install_spark.sh

## Some global parameters that you can change
VERSION=2.3.0
SPARK=spark-$VERSION-bin-hadoop2.7
INSTALL_PATH=$HOME/spark
URL=http://mirrors.ocf.berkeley.edu/apache/spark/spark-$VERSION/$SPARK.tgz

BASH_PROFILE=$HOME/.bash_profile

SPARK_PATH=$INSTALL_PATH/$SPARK
BIN_PATH=$SPARK_PATH/bin

read -p "This script is going to install Spark in $INSTALL_PATH. Proceed (y/n)? " answer

case ${answer:0:1} in
    y|Y )
        mkdir -p $INSTALL_PATH
        cd $INSTALL_PATH
        
        if ! [ -d $SPARK ]; then
            if ! [ -f $SPARK.tgz ]; then
                curl -o $SPARK.tgz $URL
            fi
            tar -zxvf $SPARK.tgz
            rm $SPARK.tgz
        fi
        echo "MASTER=local[2] PYSPARK_DRIVER_PYTHON=jupyter PYSPARK_DRIVER_PYTHON_OPTS=notebook pyspark" >> $BIN_PATH/jupyterspark
        chmod u+x $BIN_PATH/jupyterspark
        read -p "Add Spark to your Path (y/n)? " answer2
        case ${answer2:0:1} in
            y|Y )
                echo "" >> $BASH_PROFILE
                echo "# added by $SPARK installer" >> $BASH_PROFILE
                echo "export PATH=\"$INSTALL_PATH/$SPARK/bin:\$PATH\"" >> $BASH_PROFILE
                ;;
            * )
                ;;
        esac
        echo "Spark successfully installed on your system! Now close this shell and open a new one"
        ;;
    * )
        echo "This program is exiting!"
        exit;;
esac