This is an automated email from the ASF dual-hosted git repository.
alamb pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow-datafusion.git
The following commit(s) were added to refs/heads/main by this push:
new d220bf47f9 support LargeList in array_positions (#8571)
d220bf47f9 is described below
commit d220bf47f944dd019d6b1e5b2741535a3f90204f
Author: Alex Huang <[email protected]>
AuthorDate: Mon Dec 18 21:22:34 2023 +0100
support LargeList in array_positions (#8571)
---
datafusion/physical-expr/src/array_expressions.rs | 19 +++++++---
datafusion/sqllogictest/test_files/array.slt | 43 +++++++++++++++++++++++
2 files changed, 58 insertions(+), 4 deletions(-)
diff --git a/datafusion/physical-expr/src/array_expressions.rs
b/datafusion/physical-expr/src/array_expressions.rs
index 98c9aee894..cc4b2899fc 100644
--- a/datafusion/physical-expr/src/array_expressions.rs
+++ b/datafusion/physical-expr/src/array_expressions.rs
@@ -1289,12 +1289,23 @@ fn general_position<OffsetSize: OffsetSizeTrait>(
/// Array_positions SQL function
pub fn array_positions(args: &[ArrayRef]) -> Result<ArrayRef> {
- let arr = as_list_array(&args[0])?;
let element = &args[1];
- check_datatypes("array_positions", &[arr.values(), element])?;
-
- general_positions::<i32>(arr, element)
+ match &args[0].data_type() {
+ DataType::List(_) => {
+ let arr = as_list_array(&args[0])?;
+ check_datatypes("array_positions", &[arr.values(), element])?;
+ general_positions::<i32>(arr, element)
+ }
+ DataType::LargeList(_) => {
+ let arr = as_large_list_array(&args[0])?;
+ check_datatypes("array_positions", &[arr.values(), element])?;
+ general_positions::<i64>(arr, element)
+ }
+ array_type => {
+ not_impl_err!("array_positions does not support type
'{array_type:?}'.")
+ }
+ }
}
fn general_positions<OffsetSize: OffsetSizeTrait>(
diff --git a/datafusion/sqllogictest/test_files/array.slt
b/datafusion/sqllogictest/test_files/array.slt
index 640f5064ea..d148f71181 100644
--- a/datafusion/sqllogictest/test_files/array.slt
+++ b/datafusion/sqllogictest/test_files/array.slt
@@ -1832,18 +1832,33 @@ select array_positions(['h', 'e', 'l', 'l', 'o'], 'l'),
array_positions([1, 2, 3
----
[3, 4] [5] [1, 2, 3]
+query ???
+select array_positions(arrow_cast(['h', 'e', 'l', 'l', 'o'],
'LargeList(Utf8)'), 'l'), array_positions(arrow_cast([1, 2, 3, 4, 5],
'LargeList(Int64)'), 5), array_positions(arrow_cast([1, 1, 1],
'LargeList(Int64)'), 1);
+----
+[3, 4] [5] [1, 2, 3]
+
# array_positions scalar function #2 (element is list)
query ?
select array_positions(make_array([1, 2, 3], [2, 1, 3], [1, 5, 6], [2, 1, 3],
[4, 5, 6]), [2, 1, 3]);
----
[2, 4]
+query ?
+select array_positions(arrow_cast(make_array([1, 2, 3], [2, 1, 3], [1, 5, 6],
[2, 1, 3], [4, 5, 6]), 'LargeList(List(Int64))'), [2, 1, 3]);
+----
+[2, 4]
+
# list_positions scalar function #3 (function alias `array_positions`)
query ???
select list_positions(['h', 'e', 'l', 'l', 'o'], 'l'), list_positions([1, 2,
3, 4, 5], 5), list_positions([1, 1, 1], 1);
----
[3, 4] [5] [1, 2, 3]
+query ???
+select list_positions(arrow_cast(['h', 'e', 'l', 'l', 'o'],
'LargeList(Utf8)'), 'l'), list_positions(arrow_cast([1, 2, 3, 4, 5],
'LargeList(Int64)'), 5), list_positions(arrow_cast([1, 1, 1],
'LargeList(Int64)'), 1);
+----
+[3, 4] [5] [1, 2, 3]
+
# array_positions with columns #1
query ?
select array_positions(column1, column2) from arrays_values_without_nulls;
@@ -1853,6 +1868,14 @@ select array_positions(column1, column2) from
arrays_values_without_nulls;
[3]
[4]
+query ?
+select array_positions(arrow_cast(column1, 'LargeList(Int64)'), column2) from
arrays_values_without_nulls;
+----
+[1]
+[2]
+[3]
+[4]
+
# array_positions with columns #2 (element is list)
query ?
select array_positions(column1, column2) from nested_arrays;
@@ -1860,6 +1883,12 @@ select array_positions(column1, column2) from
nested_arrays;
[3]
[2, 5]
+query ?
+select array_positions(arrow_cast(column1, 'LargeList(List(Int64))'), column2)
from nested_arrays;
+----
+[3]
+[2, 5]
+
# array_positions with columns and scalars #1
query ??
select array_positions(column1, 4), array_positions(array[1, 2, 23, 13, 33,
45], column2) from arrays_values_without_nulls;
@@ -1869,6 +1898,14 @@ select array_positions(column1, 4),
array_positions(array[1, 2, 23, 13, 33, 45],
[] [3]
[] []
+query ??
+select array_positions(arrow_cast(column1, 'LargeList(Int64)'), 4),
array_positions(array[1, 2, 23, 13, 33, 45], column2) from
arrays_values_without_nulls;
+----
+[4] [1]
+[] []
+[] [3]
+[] []
+
# array_positions with columns and scalars #2 (element is list)
query ??
select array_positions(column1, make_array(4, 5, 6)),
array_positions(make_array([1, 2, 3], [11, 12, 13], [4, 5, 6]), column2) from
nested_arrays;
@@ -1876,6 +1913,12 @@ select array_positions(column1, make_array(4, 5, 6)),
array_positions(make_array
[6] []
[1] []
+query ??
+select array_positions(arrow_cast(column1, 'LargeList(List(Int64))'),
make_array(4, 5, 6)), array_positions(arrow_cast(make_array([1, 2, 3], [11, 12,
13], [4, 5, 6]), 'LargeList(List(Int64))'), column2) from nested_arrays;
+----
+[6] []
+[1] []
+
## array_replace (aliases: `list_replace`)
# array_replace scalar function #1
