You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@datafusion.apache.org by gi...@apache.org on 2026/10/03 17:06:08 UTC
(datafusion-comet) branch main updated (432523fb59 -> f980d6fb59)
This is an automated email from the ASF dual-hosted git repository.
github-merge-queue[bot] pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/datafusion-comet.git
from 432523fb59 fix: hash NaN the way Spark does in hash, xxhash64 and native shuffle (#6413)
new 2f7e1d3be1 fix: return -0.0 for signum(-0.0), as Spark does (#6556)
new f980d6fb59 fix: validate release build ref before building (#6571)
The 2 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
dev/release/build-release-comet.sh | 27 ++++-
.../expression-audits/math_funcs.md | 1 +
docs/source/contributor-guide/release_process.md | 45 ++++----
native/spark-expr/src/comet_scalar_funcs.rs | 5 +
native/spark-expr/src/lib.rs | 6 +-
native/spark-expr/src/math_funcs/mod.rs | 2 +
native/spark-expr/src/math_funcs/signum.rs | 115 +++++++++++++++++++++
.../org/apache/comet/serde/QueryPlanSerde.scala | 2 +-
.../main/scala/org/apache/comet/serde/math.scala | 25 ++++-
.../sql-tests/expressions/math/signum.sql | 28 ++++-
.../apache/comet/CometFloatSemanticsSuite.scala | 6 +-
11 files changed, 227 insertions(+), 35 deletions(-)
create mode 100644 native/spark-expr/src/math_funcs/signum.rs
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@datafusion.apache.org
For additional commands, e-mail: commits-help@datafusion.apache.org
(datafusion-comet) 02/02: fix: validate release build ref before building (#6571)
Posted by gi...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
github-merge-queue[bot] pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/datafusion-comet.git
commit f980d6fb59f51ae20c0a6916eec9771ace8e50d1
Author: Andy Grove <ag...@apache.org>
AuthorDate: Sat Oct 3 16:40:18 2026 +0000
fix: validate release build ref before building (#6571)
---
dev/release/build-release-comet.sh | 27 ++++++++++++--
docs/source/contributor-guide/release_process.md | 45 +++++++++++++-----------
2 files changed, 48 insertions(+), 24 deletions(-)
diff --git a/dev/release/build-release-comet.sh b/dev/release/build-release-comet.sh
index 495b2ece0a..4b061a4075 100755
--- a/dev/release/build-release-comet.sh
+++ b/dev/release/build-release-comet.sh
@@ -34,7 +34,7 @@ This script builds comet native binaries inside a docker image. The image is nam
Options are:
-r [repo] : git repo (default: ${REPO})
- -b [branch] : git branch (default: ${BRANCH})
+ -b [ref] : git branch or tag to build (required)
-t [tag] : tag for the spark-rm docker image to use for building (default: "latest").
EOF
exit 1
@@ -59,10 +59,10 @@ function cleanup()
trap cleanup SIGINT SIGTERM EXIT
-CLEANUP=1
+CLEANUP=0
REPO="https://github.com/apache/datafusion-comet.git"
-BRANCH="release"
+BRANCH=
MACOS_SDK=
HAS_MACOS_SDK="false"
IMGTAG=latest
@@ -77,6 +77,26 @@ while getopts "b:hr:t:" opt; do
esac
done
+if [ -z "$BRANCH" ]
+then
+ echo "Error: -b must specify the branch or tag to build"
+ usage
+fi
+
+if git ls-remote --exit-code "$REPO" "refs/heads/$BRANCH" "refs/tags/$BRANCH" >/dev/null
+then
+ echo "Found branch or tag $BRANCH in $REPO"
+else
+ STATUS=$?
+ if [ "$STATUS" -eq 2 ]
+ then
+ echo "Error: branch or tag '$BRANCH' does not exist in $REPO"
+ else
+ echo "Error: unable to check branch or tag '$BRANCH' in $REPO"
+ fi
+ exit 1
+fi
+
echo "Building binaries from $REPO/$BRANCH"
# Check Java version
@@ -135,6 +155,7 @@ pushd $COMET_HOME_DIR/native && cargo clean && popd
# Run the builder container for each architecture. The entrypoint script will build the binaries
# AMD64
+CLEANUP=1
echo "Building amd64 binary"
docker run \
--name comet-amd64-builder-container \
diff --git a/docs/source/contributor-guide/release_process.md b/docs/source/contributor-guide/release_process.md
index 9b0571bc75..cd3ee4b6d8 100644
--- a/docs/source/contributor-guide/release_process.md
+++ b/docs/source/contributor-guide/release_process.md
@@ -63,8 +63,8 @@ instructions on each step.
