[SPARK-16781][PYSPARK] java launched by PySpark as gateway may not be the same...

[SPARK-16781][PYSPARK] java launched by PySpark as gateway may not be the same java used in the spark environment

## What changes were proposed in this pull request?

Update to py4j 0.10.3 to enable JAVA_HOME support

## How was this patch tested?

Pyspark tests

Author: Sean Owen <sowen@cloudera.com>

Closes #14748 from srowen/SPARK-16781.

LICENSE

@@ -263,7 +263,7 @@ The text of each license is also included at licenses/LICENSE-[project].txt.
      (New BSD license) Protocol Buffer Java API (org.spark-project.protobuf:protobuf-java:2.4.1-shaded - http://code.google.com/p/protobuf)
      (The BSD License) Fortran to Java ARPACK (net.sourceforge.f2j:arpack_combined_all:0.1 - http://f2j.sourceforge.net)
      (The BSD License) xmlenc Library (xmlenc:xmlenc:0.52 - http://xmlenc.sourceforge.net)
-     (The New BSD License) Py4J (net.sf.py4j:py4j:0.10.1 - http://py4j.sourceforge.net/)
+     (The New BSD License) Py4J (net.sf.py4j:py4j:0.10.3 - http://py4j.sourceforge.net/)
      (Two-clause BSD-style license) JUnit-Interface (com.novocode:junit-interface:0.10 - http://github.com/szeiger/junit-interface/)
      (BSD licence) sbt and sbt-launch-lib.bash
      (BSD 3 Clause) d3.min.js (https://github.com/mbostock/d3/blob/master/LICENSE)

bin/pyspark

@@ -57,7 +57,7 @@ export PYSPARK_PYTHON
 
 # Add the PySpark classes to the Python path:
 export PYTHONPATH="${SPARK_HOME}/python/:$PYTHONPATH"
-export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.1-src.zip:$PYTHONPATH"
+export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.3-src.zip:$PYTHONPATH"
 
 # Load the PySpark shell.py script when ./pyspark is used interactively:
 export OLD_PYTHONSTARTUP="$PYTHONSTARTUP"

bin/pyspark2.cmd

@@ -30,7 +30,7 @@ if "x%PYSPARK_DRIVER_PYTHON%"=="x" (
 )
 
 set PYTHONPATH=%SPARK_HOME%\python;%PYTHONPATH%
-set PYTHONPATH=%SPARK_HOME%\python\lib\py4j-0.10.1-src.zip;%PYTHONPATH%
+set PYTHONPATH=%SPARK_HOME%\python\lib\py4j-0.10.3-src.zip;%PYTHONPATH%
 
 set OLD_PYTHONSTARTUP=%PYTHONSTARTUP%
 set PYTHONSTARTUP=%SPARK_HOME%\python\pyspark\shell.py

core/pom.xml

@@ -326,7 +326,7 @@
     <dependency>
       <groupId>net.sf.py4j</groupId>
       <artifactId>py4j</artifactId>
-      <version>0.10.1</version>
+      <version>0.10.3</version>
     </dependency>
     <dependency>
       <groupId>org.apache.spark</groupId>

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

@@ -32,7 +32,7 @@ private[spark] object PythonUtils {
     val pythonPath = new ArrayBuffer[String]
     for (sparkHome <- sys.env.get("SPARK_HOME")) {
       pythonPath += Seq(sparkHome, "python", "lib", "pyspark.zip").mkString(File.separator)
-      pythonPath += Seq(sparkHome, "python", "lib", "py4j-0.10.1-src.zip").mkString(File.separator)
+      pythonPath += Seq(sparkHome, "python", "lib", "py4j-0.10.3-src.zip").mkString(File.separator)
     }
     pythonPath ++= SparkContext.jarOfObject(this)
     pythonPath.mkString(File.pathSeparator)

dev/deps/spark-deps-hadoop-2.2

@@ -139,7 +139,7 @@ parquet-jackson-1.8.1.jar
 pmml-model-1.2.15.jar
 pmml-schema-1.2.15.jar
 protobuf-java-2.5.0.jar
-py4j-0.10.1.jar
+py4j-0.10.3.jar
 pyrolite-4.9.jar
 scala-compiler-2.11.8.jar
 scala-library-2.11.8.jar

