Merge pull request #3 from apache/branch-1.1

merge upstream changes from branch-1.1

README.md

@@ -69,7 +69,7 @@ Many of the example programs print usage help if no params are given.
 Testing first requires [building Spark](#building-spark). Once Spark is built, tests
 can be run using:
 
-    ./sbt/sbt test
+    ./dev/run-tests
 
 ## A Note About Hadoop Versions
 
@@ -118,11 +118,10 @@ If your project is built with Maven, add this to your POM file's `<dependencies>
 ## A Note About Thrift JDBC server and CLI for Spark SQL
 
 Spark SQL supports Thrift JDBC server and CLI.
-See sql-programming-guide.md for more information about those features.
-You can use those features by setting `-Phive-thriftserver` when building Spark as follows.
-
-    $ sbt/sbt -Phive-thriftserver assembly
+See sql-programming-guide.md for more information about using the JDBC server and CLI.
+You can use those features by setting `-Phive` when building Spark as follows.
 
+    $ sbt/sbt -Phive  assembly
 
 ## Configuration
 

assembly/pom.xml

@@ -163,11 +163,6 @@
           <artifactId>spark-hive_${scala.binary.version}</artifactId>
           <version>${project.version}</version>
         </dependency>
-      </dependencies>
-    </profile>
-    <profile>
-      <id>hive-thriftserver</id>
-      <dependencies>
         <dependency>
           <groupId>org.apache.spark</groupId>
           <artifactId>spark-hive-thriftserver_${scala.binary.version}</artifactId>

bin/compute-classpath.cmd

@@ -36,7 +36,8 @@ 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"
 
 rem Build up classpath
-set CLASSPATH=%FWDIR%conf
+set CLASSPATH=%SPARK_CLASSPATH%;%SPARK_SUBMIT_CLASSPATH%;%FWDIR%conf
+
 if exist "%FWDIR%RELEASE" (
   for %%d in ("%FWDIR%lib\spark-assembly*.jar") do (
     set ASSEMBLY_JAR=%%d

bin/pyspark

@@ -102,6 +102,8 @@ if [[ "$1" =~ \.py$ ]]; then
   gatherSparkSubmitOpts "$@"
   exec $FWDIR/bin/spark-submit "${SUBMISSION_OPTS[@]}" $primary "${APPLICATION_OPTS[@]}"
 else
+  # PySpark shell requires special handling downstream
+  export PYSPARK_SHELL=1
   # Only use ipython if no command line arguments were provided [SPARK-1134]
   if [[ "$IPYTHON" = "1" ]]; then
     exec ipython $IPYTHON_OPTS

bin/spark-class

@@ -17,6 +17,8 @@
 # limitations under the License.
 #
 
+# NOTE: Any changes to this file must be reflected in SparkSubmitDriverBootstrapper.scala!
+
 cygwin=false
 case "`uname`" in
     CYGWIN*) cygwin=true;;
@@ -39,7 +41,7 @@ fi
 
 if [ -n "$SPARK_MEM" ]; then
   echo -e "Warning: SPARK_MEM is deprecated, please use a more specific config option" 1>&2
-  echo -e "(e.g., spark.executor.memory or SPARK_DRIVER_MEMORY)." 1>&2
+  echo -e "(e.g., spark.executor.memory or spark.driver.memory)." 1>&2
 fi
 
 # Use SPARK_MEM or 512m as the default memory, to be overridden by specific options
@@ -73,11 +75,17 @@ case "$1" in
     OUR_JAVA_MEM=${SPARK_EXECUTOR_MEMORY:-$DEFAULT_MEM}
     ;;
 
-  # Spark submit uses SPARK_SUBMIT_OPTS and SPARK_JAVA_OPTS
-    'org.apache.spark.deploy.SparkSubmit')
-    OUR_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_SUBMIT_OPTS \
-      -Djava.library.path=$SPARK_SUBMIT_LIBRARY_PATH"
+  # Spark submit uses SPARK_JAVA_OPTS + SPARK_SUBMIT_OPTS +
+  # SPARK_DRIVER_MEMORY + SPARK_SUBMIT_DRIVER_MEMORY.
+  'org.apache.spark.deploy.SparkSubmit')
+    OUR_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_SUBMIT_OPTS"
     OUR_JAVA_MEM=${SPARK_DRIVER_MEMORY:-$DEFAULT_MEM}
+    if [ -n "$SPARK_SUBMIT_LIBRARY_PATH" ]; then
+      OUR_JAVA_OPTS="$OUR_JAVA_OPTS -Djava.library.path=$SPARK_SUBMIT_LIBRARY_PATH"
+    fi
+    if [ -n "$SPARK_SUBMIT_DRIVER_MEMORY" ]; then
+      OUR_JAVA_MEM="$SPARK_SUBMIT_DRIVER_MEMORY"
+    fi
     ;;
 
   *)
@@ -101,11 +109,12 @@ fi
 # Set JAVA_OPTS to be able to load native libraries and to set heap size
 JAVA_OPTS="-XX:MaxPermSize=128m $OUR_JAVA_OPTS"
 JAVA_OPTS="$JAVA_OPTS -Xms$OUR_JAVA_MEM -Xmx$OUR_JAVA_MEM"
+
 # Load extra JAVA_OPTS from conf/java-opts, if it exists
 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 CommandUtils.scala!
 
