Dockerfile-airflow

FROM puckel/docker-airflow:1.10.7

ARG AIRFLOW_USER_HOME=/usr/local/airflow
ENV PYTHONPATH=$PYTHONPATH:${AIRFLOW_USER_HOME}

# Python packages required for th Selenium Plugin
USER root

RUN apt-get update -yqq && \
    apt-get install -y gnupg && \
    apt-get install -y apt-transport-https ca-certificates

RUN curl https://packages.microsoft.com/keys/microsoft.asc | apt-key add -
RUN curl https://packages.microsoft.com/config/ubuntu/18.04/prod.list > /etc/apt/sources.list.d/mssql-release.list

RUN apt-get update -yqq && \
    apt-get install -y unixodbc unixodbc-dev && \
    ACCEPT_EULA=Y apt-get install -y msodbcsql17 && \
    ACCEPT_EULA=Y apt-get install mssql-tools
    

RUN pip install docker && \
    pip install selenium && \
    pip install bs4 && \
    pip install lxml && \
    pip install boto3 && \
    pip install flask-bcrypt && \
    pip install werkzeug>=0.15.0 && \
    pip install pysmb==1.1.27 && \
    pip install xlrd==1.2.0 && \
    pip install python-slugify==3.0.4 && \
    pip install pyodbc==4.0.27 && \
    pip install pandas && \
    pip install pyPDF2 && \
    pip install tqdm && \
    pip install SQLAlchemy && \
    pip install pymssql && \
    pip install petl

RUN groupadd --gid 999 docker \
   && usermod -aG docker airflow 
USER airflow

COPY config/airflow.cfg /usr/local/airflow/

RUN mkdir downloads