Merge branch 'master' into 2013-06/assembly-jar-deploy

Conflicts:
	run

Previous changes that I made to run and set-dev-classpath.sh instead
have been folded into compute-classpath.sh

1 files changed, 71 insertions, 33 deletions

diff --git a/run b/run
index 646d12c1eb..6b5bc01a51 100755
--- a/run
+++ b/run
@@ -21,29 +21,69 @@ fi
 if [ "$1" = "spark.deploy.master.Master" -o "$1" = "spark.deploy.worker.Worker" ]; then
   SPARK_MEM=${SPARK_DAEMON_MEMORY:-512m}
   SPARK_DAEMON_JAVA_OPTS="$SPARK_DAEMON_JAVA_OPTS -Dspark.akka.logLifecycleEvents=true"
-  SPARK_JAVA_OPTS=$SPARK_DAEMON_JAVA_OPTS   # Empty by default
+  # Do not overwrite SPARK_JAVA_OPTS environment variable in this script
+  OUR_JAVA_OPTS="$SPARK_DAEMON_JAVA_OPTS"   # Empty by default
+else
+  OUR_JAVA_OPTS="$SPARK_JAVA_OPTS"
 fi
 
 
 # Add java opts for master, worker, executor. The opts maybe null
 case "$1" in
   'spark.deploy.master.Master')
-    SPARK_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_MASTER_OPTS"
+    OUR_JAVA_OPTS="$OUR_JAVA_OPTS $SPARK_MASTER_OPTS"
     ;;
   'spark.deploy.worker.Worker')
-    SPARK_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_WORKER_OPTS"
+    OUR_JAVA_OPTS="$OUR_JAVA_OPTS $SPARK_WORKER_OPTS"
     ;;
   'spark.executor.StandaloneExecutorBackend')
-    SPARK_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_EXECUTOR_OPTS"
+    OUR_JAVA_OPTS="$OUR_JAVA_OPTS $SPARK_EXECUTOR_OPTS"
     ;;
   'spark.executor.MesosExecutorBackend')
-    SPARK_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_EXECUTOR_OPTS"
+    OUR_JAVA_OPTS="$OUR_JAVA_OPTS $SPARK_EXECUTOR_OPTS"
     ;;
   'spark.repl.Main')
-    SPARK_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_REPL_OPTS"
+    OUR_JAVA_OPTS="$OUR_JAVA_OPTS $SPARK_REPL_OPTS"
     ;;
 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="${SCALA_HOME}/bin/scala"
+  else
+    if [ `command -v scala` ]; then
+      RUNNER="scala"
+    else
+      echo "SCALA_HOME is not set and scala is not in PATH" >&2
+      exit 1
+    fi
+  fi
+else
+  if [ -n "${JAVA_HOME}" ]; then
+    RUNNER="${JAVA_HOME}/bin/java"
+  else
+    if [ `command -v java` ]; then
+      RUNNER="java"
+    else
+      echo "JAVA_HOME is not set" >&2
+      exit 1
+    fi
+  fi
+  if [[ ! -f "$FWDIR/RELEASE" && -z "$SCALA_LIBRARY_PATH" ]]; then
+    if [ -z "$SCALA_HOME" ]; then
+      echo "SCALA_HOME is not set" >&2
+      exit 1
+    fi
+    SCALA_LIBRARY_PATH="$SCALA_HOME/lib"
+  fi
+fi
+
 # Figure out how much memory to use per executor and set it as an environment
 # variable so that our process sees it and can report it to Mesos
 if [ -z "$SPARK_MEM" ] ; then
@@ -52,7 +92,7 @@ fi
 export SPARK_MEM
 
 # Set JAVA_OPTS to be able to load native libraries and to set heap size
-JAVA_OPTS="$SPARK_JAVA_OPTS"
+JAVA_OPTS="$OUR_JAVA_OPTS"
 JAVA_OPTS="$JAVA_OPTS -Djava.library.path=$SPARK_LIBRARY_PATH"
 JAVA_OPTS="$JAVA_OPTS -Xms$SPARK_MEM -Xmx$SPARK_MEM"
 # Load extra JAVA_OPTS from conf/java-opts, if it exists
@@ -60,37 +100,35 @@ if [ -e $FWDIR/conf/java-opts ] ; then
   JAVA_OPTS="$JAVA_OPTS `cat $FWDIR/conf/java-opts`"
 fi
 export JAVA_OPTS
+# Attention: when changing the way the JAVA_OPTS are assembled, the change must be reflected in ExecutorRunner.scala!
 
-# Check if this is a binary distribution or source distribution
-# and build up the classpath appropriately
-if [ -f "$FWDIR/RELEASE" ]; then
-  if [ `command -v java` ]; then
-    RUNNER="java"
-  else
-    if [ -z "$JAVA_HOME" ]; then
-      echo "JAVA_HOME is not set" >&2
-      exit 1
-    fi
-    RUNNER="${JAVA_HOME}/bin/java"
+if [ ! -f "$FWDIR/RELEASE" ]; then
+  CORE_DIR="$FWDIR/core"
+  EXAMPLES_DIR="$FWDIR/examples"
+  REPL_DIR="$FWDIR/repl"
+
+  # Exit if the user hasn't compiled Spark
+  if [ ! -e "$CORE_DIR/target" ]; then
+    echo "Failed to find Spark classes in $CORE_DIR/target" >&2
+    echo "You need to compile Spark before running this program" >&2
+    exit 1
+  fi
+
+  if [[ "$@" = *repl* && ! -e "$REPL_DIR/target" ]]; then
+    echo "Failed to find Spark classes in $REPL_DIR/target" >&2
+    echo "You need to compile Spark repl module before running this program" >&2
+    exit 1
   fi
+fi
 
-  CLASSPATH="$SPARK_CLASSPATH:$FWDIR/jars/*"
+# Compute classpath using external script
+CLASSPATH=`$FWDIR/bin/compute-classpath.sh`
+export CLASSPATH
 
+if [ "$SPARK_LAUNCH_WITH_SCALA" == "1" ]; then
+  EXTRA_ARGS=""     # Java options will be passed to scala as JAVA_OPTS
+else
   # The JVM doesn't read JAVA_OPTS by default so we need to pass it in
   EXTRA_ARGS="$JAVA_OPTS"
-else
-  . "$FWDIR/set-dev-classpath.sh"
 fi
-
-# Add hadoop conf dir - else FileSystem.*, etc fail !
-# Note, this assumes that there is either a HADOOP_CONF_DIR or YARN_CONF_DIR which hosts
-# the configurtion files.
-if [ "x" != "x$HADOOP_CONF_DIR" ]; then
-  CLASSPATH="$CLASSPATH:$HADOOP_CONF_DIR"
-fi
-if [ "x" != "x$YARN_CONF_DIR" ]; then
-  CLASSPATH="$CLASSPATH:$YARN_CONF_DIR"
-fi
-
-export CLASSPATH # Needed for spark-shell
 exec "$RUNNER" -cp "$CLASSPATH" $EXTRA_ARGS "$@"