Merge branch 'master' of git://git.apache.org/spark into dsl-missing-operator

Author: sarutak

.gitignore

@@ -49,7 +49,7 @@ dependency-reduced-pom.xml
 checkpoint
 derby.log
 dist/
-spark-*-bin.tar.gz
+spark-*-bin-*.tgz
 unit-tests.log
 /lib/
 rat-results.txt

core/src/main/scala/org/apache/spark/rdd/PipedRDD.scala

@@ -63,7 +63,7 @@ private[spark] class PipedRDD[T: ClassTag](
 
   /**
    * A FilenameFilter that accepts anything that isn't equal to the name passed in.
-   * @param name of file or directory to leave out
+   * @param filterName of file or directory to leave out
    */
   class NotEqualsFileNameFilter(filterName: String) extends FilenameFilter {
     def accept(dir: File, name: String): Boolean = {

core/src/main/scala/org/apache/spark/util/AkkaUtils.scala

@@ -134,9 +134,16 @@ private[spark] object AkkaUtils extends Logging {
     Duration.create(conf.getLong("spark.akka.lookupTimeout", 30), "seconds")
   }
 
+  private val AKKA_MAX_FRAME_SIZE_IN_MB = Int.MaxValue / 1024 / 1024
+
   /** Returns the configured max frame size for Akka messages in bytes. */
   def maxFrameSizeBytes(conf: SparkConf): Int = {
-    conf.getInt("spark.akka.frameSize", 10) * 1024 * 1024
+    val frameSizeInMB = conf.getInt("spark.akka.frameSize", 10)
+    if (frameSizeInMB > AKKA_MAX_FRAME_SIZE_IN_MB) {
+      throw new IllegalArgumentException("spark.akka.frameSize should not be greater than "
+        + AKKA_MAX_FRAME_SIZE_IN_MB + "MB")
+    }
+    frameSizeInMB * 1024 * 1024
   }
 
   /** Space reserved for extra data in an Akka message besides serialized task or task result. */

core/src/test/scala/org/apache/spark/ShuffleSuite.scala

@@ -270,7 +270,6 @@ object ShuffleSuite {
 
   def mergeCombineException(x: Int, y: Int): Int = {
     throw new SparkException("Exception for map-side combine.")
-    x + y
   }
 
   class NonJavaSerializableClass(val value: Int) extends Comparable[NonJavaSerializableClass] {

docs/programming-guide.md

@@ -934,6 +934,12 @@ for details.
   <td> Reshuffle the data in the RDD randomly to create either more or fewer partitions and balance it across them.
     This always shuffles all data over the network. </td>
 </tr>
+<tr>
+  <td> <b>repartitionAndSortWithinPartitions</b>(<i>partitioner</i>) </td>
+  <td> Repartition the RDD according to the given partitioner and, within each resulting partition,
+  sort records by their keys. This is more efficient than calling <code>repartition</code> and then sorting within 
+  each partition because it can push the sorting down into the shuffle machinery. </td>
+</tr>
 </table>
 
 ### Actions
@@ -1177,7 +1183,7 @@ Accumulators are variables that are only "added" to through an associative opera
 therefore be efficiently supported in parallel. They can be used to implement counters (as in
 MapReduce) or sums. Spark natively supports accumulators of numeric types, and programmers
 can add support for new types. If accumulators are created with a name, they will be
-displayed in Spark's UI. This can can be useful for understanding the progress of 
+displayed in Spark's UI. This can be useful for understanding the progress of 
 running stages (NOTE: this is not yet supported in Python).
 
 An accumulator is created from an initial value `v` by calling `SparkContext.accumulator(v)`. Tasks

docs/sql-programming-guide.md

@@ -146,7 +146,7 @@ describes the various methods for loading data into a SchemaRDD.
 
 Spark SQL supports two different methods for converting existing RDDs into SchemaRDDs.  The first
 method uses reflection to infer the schema of an RDD that contains specific types of objects.  This
-reflection based approach leads to more concise code and works well when you already know the schema 
+reflection based approach leads to more concise code and works well when you already know the schema
 while writing your Spark application.
 
