Add JRE and Pyspark

Dockerfile

-ARG FROM=cern/cc7-base
+ARG FROM=gitlab-registry.cern.ch/industrial-controls/sw-infra/cc7-openjdk:jre-8
 FROM $FROM
 
 ARG SPARK_NXCALS_URL=http://photons-resources.cern.ch/downloads/nxcals_pro/spark/spark-nxcals.zip
@@ -15,18 +15,24 @@ RUN curl -s -k -L -o /tmp/nxcals-spark.zip $SPARK_NXCALS_URL && \
     cd $SPARK_HOME && bsdtar --strip-components=1 -xvf /tmp/nxcals-spark.zip  && \
     rm /tmp/nxcals-spark.zip
 
+RUN adduser etlworker -m
 
-ENV PATH $PATH:$SPARK_HOME:$SPARK_ENV/bin
+RUN mkdir -p $SPARK_HOME/nxcals-python3-env && \
+    chown -R etlworker $SPARK_HOME
 
-RUN adduser etlworker -m
+COPY entrypoint.sh /
+RUN chmod 755 /entrypoint.sh
 
 USER etlworker
 
-RUN cd ~etlworker && python3 -m venv spark-venv && \
-    source ~etlworker/spark-venv/bin/activate && \
-    pip3 install pyspark pandas scipy
+RUN cd $SPARK_HOME && \
+    yes | source $SPARK_HOME/source-me.sh && \
+    source $SPARK_HOME/nxcals-python3-env/bin/activate && \
+    pip list && pip install pyspark==2.3.2 pandas scipy influxdb
 
 VOLUME /auth
 
-WORKDIR /work
+WORKDIR $SPARK_HOME
+
+
 ENTRYPOINT ["/entrypoint.sh"]

pom.xml

@@ -15,7 +15,7 @@
   <properties>
     <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
     <spark.nxcals.url>http://photons-resources.cern.ch/downloads/nxcals_pro/spark/spark-nxcals.zip</spark.nxcals.url>
-    <docker.from>cern/cc7-base</docker.from>
+    <docker.from>gitlab-registry.cern.ch/industrial-controls/sw-infra/cc7-openjdk:jre-8</docker.from>
   </properties>
 
   <build>