Fix computation of classpath when we launch java directly

The previous version assumed that a CLASSPATH environment variable was
set by the "run" script when launching the process that starts the
ExecutorRunner, but unfortunately this is not true in tests. Instead, we
factor the classpath calculation into an extenral script and call that.

NOTE: This includes a Windows version but hasn't yet been tested there.

6 files changed, 189 insertions, 107 deletions

diff --git a/bin/compute-classpath.cmd b/bin/compute-classpath.cmd
new file mode 100644
index 0000000000..1dff8fea22
--- /dev/null
+++ b/bin/compute-classpath.cmd
@@ -0,0 +1,52 @@
+@echo off
+
+rem This script computes Spark's classpath and prints it to stdout; it's used by both the "run"
+rem script and the ExecutorRunner in standalone cluster mode.
+
+set SCALA_VERSION=2.9.3
+
+rem Figure out where the Spark framework is installed
+set FWDIR=%~dp0\..
+
+rem Load environment variables from conf\spark-env.cmd, if it exists
+if exist "%FWDIR%conf\spark-env.cmd" call "%FWDIR%conf\spark-env.cmd"
+
+set CORE_DIR=%FWDIR%core
+set REPL_DIR=%FWDIR%repl
+set EXAMPLES_DIR=%FWDIR%examples
+set BAGEL_DIR=%FWDIR%bagel
+set STREAMING_DIR=%FWDIR%streaming
+set PYSPARK_DIR=%FWDIR%python
+
+rem Build up classpath
+set CLASSPATH=%SPARK_CLASSPATH%;%MESOS_CLASSPATH%;%FWDIR%conf;%CORE_DIR%\target\scala-%SCALA_VERSION%\classes
+set CLASSPATH=%CLASSPATH%;%CORE_DIR%\target\scala-%SCALA_VERSION%\test-classes;%CORE_DIR%\src\main\resources
+set CLASSPATH=%CLASSPATH%;%STREAMING_DIR%\target\scala-%SCALA_VERSION%\classes;%STREAMING_DIR%\target\scala-%SCALA_VERSION%\test-classes
+set CLASSPATH=%CLASSPATH%;%STREAMING_DIR%\lib\org\apache\kafka\kafka\0.7.2-spark\*
+set CLASSPATH=%CLASSPATH%;%REPL_DIR%\target\scala-%SCALA_VERSION%\classes;%EXAMPLES_DIR%\target\scala-%SCALA_VERSION%\classes
+set CLASSPATH=%CLASSPATH%;%FWDIR%lib_managed\jars\*
+set CLASSPATH=%CLASSPATH%;%FWDIR%lib_managed\bundles\*
+set CLASSPATH=%CLASSPATH%;%FWDIR%repl\lib\*
+set CLASSPATH=%CLASSPATH%;%FWDIR%python\lib\*
+set CLASSPATH=%CLASSPATH%;%BAGEL_DIR%\target\scala-%SCALA_VERSION%\classes
+
+rem Add hadoop conf dir - else FileSystem.*, etc fail
+rem Note, this assumes that there is either a HADOOP_CONF_DIR or YARN_CONF_DIR which hosts
+rem the configurtion files.
+if "x%HADOOP_CONF_DIR%"=="x" goto no_hadoop_conf_dir
+  set CLASSPATH=%CLASSPATH%;%HADOOP_CONF_DIR%
+:no_hadoop_conf_dir
+
+if "x%YARN_CONF_DIR%"=="x" goto no_yarn_conf_dir
+  set CLASSPATH=%CLASSPATH%;%YARN_CONF_DIR%
+:no_yarn_conf_dir
+
+rem Add Scala standard library
+set CLASSPATH=%CLASSPATH%;%SCALA_HOME%\lib\scala-library.jar;%SCALA_HOME%\lib\scala-compiler.jar;%SCALA_HOME%\lib\jline.jar
+
+rem A bit of a hack to allow calling this script within run2.cmd without seeing output
+if "x%DONT_PRINT_CLASSPATH%"=="x1" goto exit
+
+echo %CLASSPATH%
+
+:exit
diff --git a/bin/compute-classpath.sh b/bin/compute-classpath.sh
new file mode 100755
index 0000000000..3a78880290
--- /dev/null
+++ b/bin/compute-classpath.sh
@@ -0,0 +1,89 @@
+#!/bin/bash
+
+# This script computes Spark's classpath and prints it to stdout; it's used by both the "run"
+# script and the ExecutorRunner in standalone cluster mode.
+
+SCALA_VERSION=2.9.3
+
+# Figure out where Spark is installed
+FWDIR="$(cd `dirname $0`/..; pwd)"
+
+# Load environment variables from conf/spark-env.sh, if it exists
