diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/types/dataTypes.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/types/dataTypes.scala
index 970d98626fb7a..15a2d599450e2 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/types/dataTypes.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/types/dataTypes.scala
@@ -19,23 +19,21 @@ package org.apache.spark.sql.catalyst.types
 
 import java.sql.Timestamp
 
-import org.apache.spark.sql.catalyst.util.Metadata
-
 import scala.math.Numeric.{BigDecimalAsIfIntegral, DoubleAsIfIntegral, FloatAsIfIntegral}
 import scala.reflect.ClassTag
 import scala.reflect.runtime.universe.{TypeTag, runtimeMirror, typeTag}
 import scala.util.parsing.combinator.RegexParsers
 
-import org.json4s.JsonAST.JValue
 import org.json4s._
+import org.json4s.JsonAST.JValue
 import org.json4s.JsonDSL._
 import org.json4s.jackson.JsonMethods._
 
 import org.apache.spark.sql.catalyst.ScalaReflectionLock
 import org.apache.spark.sql.catalyst.expressions.{Attribute, AttributeReference, Expression}
+import org.apache.spark.sql.catalyst.util.Metadata
 import org.apache.spark.util.Utils
 
-
 object DataType {
   def fromJson(json: String): DataType = parseDataType(parse(json))
 
@@ -380,8 +378,7 @@ case class ArrayType(elementType: DataType, containsNull: Boolean) extends DataT
  * @param name The name of this field.
  * @param dataType The data type of this field.
  * @param nullable Indicates if values of this field can be `null` values.
- * @param metadata The metadata of this field, which is a map from string to simple type that can be
- *                 serialized to JSON automatically. The metadata should be preserved during
+ * @param metadata The metadata of this field. The metadata should be preserved during
  *                 transformation if the content of the column is not modified, e.g, in selection.
  */
 case class StructField(
diff --git a/sql/core/src/main/java/org/apache/spark/sql/api/java/DataType.java b/sql/core/src/main/java/org/apache/spark/sql/api/java/DataType.java
index cbb9b5a80d5b5..66eac38937b5e 100644
--- a/sql/core/src/main/java/org/apache/spark/sql/api/java/DataType.java
+++ b/sql/core/src/main/java/org/apache/spark/sql/api/java/DataType.java
@@ -17,10 +17,10 @@
 
 package org.apache.spark.sql.api.java;
 
-import org.apache.spark.sql.catalyst.util.Metadata;
-
 import java.util.*;
 
+import org.apache.spark.sql.catalyst.util.Metadata;
+
 /**
  * The base type of all Spark SQL data types.
  *
diff --git a/sql/core/src/main/java/org/apache/spark/sql/api/java/StructField.java b/sql/core/src/main/java/org/apache/spark/sql/api/java/StructField.java
index 3d84cda879124..264464f90fdf3 100644
--- a/sql/core/src/main/java/org/apache/spark/sql/api/java/StructField.java
+++ b/sql/core/src/main/java/org/apache/spark/sql/api/java/StructField.java
@@ -17,10 +17,10 @@
 
 package org.apache.spark.sql.api.java;
 
-import org.apache.spark.sql.catalyst.util.Metadata;
-
 import java.util.Map;
 
+import org.apache.spark.sql.catalyst.util.Metadata;
+
 /**
  * A StructField object represents a field in a StructType object.
  * A StructField object comprises three fields, {@code String name}, {@code DataType dataType},
@@ -28,11 +28,10 @@
  * The field of {@code dataType} specifies the data type of a StructField.
  * The field of {@code nullable} specifies if values of a StructField can contain {@code null}
  * values.
- * The field of {@code metadata} provides extra information of the StructField, which is a map from
- * string to simple type that can be serialized to JSON automatically
+ * The field of {@code metadata} provides extra information of the StructField.
  *
  * To create a {@link StructField},
- * {@link DataType#createStructField(String, DataType, boolean, Map)}
+ * {@link DataType#createStructField(String, DataType, boolean, Metadata)}
  * should be used.
  */
 public class StructField {
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DataTypeSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DataTypeSuite.scala
index f874d9d279098..a1d5907b357b0 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DataTypeSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DataTypeSuite.scala
@@ -17,10 +17,10 @@
 
 package org.apache.spark.sql
 
-import org.apache.spark.sql.catalyst.util.MetadataBuilder
 import org.scalatest.FunSuite
 
 import org.apache.spark.sql.catalyst.types.DataType
+import org.apache.spark.sql.catalyst.util.MetadataBuilder
 
 class DataTypeSuite extends FunSuite {
 
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/api/java/ScalaSideDataTypeConversionSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/api/java/ScalaSideDataTypeConversionSuite.scala
index 688d9ff724dfa..91bbc00be7090 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/api/java/ScalaSideDataTypeConversionSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/api/java/ScalaSideDataTypeConversionSuite.scala
@@ -17,13 +17,12 @@
 
 package org.apache.spark.sql.api.java
 
-import org.apache.spark.sql.catalyst.util.MetadataBuilder
-import org.apache.spark.sql.types.util.DataTypeConversions
 import org.scalatest.FunSuite
 
-import org.apache.spark.sql.{DataType => SDataType, StructField => SStructField}
-import org.apache.spark.sql.{StructType => SStructType}
-import DataTypeConversions._
+import org.apache.spark.sql.{DataType => SDataType, StructField => SStructField,
+  StructType => SStructType}
+import org.apache.spark.sql.catalyst.util.MetadataBuilder
+import org.apache.spark.sql.types.util.DataTypeConversions._
 
 class ScalaSideDataTypeConversionSuite extends FunSuite {