This is an automated email from the ASF dual-hosted git repository.
github-bot 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 5e36b052e9 Refactor spark `bit_get()` signature away from user defined
(#18836)
5e36b052e9 is described below
commit 5e36b052e9613ad369517c00f7d7e8fd52f40cb5
Author: Jeffrey Vo <[email protected]>
AuthorDate: Sat Nov 22 12:34:40 2025 +1100
Refactor spark `bit_get()` signature away from user defined (#18836)
## Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases.
You can link an issue to this PR using the GitHub syntax. For example
`Closes #123` indicates that this PR will close issue #123.
-->
Part of https://github.com/apache/datafusion/issues/12725
## Rationale for this change
<!--
Why are you proposing this change? If this is already explained clearly
in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand
your changes and offer better suggestions for fixes.
-->
Prefer to avoid user_defined for consistency in function definitions.
## What changes are included in this PR?
<!--
There is no need to duplicate the description in the issue here but it
is sometimes worth providing a summary of the individual changes in this
PR.
-->
Refactor signature of bit_get away from user_defined.
Various other refactors.
## Are these changes tested?
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
If tests are not included in your PR, please explain why (for example,
are they covered by existing tests)?
-->
Existing tests.
## Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
-->
<!--
If there are any breaking changes to public APIs, please add the `api
change` label.
-->
There's a public function I made private but I don't think it was ever
intended to be public.
---
datafusion/spark/src/function/bitwise/bit_get.rs | 237 +++------------------
.../test_files/spark/bitwise/bit_get.slt | 25 +++
2 files changed, 57 insertions(+), 205 deletions(-)
diff --git a/datafusion/spark/src/function/bitwise/bit_get.rs
b/datafusion/spark/src/function/bitwise/bit_get.rs
index a8562618cb..bc8d8cdbd1 100644
--- a/datafusion/spark/src/function/bitwise/bit_get.rs
+++ b/datafusion/spark/src/function/bitwise/bit_get.rs
@@ -19,25 +19,21 @@ use std::any::Any;
use std::mem::size_of;
use std::sync::Arc;
-use arrow::array::{Array, ArrayRef, ArrowPrimitiveType, AsArray,
PrimitiveArray};
-use arrow::compute::try_binary;
-use arrow::datatypes::DataType::{
- Int16, Int32, Int64, Int8, UInt16, UInt32, UInt64, UInt8,
-};
-use arrow::datatypes::{
- ArrowNativeType, DataType, Int16Type, Int32Type, Int64Type, Int8Type,
UInt16Type,
- UInt32Type, UInt64Type, UInt8Type,
+use arrow::array::{
+ downcast_integer_array, Array, ArrayRef, ArrowPrimitiveType, AsArray,
Int32Array,
+ Int8Array, PrimitiveArray,
};
-use datafusion_common::{exec_err, Result};
+use arrow::compute::try_binary;
+use arrow::datatypes::{ArrowNativeType, DataType, Int32Type, Int8Type};
+use datafusion_common::types::{logical_int32, NativeType};
+use datafusion_common::utils::take_function_args;
+use datafusion_common::{internal_err, Result};
use datafusion_expr::{
- ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility,
+ Coercion, ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature,
+ TypeSignatureClass, Volatility,
};
use datafusion_functions::utils::make_scalar_function;
-use crate::function::error_utils::{
- invalid_arg_count_exec_err, unsupported_data_type_exec_err,
-};
-
#[derive(Debug, PartialEq, Eq, Hash)]
pub struct SparkBitGet {
signature: Signature,
@@ -53,7 +49,17 @@ impl Default for SparkBitGet {
impl SparkBitGet {
pub fn new() -> Self {
Self {
- signature: Signature::user_defined(Volatility::Immutable),
+ signature: Signature::coercible(
+ vec![
+ Coercion::new_exact(TypeSignatureClass::Integer),
+ Coercion::new_implicit(
+ TypeSignatureClass::Native(logical_int32()),
+ vec![TypeSignatureClass::Integer],
+ NativeType::Int32,
+ ),
+ ],
+ Volatility::Immutable,
+ ),
aliases: vec!["getbit".to_string()],
}
}
@@ -64,34 +70,6 @@ impl ScalarUDFImpl for SparkBitGet {
self
}
- fn coerce_types(&self, arg_types: &[DataType]) -> Result<Vec<DataType>> {
- if arg_types.len() != 2 {
- return Err(invalid_arg_count_exec_err(
- "bit_get",
- (2, 2),
- arg_types.len(),
- ));
- }
- if !arg_types[0].is_integer() && !arg_types[0].is_null() {
- return Err(unsupported_data_type_exec_err(
- "bit_get",
- "Integer Type",
- &arg_types[0],
- ));
- }
- if !arg_types[1].is_integer() && !arg_types[1].is_null() {
- return Err(unsupported_data_type_exec_err(
- "bit_get",
- "Integer Type",
- &arg_types[1],
- ));
- }
- if arg_types[0].is_null() {
- return Ok(vec![Int8, Int32]);
- }
- Ok(vec![arg_types[0].clone(), Int32])
- }
-
fn name(&self) -> &str {
"bit_get"
}
@@ -105,7 +83,7 @@ impl ScalarUDFImpl for SparkBitGet {
}
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
- Ok(Int8)
+ Ok(DataType::Int8)
}
fn invoke_with_args(&self, args: ScalarFunctionArgs) ->
Result<ColumnarValue> {
@@ -115,8 +93,8 @@ impl ScalarUDFImpl for SparkBitGet {
fn spark_bit_get_inner<T: ArrowPrimitiveType>(
value: &PrimitiveArray<T>,
- pos: &PrimitiveArray<Int32Type>,
-) -> Result<PrimitiveArray<Int8Type>> {
+ pos: &Int32Array,
+) -> Result<Int8Array> {
let bit_length = (size_of::<T::Native>() * 8) as i32;
