Merge branch 'master' of https://github.com/apache/spark

Author: zhzhan

.gitignore

@@ -1,9 +1,12 @@
 *~
+*.#*
+*#*#
 *.swp
 *.ipr
 *.iml
 *.iws
 .idea/
+.idea_modules/
 sbt/*.jar
 .settings
 .cache
@@ -16,6 +19,7 @@ third_party/libmesos.so
 third_party/libmesos.dylib
 conf/java-opts
 conf/*.sh
+conf/*.cmd
 conf/*.properties
 conf/*.conf
 conf/*.xml

.rat-excludes

@@ -20,6 +20,7 @@ log4j.properties.template
 metrics.properties.template
 slaves
 spark-env.sh
+spark-env.cmd
 spark-env.sh.template
 log4j-defaults.properties
 bootstrap-tooltip.js
@@ -58,3 +59,4 @@ dist/*
 .*iws
 logs
 .*scalastyle-output.xml
+.*dependency-reduced-pom.xml

CONTRIBUTING.md

@@ -8,5 +8,5 @@ submitting any copyrighted material via pull request, email, or other means
 you agree to license the material under the project's open source license and
 warrant that you have the legal authority to do so.
 
-Please see [Contributing to Spark wiki page](https://cwiki.apache.org/SPARK/Contributing+to+Spark)
+Please see the [Contributing to Spark wiki page](https://cwiki.apache.org/SPARK/Contributing+to+Spark)
 for more information.

bin/spark-sql

@@ -24,7 +24,7 @@
 set -o posix
 
 CLASS="org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver"
-CLASS_NOT_FOUND_EXIT_STATUS=1
+CLASS_NOT_FOUND_EXIT_STATUS=101
 
 # Figure out where Spark is installed
 FWDIR="$(cd "`dirname "$0"`"/..; pwd)"

core/src/main/scala/org/apache/spark/SparkHadoopWriter.scala

@@ -116,7 +116,7 @@ class SparkHadoopWriter(@transient jobConf: JobConf)
         }
       }
     } else {
-      logWarning ("No need to commit output of task: " + taID.value)
+      logInfo ("No need to commit output of task: " + taID.value)
     }
   }
 

core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala

@@ -775,17 +775,36 @@ private[spark] object PythonRDD extends Logging {
     }.toJavaRDD()
   }
 
+  private class AutoBatchedPickler(iter: Iterator[Any]) extends Iterator[Array[Byte]] {
+    private val pickle = new Pickler()
+    private var batch = 1
+    private val buffer = new mutable.ArrayBuffer[Any]
+
+    override def hasNext(): Boolean = iter.hasNext
+
+    override def next(): Array[Byte] = {
+      while (iter.hasNext && buffer.length < batch) {
+        buffer += iter.next()
+      }
+      val bytes = pickle.dumps(buffer.toArray)
+      val size = bytes.length
+      // let  1M < size < 10M
+      if (size < 1024 * 1024) {
+        batch *= 2
+      } else if (size > 1024 * 1024 * 10 && batch > 1) {
+        batch /= 2
+      }
+      buffer.clear()
+      bytes
+    }
+  }
+
   /**
    * Convert an RDD of Java objects to an RDD of serialized Python objects, that is usable by
    * PySpark.
    */
   def javaToPython(jRDD: JavaRDD[Any]): JavaRDD[Array[Byte]] = {
-    jRDD.rdd.mapPartitions { iter =>
-      val pickle = new Pickler
-      iter.map { row =>
-        pickle.dumps(row)
-      }
-    }
+    jRDD.rdd.mapPartitions { iter => new AutoBatchedPickler(iter) }
   }
 
   /**

core/src/main/scala/org/apache/spark/api/python/SerDeUtil.scala

@@ -68,8 +68,8 @@ private[python] object SerDeUtil extends Logging {
         construct(args ++ Array(""))
       } else if (args.length == 2 && args(1).isInstanceOf[String]) {
         val typecode = args(0).asInstanceOf[String].charAt(0)
-        val data: String = args(1).asInstanceOf[String]
-        construct(typecode, machineCodes(typecode), data.getBytes("ISO-8859-1"))
+        val data: Array[Byte] = args(1).asInstanceOf[String].getBytes("ISO-8859-1")
+        construct(typecode, machineCodes(typecode), data)
       } else {
         super.construct(args)
       }

core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala

@@ -54,7 +54,7 @@ object SparkSubmit {
   private val SPARK_SHELL = "spark-shell"
   private val PYSPARK_SHELL = "pyspark-shell"
 
-  private val CLASS_NOT_FOUND_EXIT_STATUS = 1
+  private val CLASS_NOT_FOUND_EXIT_STATUS = 101
 
