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 35c1cfdd15 Introduce `TypeSignatureClass::Binary` to allow accepting
arbitrarily sized `FixedSizeBinary` arguments (#17531)
35c1cfdd15 is described below
commit 35c1cfdd151b14bbba87dbdf03d49afd9f4718a6
Author: Jeffrey Vo <[email protected]>
AuthorDate: Fri Sep 19 23:48:50 2025 +1000
Introduce `TypeSignatureClass::Binary` to allow accepting arbitrarily sized
`FixedSizeBinary` arguments (#17531)
* Introduce wildcard const for FixedSizeBinary type signature
* Add Binary to TypeSignatureClass
* Remove FIXED_SIZE_BINARY_WILDCARD
---
datafusion/common/src/types/native.rs | 5 +++++
datafusion/expr-common/src/signature.rs | 9 +++++++++
.../spark/src/function/bitmap/bitmap_count.rs | 22 +++++++++-------------
3 files changed, 23 insertions(+), 13 deletions(-)
diff --git a/datafusion/common/src/types/native.rs
b/datafusion/common/src/types/native.rs
index 223e04ba99..76ff9bd095 100644
--- a/datafusion/common/src/types/native.rs
+++ b/datafusion/common/src/types/native.rs
@@ -472,4 +472,9 @@ impl NativeType {
pub fn is_duration(&self) -> bool {
matches!(self, NativeType::Duration(_))
}
+
+ #[inline]
+ pub fn is_binary(&self) -> bool {
+ matches!(self, NativeType::Binary | NativeType::FixedSizeBinary(_))
+ }
}
diff --git a/datafusion/expr-common/src/signature.rs
b/datafusion/expr-common/src/signature.rs
index 6820b933f2..61585bd945 100644
--- a/datafusion/expr-common/src/signature.rs
+++ b/datafusion/expr-common/src/signature.rs
@@ -266,6 +266,8 @@ pub enum TypeSignatureClass {
// TODO:
// Numeric
Integer,
+ /// Encompasses both the native Binary as well as arbitrarily sized
FixedSizeBinary types
+ Binary,
}
impl Display for TypeSignatureClass {
@@ -303,6 +305,9 @@ impl TypeSignatureClass {
TypeSignatureClass::Integer => {
vec![DataType::Int64]
}
+ TypeSignatureClass::Binary => {
+ vec![DataType::Binary]
+ }
}
}
@@ -322,6 +327,7 @@ impl TypeSignatureClass {
TypeSignatureClass::Interval if logical_type.is_interval() => true,
TypeSignatureClass::Duration if logical_type.is_duration() => true,
TypeSignatureClass::Integer if logical_type.is_integer() => true,
+ TypeSignatureClass::Binary if logical_type.is_binary() => true,
_ => false,
}
}
@@ -352,6 +358,9 @@ impl TypeSignatureClass {
TypeSignatureClass::Integer if native_type.is_integer() => {
Ok(origin_type.to_owned())
}
+ TypeSignatureClass::Binary if native_type.is_binary() => {
+ Ok(origin_type.to_owned())
+ }
_ => internal_err!("May miss the matching logic in
`matches_native_type`"),
}
}
diff --git a/datafusion/spark/src/function/bitmap/bitmap_count.rs
b/datafusion/spark/src/function/bitmap/bitmap_count.rs
index e9d24585d7..e9eb751035 100644
--- a/datafusion/spark/src/function/bitmap/bitmap_count.rs
+++ b/datafusion/spark/src/function/bitmap/bitmap_count.rs
@@ -27,9 +27,10 @@ use arrow::datatypes::DataType::{
Binary, BinaryView, FixedSizeBinary, Int64, LargeBinary,
};
use datafusion_common::utils::take_function_args;
-use datafusion_common::{internal_datafusion_err, internal_err, plan_err,
Result};
+use datafusion_common::{internal_datafusion_err, 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 datafusion_functions::{downcast_arg, downcast_named_arg};
@@ -48,8 +49,10 @@ impl Default for BitmapCount {
impl BitmapCount {
pub fn new() -> Self {
Self {
- // TODO: add definitive TypeSignature after
https://github.com/apache/datafusion/issues/17291 is done
- signature: Signature::any(1, Volatility::Immutable),
+ signature: Signature::coercible(
+ vec![Coercion::new_exact(TypeSignatureClass::Binary)],
+ Volatility::Immutable,
+ ),
}
}
}
@@ -67,15 +70,8 @@ impl ScalarUDFImpl for BitmapCount {
&self.signature
}
- fn return_type(&self, arg_types: &[DataType]) -> Result<DataType> {
- match arg_types.first() {
- Some(Binary | BinaryView | FixedSizeBinary(_) | LargeBinary) =>
Ok(Int64),
- Some(data_type) => plan_err!(
- "bitmap_count expects
Binary/BinaryView/FixedSizeBinary/LargeBinary as argument, got {:?}",
- data_type
- ),
- None => internal_err!("bitmap_count does not support zero
arguments"),
- }
+ fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
+ Ok(Int64)
}
fn invoke_with_args(&self, args: ScalarFunctionArgs) ->
Result<ColumnarValue> {
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