+if [ -e $FWDIR/conf/spark-env.sh ] ; then
+  . $FWDIR/conf/spark-env.sh
+fi
+
+CORE_DIR="$FWDIR/core"
+REPL_DIR="$FWDIR/repl"
+REPL_BIN_DIR="$FWDIR/repl-bin"
+EXAMPLES_DIR="$FWDIR/examples"
+BAGEL_DIR="$FWDIR/bagel"
+STREAMING_DIR="$FWDIR/streaming"
+PYSPARK_DIR="$FWDIR/python"
+
+# Build up classpath
+CLASSPATH="$SPARK_CLASSPATH"
+CLASSPATH="$CLASSPATH:$FWDIR/conf"
+CLASSPATH="$CLASSPATH:$CORE_DIR/target/scala-$SCALA_VERSION/classes"
+if [ -n "$SPARK_TESTING" ] ; then
+  CLASSPATH="$CLASSPATH:$CORE_DIR/target/scala-$SCALA_VERSION/test-classes"
+  CLASSPATH="$CLASSPATH:$STREAMING_DIR/target/scala-$SCALA_VERSION/test-classes"
+fi
+CLASSPATH="$CLASSPATH:$CORE_DIR/src/main/resources"
+CLASSPATH="$CLASSPATH:$REPL_DIR/target/scala-$SCALA_VERSION/classes"
+CLASSPATH="$CLASSPATH:$EXAMPLES_DIR/target/scala-$SCALA_VERSION/classes"
+CLASSPATH="$CLASSPATH:$STREAMING_DIR/target/scala-$SCALA_VERSION/classes"
+CLASSPATH="$CLASSPATH:$STREAMING_DIR/lib/org/apache/kafka/kafka/0.7.2-spark/*" # <-- our in-project Kafka Jar
+if [ -e "$FWDIR/lib_managed" ]; then
+  CLASSPATH="$CLASSPATH:$FWDIR/lib_managed/jars/*"
+  CLASSPATH="$CLASSPATH:$FWDIR/lib_managed/bundles/*"
+fi
+CLASSPATH="$CLASSPATH:$REPL_DIR/lib/*"
+# Add the shaded JAR for Maven builds
+if [ -e $REPL_BIN_DIR/target ]; then
+  for jar in `find "$REPL_BIN_DIR/target" -name 'spark-repl-*-shaded-hadoop*.jar'`; do
+    CLASSPATH="$CLASSPATH:$jar"
+  done
+  # The shaded JAR doesn't contain examples, so include those separately
+  EXAMPLES_JAR=`ls "$EXAMPLES_DIR/target/spark-examples"*[0-9T].jar`
+  CLASSPATH+=":$EXAMPLES_JAR"
+fi
+CLASSPATH="$CLASSPATH:$BAGEL_DIR/target/scala-$SCALA_VERSION/classes"
+for jar in `find $PYSPARK_DIR/lib -name '*jar'`; do
+  CLASSPATH="$CLASSPATH:$jar"
+done
+
+# Figure out the JAR file that our examples were packaged into. This includes a bit of a hack
+# to avoid the -sources and -doc packages that are built by publish-local.
+if [ -e "$EXAMPLES_DIR/target/scala-$SCALA_VERSION/spark-examples"*[0-9T].jar ]; then
+  # Use the JAR from the SBT build
+  export SPARK_EXAMPLES_JAR=`ls "$EXAMPLES_DIR/target/scala-$SCALA_VERSION/spark-examples"*[0-9T].jar`
+fi
+if [ -e "$EXAMPLES_DIR/target/spark-examples"*[0-9T].jar ]; then
+  # Use the JAR from the Maven build
+  export SPARK_EXAMPLES_JAR=`ls "$EXAMPLES_DIR/target/spark-examples"*[0-9T].jar`
+fi
+
+# Add hadoop conf dir - else FileSystem.*, etc fail !
+# Note, this assumes that there is either a HADOOP_CONF_DIR or YARN_CONF_DIR which hosts
+# the configurtion files.
+if [ "x" != "x$HADOOP_CONF_DIR" ]; then
+  CLASSPATH="$CLASSPATH:$HADOOP_CONF_DIR"
+fi
+if [ "x" != "x$YARN_CONF_DIR" ]; then
+  CLASSPATH="$CLASSPATH:$YARN_CONF_DIR"
+fi
+
+# Add Scala standard library
+if [ -z "$SCALA_LIBRARY_PATH" ]; then
+  if [ -z "$SCALA_HOME" ]; then
+    echo "SCALA_HOME is not set" >&2
+    exit 1
+  fi
+  SCALA_LIBRARY_PATH="$SCALA_HOME/lib"
+fi
+CLASSPATH="$CLASSPATH:$SCALA_LIBRARY_PATH/scala-library.jar"
+CLASSPATH="$CLASSPATH:$SCALA_LIBRARY_PATH/scala-compiler.jar"
+CLASSPATH="$CLASSPATH:$SCALA_LIBRARY_PATH/jline.jar"
+
+echo "$CLASSPATH"
diff --git a/core/src/main/scala/spark/Utils.scala b/core/src/main/scala/spark/Utils.scala
index bdc1494cc9..f41efa9d29 100644
--- a/core/src/main/scala/spark/Utils.scala
+++ b/core/src/main/scala/spark/Utils.scala
@@ -522,6 +522,37 @@ private object Utils extends Logging {
     execute(command, new File("."))
   }
 
