ocrd-galley/Dockerfile-ocrd_olena

ARG GIT_COMMIT="latest"
FROM quratorspk/ocrd-galley-core:$GIT_COMMIT

ARG PIP_INSTALL="pip install --no-cache-dir"
ARG OCRD_OLENA_VERSION="1.3.0"


# Build ocrd_olena
RUN apt-get update && \
    apt-get install -y \
      imagemagick \
    && \
    apt-get clean && rm -rf /var/lib/apt/lists/*

# Install olena from .deb
RUN curl -sSL --retry 3 -O https://qurator-data.de/~mike.gerber/olena_2.1.0+ocrd-git+2-ubuntu22.04/olena-bin_2.1.0+ocrd-git+2_amd64.deb && \
    dpkg -i --force-depends olena-bin_2.1.0+ocrd-git+2_amd64.deb && \
    rm -f olena-bin_2.1.0+ocrd-git+2_amd64.deb && \
    apt-get update && \
    apt-get -f install -y && \
    apt-get clean && rm -rf /var/lib/apt/lists/* && \
    if ! scribo-cli sauvola --help >/dev/null 2>&1; then echo "Olena/scribo is not installed correctly" >&2; exit 1; fi
RUN curl -sSL --retry 3 -o ocrd_olena.tar.gz https://github.com/OCR-D/ocrd_olena/archive/v${OCRD_OLENA_VERSION}.tar.gz && \
   mkdir ocrd_olena && \
   tar xvz -C ocrd_olena --strip-components=1 -f ocrd_olena.tar.gz && \
   cd ocrd_olena && \
   sed -i 's/^install: deps/install:/' Makefile && \
   ${PIP_INSTALL} ocrd && \
   apt install xmlstarlet && \
   make install PREFIX=/usr/local && \
   cd .. && rm -rf ocrd_olena ocrd_olena.tar.gz


# Check pip dependencies
RUN pip check


# Default command
CMD ['ocrd-olena-binarize']