Dockerfile for marvin-ai

Dockerfile

FROM ubuntu:16.04

MAINTAINER Allan Batista <[email protected]>

ENV DEBIAN_FRONTEND=noninteractive

RUN apt-get update && apt-get install python2.7 libsasl2-dev python-pip vim curl git -y
RUN pip install --upgrade pip
RUN pip install virtualenvwrapper --ignore-installed six
RUN curl https://d3kbcqa49mib13.cloudfront.net/spark-2.1.1-bin-hadoop2.6.tgz -o /tmp/spark-2.1.1-bin-hadoop2.6.tgz && \
	tar -xf /tmp/spark-2.1.1-bin-hadoop2.6.tgz -C /opt/ && \
	ln -s /opt/spark-2.1.1-bin-hadoop2.6 /opt/spark && \
	echo "export SPARK_HOME=/opt/spark" >> $HOME/.bash_profile

RUN mkdir /opt/work
RUN mkdir /opt/virtual_env
WORKDIR /opt/work

ENV HOME=/opt/work
ENV WORKON_HOME=$HOME/.virtualenvs
ENV MARVIN_HOME=$HOME/marvin
ENV MARVIN_DATA_PATH=$HOME/marvin/data

RUN /bin/bash -c "source /usr/local/bin/virtualenvwrapper.sh && mkdir $MARVIN_HOME && mkdir $MARVIN_DATA_PATH && cd $MARVIN_HOME && git clone https://github.com/marvin-ai/marvin-python-toolbox.git && cd marvin-python-toolbox && mkvirtualenv python-toolbox-env && setvirtualenvproject && make marvin"

ENV MARVIN_ENGINE_HOME=$MARVIN_HOME/engine
ENV MARVIN_ENGINE_ENV=python-toolbox-env

RUN mkdir -p /opt/work/marvin/engine

RUN /bin/bash -c "cd $MARVIN_ENGINE_HOME && \
    source /usr/local/bin/virtualenvwrapper.sh && \
    mkvirtualenv $MARVIN_ENGINE_ENV"

RUN apt-get install graphviz -y

ENV SPARK_HOME=/opt/spark
ENV SPARK_CONF_DIR=$SPARK_HOME/conf
ENV HADOOP_CONF_DIR=$SPARK_CONF_DIR
ENV YARN_CONF_DIR=$SPARK_CONF_DIR

RUN apt-get install openjdk-8-jdk -y

RUN /bin/bash -c "source /usr/local/bin/virtualenvwrapper.sh && \
    workon $MARVIN_ENGINE_ENV && \
    cd /opt/work/marvin/marvin-python-toolbox && marvin test"

RUN echo "source /usr/local/bin/virtualenvwrapper.sh" >> ~/.bash_profile
RUN echo "workon $MARVIN_ENGINE_ENV" >> ~/.bash_profile

CMD /bin/bash -c "source /usr/local/bin/virtualenvwrapper.sh && \
    workon $MARVIN_ENGINE_ENV && \
    cd /opt/work/marvin/marvin-python-toolbox"