[SPARK-14451][SQL] Move encoder definition into Aggregator interface

## What changes were proposed in this pull request?
When we first introduced Aggregators, we required the user of Aggregators to (implicitly) specify the encoders. It would actually make more sense to have the encoders be specified by the implementation of Aggregators, since each implementation should have the most state about how to encode its own data type.

Note that this simplifies the Java API because Java users no longer need to explicitly specify encoders for aggregators.

## How was this patch tested?
Updated unit tests.

Author: Reynold Xin <rxin@databricks.com>

Closes #12231 from rxin/SPARK-14451.

1 files changed, 13 insertions, 4 deletions

diff --git a/sql/core/src/test/java/test/org/apache/spark/sql/sources/JavaDatasetAggregatorSuite.java b/sql/core/src/test/java/test/org/apache/spark/sql/sources/JavaDatasetAggregatorSuite.java
index 8cb174b906..0e49f871de 100644
--- a/sql/core/src/test/java/test/org/apache/spark/sql/sources/JavaDatasetAggregatorSuite.java
+++ b/sql/core/src/test/java/test/org/apache/spark/sql/sources/JavaDatasetAggregatorSuite.java
@@ -26,6 +26,7 @@ import org.junit.Test;
 
 import org.apache.spark.api.java.function.MapFunction;
 import org.apache.spark.sql.Dataset;
+import org.apache.spark.sql.Encoder;
 import org.apache.spark.sql.Encoders;
 import org.apache.spark.sql.KeyValueGroupedDataset;
 import org.apache.spark.sql.expressions.Aggregator;
@@ -39,12 +40,10 @@ public class JavaDatasetAggregatorSuite extends JavaDatasetAggregatorSuiteBase {
   public void testTypedAggregationAnonClass() {
     KeyValueGroupedDataset<String, Tuple2<String, Integer>> grouped = generateGroupedDataset();
 
-    Dataset<Tuple2<String, Integer>> agged =
-      grouped.agg(new IntSumOf().toColumn(Encoders.INT(), Encoders.INT()));
+    Dataset<Tuple2<String, Integer>> agged = grouped.agg(new IntSumOf().toColumn());
     Assert.assertEquals(Arrays.asList(tuple2("a", 3), tuple2("b", 3)), agged.collectAsList());
 
-    Dataset<Tuple2<String, Integer>> agged2 = grouped.agg(
-      new IntSumOf().toColumn(Encoders.INT(), Encoders.INT()))
+    Dataset<Tuple2<String, Integer>> agged2 = grouped.agg(new IntSumOf().toColumn())
       .as(Encoders.tuple(Encoders.STRING(), Encoders.INT()));
     Assert.assertEquals(
       Arrays.asList(
@@ -73,6 +72,16 @@ public class JavaDatasetAggregatorSuite extends JavaDatasetAggregatorSuiteBase {
     public Integer finish(Integer reduction) {
       return reduction;
     }
+
+    @Override
+    public Encoder<Integer> bufferEncoder() {
+      return Encoders.INT();
+    }
+
+    @Override
+    public Encoder<Integer> outputEncoder() {
+      return Encoders.INT();
+    }
   }
 
   @Test