dev/deps/spark-deps-hadoop-2.3

@@ -146,7 +146,7 @@ parquet-jackson-1.8.1.jar
 pmml-model-1.2.15.jar
 pmml-schema-1.2.15.jar
 protobuf-java-2.5.0.jar
-py4j-0.10.1.jar
+py4j-0.10.3.jar
 pyrolite-4.9.jar
 scala-compiler-2.11.8.jar
 scala-library-2.11.8.jar

dev/deps/spark-deps-hadoop-2.4

@@ -146,7 +146,7 @@ parquet-jackson-1.8.1.jar
 pmml-model-1.2.15.jar
 pmml-schema-1.2.15.jar
 protobuf-java-2.5.0.jar
-py4j-0.10.1.jar
+py4j-0.10.3.jar
 pyrolite-4.9.jar
 scala-compiler-2.11.8.jar
 scala-library-2.11.8.jar

dev/deps/spark-deps-hadoop-2.6

@@ -154,7 +154,7 @@ parquet-jackson-1.8.1.jar
 pmml-model-1.2.15.jar
 pmml-schema-1.2.15.jar
 protobuf-java-2.5.0.jar
-py4j-0.10.1.jar
+py4j-0.10.3.jar
 pyrolite-4.9.jar
 scala-compiler-2.11.8.jar
 scala-library-2.11.8.jar

dev/deps/spark-deps-hadoop-2.7

@@ -155,7 +155,7 @@ parquet-jackson-1.8.1.jar
 pmml-model-1.2.15.jar
 pmml-schema-1.2.15.jar
 protobuf-java-2.5.0.jar
-py4j-0.10.1.jar
+py4j-0.10.3.jar
 pyrolite-4.9.jar
 scala-compiler-2.11.8.jar
 scala-library-2.11.8.jar

python/docs/Makefile

@@ -7,7 +7,7 @@ SPHINXBUILD   ?= sphinx-build
 PAPER         ?=
 BUILDDIR      ?= _build
 
-export PYTHONPATH=$(realpath ..):$(realpath ../lib/py4j-0.10.1-src.zip)
+export PYTHONPATH=$(realpath ..):$(realpath ../lib/py4j-0.10.3-src.zip)
 
 # User-friendly check for sphinx-build
 ifeq ($(shell which $(SPHINXBUILD) >/dev/null 2>&1; echo $$?), 1)

sbin/spark-config.sh

@@ -27,4 +27,4 @@ fi
 export SPARK_CONF_DIR="${SPARK_CONF_DIR:-"${SPARK_HOME}/conf"}"
 # Add the PySpark classes to the PYTHONPATH:
 export PYTHONPATH="${SPARK_HOME}/python:${PYTHONPATH}"
-export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.1-src.zip:${PYTHONPATH}"
+export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.3-src.zip:${PYTHONPATH}"

yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala

@@ -1172,10 +1172,10 @@ private[spark] class Client(
         val pyLibPath = Seq(sys.env("SPARK_HOME"), "python", "lib").mkString(File.separator)
         val pyArchivesFile = new File(pyLibPath, "pyspark.zip")
         require(pyArchivesFile.exists(),
-          "pyspark.zip not found; cannot run pyspark application in YARN mode.")
-        val py4jFile = new File(pyLibPath, "py4j-0.10.1-src.zip")
+          s"$pyArchivesFile not found; cannot run pyspark application in YARN mode.")
+        val py4jFile = new File(pyLibPath, "py4j-0.10.3-src.zip")
         require(py4jFile.exists(),
-          "py4j-0.10.1-src.zip not found; cannot run pyspark application in YARN mode.")
+          s"$py4jFile not found; cannot run pyspark application in YARN mode.")
         Seq(pyArchivesFile.getAbsolutePath(), py4jFile.getAbsolutePath())
       }
   }

yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala

@@ -242,7 +242,7 @@ class YarnClusterSuite extends BaseYarnClusterSuite {
     // needed locations.
     val sparkHome = sys.props("spark.test.home")
     val pythonPath = Seq(
-        s"$sparkHome/python/lib/py4j-0.10.1-src.zip",
+        s"$sparkHome/python/lib/py4j-0.10.3-src.zip",
         s"$sparkHome/python")
     val extraEnvVars = Map(
       "PYSPARK_ARCHIVES_PATH" -> pythonPath.map("local:" + _).mkString(File.pathSeparator),