diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/TypeCoercion.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/TypeCoercion.scala index 0a0bef6c9953..e149bf2f4976 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/TypeCoercion.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/TypeCoercion.scala @@ -158,7 +158,7 @@ object TypeCoercion { } case (MapType(kt1, vt1, valueContainsNull1), MapType(kt2, vt2, valueContainsNull2)) => findTypeFunc(kt1, kt2) - .filter { kt => !Cast.forceNullable(kt1, kt) && !Cast.forceNullable(kt2, kt) } + .filter { kt => Cast.canCastMapKeyNullSafe(kt1, kt) && Cast.canCastMapKeyNullSafe(kt2, kt) } .flatMap { kt => findTypeFunc(vt1, vt2).map { vt => MapType(kt, vt, valueContainsNull1 || valueContainsNull2 || diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Cast.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Cast.scala index 7c4316fe0843..8177136edfd6 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Cast.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Cast.scala @@ -77,8 +77,7 @@ object Cast { resolvableNullability(fn || forceNullable(fromType, toType), tn) case (MapType(fromKey, fromValue, fn), MapType(toKey, toValue, tn)) => - canCast(fromKey, toKey) && - (!forceNullable(fromKey, toKey)) && + canCast(fromKey, toKey) && canCastMapKeyNullSafe(fromKey, toKey) && canCast(fromValue, toValue) && resolvableNullability(fn || forceNullable(fromValue, toValue), tn) @@ -98,6 +97,11 @@ object Cast { case _ => false } + def canCastMapKeyNullSafe(fromType: DataType, toType: DataType): Boolean = { + // If the original map key type is NullType, it's OK as the map must be empty. + fromType == NullType || !forceNullable(fromType, toType) + } + /** * Return true if we need to use the `timeZone` information casting `from` type to `to` type. * The patterns matched reflect the current implementation in the Cast node. diff --git a/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala index 823e432ba4b9..de0f7801a39a 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala @@ -3487,6 +3487,12 @@ class SQLQuerySuite extends QueryTest with SharedSparkSession with AdaptiveSpark } } } + + test("SPARK-31166: UNION map and other maps should not fail") { + checkAnswer( + sql("(SELECT map()) UNION ALL (SELECT map(1, 2))"), + Seq(Row(Map[Int, Int]()), Row(Map(1 -> 2)))) + } } case class Foo(bar: Option[String])