gstvg commented on code in PR #21679:
URL: https://github.com/apache/datafusion/pull/21679#discussion_r3135682024
##########
datafusion/common/src/utils/mod.rs:
##########
@@ -1280,4 +1452,166 @@ mod tests {
assert_eq!(expected, transposed);
Ok(())
}
+
+ #[test]
+ fn test_sliced_list_values() {
+ let data = vec![
+ Some(vec![Some(0), Some(1), Some(2)]),
+ None,
+ Some(vec![Some(3), None, Some(5)]),
+ Some(vec![Some(6), Some(7)]),
+ ];
+
+ let list = ListArray::from_iter_primitive::<Int32Type, _, _>(data);
+
+ assert_eq!(
+ sliced_list_values(&list).as_primitive(),
+ &Int32Array::from(vec![
+ Some(0),
+ Some(1),
+ Some(2),
+ Some(3),
+ None,
+ Some(5),
+ Some(6),
+ Some(7)
+ ])
+ );
+
+ assert_eq!(
+ sliced_list_values(&list.slice(0, 1)).as_primitive(),
+ &Int32Array::from(vec![Some(0), Some(1), Some(2)])
+ );
+
+ assert_eq!(
+ sliced_list_values(&list.slice(2, 1)).as_primitive(),
+ &Int32Array::from(vec![Some(3), None, Some(5)])
+ );
+
+ assert_eq!(
+ sliced_list_values(&list.slice(3, 1)).as_primitive(),
+ &Int32Array::from(vec![Some(6), Some(7)])
+ );
+
+ assert!(sliced_list_values(&list.slice(0, 0)).is_empty());
+ assert!(sliced_list_values(&list.slice(1, 0)).is_empty());
+ assert!(sliced_list_values(&list.slice(3, 0)).is_empty());
+ }
+
+ #[test]
+ fn test_adjust_offsets() {
+ let data = vec![
+ Some(vec![Some(0), Some(1), Some(2)]),
+ None,
+ Some(vec![Some(3), None, Some(5)]),
+ Some(vec![Some(6), Some(7)]),
+ ];
+ let list = ListArray::from_iter_primitive::<Int32Type, _, _>(data);
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list),
+ OffsetBuffer::from_lengths([3, 0, 3, 2])
+ );
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list.slice(0, 1)),
+ OffsetBuffer::from_lengths([3])
+ );
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list.slice(1, 2)),
+ OffsetBuffer::from_lengths([0, 3])
+ );
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list.slice(1, 3)),
+ OffsetBuffer::from_lengths([0, 3, 2])
+ );
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list.slice(0, 0)),
+ OffsetBuffer::from_lengths([])
+ );
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list.slice(1, 0)),
+ OffsetBuffer::from_lengths([])
+ );
+
+ assert_eq!(
+ adjust_offsets_for_slice(&list.slice(3, 0)),
+ OffsetBuffer::from_lengths([])
+ );
+ }
+
+ fn create_i32_list(
+ values: impl Into<Int32Array>,
+ offsets: OffsetBuffer<i32>,
+ nulls: Option<NullBuffer>,
+ ) -> ListArray {
+ let list_field = Arc::new(Field::new_list_field(DataType::Int32,
true));
+
+ ListArray::new(list_field, offsets, Arc::new(values.into()), nulls)
+ }
+
+ fn create_i32_fsl(
+ size: i32,
+ values: Vec<i32>,
+ nulls: Option<NullBuffer>,
+ ) -> FixedSizeListArray {
+ FixedSizeListArray::new(
+ Arc::new(Field::new_list_field(DataType::Int32, true)),
+ size,
+ Arc::new(Int32Array::from(values)),
+ nulls,
+ )
+ }
+
+ #[test]
+ fn test_remove_list_null_values_list() {
+ let list = Arc::new(create_i32_list(
+ vec![100, 20, 10, 0, 0, 0, 0, 1, 50],
+ OffsetBuffer::<i32>::from_lengths(vec![3, 4, 0, 2, 0]),
+ Some(NullBuffer::from(vec![true, false, false, true, false])),
+ )) as ArrayRef;
+
+ let res = remove_list_null_values(&list).unwrap();
+ let res = res.as_list::<i32>();
+
+ let expected = Arc::new(create_i32_list(
+ vec![100, 20, 10, 1, 50],
+ OffsetBuffer::<i32>::from_lengths(vec![3, 0, 0, 2, 0]),
+ Some(NullBuffer::from(vec![true, false, false, true, false])),
+ )) as ArrayRef;
+ let expected = expected.as_list::<i32>();
+
+ assert_eq!(res, expected);
+ // check above skips inner value of nulls
+ assert_eq!(res.values(), expected.values());
+ assert_eq!(res.offsets(), expected.offsets());
+ }
+
+ #[test]
+ fn test_remove_list_null_values_fsl() {
+ let list = Arc::new(create_i32_fsl(
+ 3,
+ vec![100, 20, 10, 0, 0, 0, 0, 0, 0],
+ Some(NullBuffer::from(vec![true, false, false])),
+ )) as ArrayRef;
+
+ let res = remove_list_null_values(&list).unwrap();
+
+ let expected = Arc::new(create_i32_fsl(
+ 3,
+ vec![100, 20, 10, 100, 20, 10, 100, 20, 10],
+ Some(NullBuffer::from(vec![true, false, false])),
+ )) as ArrayRef;
+
+ assert_eq!(&res, &expected);
+ // check above skips inner value of nulls
+ assert_eq!(
+ res.as_fixed_size_list().values(),
+ expected.as_fixed_size_list().values()
+ );
Review Comment:
Included at
https://github.com/apache/datafusion/commit/9a83018838cefda0a51bd7dd708e85dcc63c8b31#diff-433b400873e0bce183a502e2566420ec442cc2b1bc6ccc6af80e74fc4f25cad1R1596-R1677
Also changed the length of the list to differ from the sub-lists length as
your comment in
https://github.com/apache/datafusion/pull/21679#discussion_r3110232318
https://github.com/apache/datafusion/blob/9a83018838cefda0a51bd7dd708e85dcc63c8b31/datafusion/common/src/utils/mod.rs#L1594-L1677
--
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]