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

Conflicts:
	python/pyspark/worker.py

Author: davies

.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.

assembly/pom.xml

@@ -141,7 +141,9 @@
                     <include>com.google.common.**</include>
                   </includes>
                   <excludes>
-                    <exclude>com.google.common.base.Optional**</exclude>
+                    <exclude>com/google/common/base/Absent*</exclude>
+                    <exclude>com/google/common/base/Optional*</exclude>
+                    <exclude>com/google/common/base/Present*</exclude>
                   </excludes>
                 </relocation>
               </relocations>

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/pom.xml

@@ -343,7 +343,9 @@
                 <filter>
                   <artifact>com.google.guava:guava</artifact>
                   <includes>
+                    <include>com/google/common/base/Absent*</include>
                     <include>com/google/common/base/Optional*</include>
+                    <include>com/google/common/base/Present*</include>
                   </includes>
                 </filter>
               </filters>

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

@@ -1030,28 +1030,40 @@ class SparkContext(config: SparkConf) extends Logging {
   }
 
   /**
-   * Support function for API backtraces.
+   * Set the thread-local property for overriding the call sites
+   * of actions and RDDs.
    */
-  def setCallSite(site: String) {
-    setLocalProperty("externalCallSite", site)
+  def setCallSite(shortCallSite: String) {
+    setLocalProperty(CallSite.SHORT_FORM, shortCallSite)
   }
 
   /**
-   * Support function for API backtraces.
+   * Set the thread-local property for overriding the call sites
+   * of actions and RDDs.
+   */
+  private[spark] def setCallSite(callSite: CallSite) {
+    setLocalProperty(CallSite.SHORT_FORM, callSite.shortForm)
+    setLocalProperty(CallSite.LONG_FORM, callSite.longForm)
+  }
+
+  /**
+   * Clear the thread-local property for overriding the call sites
+   * of actions and RDDs.
    */
   def clearCallSite() {
-    setLocalProperty("externalCallSite", null)
+    setLocalProperty(CallSite.SHORT_FORM, null)
+    setLocalProperty(CallSite.LONG_FORM, null)
   }
 
   /**
    * Capture the current user callsite and return a formatted version for printing. If the user
-   * has overridden the call site, this will return the user's version.
+   * has overridden the call site using `setCallSite()`, this will return the user's version.
    */
   private[spark] def getCallSite(): CallSite = {
-    Option(getLocalProperty("externalCallSite")) match {
-      case Some(callSite) => CallSite(callSite, longForm = "")
-      case None => Utils.getCallSite
-    }
+    Option(getLocalProperty(CallSite.SHORT_FORM)).map { case shortCallSite =>
+      val longCallSite = Option(getLocalProperty(CallSite.LONG_FORM)).getOrElse("")
+      CallSite(shortCallSite, longCallSite)
+    }.getOrElse(Utils.getCallSite())
   }
 
   /**

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)
       }