let result: PrimitiveArray<Int8Type> = try_binary(value, pos, |value, pos|
{
@@ -130,164 +108,13 @@ fn spark_bit_get_inner<T: ArrowPrimitiveType>(
Ok(result)
}
-pub fn spark_bit_get(args: &[ArrayRef]) -> Result<ArrayRef> {
- if args.len() != 2 {
- return exec_err!("`bit_get` expects exactly two arguments");
- }
-
- if args[1].data_type() != &Int32 {
- return exec_err!("`bit_get` expects Int32 as the second argument");
- }
-
- let pos_arg = args[1].as_primitive::<Int32Type>();
-
- let ret = match &args[0].data_type() {
- Int64 => {
- let value_arg = args[0].as_primitive::<Int64Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- Int32 => {
- let value_arg = args[0].as_primitive::<Int32Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- Int16 => {
- let value_arg = args[0].as_primitive::<Int16Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- Int8 => {
- let value_arg = args[0].as_primitive::<Int8Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- UInt64 => {
- let value_arg = args[0].as_primitive::<UInt64Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- UInt32 => {
- let value_arg = args[0].as_primitive::<UInt32Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- UInt16 => {
- let value_arg = args[0].as_primitive::<UInt16Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- UInt8 => {
- let value_arg = args[0].as_primitive::<UInt8Type>();
- spark_bit_get_inner(value_arg, pos_arg)
- }
- _ => {
- exec_err!(
- "`bit_get` expects Int64, Int32, Int16, or Int8 as the first
argument"
- )
- }
- }?;
+fn spark_bit_get(args: &[ArrayRef]) -> Result<ArrayRef> {
+ let [value, position] = take_function_args("bit_get", args)?;
+ let pos_arg = position.as_primitive::<Int32Type>();
+ let ret = downcast_integer_array!(
+ value => spark_bit_get_inner(value, pos_arg),
+ DataType::Null => Ok(Int8Array::new_null(value.len())),
+ d => internal_err!("Unsupported datatype for bit_get: {d}"),
+ )?;
Ok(Arc::new(ret))
}
-
-#[cfg(test)]
-mod tests {
- use arrow::array::{Int32Array, Int64Array};
-
- use super::*;
-
- #[test]
- fn test_bit_get_basic() {
- // Test bit_get(11, 0) - 11 = 1011 in binary, bit 0 = 1
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![11])),
- Arc::new(Int32Array::from(vec![0])),
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 1);
-
- // Test bit_get(11, 2) - 11 = 1011 in binary, bit 2 = 0
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![11])),
- Arc::new(Int32Array::from(vec![2])),
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 0);
-
- // Test bit_get(11, 3) - 11 = 1011 in binary, bit 3 = 1
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![11])),
- Arc::new(Int32Array::from(vec![3])),
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 1);
- }
-
- #[test]
- fn test_bit_get_edge_cases() {
- // Test with 0
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![0])),
- Arc::new(Int32Array::from(vec![0])),
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 0);
-
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![11])),
- Arc::new(Int32Array::from(vec![-1])),
- ]);
- assert_eq!(
- result.unwrap_err().message().lines().next().unwrap(),
- "Compute error: bit_get: position -1 is out of bounds. Expected
pos < 64 and pos >= 0"
- );
-
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![11])),
- Arc::new(Int32Array::from(vec![64])),
- ]);
-
- assert_eq!(
- result.unwrap_err().message().lines().next().unwrap(),
- "Compute error: bit_get: position 64 is out of bounds. Expected
pos < 64 and pos >= 0"
- );
- }
-
- #[test]
- fn test_bit_get_null_inputs() {
- // Test with NULL value
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![None])),
- Arc::new(Int32Array::from(vec![0])),
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 0);
-
- // Test with NULL position
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![11])),
- Arc::new(Int32Array::from(vec![None])),
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 0);
- }
-
- #[test]
- fn test_bit_get_large_numbers() {
- // Test with larger number
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![255])), // 11111111 in binary
- Arc::new(Int32Array::from(vec![7])), // bit 7 = 1
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 1);
-
- let result = spark_bit_get(&[
- Arc::new(Int64Array::from(vec![255])), // 11111111 in binary
- Arc::new(Int32Array::from(vec![8])), // bit 8 = 0
- ])
- .unwrap();
-
- assert_eq!(result.as_primitive::<Int8Type>().value(0), 0);
- }
-}
diff --git a/datafusion/sqllogictest/test_files/spark/bitwise/bit_get.slt
b/datafusion/sqllogictest/test_files/spark/bitwise/bit_get.slt
index 6a2b244d58..faba0b66c4 100644
--- a/datafusion/sqllogictest/test_files/spark/bitwise/bit_get.slt
+++ b/datafusion/sqllogictest/test_files/spark/bitwise/bit_get.slt
@@ -69,7 +69,32 @@ SELECT bit_get(NULL, 0);
----
NULL
+query I
+SELECT bit_get(NULL::int, 0);
+----
+NULL
+
query I
SELECT bit_get(11, NULL);
----
NULL
+
+query I
+SELECT bit_get(11, NULL::int);
+----
+NULL
+
+query I
+SELECT bit_get(11::tinyint, 0);
+----
+1
+
+query I
+SELECT bit_get(11::bigint, 0);
+----
+1
+
+query I
+SELECT bit_get(11, 3::bigint);
+----
+1
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