+  /**
+   * Execute a command and get its output, throwing an exception if it yields a code other than 0.
+   */
+  def executeAndGetOutput(command: Seq[String], workingDir: File = new File(".")): String = {
+    val process = new ProcessBuilder(command: _*)
+        .directory(workingDir)
+        .start()
+    new Thread("read stderr for " + command(0)) {
+      override def run() {
+        for (line <- Source.fromInputStream(process.getErrorStream).getLines) {
+          System.err.println(line)
+        }
+      }
+    }.start()
+    val output = new StringBuffer
+    val stdoutThread = new Thread("read stdout for " + command(0)) {
+      override def run() {
+        for (line <- Source.fromInputStream(process.getInputStream).getLines) {
+          output.append(line)
+        }
+      }
+    }
+    stdoutThread.start()
+    val exitCode = process.waitFor()
+    stdoutThread.join()   // Wait for it to finish reading output
+    if (exitCode != 0) {
+      throw new SparkException("Process " + command + " exited with code " + exitCode)
+    }
+    output.toString
+  }
+
   private[spark] class CallSiteInfo(val lastSparkMethod: String, val firstUserFile: String,
                                     val firstUserLine: Int, val firstUserClass: String)
   /**
diff --git a/core/src/main/scala/spark/deploy/worker/ExecutorRunner.scala b/core/src/main/scala/spark/deploy/worker/ExecutorRunner.scala
index db580e39ab..4f8e1dcb26 100644
--- a/core/src/main/scala/spark/deploy/worker/ExecutorRunner.scala
+++ b/core/src/main/scala/spark/deploy/worker/ExecutorRunner.scala
@@ -107,21 +107,9 @@ private[spark] class ExecutorRunner(
 
     val memoryOpts = Seq("-Xms" + memory + "M", "-Xmx" + memory + "M")
 