 TOOLS_DIR="$FWDIR"/tools
@@ -146,10 +155,28 @@ if $cygwin; then
 fi
 export CLASSPATH
 
-if [ "$SPARK_PRINT_LAUNCH_COMMAND" == "1" ]; then
-  echo -n "Spark Command: " 1>&2
-  echo "$RUNNER" -cp "$CLASSPATH" $JAVA_OPTS "$@" 1>&2
-  echo -e "========================================\n" 1>&2
+# In Spark submit client mode, the driver is launched in the same JVM as Spark submit itself.
+# Here we must parse the properties file for relevant "spark.driver.*" configs before launching
+# the driver JVM itself. Instead of handling this complexity in Bash, we launch a separate JVM
+# to prepare the launch environment of this driver JVM.
+
+if [ -n "$SPARK_SUBMIT_BOOTSTRAP_DRIVER" ]; then
+  # This is used only if the properties file actually contains these special configs
+  # Export the environment variables needed by SparkSubmitDriverBootstrapper
+  export RUNNER
+  export CLASSPATH
+  export JAVA_OPTS
+  export OUR_JAVA_MEM
+  export SPARK_CLASS=1
+  shift # Ignore main class (org.apache.spark.deploy.SparkSubmit) and use our own
+  exec "$RUNNER" org.apache.spark.deploy.SparkSubmitDriverBootstrapper "$@"
+else
+  # Note: The format of this command is closely echoed in SparkSubmitDriverBootstrapper.scala
+  if [ -n "$SPARK_PRINT_LAUNCH_COMMAND" ]; then
+    echo -n "Spark Command: " 1>&2
+    echo "$RUNNER" -cp "$CLASSPATH" $JAVA_OPTS "$@" 1>&2
+    echo -e "========================================\n" 1>&2
+  fi
+  exec "$RUNNER" -cp "$CLASSPATH" $JAVA_OPTS "$@"
 fi
 
-exec "$RUNNER" -cp "$CLASSPATH" $JAVA_OPTS "$@"

bin/spark-class2.cmd

@@ -17,6 +17,8 @@ rem See the License for the specific language governing permissions and
 rem limitations under the License.
 rem
 
+rem Any changes to this file must be reflected in SparkSubmitDriverBootstrapper.scala!
+
 setlocal enabledelayedexpansion
 
 set SCALA_VERSION=2.10
@@ -38,7 +40,7 @@ if not "x%1"=="x" goto arg_given
 
 if not "x%SPARK_MEM%"=="x" (
   echo Warning: SPARK_MEM is deprecated, please use a more specific config option
-  echo e.g., spark.executor.memory or SPARK_DRIVER_MEMORY.
+  echo e.g., spark.executor.memory or spark.driver.memory.
 )
 
 rem Use SPARK_MEM or 512m as the default memory, to be overridden by specific options
@@ -67,18 +69,26 @@ rem Executors use SPARK_JAVA_OPTS + SPARK_EXECUTOR_MEMORY.
   set OUR_JAVA_OPTS=%SPARK_JAVA_OPTS% %SPARK_EXECUTOR_OPTS%
   if not "x%SPARK_EXECUTOR_MEMORY%"=="x" set OUR_JAVA_MEM=%SPARK_EXECUTOR_MEMORY%
 
