#!/bin/bash
: '
 @vendor    Chorke Academia, Inc.
 @web       https://cdn.chorke.org/docs/academia
 @version   1.0.00
 @since     1.0.00
'

:<<EOF
if [ -x "$(command -v curl)" ];then \
sudo dnf -q clean all;\
export SSH_PUB_KEY_TAG="hadoop@${HOSTNAME}";\
export HADOOP_HOME_DIR='/var/hadoop/hadoop-3.3.4';\
export JAVA_HOME_DIR='/usr/lib/jvm/java-11-openjdk-11.0.17.0.8-1.fc37.x86_64';\
bash <(curl -s 'https://cdn.chorke.org/exec/cli/bash/install/apache-hadoop/3.3.4-fedora.sh.txt');\
else printf 'curl \033[0;31mnot found! \033[0m:(\n';fi
EOF


:<<EOF
if [ -x "$(command -v curl)" ];then \
sudo rpm-ostree cleanup --base;\
export SSH_PUB_KEY_TAG="hadoop@${HOSTNAME}";\
export HADOOP_HOME_DIR='/var/hadoop/hadoop-3.3.4';\
export JAVA_HOME_DIR='/usr/lib/jvm/java-11-openjdk';\
bash <(curl -s 'https://cdn.chorke.org/exec/cli/bash/install/apache-hadoop/3.3.4-fedora.sh.txt');\
else printf 'curl \033[0;31mnot found! \033[0m:(\n';fi
EOF


function validate(){
    # check existence of wget
    if [ ! -x "$(command -v wget)" ];then
        printf 'wget \033[0;31mnot found! \033[0m:(\n';exit 1
    fi

    # check existence of java runtime virtual machine
    if [ ! -f "${JAVA_HOME_DIR}/bin/java" ]||[ ! -x "$(command -v java)" ];then
        printf 'java \033[0;31mnot found! \033[0m:(\n';exit 1
    fi

    # resolve hadoop ssh key comment
    if [[ "${SSH_PUB_KEY_TAG}/env/set" == '/env/set' ]];then
        export SSH_PUB_KEY_TAG="hadoop@${HOSTNAME}"
    fi

    # resolve hadoop user home if not setted up
    if [[ "${HADOOP_USER_HOME}/env/set" == '/env/set' ]];then
        export HADOOP_USER_HOME='/home/hadoop'
    fi

    # resolve hadoop home if not setted up
    if [[ "${HADOOP_HOME_DIR}/env/set" == '/env/set' ]];then
        export HADOOP_HOME_DIR='/var/hadoop/hadoop-3.3.4'
    fi

    # resovle and check hadoop base from it's home
    export HADOOP_BASE="$(dirname "${HADOOP_HOME_DIR}")"
    if [ "${HADOOP_BASE}" == '.' ];then
        printf 'HADOOP_BASE \033[0;31mpath is invalid \033[0m:)\n';exit 1
    fi

    # check existence of hadoop
    if [ -d "${HADOOP_HOME_DIR}" ];then
        if [ "$(ls -A "${HADOOP_HOME_DIR}")" ];then
            printf 'HADOOP_HOME \033[0;31malready exists \033[0m:)\n';exit 1
        else rm -rf "${HADOOP_HOME_DIR}";fi
    fi

    # check existence of hadoop.service
    if [ -f '/etc/systemd/system/hadoop.service' ];then
        printf 'hadoop.service     \033[0;31malready installed \033[0m:)\n';exit 1
    fi
}

function download(){
    if [ ! -d "${HADOOP_HOME_DIR}" ];then
        if [ ! -d "${HADOOP_BASE}" ];then sudo mkdir -p "${HADOOP_BASE}";fi
        sudo wget -cq https://archive.apache.org/dist/hadoop/common/hadoop-3.3.4/hadoop-3.3.4.tar.gz -P "${HADOOP_BASE}"
        sudo tar -xzf "${HADOOP_BASE}/hadoop-3.3.4.tar.gz" -C "${HADOOP_BASE}"
        sudo rm  -rf  "${HADOOP_BASE}/hadoop-3.3.4.tar.gz"
    fi
}

function approve(){
    if ! sudo id -g hadoop &>/dev/null;then sudo groupadd --system hadoop;fi
    if ! sudo id -u hadoop &>/dev/null;then
        sudo useradd --system hadoop --gid hadoop --create-home --home-dir ${HADOOP_USER_HOME};
    fi

