This is an automated email from the ASF dual-hosted git repository.
dheres pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow-datafusion.git
The following commit(s) were added to refs/heads/master by this push:
new 15ea3b9 Support date datatypes in max/min (#820)
15ea3b9 is described below
commit 15ea3b9e6ac485ec8b8604fb3409f19aaa295d5c
Author: Liang-Chi Hsieh <[email protected]>
AuthorDate: Tue Aug 3 00:02:31 2021 -0700
Support date datatypes in max/min (#820)
* Support date datatype in max/min.
* fix format.
---
datafusion/src/physical_plan/aggregates.rs | 3 +
.../src/physical_plan/expressions/min_max.rs | 70 ++++++++++++++++++++--
2 files changed, 69 insertions(+), 4 deletions(-)
diff --git a/datafusion/src/physical_plan/aggregates.rs
b/datafusion/src/physical_plan/aggregates.rs
index c297a95..57c9b61 100644
--- a/datafusion/src/physical_plan/aggregates.rs
+++ b/datafusion/src/physical_plan/aggregates.rs
@@ -188,6 +188,8 @@ static TIMESTAMPS: &[DataType] = &[
DataType::Timestamp(TimeUnit::Nanosecond, None),
];
+static DATES: &[DataType] = &[DataType::Date32, DataType::Date64];
+
/// the signatures supported by the function `fun`.
pub fn signature(fun: &AggregateFunction) -> Signature {
// note: the physical expression must accept the type returned by this
function or the execution panics.
@@ -198,6 +200,7 @@ pub fn signature(fun: &AggregateFunction) -> Signature {
.iter()
.chain(NUMERICS.iter())
.chain(TIMESTAMPS.iter())
+ .chain(DATES.iter())
.cloned()
.collect::<Vec<_>>();
Signature::Uniform(1, valid)
diff --git a/datafusion/src/physical_plan/expressions/min_max.rs
b/datafusion/src/physical_plan/expressions/min_max.rs
index 46e41f4..6bb4c5b 100644
--- a/datafusion/src/physical_plan/expressions/min_max.rs
+++ b/datafusion/src/physical_plan/expressions/min_max.rs
@@ -28,10 +28,10 @@ use arrow::compute;
use arrow::datatypes::{DataType, TimeUnit};
use arrow::{
array::{
- ArrayRef, Float32Array, Float64Array, Int16Array, Int32Array,
Int64Array,
- Int8Array, LargeStringArray, StringArray, TimestampMicrosecondArray,
- TimestampMillisecondArray, TimestampNanosecondArray,
TimestampSecondArray,
- UInt16Array, UInt32Array, UInt64Array, UInt8Array,
+ ArrayRef, Date32Array, Date64Array, Float32Array, Float64Array,
Int16Array,
+ Int32Array, Int64Array, Int8Array, LargeStringArray, StringArray,
+ TimestampMicrosecondArray, TimestampMillisecondArray,
TimestampNanosecondArray,
+ TimestampSecondArray, UInt16Array, UInt32Array, UInt64Array,
UInt8Array,
},
datatypes::Field,
};
@@ -158,6 +158,8 @@ macro_rules! min_max_batch {
TimestampNanosecond,
$OP
),
+ DataType::Date32 => typed_min_max_batch!($VALUES, Date32Array,
Date32, $OP),
+ DataType::Date64 => typed_min_max_batch!($VALUES, Date64Array,
Date64, $OP),
other => {
// This should have been handled before
return Err(DataFusionError::Internal(format!(
@@ -280,6 +282,18 @@ macro_rules! min_max {
) => {
typed_min_max!(lhs, rhs, TimestampNanosecond, $OP)
}
+ (
+ ScalarValue::Date32(lhs),
+ ScalarValue::Date32(rhs),
+ ) => {
+ typed_min_max!(lhs, rhs, Date32, $OP)
+ }
+ (
+ ScalarValue::Date64(lhs),
+ ScalarValue::Date64(rhs),
+ ) => {
+ typed_min_max!(lhs, rhs, Date64, $OP)
+ }
e => {
return Err(DataFusionError::Internal(format!(
"MIN/MAX is not expected to receive scalars of
incompatible types {:?}",
@@ -668,4 +682,52 @@ mod tests {
DataType::Float64
)
}
+
+ #[test]
+ fn min_date32() -> Result<()> {
+ let a: ArrayRef = Arc::new(Date32Array::from(vec![1, 2, 3, 4, 5]));
+ generic_test_op!(
+ a,
+ DataType::Date32,
+ Min,
+ ScalarValue::Date32(Some(1)),
+ DataType::Date32
+ )
+ }
+
+ #[test]
+ fn min_date64() -> Result<()> {
+ let a: ArrayRef = Arc::new(Date64Array::from(vec![1, 2, 3, 4, 5]));
+ generic_test_op!(
+ a,
+ DataType::Date64,
+ Min,
+ ScalarValue::Date64(Some(1)),
+ DataType::Date64
+ )
+ }
+
+ #[test]
+ fn max_date32() -> Result<()> {
+ let a: ArrayRef = Arc::new(Date32Array::from(vec![1, 2, 3, 4, 5]));
+ generic_test_op!(
+ a,
+ DataType::Date32,
+ Max,
+ ScalarValue::Date32(Some(5)),
+ DataType::Date32
+ )
+ }
+
+ #[test]
+ fn max_date64() -> Result<()> {
+ let a: ArrayRef = Arc::new(Date64Array::from(vec![1, 2, 3, 4, 5]));
+ generic_test_op!(
+ a,
+ DataType::Date64,
+ Max,
+ ScalarValue::Date64(Some(5)),
+ DataType::Date64
+ )
+ }
}