-rem All drivers use SPARK_JAVA_OPTS + SPARK_DRIVER_MEMORY. The repl also uses SPARK_REPL_OPTS.
-) else if "%1"=="org.apache.spark.repl.Main" (
-  set OUR_JAVA_OPTS=%SPARK_JAVA_OPTS% %SPARK_REPL_OPTS%
+rem Spark submit uses SPARK_JAVA_OPTS + SPARK_SUBMIT_OPTS +
+rem SPARK_DRIVER_MEMORY + SPARK_SUBMIT_DRIVER_MEMORY.
+rem The repl also uses SPARK_REPL_OPTS.
+) else if "%1"=="org.apache.spark.deploy.SparkSubmit" (
+  set OUR_JAVA_OPTS=%SPARK_JAVA_OPTS% %SPARK_SUBMIT_OPTS% %SPARK_REPL_OPTS%
+  if not "x%SPARK_SUBMIT_LIBRARY_PATH%"=="x" (
+    set OUR_JAVA_OPTS=!OUR_JAVA_OPTS! -Djava.library.path=%SPARK_SUBMIT_LIBRARY_PATH%
+  ) else if not "x%SPARK_LIBRARY_PATH%"=="x" (
+    set OUR_JAVA_OPTS=!OUR_JAVA_OPTS! -Djava.library.path=%SPARK_LIBRARY_PATH%
+  )
   if not "x%SPARK_DRIVER_MEMORY%"=="x" set OUR_JAVA_MEM=%SPARK_DRIVER_MEMORY%
+  if not "x%SPARK_SUBMIT_DRIVER_MEMORY%"=="x" set OUR_JAVA_MEM=%SPARK_SUBMIT_DRIVER_MEMORY%
 ) else (
   set OUR_JAVA_OPTS=%SPARK_JAVA_OPTS%
   if not "x%SPARK_DRIVER_MEMORY%"=="x" set OUR_JAVA_MEM=%SPARK_DRIVER_MEMORY%
 )
 
-rem Set JAVA_OPTS to be able to load native libraries and to set heap size
-set JAVA_OPTS=-XX:MaxPermSize=128m %OUR_JAVA_OPTS% -Djava.library.path=%SPARK_LIBRARY_PATH% -Xms%OUR_JAVA_MEM% -Xmx%OUR_JAVA_MEM%
 rem Attention: when changing the way the JAVA_OPTS are assembled, the change must be reflected in CommandUtils.scala!
+rem Set JAVA_OPTS to be able to load native libraries and to set heap size
+set JAVA_OPTS=-XX:MaxPermSize=128m %OUR_JAVA_OPTS% -Xms%OUR_JAVA_MEM% -Xmx%OUR_JAVA_MEM%
 
 rem Test whether the user has built Spark
 if exist "%FWDIR%RELEASE" goto skip_build_test
@@ -109,5 +119,27 @@ rem Figure out where java is.
 set RUNNER=java
 if not "x%JAVA_HOME%"=="x" set RUNNER=%JAVA_HOME%\bin\java
 
-"%RUNNER%" -cp "%CLASSPATH%" %JAVA_OPTS% %*
+rem In Spark submit client mode, the driver is launched in the same JVM as Spark submit itself.
+rem Here we must parse the properties file for relevant "spark.driver.*" configs before launching
+rem the driver JVM itself. Instead of handling this complexity here, we launch a separate JVM
+rem to prepare the launch environment of this driver JVM.
+
+rem In this case, leave out the main class (org.apache.spark.deploy.SparkSubmit) and use our own.
+rem Leaving out the first argument is surprisingly difficult to do in Windows. Note that this must
+rem be done here because the Windows "shift" command does not work in a conditional block.
+set BOOTSTRAP_ARGS=
+shift
+:start_parse
+if "%~1" == "" goto end_parse
+set BOOTSTRAP_ARGS=%BOOTSTRAP_ARGS% %~1
+shift
+goto start_parse
+:end_parse
+
+if not [%SPARK_SUBMIT_BOOTSTRAP_DRIVER%] == [] (
+  set SPARK_CLASS=1
+  "%RUNNER%" org.apache.spark.deploy.SparkSubmitDriverBootstrapper %BOOTSTRAP_ARGS%
+) else (
+  "%RUNNER%" -cp "%CLASSPATH%" %JAVA_OPTS% %*
+)
 :exit

bin/spark-shell

@@ -22,7 +22,7 @@
 
 cygwin=false
 case "`uname`" in
-    CYGWIN*) cygwin=true;;
+  CYGWIN*) cygwin=true;;
 esac
 
 # Enter posix mode for bash
@@ -32,9 +32,9 @@ set -o posix
 FWDIR="$(cd `dirname $0`/..; pwd)"
 
 function usage() {
-    echo "Usage: ./bin/spark-shell [options]"
-    $FWDIR/bin/spark-submit --help 2>&1 | grep -v Usage 1>&2
-    exit 0
+  echo "Usage: ./bin/spark-shell [options]"
+  $FWDIR/bin/spark-submit --help 2>&1 | grep -v Usage 1>&2
+  exit 0
 }
 
 if [[ "$@" = *--help ]] || [[ "$@" = *-h ]]; then
@@ -46,20 +46,20 @@ SUBMIT_USAGE_FUNCTION=usage
 gatherSparkSubmitOpts "$@"
 
