Strict argument checks for array() and map()

Author: maropu

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/complexTypeCreator.scala

@@ -41,8 +41,13 @@ case class CreateArray(children: Seq[Expression]) extends Expression {
 
   override def foldable: Boolean = children.forall(_.foldable)
 
-  override def checkInputDataTypes(): TypeCheckResult =
-    TypeUtils.checkForSameTypeInputExpr(children.map(_.dataType), "function array")
+  override def checkInputDataTypes(): TypeCheckResult = {
+    if (children == Nil) {
+      TypeCheckResult.TypeCheckFailure("input to function coalesce cannot be empty")
+    } else {
+      TypeUtils.checkForSameTypeInputExpr(children.map(_.dataType), "function array")
+    }
+  }
 
   override def dataType: ArrayType = {
     ArrayType(
@@ -168,7 +173,9 @@ case class CreateMap(children: Seq[Expression]) extends Expression {
   override def foldable: Boolean = children.forall(_.foldable)
 
   override def checkInputDataTypes(): TypeCheckResult = {
-    if (children.size % 2 != 0) {
+    if (children == Nil) {
+      TypeCheckResult.TypeCheckFailure("input to function coalesce cannot be empty")
+    } else if (children.size % 2 != 0) {
       TypeCheckResult.TypeCheckFailure(s"$prettyName expects a positive even number of arguments.")
     } else if (keys.map(_.dataType).distinct.length > 1) {
       TypeCheckResult.TypeCheckFailure("The given keys of function map should all be the same " +

sql/core/src/test/scala/org/apache/spark/sql/DataFrameFunctionsSuite.scala

@@ -448,6 +448,43 @@ class DataFrameFunctionsSuite extends QueryTest with SharedSQLContext {
       rand(Random.nextLong()), randn(Random.nextLong())
     ).foreach(assertValuesDoNotChangeAfterCoalesceOrUnion(_))
   }
+
+  test("SPARK-21281 fails if functions have no argument") {
+    var errMsg = intercept[AnalysisException] {
+      spark.range(1).select(array())
+    }.getMessage
+    assert(errMsg.contains("due to data type mismatch: input to function coalesce cannot be empty"))
+
+    errMsg = intercept[AnalysisException] {
+      spark.range(1).select(map())
+    }.getMessage
+    assert(errMsg.contains("due to data type mismatch: input to function coalesce cannot be empty"))
+
+    // spark.range(1).select(coalesce())
+    errMsg = intercept[AnalysisException] {
+      spark.range(1).select(coalesce())
+    }.getMessage
+    assert(errMsg.contains("due to data type mismatch: input to function coalesce cannot be empty"))
+
+    // This hits java.lang.AssertionError
+    // spark.range(1).select(struct())
+
+    errMsg = intercept[IllegalArgumentException] {
+      spark.range(1).select(greatest())
+    }.getMessage
+    assert(errMsg.contains("requirement failed: greatest requires at least 2 arguments"))
+
+    errMsg = intercept[IllegalArgumentException] {
+      spark.range(1).select(least())
+    }.getMessage
+    assert(errMsg.contains("requirement failed: least requires at least 2 arguments"))
+
+    errMsg = intercept[AnalysisException] {
+      spark.range(1).select(hash())
+    }.getMessage
+    assert(errMsg.contains(
+      "due to data type mismatch: function hash requires at least one argument"))
+  }
 }
 
 object DataFrameFunctionsSuite {