- [ ] Check for fixes missing from the release branches
- [ ] Generate the change log and PR it against the release branch
- [ ] Run the full CI suite on the release branch
- - [ ] Build the jars
- [ ] Tag the release candidate
+ - [ ] Build the jars from the release candidate tag
- [ ] Update documentation for the new release
- [ ] Publish Maven artifacts to staging
- [ ] Create the release candidate tarball
@@ -327,6 +327,21 @@ done
Both runs must be green at the commit you are about to tag. If anything merges to the release branch after the
runs start, run them again. Repeat this for every release candidate.
+### Tag the Release Candidate
+
+Ensure that the Maven version update and change log have been merged to the release branch, and that the
+[full CI run](#run-the-full-ci-suite) is green at the commit you are tagging, before tagging.
+
+Tag the release branch with `0.13.0-rc1` and push to the `apache` repo:
+
+```shell
+git fetch apache
+git checkout branch-0.13
+git reset --hard apache/branch-0.13
+git tag 0.13.0-rc1
+git push apache 0.13.0-rc1
+```
+
### Build the jars
#### A note on workspace cleanliness
@@ -363,8 +378,10 @@ $ docker context use desktop-linux
The `build-release-comet.sh` script will create a docker image for each architecture and use the image
to build the platform specific binaries. These builder images are created every time this script is run.
-The script optionally allows overriding of the repository and branch to build the binaries from (Note that
-the local git repo is not used in the building of the binaries, but it is used to build the final uber jar).
+The branch or tag to build is required, and the script optionally allows overriding the repository. For a release,
+always build from the release candidate tag created in the previous step. The remote tag is used to build the native
+binaries, while the local checkout is used to build the final uber jar, so check out the same tag locally before
+running the script.
```shell
Usage: build-release-comet.sh [options]
@@ -375,36 +392,22 @@ This script builds comet native binaries inside a docker image. The image is nam
Options are:
-r [repo] : git repo (default: https://github.com/apache/datafusion-comet.git)
--b [branch] : git branch (default: release)
+-b [ref] : git branch or tag to build (required)
-t [tag] : tag for the spark-rm docker image to use for building (default: "latest").
```
Example:
```shell
-cd dev/release && ./build-release-comet.sh && cd ../..
+git checkout 0.13.0-rc1
+cd dev/release && ./build-release-comet.sh -b 0.13.0-rc1 && cd ../..
```
#### Build output
The build output is installed to a temporary local maven repository. The build script will print the name of the
repository location at the end. This location will be required at the time of deploying the artifacts to a staging
-repository
-
-### Tag the Release Candidate
-
-Ensure that the Maven version update and change log have been merged to the release branch, and that the
-[full CI run](#run-the-full-ci-suite) is green at the commit you are tagging, before tagging.
-
-Tag the release branch with `0.13.0-rc1` and push to the `apache` repo
-
-```shell
-git fetch apache
-git checkout branch-0.13
-git reset --hard apache/branch-0.13
-git tag 0.13.0-rc1
-git push apache 0.13.0-rc1
-```
+repository.
### Publishing Documentation
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@datafusion.apache.org
For additional commands, e-mail: commits-help@datafusion.apache.org
(datafusion-comet) 01/02: fix: return -0.0 for signum(-0.0), as Spark does (#6556)
Posted by gi...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
github-merge-queue[bot] pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/datafusion-comet.git
commit 2f7e1d3be1e31b5dca3e73ddde9c5cef6da63973
Author: Feng Zhang <fe...@apache.org>
AuthorDate: Sat Oct 3 15:58:21 2026 +0000
fix: return -0.0 for signum(-0.0), as Spark does (#6556)
* fix: return -0.0 for signum(-0.0), as Spark does
DataFusion's signum returns 0.0 for both zeros, where Math.signum returns
the zero it is given. Use a Comet kernel instead. Interval inputs, which
failed at execution, now fall back.