 The second method for creating SchemaRDDs is through a programmatic interface that allows you to
@@ -566,7 +566,7 @@ for teenName in teenNames.collect():
 
 ### Configuration
 
-Configuration of Parquet can be done using the `setConf` method on SQLContext or by running 
+Configuration of Parquet can be done using the `setConf` method on SQLContext or by running
 `SET key=value` commands using SQL.
 
 <table class="table">
@@ -575,8 +575,8 @@ Configuration of Parquet can be done using the `setConf` method on SQLContext or
   <td><code>spark.sql.parquet.binaryAsString</code></td>
   <td>false</td>
   <td>
-    Some other Parquet-producing systems, in particular Impala and older versions of Spark SQL, do 
-    not differentiate between binary data and strings when writing out the Parquet schema.  This 
+    Some other Parquet-producing systems, in particular Impala and older versions of Spark SQL, do
+    not differentiate between binary data and strings when writing out the Parquet schema.  This
     flag tells Spark SQL to interpret binary data as a string to provide compatibility with these systems.
   </td>
 </tr>
@@ -591,10 +591,20 @@ Configuration of Parquet can be done using the `setConf` method on SQLContext or
   <td><code>spark.sql.parquet.compression.codec</code></td>
   <td>gzip</td>
   <td>
-    Sets the compression codec use when writing Parquet files. Acceptable values include: 
+    Sets the compression codec use when writing Parquet files. Acceptable values include:
     uncompressed, snappy, gzip, lzo.
   </td>
 </tr>
+<tr>
+  <td><code>spark.sql.parquet.filterPushdown</code></td>
+  <td>false</td>
+  <td>
+    Turn on Parquet filter pushdown optimization. This feature is turned off by default because of a known
+    bug in Paruet 1.6.0rc3 (<a href="https://issues.apache.org/jira/browse/PARQUET-136">PARQUET-136</a>).
+    However, if your table doesn't contain any nullable string or binary columns, it's still safe to turn
+    this feature on.
+  </td>
+</tr>
 <tr>
   <td><code>spark.sql.hive.convertMetastoreParquet</code></td>
   <td>true</td>
@@ -900,7 +910,6 @@ export HIVE_SERVER2_THRIFT_BIND_HOST=<listening-host>
 ./sbin/start-thriftserver.sh \
   --master <master-uri> \
   ...
-```
 {% endhighlight %}
 
 or system properties:
@@ -911,7 +920,6 @@ or system properties:
   --hiveconf hive.server2.thrift.bind.host=<listening-host> \
   --master <master-uri>
   ...
-```
 {% endhighlight %}
 
 Now you can use beeline to test the Thrift JDBC/ODBC server:
@@ -947,7 +955,7 @@ options.
 
 ## Migration Guide for Shark User
 
-### Scheduling 
+### Scheduling
 To set a [Fair Scheduler](job-scheduling.html#fair-scheduler-pools) pool for a JDBC client session,
 users can set the `spark.sql.thriftserver.scheduler.pool` variable:
 
@@ -994,7 +1002,7 @@ Several caching related features are not supported yet:
 ## Compatibility with Apache Hive
 
 Spark SQL is designed to be compatible with the Hive Metastore, SerDes and UDFs.  Currently Spark
-SQL is based on Hive 0.12.0.
+SQL is based on Hive 0.12.0 and 0.13.1.
 
