Dockerfile_singleuser

FROM jupyterhub/singleuser:5.4

ARG HTTP_PROXY
ARG HTTPS_PROXY
ARG NO_PROXY
ARG http_proxy
ARG https_proxy
ARG no_proxy

# set to "true" for the GPU build
ARG GPU_BUILD=false

ENV HTTP_PROXY=$HTTP_PROXY
ENV HTTPS_PROXY=$HTTPS_PROXY
ENV NO_PROXY=$NO_PROXY
ENV http_proxy=$HTTP_PROXY
ENV https_proxy=$HTTPS_PROXY
ENV no_proxy=$NO_PROXY

ENV DEBIAN_FRONTEND=noninteractive
ENV DEBIAN_PRIORITY=critical

ARG CPU_ARCHITECTURE
ENV CPU_ARCHITECTURE=${CPU_ARCHITECTURE}

ENV NVIDIA_DRIVER_VERSION=510

# nvidia-container-runtime
ENV NVIDIA_VISIBLE_DEVICES=all
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility


# prefer prebuilt wheels, never build from sdist for known heavy libs
ENV PIP_PREFER_BINARY=1
ENV PIP_DISABLE_PIP_VERSION_CHECK=1
ENV PIP_NO_CACHE_DIR=1
# allow python to write byte code
ENV PYTHONDONTWRITEBYTECODE=0

USER root

RUN apt-get update && apt-get upgrade -y && \
     apt-get install -y software-properties-common

# add extra repos
RUN apt-add-repository multiverse && \
    apt-add-repository universe && \
    add-apt-repository ppa:graphics-drivers/ppa && \
    add-apt-repository ppa:deadsnakes/ppa && \
    apt-get update && apt-get upgrade -y && \
    apt-get --force-yes -o Dpkg::Options::="--force-confold" --force-yes -o Dpkg::Options::="--force-confdef" -fuy full-upgrade && \
    apt-get autoremove -y && \
    apt-get clean -y && \
    apt-get install -y \
    wget \
    curl \
    ca-certificates \
    apt-transport-https

# Microsoft repos
RUN wget -q -O- https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor -o /etc/apt/trusted.gpg.d/packages.microsoft.gpg
RUN echo "deb [arch=amd64,armhf,arm64] https://packages.microsoft.com/ubuntu/22.04/prod jammy main" | tee -a /etc/apt/sources.list

# Add latest ubuntu repos to sources.list
# add arch
RUN if [ "${CPU_ARCHITECTURE}" = "amd64" ]; then \
        echo "deb http://archive.ubuntu.com/ubuntu/ jammy main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src http://archive.ubuntu.com/ubuntu/ jammy main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb http://archive.ubuntu.com/ubuntu/ jammy-security main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src http://archive.ubuntu.com/ubuntu/ jammy-security main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb http://archive.ubuntu.com/ubuntu/ jammy-backports main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src http://archive.ubuntu.com/ubuntu/ jammy-backports main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb http://archive.canonical.com/ubuntu/ jammy partner" | tee -a /etc/apt/sources.list ; \
        echo "deb-src http://archive.canonical.com/ubuntu/ jammy partner" | tee -a /etc/apt/sources.list ; \
    elif [ "${CPU_ARCHITECTURE}" = "arm64" ]; then \
        echo "deb [arch=arm64] http://archive.canonical.com/ubuntu/ jammy partner" | tee -a /etc/apt/sources.list ; \
        echo "deb-src [arch=arm64] http://archive.canonical.com/ubuntu/ jammy partner" | tee -a /etc/apt/sources.list ; \
        echo "deb [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy-updates main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy-backports main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy-security main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy-security main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy-updates main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy-backports main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
        echo "deb-src [arch=arm64] http://ports.ubuntu.com/ubuntu-ports jammy main restricted universe multiverse" | tee -a /etc/apt/sources.list ; \
    else \
        echo "No valid CPU_ARCHITECTURE specified"; \
    fi

# run updates
RUN apt-get update && apt-get upgrade -y && dpkg --configure -a && \
    apt-get --force-yes -o Dpkg::Options::="--force-confold" --force-yes -o Dpkg::Options::="--force-confdef" -fuy full-upgrade && \
    apt-get autoremove -y && \
    apt-get clean -y && \
    apt-get install -y --no-install-recommends \
    gnupg \
    ssl-cert \
    libevent-dev \
    libsqlite3-dev \
    libxml2-dev \
    libbz2-dev \
    libxslt1-dev \
    zlib1g-dev \
    libssl-dev \
    gnupg-agent \
    dirmngr \
    apt-utils \
    fonts-dejavu \
    nano \
    iputils-ping \
    cmake \
    make \
    build-essential \
    unixodbc \
    unixodbc-dev \
    r-cran-rodbc \
    gfortran \
    gcc \
    g++ \
    git \
    ssh \
    jq \
    htop \
    libglx-mesa0 libegl1 libxrandr2 libxss1 \
    libxcursor1 libxcomposite1 libasound2-dev libxi6 libxtst6 \
    r-base \
    libopenblas-dev

# remove newer python version, freeze other versions for max compatibility before updates
# RUN apt --fix-missing purge $(dpkg -l | grep 'python3\.1[01]' | awk '{print $2}')

################################################################################################################################################################
# GPU requirements section

