[SPARK-LLAP-135] Bump to HDP 2.6.1. (#136)

## What changes were proposed in this pull request?

This bumps the Hive dependency to HDP 2.6.1.0 for Spark-LLAP master/branch-2.1.

## How was this patch tested?

- [x] Travis CI should passed.
- [x] Ranger test should passed.

```
[hive@hdp26-6 python]$ python spark-ranger-secure-test.py
........................
----------------------------------------------------------------------
Ran 24 tests in 3723.090s

OK
```

This closes #135 .
(cherry picked from commit c0cafa9)

Signed-off-by: Dongjoon Hyun <[email protected]>

Signed-off-by: Dongjoon Hyun <[email protected]>

Author: dongjoon-hyun

build.sbt

@@ -7,8 +7,8 @@ val scalatestVersion = "2.2.6"
 
 sparkVersion := sys.props.getOrElse("spark.version", "2.1.1")
 
-val hadoopVersion = sys.props.getOrElse("hadoop.version", "2.7.3")
-val hiveVersion = sys.props.getOrElse("hive.version", "2.1.0.2.6.0.3-8")
+val hadoopVersion = sys.props.getOrElse("hadoop.version", "2.7.3.2.6.1.0-129")
+val hiveVersion = sys.props.getOrElse("hive.version", "2.1.0.2.6.1.0-129")
 val log4j2Version = sys.props.getOrElse("log4j2.version", "2.4.1")
 val tezVersion = sys.props.getOrElse("tez.version", "0.8.4")
 val thriftVersion = sys.props.getOrElse("thrift.version", "0.9.3")

src/main/scala/com/hortonworks/spark/sql/hive/llap/HS2JDBCWrapper.scala

@@ -21,15 +21,14 @@ import java.net.URI
 import java.sql.{Connection, DatabaseMetaData, Driver, DriverManager, ResultSet, ResultSetMetaData,
   SQLException}
 
-import collection.JavaConverters._
+import scala.collection.JavaConverters._
 import scala.collection.mutable.ArrayBuffer
 
-import org.slf4j.LoggerFactory
-
-import org.apache.hadoop.hive.llap.FieldDesc
 import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector.Category
 import org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector.PrimitiveCategory
 import org.apache.hadoop.hive.serde2.typeinfo._
+import org.slf4j.LoggerFactory
+
 import org.apache.spark.sql.types._
 
 
@@ -259,43 +258,48 @@ class JDBCWrapper {
 
   private def getCatalystType(typeInfo: TypeInfo) : DataType = {
     typeInfo.getCategory match {
-      case Category.PRIMITIVE => getCatalystType(typeInfo.asInstanceOf[PrimitiveTypeInfo])
-      case Category.LIST      => ArrayType(
-          getCatalystType(typeInfo.asInstanceOf[ListTypeInfo].getListElementTypeInfo))
-      case Category.MAP       => MapType(
+      case Category.PRIMITIVE =>
+        getCatalystType(typeInfo.asInstanceOf[PrimitiveTypeInfo])
+      case Category.LIST =>
+        ArrayType(getCatalystType(typeInfo.asInstanceOf[ListTypeInfo].getListElementTypeInfo))
+      case Category.MAP =>
+        MapType(
           getCatalystType(typeInfo.asInstanceOf[MapTypeInfo].getMapKeyTypeInfo),
           getCatalystType(typeInfo.asInstanceOf[MapTypeInfo].getMapValueTypeInfo))
-      case Category.STRUCT    => StructType(getCatalystStructFields(typeInfo.asInstanceOf[StructTypeInfo]))
-      case _                  => throw new SQLException("Unsupported type " + typeInfo)
+      case Category.STRUCT =>
+        StructType(getCatalystStructFields(typeInfo.asInstanceOf[StructTypeInfo]))
+      case _ =>
+        throw new SQLException("Unsupported type " + typeInfo)
     }
   }
 