 #### Deploying in Existing Hive Warehouses
 
@@ -1033,6 +1041,7 @@ Spark SQL supports the vast majority of Hive features, such as:
 * Sampling
 * Explain
 * Partitioned tables
+* View
 * All Hive DDL Functions, including:
   * `CREATE TABLE`
   * `CREATE TABLE AS SELECT`
@@ -1048,6 +1057,7 @@ Spark SQL supports the vast majority of Hive features, such as:
   * `STRING`
   * `BINARY`
   * `TIMESTAMP`
+  * `DATE`
   * `ARRAY<>`
   * `MAP<>`
   * `STRUCT<>`
@@ -1148,6 +1158,7 @@ evaluated by the SQL execution engine.  A full list of the functions supported c
 * Datetime type
     - `TimestampType`: Represents values comprising values of fields year, month, day,
     hour, minute, and second.
+    - `DateType`: Represents values comprising values of fields year, month, day.
 * Complex types
     - `ArrayType(elementType, containsNull)`: Represents values comprising a sequence of
     elements with the type of `elementType`. `containsNull` is used to indicate if
@@ -1255,6 +1266,13 @@ import  org.apache.spark.sql._
   TimestampType
   </td>
 </tr>
+<tr>
+  <td> <b>DateType</b> </td>
+  <td> java.sql.Date </td>
+  <td>
+  DateType
+  </td>
+</tr>
 <tr>
   <td> <b>ArrayType</b> </td>
   <td> scala.collection.Seq </td>
@@ -1381,6 +1399,13 @@ please use factory methods provided in
   DataType.TimestampType
   </td>
 </tr>
+<tr>
+  <td> <b>DateType</b> </td>
+  <td> java.sql.Date </td>
+  <td>
+  DataType.DateType
+  </td>
+</tr>
 <tr>
   <td> <b>ArrayType</b> </td>
   <td> java.util.List </td>
@@ -1528,6 +1553,13 @@ from pyspark.sql import *
   TimestampType()
   </td>
 </tr>
+<tr>
+  <td> <b>DateType</b> </td>
+  <td> datetime.date </td>
+  <td>
+  DateType()
+  </td>
+</tr>
 <tr>
   <td> <b>ArrayType</b> </td>
   <td> list, tuple, or array </td>

examples/src/main/scala/org/apache/spark/examples/sql/hive/HiveFromSpark.scala

@@ -29,9 +29,10 @@ object HiveFromSpark {
     val sc = new SparkContext(sparkConf)
     val path = s"${System.getenv("SPARK_HOME")}/examples/src/main/resources/kv1.txt"
 
-    // A local hive context creates an instance of the Hive Metastore in process, storing 
-    // the warehouse data in the current directory.  This location can be overridden by
-    // specifying a second parameter to the constructor.
+    // A hive context adds support for finding tables in the MetaStore and writing queries
+    // using HiveQL. Users who do not have an existing Hive deployment can still create a
+    // HiveContext. When not configured by the hive-site.xml, the context automatically
+    // creates metastore_db and warehouse in the current directory.
     val hiveContext = new HiveContext(sc)
     import hiveContext._
 

external/mqtt/pom.xml

@@ -43,8 +43,8 @@
     </dependency>
     <dependency>
       <groupId>org.eclipse.paho</groupId>
-      <artifactId>mqtt-client</artifactId>
-       <version>0.4.0</version>
+      <artifactId>org.eclipse.paho.client.mqttv3</artifactId>
+      <version>1.0.1</version>
     </dependency>
     <dependency>
       <groupId>org.scalatest</groupId>

make-distribution.sh

@@ -201,6 +201,9 @@ if [ -e "$FWDIR"/CHANGES.txt ]; then
   cp "$FWDIR/CHANGES.txt" "$DISTDIR"
 fi
 
+# Copy data files
+cp -r "$FWDIR/data" "$DISTDIR"
+
 # Copy other things
 mkdir "$DISTDIR"/conf
 cp "$FWDIR"/conf/*.template "$DISTDIR"/conf

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SparkSQLParser.scala

@@ -97,10 +97,10 @@ class SqlLexical(val keywords: Seq[String]) extends StdLexical {
 
   /** Generate all variations of upper and lower case of a given string */
   def allCaseVersions(s: String, prefix: String = ""): Stream[String] = {
-    if (s == "") {
+    if (s.isEmpty) {
       Stream(prefix)
     } else {
-      allCaseVersions(s.tail, prefix + s.head.toLower) ++
+      allCaseVersions(s.tail, prefix + s.head.toLower) #:::
         allCaseVersions(s.tail, prefix + s.head.toUpper)
     }
   }