 function main() {
-    if $cygwin; then
-        # Workaround for issue involving JLine and Cygwin
-        # (see http://sourceforge.net/p/jline/bugs/40/).
-        # If you're using the Mintty terminal emulator in Cygwin, may need to set the
-        # "Backspace sends ^H" setting in "Keys" section of the Mintty options
-        # (see https://github.com/sbt/sbt/issues/562).
-        stty -icanon min 1 -echo > /dev/null 2>&1
-        export SPARK_SUBMIT_OPTS="$SPARK_SUBMIT_OPTS -Djline.terminal=unix"
-        $FWDIR/bin/spark-submit --class org.apache.spark.repl.Main "${SUBMISSION_OPTS[@]}" spark-shell "${APPLICATION_OPTS[@]}"
-        stty icanon echo > /dev/null 2>&1
-    else
-        export SPARK_SUBMIT_OPTS
-        $FWDIR/bin/spark-submit --class org.apache.spark.repl.Main "${SUBMISSION_OPTS[@]}" spark-shell "${APPLICATION_OPTS[@]}"
-    fi
+  if $cygwin; then
+    # Workaround for issue involving JLine and Cygwin
+    # (see http://sourceforge.net/p/jline/bugs/40/).
+    # If you're using the Mintty terminal emulator in Cygwin, may need to set the
+    # "Backspace sends ^H" setting in "Keys" section of the Mintty options
+    # (see https://github.com/sbt/sbt/issues/562).
+    stty -icanon min 1 -echo > /dev/null 2>&1
+    export SPARK_SUBMIT_OPTS="$SPARK_SUBMIT_OPTS -Djline.terminal=unix"
+    $FWDIR/bin/spark-submit --class org.apache.spark.repl.Main "${SUBMISSION_OPTS[@]}" spark-shell "${APPLICATION_OPTS[@]}"
+    stty icanon echo > /dev/null 2>&1
+  else
+    export SPARK_SUBMIT_OPTS
+    $FWDIR/bin/spark-submit --class org.apache.spark.repl.Main "${SUBMISSION_OPTS[@]}" spark-shell "${APPLICATION_OPTS[@]}"
+  fi
 }
 
 # Copy restore-TTY-on-exit functions from Scala script so spark-shell exits properly even in

bin/spark-sql

@@ -24,6 +24,7 @@
 set -o posix
 
 CLASS="org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver"
+CLASS_NOT_FOUND_EXIT_STATUS=1
 
 # Figure out where Spark is installed
 FWDIR="$(cd `dirname $0`/..; pwd)"
@@ -43,52 +44,22 @@ function usage {
   $FWDIR/bin/spark-class $CLASS --help 2>&1 | grep -v "$pattern" 1>&2
 }
 
-function ensure_arg_number {
-  arg_number=$1
-  at_least=$2
-
-  if [[ $arg_number -lt $at_least ]]; then
-    usage
-    exit 1
-  fi
-}
-
-if [[ "$@" = --help ]] || [[ "$@" = -h ]]; then
+if [[ "$@" = *--help ]] || [[ "$@" = *-h ]]; then
   usage
   exit 0
 fi
 
-CLI_ARGS=()
-SUBMISSION_ARGS=()
-
-while (($#)); do
-  case $1 in
-    -d | --define | --database | -f | -h | --hiveconf | --hivevar | -i | -p)
-      ensure_arg_number $# 2
-      CLI_ARGS+=("$1"); shift
-      CLI_ARGS+=("$1"); shift
-      ;;
+source $FWDIR/bin/utils.sh
+SUBMIT_USAGE_FUNCTION=usage
+gatherSparkSubmitOpts "$@"
 
-    -e)
-      ensure_arg_number $# 2
-      CLI_ARGS+=("$1"); shift
-      CLI_ARGS+=("$1"); shift
-      ;;
+"$FWDIR"/bin/spark-submit --class $CLASS "${SUBMISSION_OPTS[@]}" spark-internal "${APPLICATION_OPTS[@]}"
+exit_status=$?
 
-    -s | --silent)
-      CLI_ARGS+=("$1"); shift
-      ;;
-
-    -v | --verbose)
-      # Both SparkSubmit and SparkSQLCLIDriver recognizes -v | --verbose
-      CLI_ARGS+=("$1")
-      SUBMISSION_ARGS+=("$1"); shift
-      ;;
-
-    *)
-      SUBMISSION_ARGS+=("$1"); shift
-      ;;
-  esac
-done
+if [[ exit_status -eq CLASS_NOT_FOUND_EXIT_STATUS ]]; then
+  echo
+  echo "Failed to load Spark SQL CLI main class $CLASS."
+  echo "You need to build Spark with -Phive."
+fi
 
-exec "$FWDIR"/bin/spark-submit --class $CLASS "${SUBMISSION_ARGS[@]}" spark-internal "${CLI_ARGS[@]}"
+exit $exit_status