-    var classPath = System.getenv("CLASSPATH")
-    if (System.getenv("SPARK_LAUNCH_WITH_SCALA") == "1") {
-      // Add the Scala library JARs to the classpath; this is needed when the ExecutorRunner
-      // was launched with "scala" as the runner (e.g. in spark-shell in local-cluster mode)
-      // and the Scala libraries won't be in the CLASSPATH environment variable by defalt.
-      if (System.getenv("SCALA_LIBRARY_PATH") == null && System.getenv("SCALA_HOME") == null) {
-        logError("Cloud not launch executors: neither SCALA_LIBRARY_PATH nor SCALA_HOME are set")
-        System.exit(1)
-      }
-      val scalaLib = Option(System.getenv("SCALA_LIBRARY_PATH")).getOrElse(
-        System.getenv("SCALA_HOME") + "/lib")
-      classPath += ":" + scalaLib + "/scala-library.jar" +
-                   ":" + scalaLib + "/scala-compiler.jar" +
-                   ":" + scalaLib + "/jline.jar"
-    }
+    // Figure out our classpath with the external compute-classpath script
+    val ext = if (System.getProperty("os.name").startsWith("Windows")) ".cmd" else ".sh"
+    val classPath = Utils.executeAndGetOutput(Seq(sparkHome + "/bin/compute-classpath" + ext))
 
     Seq("-cp", classPath) ++ libraryOpts ++ userOpts ++ memoryOpts
   }
@@ -154,7 +142,6 @@ private[spark] class ExecutorRunner(
 
       // Launch the process
       val command = buildCommandSeq()
-      println("COMMAND: " + command.mkString(" "))
       val builder = new ProcessBuilder(command: _*).directory(executorDir)
       val env = builder.environment()
       for ((key, value) <- appDesc.command.environment) {
diff --git a/run b/run
index 0fb15f8b24..7c06a55062 100755
--- a/run
+++ b/run
@@ -49,6 +49,12 @@ case "$1" in
     ;;
 esac
 
+# Figure out whether to run our class with java or with the scala launcher.
+# In most cases, we'd prefer to execute our process with java because scala
+# creates a shell script as the parent of its Java process, which makes it
+# hard to kill the child with stuff like Process.destroy(). However, for
+# the Spark shell, the wrapper is necessary to properly reset the terminal
+# when we exit, so we allow it to set a variable to launch with scala.
 if [ "$SPARK_LAUNCH_WITH_SCALA" == "1" ]; then
   if [ "$SCALA_HOME" ]; then
     RUNNER="${SCALA_HOME}/bin/scala"
@@ -98,12 +104,8 @@ export JAVA_OPTS
 # Attention: when changing the way the JAVA_OPTS are assembled, the change must be reflected in ExecutorRunner.scala!
 
 CORE_DIR="$FWDIR/core"
-REPL_DIR="$FWDIR/repl"
-REPL_BIN_DIR="$FWDIR/repl-bin"
 EXAMPLES_DIR="$FWDIR/examples"
-BAGEL_DIR="$FWDIR/bagel"
-STREAMING_DIR="$FWDIR/streaming"
-PYSPARK_DIR="$FWDIR/python"
+REPL_DIR="$FWDIR/repl"
 
 # Exit if the user hasn't compiled Spark
 if [ ! -e "$CORE_DIR/target" ]; then
@@ -118,37 +120,9 @@ if [[ "$@" = *repl* && ! -e "$REPL_DIR/target" ]]; then
   exit 1
 fi
 
