
Arc::clone(&second)];
- if descending {
- let reverse =
- UInt32Array::from_iter_values((0..rows as
u32).rev());
- input = input
- .iter()
- .map(|array| take(array, &reverse, None))
- .collect::<std::result::Result<_, _>>()?;
- }
- let schema = Arc::new(Schema::new(vec![
- Field::new("a", DataType::Int32, nullable),
- Field::new("b", key_type.clone(), nullable),
- ]));
- assert!(GroupValuesClustered::supports_schema(&schema),
"{key_type}");
- for batch_size in [1, 2, 7, 8, 63, 1024] {
+ for batch_size in [1, 2, 7, 8, 63, 1024] {
+ // Small batches need only enough rows to cross key runs
and null
+ // bitmap boundaries. The 1024-row batch still uses 1025
rows.
Review Comment:
```suggestion
```
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