    sudo mkdir -p  ${HADOOP_HOME_DIR}/{etc,logs}
    sudo chown -R hadoop:hadoop ${HADOOP_HOME_DIR}
}

function setting(){
    # add property for default file system
    CORE_SITE_FILE="${HADOOP_HOME_DIR}/etc/hadoop/core-site.xml"
    CONFIGURATION_FIND=$(cat <<EOF
<configuration>\n\
</configuration>
EOF
)
    CONFIGURATION_FILL=$(cat <<EOF
<configuration>\n\
  <property>\n\
    <name>fs.defaultFS</name>\n\
    <value>hdfs://localhost:9000</value>\n\
  </property>\n\
</configuration>
EOF
)
    sudo sed -z "s|${CONFIGURATION_FIND}|${CONFIGURATION_FILL}|" -i "${CORE_SITE_FILE}"


    # add property for replication, namenode & datanode
    HDFS_SITE_FILE="${HADOOP_HOME_DIR}/etc/hadoop/hdfs-site.xml"
    CONFIGURATION_FIND=$(cat <<EOF
<configuration>\n\n\
</configuration>
EOF
)
    CONFIGURATION_FILL=$(cat <<EOF
<configuration>\n\
  <property>\n\
    <name>dfs.replication</name>\n\
    <value>1</value>\n\
  </property>\n\
  <property>\n\
    <name>dfs.namenode.name.dir</name>\n\
    <value>${HADOOP_USER_HOME}/hdfs/namenode/</value>\n\
  </property>\n\
  <property>\n\
    <name>dfs.datanode.data.dir</name>\n\
    <value>${HADOOP_USER_HOME}/hdfs/datanode/</value>\n\
  </property>\n\
</configuration>
EOF
)
    sudo sed -z "s|${CONFIGURATION_FIND}|${CONFIGURATION_FILL}|" -i "${HDFS_SITE_FILE}"


    # add java home to hadoop env
    JAVA_HOME_FIND='# export JAVA_HOME='
    JAVA_HOME_FILL="export JAVA_HOME=${JAVA_HOME_DIR}"
    HADOOP_ENV_FILE="${HADOOP_HOME_DIR}/etc/hadoop/hadoop-env.sh"
    sudo sed -i "s|${JAVA_HOME_FIND}|${JAVA_HOME_FILL}|" "${HADOOP_ENV_FILE}"

    sudo -i -u hadoop bash << EOF
# geneate ssh key for password less shell access 
ssh-keygen -q -b 4096 -t rsa -f ${HADOOP_USER_HOME}/.ssh/id_rsa -C ${SSH_PUB_KEY_TAG} -N ''
cat ${HADOOP_USER_HOME}/.ssh/id_rsa.pub >> ${HADOOP_USER_HOME}/.ssh/authorized_keys

# format hadoop file system
mkdir -p ${HADOOP_USER_HOME}/hdfs/{datanode,namenode}
${HADOOP_HOME_DIR}/bin/hdfs namenode -format
EOF
}

function install(){
    sudo chown -R hadoop:hadoop ${HADOOP_HOME_DIR}
    sudo tee -a /etc/systemd/system/hadoop.service >/dev/null <<EOF
[Unit]
Description=Apache Hadoop
Documentation=https://hadoop.apache.org/docs/r3.3.4
Wants=network-online.target
After=network-online.target

[Service]
User=hadoop
Group=hadoop
Type=forking
SuccessExitStatus=143

WorkingDirectory=${HADOOP_HOME_DIR}/sbin
ExecStartPost=/bin/bash -c "echo \$MAINPID > ${HADOOP_HOME_DIR}/etc/hadoop.pid"
ExecStart=${HADOOP_HOME_DIR}/sbin/start-all.sh
ExecStop=${HADOOP_HOME_DIR}/sbin/stop-all.sh

[Install]
WantedBy=multi-user.target
EOF
}

function launch(){
    sudo systemctl daemon-reload
    sudo systemctl enable hadoop.service
    sudo systemctl start  hadoop.service
    sudo systemctl status hadoop.service
}

function init(){
    validate
    download
    approve
    setting
    install
    launch
}

init