Dockerfile-airflow

FROM apache/airflow:3.1.7-python3.12
# See https://airflow.apache.org/docs/docker-stack/build.html
USER root

RUN apt-get update && \
  apt-get install -y --no-install-recommends \
    libpq-dev \
    libpq5 \
    apt-utils \
    build-essential \
    ca-certificates \
    curl \
    git \
    gnupg \
    lsb-release \
    wget \
    unzip \
    libkrb5-dev \
  && apt-get autoremove -yqq --purge \
  && apt-get clean \
  && rm -rf /var/lib/apt/lists/*

USER airflow
WORKDIR /home/airflow/

COPY --chown=airflow:airflow pyproject.toml ./pyproject.toml
COPY --chown=airflow:airflow databridge_etl_tools ./databridge_etl_tools
COPY --chown=airflow:airflow tests/ ./tests/

# Python syntax check
RUN python -m compileall ./databridge_etl_tools

# Dedicated sandbox for Openmetadata library
RUN python -m venv /home/airflow/venvs/om && \
    /home/airflow/venvs/om/bin/pip install "openmetadata-ingestion==1.12.1" "cachetools" "pendulum"

# Dedicated sandbox for ArcGIS library
RUN python -m venv /home/airflow/venvs/arcgis && \
    /home/airflow/venvs/arcgis/bin/pip install "pendulum" "arcgis"

# Install databridge-etl-tools using pyproject.toml
RUN pip install --no-cache-dir "apache-airflow==${AIRFLOW_VERSION}" \
  "apache-airflow-providers-amazon[s3fs]" .
  # --constraint "https://raw.githubusercontent.com/apache/airflow/constraints-${AIRFLOW_VERSION}/constraints-3.12.txt" .