# ===== GPU-only system bits (optional) =====
# Keep this lightweight; most CUDA libs come from host via NVIDIA Container Runtime.
# If you need CuDNN or CUDA user-space libs inside the container, add them here.
# This block only runs for amd64 when GPU_BUILD=true.
RUN if [ "$GPU_BUILD" = "true" ] && [ "$CPU_ARCHITECTURE" = "amd64" ]; then \
      install -d -m 0755 /etc/apt/keyrings; \
      curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey \
        | gpg --dearmor -o /etc/apt/keyrings/nvidia-container-toolkit.gpg; \
      curl -fsSL https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \
        | sed 's#deb https://#deb [signed-by=/etc/apt/keyrings/nvidia-container-toolkit.gpg] https://#' \
        | tee /etc/apt/sources.list.d/nvidia-container-toolkit.list > /dev/null; \
      apt-get update; \
      # OPTIONAL: only if your code needs OpenCL userspace
      apt-get install -y --no-install-recommends \
        ocl-icd-libopencl1 ocl-icd-dev opencl-headers libtiff-dev; \
      rm -rf /var/lib/apt/lists/*; \
    else \
      echo "CPU build or non-amd64: skipping NVIDIA repo setup"; \
    fi

# Not necessary, will corrupt nvidia-doker on install, use with care
# RUN DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends apnvidia-driver-$NVIDIA_DRIVER_VERSION pocl-opencl-icd intel-opencl-icd \
#   mesa-opencl-icd nvidia-opencl-dev libpocl-dev beignet-opencl-icd libnvidia-egl-wayland-dev nvidia-cuda-dev nvidia-cuda-toolkit nvidia-headless-$NVIDIA_DRIVER_VERSION

RUN apt --fix-broken install

# Install Python 3.11 system-wide
RUN apt-get update && apt-get install -y python3.11-full python3.11-dev python3.11-venv python3-pip \
 && update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.11 2 \
 && update-alternatives --install /usr/bin/pip pip /usr/bin/pip3 2

# install remaining packages
RUN apt-get update && apt-get upgrade -y
RUN apt-get -y --no-install-recommends install unixodbc unixodbc-dev odbcinst
RUN ACCEPT_EULA=Y apt-get install -y msodbcsql18
RUN ACCEPT_EULA=Y apt-get install -y mssql-tools18
RUN echo 'export PATH="$PATH:/opt/mssql-tools18/bin"' >> ~/.bashrc

# deps for psycopg2
RUN apt-get install -y --no-install-recommends postgresql-server-dev-all

# clean up apt
RUN apt-get clean autoclean && apt-get autoremove --purge -y

######################################## Python / PIP SECTION ########################################
RUN conda update conda

# Create a Python 3.11 env and make it the default for builds & kernels
RUN conda create -y -n py311 python=3.11 ipykernel && conda clean --all -f -y
ENV CONDA_DEFAULT_ENV=py311
ENV PATH=/opt/conda/envs/py311/bin:$PATH

RUN conda install --quiet --yes \
    ipykernel \
    pip

RUN conda clean --all -f -y

RUN fix-permissions $CONDA_DIR && fix-permissions /home/$NB_USER

# bust cache
ENV UV_VERSION=latest

# install rust, medcat requirement, install UV
ENV HOME=/root
ENV PATH="/root/.cargo/bin:${PATH}"

RUN curl -sSf https://sh.rustup.rs -o /tmp/rustup-init.sh \
 && chmod +x /tmp/rustup-init.sh \
 && /tmp/rustup-init.sh -y \
 && rm /tmp/rustup-init.sh

RUN curl -Ls https://astral.sh/uv/install.sh -o /tmp/install_uv.sh \
 && bash /tmp/install_uv.sh

RUN UV_PATH=$(find / -name uv -type f | head -n1) && \
    ln -s "$UV_PATH" /usr/local/bin/uv

# update pip and install basic utilities
RUN uv pip install --upgrade --system pip setuptools wheel

# install the rest of the packages including medcat
COPY ./requirements.txt /srv/jupyterhub/

# https://download.pytorch.org/whl/cpu is out of date on packages, using alternative URL for CPU
RUN if [ "$GPU_BUILD" = "true" ] && [ "$CPU_ARCHITECTURE" = "amd64" ]; then \
      uv pip install --system --no-cache-dir -r /srv/jupyterhub/requirements.txt && \
      uv run python -m medcat download-scripts /srv/jupyterhub/medcat-scripts ; \
    else \
      uv pip install --system --no-cache-dir -r /srv/jupyterhub/requirements.txt --index-url https://pypi.org/simple && \
      uv run python -m medcat download-scripts /srv/jupyterhub/medcat-scripts ; \
    fi

RUN uv pip install --no-cache-dir --system -r /srv/jupyterhub/medcat-scripts/requirements.txt

# Copy medcat-scripts to default location (for Docker usage)
# Keep original at /srv/jupyterhub/medcat-scripts for K8s usage
RUN cp -r /srv/jupyterhub/medcat-scripts /home/jovyan/work/medcat-scripts

#######################################################################################################

# install R and other dependencies
COPY ./scripts/r_kernel_install.sh /srv/jupyterhub/
RUN Rscript /srv/jupyterhub/r_kernel_install.sh

# create jupyterhub shared folder
RUN mkdir -p /home/jovyan/scratch
RUN chmod -R 777 /home/jovyan/scratch
RUN chmod g+s /home/jovyan/scratch
RUN chmod 0777 /home/jovyan/scratch

# copy scripts and config files
COPY config/jupyter_notebook_config.py /home/jovyan/.jupyter/

# copy notebooks
COPY notebooks /home/jovyan/work/
# Kubernetes does not automatically copy the contents of /home/joyvan before mounting. Therefore also copying notebooks to /srv/jupyterhub for usage by k8s
COPY notebooks /srv/jupyterhub/notebooks/

WORKDIR /home/jovyan
USER jovyan

RUN echo "cd /home/jovyan/work" >> /home/jovyan/.bashrc

# Configure container startup
ENTRYPOINT ["tini", "-g", "--"]

CMD ["start-singleuser.py"]