Closes #6522
* add -0.0 literal to the query example
---
.../expression-audits/math_funcs.md | 1 +
native/spark-expr/src/comet_scalar_funcs.rs | 5 +
native/spark-expr/src/lib.rs | 6 +-
native/spark-expr/src/math_funcs/mod.rs | 2 +
native/spark-expr/src/math_funcs/signum.rs | 115 +++++++++++++++++++++
.../org/apache/comet/serde/QueryPlanSerde.scala | 2 +-
.../main/scala/org/apache/comet/serde/math.scala | 25 ++++-
.../sql-tests/expressions/math/signum.sql | 28 ++++-
.../apache/comet/CometFloatSemanticsSuite.scala | 6 +-
9 files changed, 179 insertions(+), 11 deletions(-)
diff --git a/docs/source/contributor-guide/expression-audits/math_funcs.md b/docs/source/contributor-guide/expression-audits/math_funcs.md
index 984a0468fd..2810abac90 100644
--- a/docs/source/contributor-guide/expression-audits/math_funcs.md
+++ b/docs/source/contributor-guide/expression-audits/math_funcs.md
@@ -234,6 +234,7 @@ Internal fused expression that replaces the `CheckOverflow(Cast(expr, Decimal128
## signum
- Spark 3.4.3, 3.5.8, 4.0.1, 4.1.1 (audited 2026-05-27): `Signum(child)` over `DoubleType`. Spark also restricts to the two interval types via `inputTypes`; Comet handles only the `Double` case via DataFusion `signum`.
+- Spark 3.4, 3.5, 4.0, 4.1, 4.2 (2026-10-02, [#6522](https://github.com/apache/datafusion-comet/issues/6522)): `Math.signum` returns the zero it is given, so `signum(-0.0)` is `-0.0`, where DataFusion's `signum` returns `0.0`. Comet now uses its own `spark_signum` kernel for the `Double` case. Interval inputs fall back; DataFusion's `signum` failed on them at execution.
## sin
diff --git a/native/spark-expr/src/comet_scalar_funcs.rs b/native/spark-expr/src/comet_scalar_funcs.rs
index 7b3c980639..4ed5e561f7 100644
--- a/native/spark-expr/src/comet_scalar_funcs.rs
+++ b/native/spark-expr/src/comet_scalar_funcs.rs
@@ -25,6 +25,7 @@ use crate::math_funcs::checked_arithmetic::{checked_add, checked_div, checked_mu
use crate::math_funcs::log::spark_log;
use crate::math_funcs::modulo_expr::spark_modulo;
use crate::math_funcs::pow::spark_pow;
+use crate::math_funcs::signum::spark_signum;
use crate::math_funcs::sqrt::spark_sqrt;
use crate::{
spark_ceil, spark_day_name, spark_decimal_div, spark_decimal_integral_div, spark_floor,
@@ -224,6 +225,10 @@ pub fn create_comet_physical_fun_with_eval_mode(
let func = Arc::new(spark_pow);
make_comet_scalar_udf!("pow", func, without data_type)
}
+ "spark_signum" => {
+ let func = Arc::new(spark_signum);
+ make_comet_scalar_udf!("spark_signum", func, without data_type)
+ }
"spark_sqrt" => {
let func = Arc::new(spark_sqrt);
make_comet_scalar_udf!("spark_sqrt", func, without data_type)
diff --git a/native/spark-expr/src/lib.rs b/native/spark-expr/src/lib.rs
index 991fd5e52e..b1447e37d6 100644
--- a/native/spark-expr/src/lib.rs
+++ b/native/spark-expr/src/lib.rs
@@ -97,9 +97,9 @@ pub use json_funcs::{FromJson, JsonArrayLength, ToJson};
pub use math_funcs::{
abs, checked_add, checked_div, checked_mul, checked_sub, create_modulo_expr,
create_negate_expr, spark_ceil, spark_decimal_div, spark_decimal_integral_div, spark_floor,
- spark_log, spark_make_decimal, spark_modulo, spark_pow, spark_round, spark_sqrt, spark_unhex,
- spark_unscaled_value, CheckOverflow, CheckedBinaryExpr, DecimalRescaleCheckOverflow,
- NegativeExpr, WideDecimalBinaryExpr, WideDecimalOp,
+ spark_log, spark_make_decimal, spark_modulo, spark_pow, spark_round, spark_signum, spark_sqrt,