   // Exposed for testing
   private[spark] var exitFn: () => Unit = () => System.exit(-1)
@@ -172,7 +172,7 @@ object SparkSubmit {
       // All cluster managers
       OptionAssigner(args.master, ALL_CLUSTER_MGRS, ALL_DEPLOY_MODES, sysProp = "spark.master"),
       OptionAssigner(args.name, ALL_CLUSTER_MGRS, ALL_DEPLOY_MODES, sysProp = "spark.app.name"),
-      OptionAssigner(args.jars, ALL_CLUSTER_MGRS, ALL_DEPLOY_MODES, sysProp = "spark.jars"),
+      OptionAssigner(args.jars, ALL_CLUSTER_MGRS, CLIENT, sysProp = "spark.jars"),
       OptionAssigner(args.driverMemory, ALL_CLUSTER_MGRS, CLIENT,
         sysProp = "spark.driver.memory"),
       OptionAssigner(args.driverExtraClassPath, ALL_CLUSTER_MGRS, ALL_DEPLOY_MODES,
@@ -183,6 +183,7 @@ object SparkSubmit {
         sysProp = "spark.driver.extraLibraryPath"),
 
       // Standalone cluster only
+      OptionAssigner(args.jars, STANDALONE, CLUSTER, sysProp = "spark.jars"),
       OptionAssigner(args.driverMemory, STANDALONE, CLUSTER, clOption = "--memory"),
       OptionAssigner(args.driverCores, STANDALONE, CLUSTER, clOption = "--cores"),
 
@@ -261,7 +262,7 @@ object SparkSubmit {
     }
 
     // In yarn-cluster mode, use yarn.Client as a wrapper around the user class
-    if (clusterManager == YARN && deployMode == CLUSTER) {
+    if (isYarnCluster) {
       childMainClass = "org.apache.spark.deploy.yarn.Client"
       if (args.primaryResource != SPARK_INTERNAL) {
         childArgs += ("--jar", args.primaryResource)
@@ -279,7 +280,7 @@ object SparkSubmit {
     }
 
     // Read from default spark properties, if any
-    for ((k, v) <- args.getDefaultSparkProperties) {
+    for ((k, v) <- args.defaultSparkProperties) {
       sysProps.getOrElseUpdate(k, v)
     }
 

core/src/main/scala/org/apache/spark/deploy/SparkSubmitArguments.scala

@@ -57,12 +57,8 @@ private[spark] class SparkSubmitArguments(args: Seq[String]) {
   var pyFiles: String = null
   val sparkProperties: HashMap[String, String] = new HashMap[String, String]()
 
-  parseOpts(args.toList)
-  mergeSparkProperties()
-  checkRequiredArguments()
-
-  /** Return default present in the currently defined defaults file. */
-  def getDefaultSparkProperties = {
+  /** Default properties present in the currently defined defaults file. */
+  lazy val defaultSparkProperties: HashMap[String, String] = {
     val defaultProperties = new HashMap[String, String]()
     if (verbose) SparkSubmit.printStream.println(s"Using properties file: $propertiesFile")
     Option(propertiesFile).foreach { filename =>
@@ -79,6 +75,10 @@ private[spark] class SparkSubmitArguments(args: Seq[String]) {
     defaultProperties
   }
 
+  parseOpts(args.toList)
+  mergeSparkProperties()
+  checkRequiredArguments()
+
   /**
    * Fill in any undefined values based on the default properties file or options passed in through
    * the '--conf' flag.
@@ -107,7 +107,8 @@ private[spark] class SparkSubmitArguments(args: Seq[String]) {
       }
     }
 
-    val properties = getDefaultSparkProperties
+    val properties = HashMap[String, String]()
+    properties.putAll(defaultSparkProperties)
     properties.putAll(sparkProperties)
 
     // Use properties file as fallback for values which have a direct analog to
@@ -213,7 +214,7 @@ private[spark] class SparkSubmitArguments(args: Seq[String]) {
     |  verbose                 $verbose
     |
     |Default properties from $propertiesFile:
-    |${getDefaultSparkProperties.mkString("  ", "\n  ", "\n")}
+    |${defaultSparkProperties.mkString("  ", "\n  ", "\n")}
     """.stripMargin
   }
 

core/src/main/scala/org/apache/spark/deploy/history/HistoryPage.scala

@@ -67,6 +67,7 @@ private[spark] class HistoryPage(parent: HistoryServer) extends WebUIPage("") {
   }
 
   private val appHeader = Seq(
+    "App ID",
     "App Name",
     "Started",
     "Completed",
@@ -81,7 +82,8 @@ private[spark] class HistoryPage(parent: HistoryServer) extends WebUIPage("") {
     val duration = UIUtils.formatDuration(info.endTime - info.startTime)
     val lastUpdated = UIUtils.formatDate(info.lastUpdated)
     <tr>
-      <td><a href={uiAddress}>{info.name}</a></td>
+      <td><a href={uiAddress}>{info.id}</a></td>
+      <td>{info.name}</td>
       <td>{startTime}</td>
       <td>{endTime}</td>
       <td>{duration}</td>