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/datafusion.git
The following commit(s) were added to refs/heads/main by this push:
new b9961c3001 Add native stringview support for LEFT (#11976)
b9961c3001 is described below
commit b9961c30014719818679e6bbce80ed4fa4b09f27
Author: kf zheng <[email protected]>
AuthorDate: Thu Aug 15 05:40:53 2024 +0800
Add native stringview support for LEFT (#11976)
* add string view support for left
* add tests for stringview support of LEFT
---
datafusion/functions/src/unicode/left.rs | 43 ++++++++++++++++++----
datafusion/sqllogictest/test_files/string_view.slt | 23 ++++++++++++
2 files changed, 58 insertions(+), 8 deletions(-)
diff --git a/datafusion/functions/src/unicode/left.rs
b/datafusion/functions/src/unicode/left.rs
index 7d456f5f1e..c49784948d 100644
--- a/datafusion/functions/src/unicode/left.rs
+++ b/datafusion/functions/src/unicode/left.rs
@@ -19,10 +19,15 @@ use std::any::Any;
use std::cmp::Ordering;
use std::sync::Arc;
-use arrow::array::{ArrayRef, GenericStringArray, OffsetSizeTrait};
+use arrow::array::{
+ Array, ArrayAccessor, ArrayIter, ArrayRef, GenericStringArray, Int64Array,
+ OffsetSizeTrait,
+};
use arrow::datatypes::DataType;
-use datafusion_common::cast::{as_generic_string_array, as_int64_array};
+use datafusion_common::cast::{
+ as_generic_string_array, as_int64_array, as_string_view_array,
+};
use datafusion_common::exec_err;
use datafusion_common::Result;
use datafusion_expr::TypeSignature::Exact;
@@ -46,7 +51,11 @@ impl LeftFunc {
use DataType::*;
Self {
signature: Signature::one_of(
- vec![Exact(vec![Utf8, Int64]), Exact(vec![LargeUtf8, Int64])],
+ vec![
+ Exact(vec![Utf8View, Int64]),
+ Exact(vec![Utf8, Int64]),
+ Exact(vec![LargeUtf8, Int64]),
+ ],
Volatility::Immutable,
),
}
@@ -72,9 +81,14 @@ impl ScalarUDFImpl for LeftFunc {
fn invoke(&self, args: &[ColumnarValue]) -> Result<ColumnarValue> {
match args[0].data_type() {
- DataType::Utf8 => make_scalar_function(left::<i32>, vec![])(args),
+ DataType::Utf8 | DataType::Utf8View => {
+ make_scalar_function(left::<i32>, vec![])(args)
+ }
DataType::LargeUtf8 => make_scalar_function(left::<i64>,
vec![])(args),
- other => exec_err!("Unsupported data type {other:?} for function
left"),
+ other => exec_err!(
+ "Unsupported data type {other:?} for function left,\
+ expected Utf8View, Utf8 or LargeUtf8."
+ ),
}
}
}
@@ -83,10 +97,23 @@ impl ScalarUDFImpl for LeftFunc {
/// left('abcde', 2) = 'ab'
/// The implementation uses UTF-8 code points as characters
pub fn left<T: OffsetSizeTrait>(args: &[ArrayRef]) -> Result<ArrayRef> {
- let string_array = as_generic_string_array::<T>(&args[0])?;
let n_array = as_int64_array(&args[1])?;
- let result = string_array
- .iter()
+
+ if args[0].data_type() == &DataType::Utf8View {
+ let string_array = as_string_view_array(&args[0])?;
+ left_impl::<T, _>(string_array, n_array)
+ } else {
+ let string_array = as_generic_string_array::<T>(&args[0])?;
+ left_impl::<T, _>(string_array, n_array)
+ }
+}
+
+fn left_impl<'a, T: OffsetSizeTrait, V: ArrayAccessor<Item = &'a str>>(
+ string_array: V,
+ n_array: &Int64Array,
+) -> Result<ArrayRef> {
+ let iter = ArrayIter::new(string_array);
+ let result = iter
.zip(n_array.iter())
.map(|(string, n)| match (string, n) {
(Some(string), Some(n)) => match n.cmp(&0) {
diff --git a/datafusion/sqllogictest/test_files/string_view.slt
b/datafusion/sqllogictest/test_files/string_view.slt
index 37ca89cf2d..e1d4a96620 100644
--- a/datafusion/sqllogictest/test_files/string_view.slt
+++ b/datafusion/sqllogictest/test_files/string_view.slt
@@ -917,6 +917,29 @@ eng (empty) ngpeng
ael (empty) hael
NULL NULL NULL
+## Ensure no casts for LEFT
+query TT
+EXPLAIN SELECT
+ LEFT(column1_utf8view, 3) as c2
+FROM test;
+----
+logical_plan
+01)Projection: left(test.column1_utf8view, Int64(3)) AS c2
+02)--TableScan: test projection=[column1_utf8view]
+
+# Test outputs of LEFT
+query TTT
+SELECT
+ LEFT(column1_utf8view, 3) as c1,
+ LEFT(column1_utf8view, 0) as c2,
+ LEFT(column1_utf8view, -3) as c3
+FROM test;
+----
+And (empty) And
+Xia (empty) Xiangp
+Rap (empty) Raph
+NULL NULL NULL
+
## Ensure no casts for RPAD
## TODO file ticket
query TT
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]