Dockerfile-debian

# Multi-stage Dockerfile for building RDKit artifacts.
# The build image is a heavyweight image containing all aspects of RDKit plus the build system.
# It's artifacts are then used in the subsequent images to create lightweight images.

################################# base image ########################################
# This image contains the artifacts used by the later stages.
# The image is NOT pushed, just the cached layers used.

# also change the base image in the other sections
FROM debian:trixie as build
LABEL maintainer="Tim Dudgeon<tdudgeon@informaticsmatters.com>"

ARG GIT_REPO=https://github.com/rdkit/rdkit.git
ARG GIT_BRANCH=master
ARG GIT_TAG
# also change the POSTGRES_VER variable in the cartridge image section
ARG POSTGRES_VER=17
# also change the BOOST_VER variable in the other sections
ARG BOOST_VER=1.83
# also change the JAVA_VER variable in the java image section
ARG JAVA_VER=17

RUN apt-get update &&\
  apt-get install -y --no-install-recommends \
  build-essential\
  python3-dev\
  python3-numpy\
  python3-pip\
  cmake\
  sqlite3\
  libsqlite3-dev\
  libboost$BOOST_VER\
  libboost$BOOST_VER-dev\
  libboost-system$BOOST_VER\
  libboost-thread$BOOST_VER\
  libboost-serialization$BOOST_VER\
  libboost-python$BOOST_VER\
  libboost-regex$BOOST_VER\
  libboost-iostreams$BOOST_VER\
  zlib1g-dev\
  swig\
  libeigen3-dev\
  git\
  wget\
  openjdk-$JAVA_VER-jdk\
  postgresql-$POSTGRES_VER\
  postgresql-server-dev-$POSTGRES_VER\
  postgresql-plpython3-$POSTGRES_VER\
  zip\
  unzip\
  libfreetype6-dev &&\
  apt-get clean -y


RUN if [ $GIT_TAG ]; then echo "Checking out tag $GIT_TAG from repo $GIT_REPO branch $GIT_BRANCH"; else echo "Checking out repo $GIT_REPO branch $GIT_BRANCH"; fi
RUN git clone -b $GIT_BRANCH --single-branch $GIT_REPO &&\
  if [ $GIT_TAG ]; then cd rdkit && git fetch --tags && git checkout $GIT_TAG; fi

ENV RDBASE=/rdkit
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$RDBASE/lib:$RDBASE/Code/JavaWrappers/gmwrapper:/usr/lib/x86_64-linux-gnu:/usr/lib/aarch64-linux-gnu/
ENV PYTHONPATH=$PYTHONPATH:$RDBASE
RUN ln -s $(ls -d /usr/lib/jvm/java-$JAVA_VER-openjdk-*) /usr/lib/jvm/java-$JAVA_VER-openjdk
ENV JAVA_HOME=/usr/lib/jvm/java-$JAVA_VER-openjdk
ENV CLASSPATH=$RDBASE/Code/JavaWrappers/gmwrapper/org.RDKit.jar

RUN mkdir $RDBASE/build
WORKDIR $RDBASE/build

RUN cmake -Wno-dev\
  -DPYTHON_EXECUTABLE=/usr/bin/python3\
  -DRDK_INSTALL_INTREE=OFF\
  -DRDK_BUILD_INCHI_SUPPORT=ON\
  -DRDK_BUILD_AVALON_SUPPORT=ON\
  -DRDK_BUILD_PYTHON_WRAPPERS=ON\
  -DRDK_BUILD_SWIG_WRAPPERS=ON\
  -DRDK_BUILD_PGSQL=ON\
  -DPostgreSQL_ROOT=/usr/lib/postgresql/$POSTGRES_VER\
  -DPostgreSQL_TYPE_INCLUDE_DIR=/usr/include/postgresql/$POSTGRES_VER/server\
  -DCMAKE_INSTALL_PREFIX=/usr\
  -DCPACK_PACKAGE_RELOCATABLE=OFF\
  ..