+ spark_unhex, spark_unscaled_value, CheckOverflow, CheckedBinaryExpr,
+ DecimalRescaleCheckOverflow, NegativeExpr, WideDecimalBinaryExpr, WideDecimalOp,
};
pub use query_context::{create_query_context_map, QueryContext, QueryContextMap};
pub use string_funcs::*;
diff --git a/native/spark-expr/src/math_funcs/mod.rs b/native/spark-expr/src/math_funcs/mod.rs
index 1c993cd2fe..9b72966dc6 100644
--- a/native/spark-expr/src/math_funcs/mod.rs
+++ b/native/spark-expr/src/math_funcs/mod.rs
@@ -27,6 +27,7 @@ pub mod modulo_expr;
mod negative;
pub(crate) mod pow;
mod round;
+pub(crate) mod signum;
pub(crate) mod sqrt;
pub(crate) mod unhex;
mod utils;
@@ -45,6 +46,7 @@ pub use modulo_expr::{create_modulo_expr, spark_modulo};
pub use negative::{create_negate_expr, NegativeExpr};
pub use pow::spark_pow;
pub use round::spark_round;
+pub use signum::spark_signum;
pub use sqrt::spark_sqrt;
pub use unhex::spark_unhex;
pub use wide_decimal_binary_expr::{WideDecimalBinaryExpr, WideDecimalOp};
diff --git a/native/spark-expr/src/math_funcs/signum.rs b/native/spark-expr/src/math_funcs/signum.rs
new file mode 100644
index 0000000000..3f5670e4d0
--- /dev/null
+++ b/native/spark-expr/src/math_funcs/signum.rs
@@ -0,0 +1,115 @@
+// Licensed to the Apache Software Foundation (ASF) under one
+// or more contributor license agreements. See the NOTICE file
+// distributed with this work for additional information
+// regarding copyright ownership. The ASF licenses this file
+// to you under the Apache License, Version 2.0 (the
+// "License"); you may not use this file except in compliance
+// with the License. You may obtain a copy of the License at
+//
+// http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing,
+// software distributed under the License is distributed on an
+// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied. See the License for the
+// specific language governing permissions and limitations
+// under the License.
+
+use arrow::array::{ArrayRef, AsArray, Float64Array};
+use arrow::datatypes::Float64Type;
+use datafusion::common::{DataFusionError, ScalarValue};
+use datafusion::physical_plan::ColumnarValue;
+use std::sync::Arc;
+
+/// Spark-compatible `signum`, matching `java.lang.Math.signum`: a zero is returned as it is, so
+/// `-0.0` keeps its sign. DataFusion's own `signum` returns `0.0` for both zeros.
+pub fn spark_signum(args: &[ColumnarValue]) -> Result<ColumnarValue, DataFusionError> {
+ if args.len() != 1 {
+ return Err(DataFusionError::Internal(format!(
+ "spark_signum requires 1 argument, got {}",
+ args.len()
+ )));
+ }
+
+ match &args[0] {
+ ColumnarValue::Array(array) => {
+ let values = array.as_primitive_opt::<Float64Type>().ok_or_else(|| {
+ DataFusionError::Internal(format!(
+ "spark_signum expected Float64, got {:?}",
+ array.data_type()
+ ))
+ })?;
+ let result: Float64Array = values.unary(signum);
+ Ok(ColumnarValue::Array(Arc::new(result) as ArrayRef))
+ }
+ ColumnarValue::Scalar(ScalarValue::Float64(v)) => {
+ Ok(ColumnarValue::Scalar(ScalarValue::Float64(v.map(signum))))
+ }
+ ColumnarValue::Scalar(other) => Err(DataFusionError::Internal(format!(
+ "spark_signum expected Float64 scalar, got {other:?}",
+ ))),
+ }
+}
+
+/// `f64::signum` returns `1.0` and `-1.0` for the two zeros, so a zero is passed through instead.
+/// NaN is not equal to zero, and `f64::signum` returns NaN for it.
+#[inline]
+fn signum(v: f64) -> f64 {
+ if v == 0.0 {
+ v
+ } else {
+ v.signum()
+ }
+}
+
+#[cfg(test)]
+mod test {
+ use super::*;
+ use arrow::array::Array;
+
+ #[test]
+ fn test_spark_signum_keeps_the_sign_of_zero() {
+ let input = Float64Array::from(vec![
+ Some(5.5),
+ Some(-5.5),
+ Some(0.0),
+ Some(-0.0),
+ Some(f64::NAN),
+ Some(f64::INFINITY),
+ Some(f64::NEG_INFINITY),
+ None,
+ ]);
+ let result = spark_signum(&[ColumnarValue::Array(Arc::new(input))]).unwrap();
+ let ColumnarValue::Array(result) = result else {
+ unreachable!()
+ };
+ let result = result.as_primitive::<Float64Type>();
+ assert_eq!(result.value(0), 1.0);
+ assert_eq!(result.value(1), -1.0);
+ // `0.0 == -0.0`, so the zeros are compared by their bits.
+ assert_eq!(result.value(2).to_bits(), 0.0_f64.to_bits());
+ assert_eq!(result.value(3).to_bits(), (-0.0_f64).to_bits());
+ assert!(result.value(4).is_nan());
+ assert_eq!(result.value(5), 1.0);
+ assert_eq!(result.value(6), -1.0);
+ assert!(result.is_null(7));
+ }
+
+ #[test]
+ fn test_spark_signum_scalar_negative_zero() {
+ let result =
+ spark_signum(&[ColumnarValue::Scalar(ScalarValue::Float64(Some(-0.0)))]).unwrap();
+ let ColumnarValue::Scalar(ScalarValue::Float64(Some(result))) = result else {
+ unreachable!()
+ };
+ assert_eq!(result.to_bits(), (-0.0_f64).to_bits());
+ }
+
+ #[test]
+ fn test_spark_signum_scalar_null() {
+ let result = spark_signum(&[ColumnarValue::Scalar(ScalarValue::Float64(None))]).unwrap();
+ let ColumnarValue::Scalar(ScalarValue::Float64(None)) = result else {
+ unreachable!()
+ };
+ }
+}
diff --git a/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala b/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala
index b41b37a011..bcdb37122b 100644
--- a/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala
+++ b/spark/src/main/scala/org/apache/comet/serde/QueryPlanSerde.scala
@@ -151,7 +151,7 @@ object QueryPlanSerde extends Logging with CometExprShim with CometTypeShim {
classOf[Rint] -> CometScalarFunction("rint"),
classOf[Round] -> CometRound,
classOf[Sec] -> CometScalarFunction("sec"),
- classOf[Signum] -> CometScalarFunction("signum"),
+ classOf[Signum] -> CometSignum,
classOf[Sin] -> CometScalarFunction("sin"),
classOf[Sinh] -> CometScalarFunction("sinh"),
classOf[Sqrt] -> CometSqrt,
diff --git a/spark/src/main/scala/org/apache/comet/serde/math.scala b/spark/src/main/scala/org/apache/comet/serde/math.scala
index ed0efb1637..e1d305cadf 100644
--- a/spark/src/main/scala/org/apache/comet/serde/math.scala
+++ b/spark/src/main/scala/org/apache/comet/serde/math.scala
@@ -19,7 +19,7 @@
package org.apache.comet.serde
-import org.apache.spark.sql.catalyst.expressions.{Abs, Add, Atan2, Attribute, BRound, Ceil, CheckOverflow, Conv, Expression, Floor, Greatest, Hex, Hypot, If, Least, LessThanOrEqual, Literal, Log, Log10, Log1p, Log2, Logarithm, NaNvl, Pmod, Pow, Sqrt, UnaryPositive, Unhex, WidthBucket}
+import org.apache.spark.sql.catalyst.expressions.{Abs, Add, Atan2, Attribute, BRound, Ceil, CheckOverflow, Conv, Expression, Floor, Greatest, Hex, Hypot, If, Least, LessThanOrEqual, Literal, Log, Log10, Log1p, Log2, Logarithm, NaNvl, Pmod, Pow, Signum, Sqrt, UnaryPositive, Unhex, WidthBucket}
import org.apache.spark.sql.types.{DecimalType, DoubleType, NumericType}
import org.apache.comet.serde.QueryPlanSerde.{exprToProtoInternal, scalarFunctionExprToProto, scalarFunctionExprToProtoWithReturnType, serializeDataType}
@@ -239,6 +239,29 @@ object CometSqrt extends CometExpressionSerde[Sqrt] {
}
}
+// Uses a custom spark_signum UDF because DataFusion's own `signum` returns 0.0 for both zeros,
+// while Spark's Signum (java.lang.Math.signum) returns the zero it is given, so -0.0 keeps its
+// sign. The kernel takes doubles only. Spark's Signum also accepts the two interval types, which
+// DataFusion's `signum` failed on at execution, so those fall back. As with CometSqrt, the
+// Comet-only name needs its return type set explicitly.
+object CometSignum extends CometExpressionSerde[Signum] {
+
+ override def getUnsupportedReasons(): Seq[String] = Seq("Only `DoubleType` is supported")
+
+ override def getSupportLevel(expr: Signum): SupportLevel = expr.child.dataType match {
+ case DoubleType => Compatible()
+ case other => Unsupported(Some(s"signum does not support input type $other"))
+ }
+
+ override def convert(
+ expr: Signum,
+ inputs: Seq[Attribute],
+ binding: Boolean): Option[ExprOuterClass.Expr] = {
+ val childExpr = exprToProtoInternal(expr.child, inputs, binding)
+ scalarFunctionExprToProtoWithReturnType("spark_signum", DoubleType, false, childExpr)
+ }
+}
+
sealed trait MathExprBase {
protected def nullIfNegative(expression: Expression): Expression = {
val zero = Literal.default(expression.dataType)
diff --git a/spark/src/test/resources/sql-tests/expressions/math/signum.sql b/spark/src/test/resources/sql-tests/expressions/math/signum.sql
index 04db8c3461..c8f40887c9 100644
--- a/spark/src/test/resources/sql-tests/expressions/math/signum.sql
+++ b/spark/src/test/resources/sql-tests/expressions/math/signum.sql
@@ -26,4 +26,30 @@ SELECT signum(d) FROM test_signum
-- literal arguments
query
-SELECT signum(-5.0), signum(5.0), signum(0.0), signum(NULL)
+SELECT signum(-5.0), signum(5.0), signum(0.0), signum(-0.0), signum(NULL)
+
+-- java.lang.Math.signum returns the zero it is given, so -0.0 keeps its sign
+-- (https://github.com/apache/datafusion-comet/issues/6522)
+statement
+CREATE TABLE test_signum_zero(d double, f float) USING parquet
+
+statement
+INSERT INTO test_signum_zero VALUES (double('-0.0'), float('-0.0')), (double('0.0'), float('0.0')), (-1.5, float('-1.5')), (2.5, float('2.5')), (NULL, NULL)
+
+query
+SELECT signum(d), signum(f) FROM test_signum_zero
+
+-- the string form shows the sign of a zero
+query
+SELECT CAST(signum(d) AS STRING), CAST(signum(f) AS STRING) FROM test_signum_zero
+
+-- Spark's Signum also accepts the two interval types. The native kernel takes doubles only, so
+-- these fall back.
+statement
+CREATE TABLE test_signum_iv(m int) USING parquet
+
+statement
+INSERT INTO test_signum_iv VALUES (18), (-18), (0), (NULL)
+
+query expect_fallback(signum does not support input type)
+SELECT signum(make_ym_interval(0, m)), signum(make_dt_interval(m, 0, 0, 0)) FROM test_signum_iv
diff --git a/spark/src/test/scala/org/apache/comet/CometFloatSemanticsSuite.scala b/spark/src/test/scala/org/apache/comet/CometFloatSemanticsSuite.scala
index 89c7d6ee81..8715e06cc8 100644
--- a/spark/src/test/scala/org/apache/comet/CometFloatSemanticsSuite.scala
+++ b/spark/src/test/scala/org/apache/comet/CometFloatSemanticsSuite.scala
@@ -358,9 +358,5 @@ object CometFloatSemanticsSuite {
KnownGap(
issue(5312),
"collect_set before Spark 4.2 treats -0.0 and 0.0 as one value and NaNs as distinct.",
- c => in("aggregate", "collect_set")(c) && !CometSparkSessionExtensions.isSpark42Plus),
- KnownGap(
- issue(6522),
- "signum(-0.0) returns 0.0, where Spark returns -0.0.",
- in("expression", "signum")))
+ c => in("aggregate", "collect_set")(c) && !CometSparkSessionExtensions.isSpark42Plus))
}
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@datafusion.apache.org
For additional commands, e-mail: commits-help@datafusion.apache.org