[SPARK-15597][SQL] Add SparkSession.emptyDataset

## What changes were proposed in this pull request?
This patch adds a new function emptyDataset to SparkSession, for creating an empty dataset.

## How was this patch tested?
Added a test case.

Author: Reynold Xin <rxin@databricks.com>

Closes #13344 from rxin/SPARK-15597.

2 files changed, 18 insertions, 0 deletions

diff --git a/sql/core/src/main/scala/org/apache/spark/sql/SparkSession.scala b/sql/core/src/main/scala/org/apache/spark/sql/SparkSession.scala
index aa60048405..c9276cf140 100644
--- a/sql/core/src/main/scala/org/apache/spark/sql/SparkSession.scala
+++ b/sql/core/src/main/scala/org/apache/spark/sql/SparkSession.scala
@@ -224,6 +224,18 @@ class SparkSession private(
 
   /**
    * :: Experimental ::
+   * Creates a new [[Dataset]] of type T containing zero elements.
+   *
+   * @return 2.0.0
+   */
+  @Experimental
+  def emptyDataset[T: Encoder]: Dataset[T] = {
+    val encoder = implicitly[Encoder[T]]
+    new Dataset(self, LocalRelation(encoder.schema.toAttributes), encoder)
+  }
+
+  /**
+   * :: Experimental ::
    * Creates a [[DataFrame]] from an RDD of Product (e.g. case classes, tuples).
    *
    * @group dataframes
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala
index 2a6591653e..e395007999 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala
@@ -46,6 +46,12 @@ class DatasetSuite extends QueryTest with SharedSQLContext {
       1, 1, 1)
   }
 
+  test("emptyDataset") {
+    val ds = spark.emptyDataset[Int]
+    assert(ds.count() == 0L)
+    assert(ds.collect() sameElements Array.empty[Int])
+  }
+
   test("range") {
     assert(spark.range(10).map(_ + 1).reduce(_ + _) == 55)
     assert(spark.range(10).map{ case i: java.lang.Long => i + 1 }.reduce(_ + _) == 55)