RUN nproc=$(getconf _NPROCESSORS_ONLN)\
  && make -j $(( nproc > 2 ? nproc - 2 : 1 ))
RUN make install
RUN sh Code/PgSQL/rdkit/pgsql_install.sh
RUN cpack -G DEB
RUN cd /rdkit/Code/JavaWrappers/gmwrapper && tar cvfz javadoc.tgz doc

WORKDIR $RDBASE

################################# Python image ######################################

FROM debian:trixie AS python
LABEL maintainer="Tim Dudgeon<tdudgeon@informaticsmatters.com>"

ARG BOOST_VER=1.83

RUN apt-get update &&\
  apt-get upgrade -y &&\ 
  apt-get install -y --no-install-recommends\
  python3\
  python3-dev\
  python3-numpy\
  python3-pip\
  python3-setuptools\
  python3-wheel\
  python3-six\
  python-is-python3\
  gcc\
  libboost-system$BOOST_VER\
  libboost-thread$BOOST_VER\
  libboost-serialization$BOOST_VER\
  libboost-python$BOOST_VER\
  libboost-regex$BOOST_VER\
  libboost-chrono$BOOST_VER\
  libboost-date-time$BOOST_VER\
  libboost-atomic$BOOST_VER\
  libboost-iostreams$BOOST_VER\
  sqlite3\
  wget\
  zip\
  libfreetype6 &&\
  apt-get clean -y


COPY --from=build /rdkit/build/RDKit-*Linux-Runtime.deb /rdkit/build/RDKit-*Linux-Python.deb /tmp/
RUN dpkg -i /tmp/RDKit-*.deb && rm -f /tmp/*.deb

ENV RDBASE=/usr/share/RDKit

# since the 2023 builds cpack seems to putting things under site-packages rather than dist-packages 
ENV PYTHONPATH=/usr/lib/python3.11/site-packages

WORKDIR /

# add the rdkit user
RUN useradd -u 1000 -g 0 -m rdkit
USER 1000


################################# Java image ########################################

# Dockerfile for Java based RDKit implementation
# Based on Debian.

FROM debian:trixie AS java
LABEL maintainer="Tim Dudgeon<tdudgeon@informaticsmatters.com>"

ARG BOOST_VER=1.83
ARG JAVA_VER=17

RUN apt-get update &&\
  apt-get upgrade -y &&\
  apt-get install -y --no-install-recommends\
  openjdk-$JAVA_VER-jre-headless\
  libboost$BOOST_VER\
  libboost$BOOST_VER-dev\
  libboost-system$BOOST_VER\
  libboost-thread$BOOST_VER\
  libboost-serialization$BOOST_VER\
  libboost-regex$BOOST_VER\
  libboost-chrono$BOOST_VER\
  libboost-date-time$BOOST_VER\
  libboost-atomic$BOOST_VER\
  libboost-iostreams$BOOST_VER\
  libfreetype6 &&\
  apt-get clean -y

COPY --from=build /rdkit/Code/JavaWrappers/gmwrapper/org.RDKit.jar /rdkit/Code/JavaWrappers/gmwrapper/libGraphMolWrap.so /rdkit/gmwrapper/
COPY --from=build /rdkit/build/RDKit-*-Linux-Runtime.deb /tmp/
RUN dpkg -i /tmp/*.deb && rm -f /tmp/*.deb

WORKDIR /

RUN ln -s $(ls -d /usr/lib/jvm/java-$JAVA_VER-openjdk-*) /usr/lib/jvm/java-$JAVA_VER-openjdk
ENV JAVA_HOME=/usr/lib/jvm/java-$JAVA_VER-openjdk
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/rdkit/gmwrapper
ENV CLASSPATH=/rdkit/gmwrapper/org.RDKit.jar
ENV RDBASE=/usr/share/RDKit

# add the rdkit user
RUN useradd -u 1000 -U -G 0 -m rdkit
USER 1000


################################# Tomcat image ######################################

FROM debian:trixie AS tomcat
LABEL maintainer="Tim Dudgeon<tdudgeon@informaticsmatters.com>"

ARG BOOST_VER=1.83

ENV TOMCAT_MAJOR 10
ENV TOMCAT_VERSION 10.1.33
ENV ARCHIVE apache-tomcat-$TOMCAT_VERSION.tar.gz
ENV JAVA_HOME /usr/lib/jvm/java-17-openjdk-amd64

USER root

RUN apt-get update &&\
 apt-get upgrade -y &&\
 apt-get install -y --no-install-recommends\
 openjdk-17-jre-headless\
 curl\
 unzip\
 libboost-system$BOOST_VER\
 libboost-thread$BOOST_VER\
 libboost-serialization$BOOST_VER\
 libboost-regex$BOOST_VER\
 libboost-chrono$BOOST_VER\
 libboost-date-time$BOOST_VER\
 libboost-atomic$BOOST_VER\
 libboost-iostreams$BOOST_VER\
 libfreetype6-dev &&\
 apt-get clean -y

COPY --from=build /rdkit/Code/JavaWrappers/gmwrapper/org.RDKit.jar /rdkit/Code/JavaWrappers/gmwrapper/libGraphMolWrap.so /rdkit/gmwrapper/
COPY --from=build /rdkit/build/RDKit-*-Linux-Runtime.deb /tmp/
RUN dpkg -i /tmp/*.deb && rm -f /tmp/*.deb