-# Build up classpath
-CLASSPATH="$SPARK_CLASSPATH"
-CLASSPATH="$CLASSPATH:$FWDIR/conf"
-CLASSPATH="$CLASSPATH:$CORE_DIR/target/scala-$SCALA_VERSION/classes"
-if [ -n "$SPARK_TESTING" ] ; then
-  CLASSPATH="$CLASSPATH:$CORE_DIR/target/scala-$SCALA_VERSION/test-classes"
-  CLASSPATH="$CLASSPATH:$STREAMING_DIR/target/scala-$SCALA_VERSION/test-classes"
-fi
-CLASSPATH="$CLASSPATH:$CORE_DIR/src/main/resources"
-CLASSPATH="$CLASSPATH:$REPL_DIR/target/scala-$SCALA_VERSION/classes"
-CLASSPATH="$CLASSPATH:$EXAMPLES_DIR/target/scala-$SCALA_VERSION/classes"
-CLASSPATH="$CLASSPATH:$STREAMING_DIR/target/scala-$SCALA_VERSION/classes"
-CLASSPATH="$CLASSPATH:$STREAMING_DIR/lib/org/apache/kafka/kafka/0.7.2-spark/*" # <-- our in-project Kafka Jar
-if [ -e "$FWDIR/lib_managed" ]; then
-  CLASSPATH="$CLASSPATH:$FWDIR/lib_managed/jars/*"
-  CLASSPATH="$CLASSPATH:$FWDIR/lib_managed/bundles/*"
-fi
-CLASSPATH="$CLASSPATH:$REPL_DIR/lib/*"
-# Add the shaded JAR for Maven builds
-if [ -e $REPL_BIN_DIR/target ]; then
-  for jar in `find "$REPL_BIN_DIR/target" -name 'spark-repl-*-shaded-hadoop*.jar'`; do
-    CLASSPATH="$CLASSPATH:$jar"
-  done
-  # The shaded JAR doesn't contain examples, so include those separately
-  EXAMPLES_JAR=`ls "$EXAMPLES_DIR/target/spark-examples"*[0-9T].jar`
-  CLASSPATH+=":$EXAMPLES_JAR"
-fi
-CLASSPATH="$CLASSPATH:$BAGEL_DIR/target/scala-$SCALA_VERSION/classes"
-for jar in `find $PYSPARK_DIR/lib -name '*jar'`; do
-  CLASSPATH="$CLASSPATH:$jar"
-done
+# Compute classpath using external script
+CLASSPATH=`$FWDIR/bin/compute-classpath.sh`
+export CLASSPATH
 
 # Figure out the JAR file that our examples were packaged into. This includes a bit of a hack
 # to avoid the -sources and -doc packages that are built by publish-local.
@@ -161,32 +135,11 @@ if [ -e "$EXAMPLES_DIR/target/spark-examples"*[0-9T].jar ]; then
   export SPARK_EXAMPLES_JAR=`ls "$EXAMPLES_DIR/target/spark-examples"*[0-9T].jar`
 fi
 
-# Add hadoop conf dir - else FileSystem.*, etc fail !
-# Note, this assumes that there is either a HADOOP_CONF_DIR or YARN_CONF_DIR which hosts
-# the configurtion files.
-if [ "x" != "x$HADOOP_CONF_DIR" ]; then
-  CLASSPATH="$CLASSPATH:$HADOOP_CONF_DIR"
-fi
-if [ "x" != "x$YARN_CONF_DIR" ]; then
-  CLASSPATH="$CLASSPATH:$YARN_CONF_DIR"
-fi
-
-
-# Figure out whether to run our class with java or with the scala launcher.
-# In most cases, we'd prefer to execute our process with java because scala
-# creates a shell script as the parent of its Java process, which makes it
-# hard to kill the child with stuff like Process.destroy(). However, for
-# the Spark shell, the wrapper is necessary to properly reset the terminal
-# when we exit, so we allow it to set a variable to launch with scala.
 if [ "$SPARK_LAUNCH_WITH_SCALA" == "1" ]; then
   EXTRA_ARGS=""     # Java options will be passed to scala as JAVA_OPTS
 else
-  CLASSPATH="$CLASSPATH:$SCALA_LIBRARY_PATH/scala-library.jar"
-  CLASSPATH="$CLASSPATH:$SCALA_LIBRARY_PATH/scala-compiler.jar"
-  CLASSPATH="$CLASSPATH:$SCALA_LIBRARY_PATH/jline.jar"
   # The JVM doesn't read JAVA_OPTS by default so we need to pass it in
   EXTRA_ARGS="$JAVA_OPTS"
 fi
 
