This is an automated email from the ASF dual-hosted git repository.
dheres pushed a commit to branch filter_empty_record_batch
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git
The following commit(s) were added to refs/heads/filter_empty_record_batch by
this push:
new e1e1197454 Support `concat_batches` for 0 columns
e1e1197454 is described below
commit e1e119745423cc3065ba664cd3b5adb291507c37
Author: Daniƫl Heres <[email protected]>
AuthorDate: Mon Aug 7 22:24:58 2023 +0200
Support `concat_batches` for 0 columns
---
arrow-select/src/concat.rs | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
diff --git a/arrow-select/src/concat.rs b/arrow-select/src/concat.rs
index 0bf4c97ff8..ac0405a156 100644
--- a/arrow-select/src/concat.rs
+++ b/arrow-select/src/concat.rs
@@ -97,6 +97,13 @@ pub fn concat_batches<'a>(
schema: &SchemaRef,
input_batches: impl IntoIterator<Item = &'a RecordBatch>,
) -> Result<RecordBatch, ArrowError> {
+ if schema.fields().is_empty() {
+ let num_rows: usize =
input_batches.into_iter().map(RecordBatch::num_rows).sum();
+ let mut options = RecordBatchOptions::default();
+ options.row_count = Some(num_rows);
+ return RecordBatch::try_new_with_options(schema.clone(), vec![],
&options);
+ }
+
let batches: Vec<&RecordBatch> = input_batches.into_iter().collect();
if batches.is_empty() {
return Ok(RecordBatch::new_empty(schema.clone()));
@@ -142,6 +149,21 @@ mod tests {
assert!(re.is_err());
}
+ #[test]
+ fn test_concat_batches_no_columns() {
+ // Test concat using empty schema / batches without columns
+ let schema = Arc::new(Schema::empty());
+
+ let mut options = RecordBatchOptions::default();
+ options.row_count = Some(100);
+ let batch =
+ RecordBatch::try_new_with_options(schema.clone(), vec![],
&options).unwrap();
+ // put in 2 batches of 100 rows each
+ let re = concat_batches(&schema, &[batch.clone(), batch]).unwrap();
+
+ assert_eq!(re.num_rows(), 200);
+ }
+
#[test]
fn test_concat_one_element_vec() {
let arr = Arc::new(PrimitiveArray::<Int64Type>::from(vec![