Revert "[SPARK-29462] The data type of "array()" should be array<null>"

This reverts commit 0dcd739.

Author: HyukjinKwon

docs/sql-migration-guide.md

@@ -217,8 +217,6 @@ license: |
   For example `SELECT timestamp 'tomorrow';`.
 
   - Since Spark 3.0, the `size` function returns `NULL` for the `NULL` input. In Spark version 2.4 and earlier, this function gives `-1` for the same input. To restore the behavior before Spark 3.0, you can set `spark.sql.legacy.sizeOfNull` to `true`.
-  
-  - Since Spark 3.0, when `array` function is called without parameters, it returns an empty array with `NullType` data type. In Spark version 2.4 and earlier, the data type of the result is `StringType`.
 
   - Since Spark 3.0, the interval literal syntax does not allow multiple from-to units anymore. For example, `SELECT INTERVAL '1-1' YEAR TO MONTH '2-2' YEAR TO MONTH'` throws parser exception.
 

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

@@ -47,7 +47,7 @@ case class CreateArray(children: Seq[Expression]) extends Expression {
   override def dataType: ArrayType = {
     ArrayType(
       TypeCoercion.findCommonTypeDifferentOnlyInNullFlags(children.map(_.dataType))
-        .getOrElse(NullType),
+        .getOrElse(StringType),
       containsNull = children.exists(_.nullable))
   }
 

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

@@ -3400,9 +3400,12 @@ class DataFrameFunctionsSuite extends QueryTest with SharedSparkSession {
     ).foreach(assertValuesDoNotChangeAfterCoalesceOrUnion(_))
   }
 
-  test("SPARK-21281 use string types by default if map have no argument") {
+  test("SPARK-21281 use string types by default if array and map have no argument") {
     val ds = spark.range(1)
     var expectedSchema = new StructType()
+      .add("x", ArrayType(StringType, containsNull = false), nullable = false)
+    assert(ds.select(array().as("x")).schema == expectedSchema)
+    expectedSchema = new StructType()
       .add("x", MapType(StringType, StringType, valueContainsNull = false), nullable = false)
     assert(ds.select(map().as("x")).schema == expectedSchema)
   }
@@ -3460,13 +3463,6 @@ class DataFrameFunctionsSuite extends QueryTest with SharedSparkSession {
     checkAnswer(df.select("x").filter("exists(i, x -> x % d == 0)"),
       Seq(Row(1)))
   }
-
-  test("SPARK-29462: Use null type by default if array have no argument") {
-    val ds = spark.range(1)
-    var expectedSchema = new StructType()
-      .add("x", ArrayType(NullType, containsNull = false), nullable = false)
-    assert(ds.select(array().as("x")).schema == expectedSchema)
-  }
 }
 
 object DataFrameFunctionsSuite {