-export CLASSPATH # Needed for spark-shell
 exec "$RUNNER" -cp "$CLASSPATH" $EXTRA_ARGS "$@"
diff --git a/run2.cmd b/run2.cmd
index bf76844d11..25e4f3b57c 100644
--- a/run2.cmd
+++ b/run2.cmd
@@ -33,51 +33,21 @@ if not "x%SCALA_HOME%"=="x" goto scala_exists
   goto exit
 :scala_exists
 
-rem If the user specifies a Mesos JAR, put it before our included one on the classpath
-set MESOS_CLASSPATH=
-if not "x%MESOS_JAR%"=="x" set MESOS_CLASSPATH=%MESOS_JAR%
-
 rem Figure out how much memory to use per executor and set it as an environment
 rem variable so that our process sees it and can report it to Mesos
 if "x%SPARK_MEM%"=="x" set SPARK_MEM=512m
 
 rem Set JAVA_OPTS to be able to load native libraries and to set heap size
 set JAVA_OPTS=%OUR_JAVA_OPTS% -Djava.library.path=%SPARK_LIBRARY_PATH% -Xms%SPARK_MEM% -Xmx%SPARK_MEM%
-rem Load extra JAVA_OPTS from conf/java-opts, if it exists
-if exist "%FWDIR%conf\java-opts.cmd" call "%FWDIR%conf\java-opts.cmd"
 rem Attention: when changing the way the JAVA_OPTS are assembled, the change must be reflected in ExecutorRunner.scala!
 
 set CORE_DIR=%FWDIR%core
-set REPL_DIR=%FWDIR%repl
 set EXAMPLES_DIR=%FWDIR%examples
-set BAGEL_DIR=%FWDIR%bagel
-set STREAMING_DIR=%FWDIR%streaming
-set PYSPARK_DIR=%FWDIR%python
-
-rem Build up classpath
-set CLASSPATH=%SPARK_CLASSPATH%;%MESOS_CLASSPATH%;%FWDIR%conf;%CORE_DIR%\target\scala-%SCALA_VERSION%\classes
-set CLASSPATH=%CLASSPATH%;%CORE_DIR%\target\scala-%SCALA_VERSION%\test-classes;%CORE_DIR%\src\main\resources
-set CLASSPATH=%CLASSPATH%;%STREAMING_DIR%\target\scala-%SCALA_VERSION%\classes;%STREAMING_DIR%\target\scala-%SCALA_VERSION%\test-classes
-set CLASSPATH=%CLASSPATH%;%STREAMING_DIR%\lib\org\apache\kafka\kafka\0.7.2-spark\*
-set CLASSPATH=%CLASSPATH%;%REPL_DIR%\target\scala-%SCALA_VERSION%\classes;%EXAMPLES_DIR%\target\scala-%SCALA_VERSION%\classes
-set CLASSPATH=%CLASSPATH%;%FWDIR%lib_managed\jars\*
-set CLASSPATH=%CLASSPATH%;%FWDIR%lib_managed\bundles\*
-set CLASSPATH=%CLASSPATH%;%FWDIR%repl\lib\*
-set CLASSPATH=%CLASSPATH%;%FWDIR%python\lib\*
-set CLASSPATH=%CLASSPATH%;%BAGEL_DIR%\target\scala-%SCALA_VERSION%\classes
-
-rem Add hadoop conf dir - else FileSystem.*, etc fail
-rem Note, this assumes that there is either a HADOOP_CONF_DIR or YARN_CONF_DIR which hosts
-rem the configurtion files.
-if "x%HADOOP_CONF_DIR%"=="x" goto no_hadoop_conf_dir
-  set CLASSPATH=%CLASSPATH%;%HADOOP_CONF_DIR%
-:no_hadoop_conf_dir
-
-if "x%YARN_CONF_DIR%"=="x" goto no_yarn_conf_dir
-  set CLASSPATH=%CLASSPATH%;%YARN_CONF_DIR%
-:no_yarn_conf_dir
-
+set REPL_DIR=%FWDIR%repl
 
+rem Compute classpath using external script
+set DONT_PRINT_CLASSPATH=1
+call "%FWDIR%bin\compute-classpath.cmd"
 
 rem Figure out the JAR file that our examples were packaged into.
 rem First search in the build path from SBT: