This is an automated email from the ASF dual-hosted git repository.
yangjie01 pushed a commit to branch branch-4.0
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/branch-4.0 by this push:
new 8fac87e8cbe9 [SPARK-52945][SQL][TESTS] Split
`CastSuiteBase#checkInvalidCastFromNumericType` into three methods and
guarantee assertions are valid
8fac87e8cbe9 is described below
commit 8fac87e8cbe952c4965bbcaf363bf0af10c4dc18
Author: yangjie01 <yangji...@baidu.com>
AuthorDate: Mon Jul 28 20:20:59 2025 +0800
[SPARK-52945][SQL][TESTS] Split
`CastSuiteBase#checkInvalidCastFromNumericType` into three methods and
guarantee assertions are valid
### What changes were proposed in this pull request?
Due to the absence of `assert` statements, the
`CastSuiteBase#checkInvalidCastFromNumericType` method previously performed no
assertion checks.
Additionally, since `checkInvalidCastFromNumericType` had significant
variations in target type validation logic across different `EvalMode`
contexts, this pr refactors the method into three specialized methods to ensure
robust assertion enforcement:
- `checkInvalidCastFromNumericTypeToDateType`
- `checkInvalidCastFromNumericTypeToTimestampNTZType`
- `checkInvalidCastFromNumericTypeToBinaryType`
### Why are the changes needed?
To address the missing assertion validation in `CastSuiteBase`.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
- Pass GitHub Actions
### Was this patch authored or co-authored using generative AI tooling?
No
Closes #51668 from LuciferYang/SPARK-52945.
Authored-by: yangjie01 <yangji...@baidu.com>
Signed-off-by: yangjie01 <yangji...@baidu.com>
(cherry picked from commit 9a452f81dbddb765f55d0610e0e1691bd2ca6e96)
Signed-off-by: yangjie01 <yangji...@baidu.com>
---
.../sql/catalyst/expressions/CastSuiteBase.scala | 95 +++++++++-------------
.../catalyst/expressions/CastWithAnsiOnSuite.scala | 32 +++++++-
.../sql/catalyst/expressions/TryCastSuite.scala | 9 ++
3 files changed, 77 insertions(+), 59 deletions(-)
diff --git
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastSuiteBase.scala
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastSuiteBase.scala
index cec49a5ae1de..869148165397 100644
---
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastSuiteBase.scala
+++
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastSuiteBase.scala
@@ -545,61 +545,42 @@ abstract class CastSuiteBase extends SparkFunSuite with
ExpressionEvalHelper {
checkCast("0", false)
}
- protected def checkInvalidCastFromNumericType(to: DataType): Unit = {
- cast(1.toByte, to).checkInputDataTypes() ==
- DataTypeMismatch(
- errorSubClass = "CAST_WITH_FUNC_SUGGESTION",
- messageParameters = Map(
- "srcType" -> toSQLType(Literal(1.toByte).dataType),
- "targetType" -> toSQLType(to),
- "functionNames" -> "`DATE_FROM_UNIX_DATE`"
- )
- )
- cast(1.toShort, to).checkInputDataTypes() ==
- DataTypeMismatch(
- errorSubClass = "CAST_WITH_FUNC_SUGGESTION",
- messageParameters = Map(
- "srcType" -> toSQLType(Literal(1.toShort).dataType),
- "targetType" -> toSQLType(to),
- "functionNames" -> "`DATE_FROM_UNIX_DATE`"
- )
- )
- cast(1, to).checkInputDataTypes() ==
- DataTypeMismatch(
- errorSubClass = "CAST_WITH_FUNC_SUGGESTION",
- messageParameters = Map(
- "srcType" -> toSQLType(Literal(1).dataType),
- "targetType" -> toSQLType(to),
- "functionNames" -> "`DATE_FROM_UNIX_DATE`"
- )
- )
- cast(1L, to).checkInputDataTypes() ==
- DataTypeMismatch(
- errorSubClass = "CAST_WITH_FUNC_SUGGESTION",
- messageParameters = Map(
- "srcType" -> toSQLType(Literal(1L).dataType),
- "targetType" -> toSQLType(to),
- "functionNames" -> "`DATE_FROM_UNIX_DATE`"
- )
- )
- cast(1.0.toFloat, to).checkInputDataTypes() ==
- DataTypeMismatch(
- errorSubClass = "CAST_WITH_FUNC_SUGGESTION",
- messageParameters = Map(
- "srcType" -> toSQLType(Literal(1.0.toFloat).dataType),
- "targetType" -> toSQLType(to),
- "functionNames" -> "`DATE_FROM_UNIX_DATE`"
- )
- )
- cast(1.0, to).checkInputDataTypes() ==
- DataTypeMismatch(
- errorSubClass = "CAST_WITH_FUNC_SUGGESTION",
- messageParameters = Map(
- "srcType" -> toSQLType(Literal(1.0).dataType),
- "targetType" -> toSQLType(to),
- "functionNames" -> "`DATE_FROM_UNIX_DATE`"
- )
- )
+ protected def createCastMismatch(
+ srcType: DataType,
+ targetType: DataType,
+ errorSubClass: String,
+ extraParams: Map[String, String] = Map.empty): DataTypeMismatch = {
+ val baseParams = Map(
+ "srcType" -> toSQLType(srcType),
+ "targetType" -> toSQLType(targetType)
+ )
+ DataTypeMismatch(errorSubClass, baseParams ++ extraParams)
+ }
+
+ protected def checkInvalidCastFromNumericTypeToDateType(): Unit = {
+ val errorSubClass = if (evalMode == EvalMode.LEGACY) {
+ "CAST_WITHOUT_SUGGESTION"
+ } else {
+ "CAST_WITH_FUNC_SUGGESTION"
+ }
+ val funcParams = if (evalMode == EvalMode.LEGACY) {
+ Map.empty[String, String]
+ } else {
+ Map("functionNames" -> "`DATE_FROM_UNIX_DATE`")
+ }
+ Seq(1.toByte, 1.toShort, 1, 1L, 1.0.toFloat, 1.0).foreach { testValue =>
+ val expectedError =
+ createCastMismatch(Literal(testValue).dataType, DateType,
errorSubClass, funcParams)
+ assert(cast(testValue, DateType).checkInputDataTypes() == expectedError)
+ }
+ }
+ protected def checkInvalidCastFromNumericTypeToTimestampNTZType(): Unit = {
+ // All numeric types: `CAST_WITHOUT_SUGGESTION`
+ Seq(1.toByte, 1.toShort, 1, 1L, 1.0.toFloat, 1.0).foreach { testValue =>
+ val expectedError =
+ createCastMismatch(Literal(testValue).dataType, TimestampNTZType,
"CAST_WITHOUT_SUGGESTION")
+ assert(cast(testValue, TimestampNTZType).checkInputDataTypes() ==
expectedError)
+ }
}
test("SPARK-16729 type checking for casting to date type") {
@@ -614,7 +595,7 @@ abstract class CastSuiteBase extends SparkFunSuite with
ExpressionEvalHelper {
)
)
)
- checkInvalidCastFromNumericType(DateType)
+ checkInvalidCastFromNumericTypeToDateType()
}
test("SPARK-20302 cast with same structure") {
@@ -998,7 +979,7 @@ abstract class CastSuiteBase extends SparkFunSuite with
ExpressionEvalHelper {
test("disallow type conversions between Numeric types and Timestamp without
time zone type") {
import DataTypeTestUtils.numericTypes
- checkInvalidCastFromNumericType(TimestampNTZType)
+ checkInvalidCastFromNumericTypeToTimestampNTZType()
verifyCastFailure(
cast(Literal(0L), TimestampNTZType),
DataTypeMismatch(
diff --git
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastWithAnsiOnSuite.scala
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastWithAnsiOnSuite.scala
index f0709db7259d..5bb726e09afe 100644
---
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastWithAnsiOnSuite.scala
+++
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastWithAnsiOnSuite.scala
@@ -29,6 +29,7 @@ import
org.apache.spark.sql.catalyst.util.DateTimeConstants.MILLIS_PER_SECOND
import org.apache.spark.sql.catalyst.util.DateTimeTestUtils
import
org.apache.spark.sql.catalyst.util.DateTimeTestUtils.{withDefaultTimeZone, UTC}
import org.apache.spark.sql.errors.QueryErrorsBase
+import org.apache.spark.sql.internal.SQLConf
import org.apache.spark.sql.types._
import org.apache.spark.unsafe.types.{UTF8String, VariantVal}
@@ -39,6 +40,33 @@ class CastWithAnsiOnSuite extends CastSuiteBase with
QueryErrorsBase {
override def evalMode: EvalMode.Value = EvalMode.ANSI
+ protected def checkInvalidCastFromNumericTypeToBinaryType(): Unit = {
+ def checkNumericTypeCast(
+ testValue: Any,
+ srcType: DataType,
+ to: DataType,
+ expectedErrorClass: String,
+ extraParams: Map[String, String] = Map.empty): Unit = {
+ val expectedError = createCastMismatch(srcType, to, expectedErrorClass,
extraParams)
+ assert(cast(testValue, to).checkInputDataTypes() == expectedError)
+ }
+
+ // Integer types: suggest config change
+ val configParams = Map(
+ "config" -> toSQLConf(SQLConf.ANSI_ENABLED.key),
+ "configVal" -> toSQLValue("false", StringType)
+ )
+ checkNumericTypeCast(1.toByte, ByteType, BinaryType,
"CAST_WITH_CONF_SUGGESTION", configParams)
+ checkNumericTypeCast(
+ 1.toShort, ShortType, BinaryType, "CAST_WITH_CONF_SUGGESTION",
configParams)
+ checkNumericTypeCast(1, IntegerType, BinaryType,
"CAST_WITH_CONF_SUGGESTION", configParams)
+ checkNumericTypeCast(1L, LongType, BinaryType,
"CAST_WITH_CONF_SUGGESTION", configParams)
+
+ // Floating types: no suggestion
+ checkNumericTypeCast(1.0.toFloat, FloatType, BinaryType,
"CAST_WITHOUT_SUGGESTION")
+ checkNumericTypeCast(1.0, DoubleType, BinaryType,
"CAST_WITHOUT_SUGGESTION")
+ }
+
private def isTryCast = evalMode == EvalMode.TRY
private def testIntMaxAndMin(dt: DataType): Unit = {
@@ -142,7 +170,7 @@ class CastWithAnsiOnSuite extends CastSuiteBase with
QueryErrorsBase {
test("ANSI mode: disallow type conversions between Numeric types and Date
type") {
import DataTypeTestUtils.numericTypes
- checkInvalidCastFromNumericType(DateType)
+ checkInvalidCastFromNumericTypeToDateType()
verifyCastFailure(
cast(Literal(0L), DateType),
DataTypeMismatch(
@@ -168,7 +196,7 @@ class CastWithAnsiOnSuite extends CastSuiteBase with
QueryErrorsBase {
test("ANSI mode: disallow type conversions between Numeric types and Binary
type") {
import DataTypeTestUtils.numericTypes
- checkInvalidCastFromNumericType(BinaryType)
+ checkInvalidCastFromNumericTypeToBinaryType()
val binaryLiteral = Literal(new Array[Byte](1.toByte), BinaryType)
numericTypes.foreach { numericType =>
assert(cast(binaryLiteral, numericType).checkInputDataTypes() ==
diff --git
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/TryCastSuite.scala
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/TryCastSuite.scala
index 446514de91d6..312b05755507 100644
---
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/TryCastSuite.scala
+++
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/TryCastSuite.scala
@@ -61,6 +61,15 @@ class TryCastSuite extends CastWithAnsiOnSuite {
checkEvaluation(cast(l, to), tryCastResult, InternalRow(l.value))
}
+ override protected def checkInvalidCastFromNumericTypeToBinaryType(): Unit =
{
+ // All numeric types: `CAST_WITHOUT_SUGGESTION`
+ Seq(1.toByte, 1.toShort, 1, 1L, 1.0.toFloat, 1.0).foreach { testValue =>
+ val expectedError =
+ createCastMismatch(Literal(testValue).dataType, BinaryType,
"CAST_WITHOUT_SUGGESTION")
+ assert(cast(testValue, BinaryType).checkInputDataTypes() ==
expectedError)
+ }
+ }
+
test("print string") {
assert(cast(Literal("1"), IntegerType).toString == "try_cast(1 as int)")
assert(cast(Literal("1"), IntegerType).sql == "TRY_CAST('1' AS INT)")
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org