   private def getCatalystType(primitiveTypeInfo: PrimitiveTypeInfo) : DataType = {
     primitiveTypeInfo.getPrimitiveCategory match {
-      case PrimitiveCategory.BOOLEAN   => BooleanType
-      case PrimitiveCategory.BYTE      => ByteType
-      case PrimitiveCategory.SHORT     => ShortType
-      case PrimitiveCategory.INT       => IntegerType
-      case PrimitiveCategory.LONG      => LongType
-      case PrimitiveCategory.FLOAT     => FloatType
-      case PrimitiveCategory.DOUBLE    => DoubleType
-      case PrimitiveCategory.STRING    => StringType
-      case PrimitiveCategory.CHAR      => StringType
-      case PrimitiveCategory.VARCHAR   => StringType
-      case PrimitiveCategory.DATE      => DateType
+      case PrimitiveCategory.BOOLEAN => BooleanType
+      case PrimitiveCategory.BYTE => ByteType
+      case PrimitiveCategory.SHORT => ShortType
+      case PrimitiveCategory.INT => IntegerType
+      case PrimitiveCategory.LONG => LongType
+      case PrimitiveCategory.FLOAT => FloatType
+      case PrimitiveCategory.DOUBLE => DoubleType
+      case PrimitiveCategory.STRING => StringType
+      case PrimitiveCategory.CHAR => StringType
+      case PrimitiveCategory.VARCHAR => StringType
+      case PrimitiveCategory.DATE => DateType
       case PrimitiveCategory.TIMESTAMP => TimestampType
-      case PrimitiveCategory.BINARY    => BinaryType
-      case PrimitiveCategory.DECIMAL   => DecimalType(
+      case PrimitiveCategory.BINARY => BinaryType
+      case PrimitiveCategory.DECIMAL => DecimalType(
           primitiveTypeInfo.asInstanceOf[DecimalTypeInfo].getPrecision,
           primitiveTypeInfo.asInstanceOf[DecimalTypeInfo].getScale)
       case _ => throw new SQLException("Unsupported type " + primitiveTypeInfo)
     }
   }
 
   private def getCatalystStructFields(structTypeInfo: StructTypeInfo) : Array[StructField] = {
-    structTypeInfo.getAllStructFieldNames.asScala.zip(structTypeInfo.getAllStructFieldTypeInfos.asScala).map(
+    structTypeInfo.getAllStructFieldNames.asScala
+      .zip(structTypeInfo.getAllStructFieldTypeInfos.asScala).map(
         { case (fieldName, fieldType) => new StructField(fieldName, getCatalystType(fieldType)) }
-    ).toArray
+      ).toArray
   }
 
   private def getCatalystType(typeName: String) : DataType = {

src/main/scala/com/hortonworks/spark/sql/hive/llap/RowConverter.scala

@@ -16,10 +16,12 @@
  */
 package com.hortonworks.spark.sql.hive.llap
 
-import collection.JavaConverters._
-import org.apache.hadoop.hive.llap.{Schema}
+import scala.collection.JavaConverters._
+
+import org.apache.hadoop.hive.llap.Schema
 import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector.Category
 import org.apache.hadoop.hive.serde2.typeinfo._
+
 import org.apache.spark.sql.Row
 
 object RowConverter {
@@ -44,14 +46,13 @@ object RowConverter {
         listElement => convertValue(
             listElement,
             colType.asInstanceOf[ListTypeInfo].getListElementTypeInfo))
-      case Category.MAP => {
+      case Category.MAP =>
         // Try LinkedHashMap to preserve order of elements - is that necessary?
-        var convertedMap = scala.collection.mutable.LinkedHashMap.empty[Any, Any]
+        var map = scala.collection.mutable.LinkedHashMap.empty[Any, Any]
         value.asInstanceOf[java.util.Map[Any, Any]].asScala.foreach((tuple) =>
-          convertedMap(convertValue(tuple._1, colType.asInstanceOf[MapTypeInfo].getMapKeyTypeInfo)) =
+          map(convertValue(tuple._1, colType.asInstanceOf[MapTypeInfo].getMapKeyTypeInfo)) =
             convertValue(tuple._2, colType.asInstanceOf[MapTypeInfo].getMapValueTypeInfo))
-        convertedMap
-      }
+        map
       case Category.STRUCT =>
         // Struct value is just a list of values. Convert each value based on corresponding typeinfo
         Row.fromSeq(

src/main/scala/org/apache/spark/sql/hive/llap/LlapExternalCatalog.scala

@@ -229,7 +229,8 @@ private[spark] class LlapExternalCatalog(
               val nullable = true // Hive cols nullable
               val isSigned = true
               val columnType =
-                DefaultJDBCWrapper.getCatalystType(dataType, typeName, fieldSize, fieldScale, isSigned)
+                DefaultJDBCWrapper.getCatalystType(
+                  dataType, typeName, fieldSize, fieldScale, isSigned)
               schema.add(columnName, columnType, nullable)
             }
             CatalogTable(