RUN useradd -u 501 -U -G 0 -m tomcat

WORKDIR /usr/local

RUN curl -s -o tomcat.tar.gz -L https://www.apache.org/dist/tomcat/tomcat-$TOMCAT_MAJOR/v$TOMCAT_VERSION/bin/$ARCHIVE &&\
 tar xfz tomcat.tar.gz &&\
 rm tomcat.tar.gz &&\
 mv apache-tomcat-$TOMCAT_VERSION tomcat &&\
 rm tomcat/bin/*.bat &&\
 rm -rf tomcat/webapps/* &&\
 chown -R 501:0 tomcat &&\
 chmod -R g+r tomcat &&\
 chmod g+w tomcat tomcat/logs tomcat/webapps tomcat/temp tomcat/work tomcat/conf tomcat/bin tomcat/lib

WORKDIR /usr/local/tomcat
ENV CATALINA_HOME /usr/local/tomcat
EXPOSE 8080
USER 501

WORKDIR $CATALINA_HOME
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/rdkit/gmwrapper
ENV CLASSPATH=/rdkit/gmwrapper/org.RDKit.jar
ENV RDBASE=/usr/share/RDKit/

RUN echo "CLASSPATH=/rdkit/gmwrapper/org.RDKit.jar\n" > /usr/local/tomcat/bin/setenv.sh

CMD ["$CATALINA_HOME/bin/catalina.sh", "run"]

################################# Cartridge image ###################################
# Dockerfile for RDKit cartridge based on Debian.
# This image is inspired by the docker.io:/library/postgres image, and uses the docker-entrypoint.sh 
# file from there, but there is a need to have the exact same versions of libraries that were used to 
# build the deb packages so we need to install the packages using apt to have tight control over this.
#

FROM debian:trixie AS cartridge
LABEL maintainer="Tim Dudgeon<tdudgeon@informaticsmatters.com>"

ENV POSTGRES_VER=17
ENV BOOST_VER=1.83

RUN apt-get update &&\
 apt-get upgrade -y &&\
 apt-get install -y --no-install-recommends\
 python3\
 python3-numpy\
 python-is-python3\
 libboost-system$BOOST_VER\
 libboost-thread$BOOST_VER\
 libboost-serialization$BOOST_VER\
 libboost-python$BOOST_VER\
 libboost-regex$BOOST_VER\
 libboost-chrono$BOOST_VER\
 libboost-date-time$BOOST_VER\
 libboost-atomic$BOOST_VER\
 libboost-iostreams$BOOST_VER\
 libfreetype6\
 postgresql-$POSTGRES_VER\
 postgresql-client-$POSTGRES_VER\
 postgresql-plpython3-$POSTGRES_VER\
 pgtop\
 gosu\
 wget\
 zip &&\
 apt-get clean -y

COPY --from=build \
  /rdkit/build/RDKit-*Linux-Runtime.deb \
  /rdkit/build/RDKit-*Linux-Python.deb \
  /rdkit/build/RDKit-*-Linux-PgSQL.deb \
  /tmp/
RUN dpkg -i /tmp/RDKit-*.deb && rm -f /tmp/*.deb

WORKDIR /

# make the sample config easier to munge (and "correct by default")
RUN mv -v "/usr/share/postgresql/$POSTGRES_VER/postgresql.conf.sample" /usr/share/postgresql/ \
  && ln -sv ../postgresql.conf.sample "/usr/share/postgresql/$POSTGRES_VER/" \
  && sed -ri "s!^#?(listen_addresses)\s*=\s*\S+.*!\1 = '*'!" /usr/share/postgresql/postgresql.conf.sample

RUN mkdir -p /var/run/postgresql && chown -R postgres:postgres /var/run/postgresql && chmod 2777 /var/run/postgresql

ENV RDBASE=/usr/share/RDKit
ENV PGDATA /var/lib/postgresql/data
ENV PATH $PATH:/usr/lib/postgresql/$POSTGRES_VER/bin
# this 777 will be replaced by 700 at runtime (allows semi-arbitrary "--user" values)
RUN mkdir -p "$PGDATA" && chown -R postgres:postgres "$PGDATA" && chmod 777 "$PGDATA"
VOLUME /var/lib/postgresql/data

ENV PYTHONPATH=/usr/lib/python3.11/site-packages

COPY docker-entrypoint-cartridge.sh /usr/local/bin/docker-entrypoint.sh
RUN ln -s usr/local/bin/docker-entrypoint.sh / # backwards compat
ENTRYPOINT ["docker-entrypoint.sh"]

EXPOSE 5432
CMD ["postgres"]