GRASS¶

If you want to use GRASS GIS together with its jupyter companion (grass.jupyter package), we prepared a Dockerfile for you. The image extends the base image.

We occasionally push docker images for the GRASS tag to our registry (https://registry.gitlab.vgiscience.org/lbsn/tools/jupyterlab:grass). If you want the latest or specific versions, you have to build the image yourself.

Use the following docker-compose.grass.yml:

services:
  jupyterlab:
    image: registry.gitlab.vgiscience.org/lbsn/tools/jupyterlab:grass
    build:
      context: ./grass
      args:
        - APP_VERSION=${APP_VERSION:-latest}
    container_name: ${CONTAINER_NAME:-lbsn-jupyterlab}
    restart: "no"
    ports:
      - 127.0.0.1:${JUPYTER_WEBPORT:-8888}:8888
    volumes:
      - ${JUPYTER_NOTEBOOKS:-~/notebooks}:/home/jovyan/work
      - ${CONDA_ENVS:-~/envs}:/envs
    environment:
      - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-eX4mP13p455w0Rd}
      - JUPYTER_PASSWORD=${JUPYTER_PASSWORD:-eX4mP13p455w0Rd}
      - JUPYTER_WEBURL=${JUPYTER_WEBURL:-http://localhost:8888}
    env_file: ".env"
    networks:
      - lbsn-network

networks:
  lbsn-network:
    name: ${NETWORK_NAME:-lbsn-network}
    external: true

See the grass/Dockerfile file for the list of Grass dependencies.

ARG APP_VERSION=latest

FROM registry.gitlab.vgiscience.org/lbsn/tools/jupyterlab:$APP_VERSION

ENV PYTHON_BINDINGS=" \
    gdal-bin \
    python3 \
    python3-pip \
    python3-psycopg2 \
    python3-yaml \
    grass" \
    ## GRASS GIS: Make sure the distro's python is used
    GRASS_PYTHON=/usr/bin/python3

RUN apt-get update \
    && apt install -y --no-install-recommends \
        $PYTHON_BINDINGS

ENV GRASS_BINDINGS=" \
    autoconf2.13 \
    autotools-dev \
    bison \
    flex \
    g++ \
    gettext \
    git \
    imagemagick \
    libblas-dev \
    libbz2-dev \
    libcairo2-dev \
    libfftw3-dev \
    libfreetype6-dev \
    libgdal-dev \
    libgeos-dev \
    libglu1-mesa-dev \
    libjpeg-dev \
    liblapack-dev \
    libncurses5-dev \
    libnetcdf-dev \
    libpng-dev \
    libpq-dev \
    libproj-dev \
    libreadline-dev \
    libsqlite3-dev \
    libtiff-dev \
    libxmu-dev \
    libzstd-dev \
    make \
    netcdf-bin \
    p7zip \
    proj-bin \
    sqlite3 \
    unixodbc-dev \
    xvfb \
    zlib1g-dev \
    parallel"

RUN apt-get update \
    && apt install -y --no-install-recommends \
        $GRASS_BINDINGS

RUN cd ~ \
    && git clone --depth 1 https://github.com/libLAS/libLAS.git liblas \
    && cd liblas \
    && mkdir makefiles \
    && cd makefiles \
    && apt install cmake -y --no-install-recommends \
    && apt install -y --no-install-recommends build-essential cmake libboost-dev libboost-serialization-dev \
        libboost-filesystem-dev libboost-iostreams-dev libboost-program-options-dev \
        zlib1g-dev rapidjson-dev libquadmath0 \
    && apt-get install libboost-thread-dev -y --no-install-recommends \
    && cmake -G "Unix Makefiles" ../ \
    && make \
    && make install

RUN /opt/conda/envs/worker_env/bin/python -m pip install folium

Building the image¶

Build:

docker compose -f docker-compose.grass.yml build \
        --no-cache --progress=plain \
    && docker compose -f docker-compose.grass.yml up -d

Optionally push to a registry:

docker compose -f docker-compose.grass.yml push

Run the image¶

Run:

docker compose -f docker-compose.grass.yml up -d

Example Jupyter notebook¶

See this Jupyter notebook for an example of working with GRASS Gis in Carto-Lab Docker (a viewshed analysis).