Dockerfile

# The base image will be used by additional layers
# Only add tools, libraries, etc that will be
# useful in all of the images.
# This is the base layer for the Production Image
ARG hit_master_registry="066561413325.dkr.ecr.us-east-2.amazonaws.com"

FROM ubuntu:22.04 AS baseimage

# Install the packages we'll need in all images
RUN DEBIAN_FRONTEND=noninteractive \
    apt-get update && \
    apt-get dist-upgrade -y && \
    apt-get autoremove -y && \
    apt-get install --no-install-recommends -y \
    python3-pip \
    python3.10-venv

WORKDIR /app
COPY src/ /app/src/

# The builder image contains additional tools we'll need for building requirements.
# This image will be thrown away after we copy needed build artifacts out.
FROM baseimage AS builder

COPY requirements.txt /app/

RUN /usr/bin/python3 -m venv --copies /opt/venv &&       \
    /opt/venv/bin/pip3 install setuptools wheel &&                  \
    /opt/venv/bin/pip3 install --no-cache -r requirements.txt

# Add CA Certs to Certifi CA Store
COPY .build/certificates/HIT-CA-Chain.pem /tmp/
RUN cat /tmp/HIT-CA-Chain.pem >> \
    $(/opt/venv/bin/pip3 show certifi | awk '/Location/ {print $2}')/certifi/cacert.pem

# Start with the Production Image
# This has the baseline tools we need, the source code,
# and copies in the venv with the prod modules
FROM baseimage AS prodimage
COPY --from=builder /opt/venv /opt/venv/

# Set the environment variable
ENV EASYOCR_MODULE_PATH=/tmp/.EasyOCR
ENV HF_HOME=/tmp/docling_cache/huggingface
ENV UV_CACHE_DIR=/tmp/docling_cache/uv
ENV DOCLING_SERVE_ARTIFACTS_PATH=/tmp/docling_cache/docling


# Create a folder (if it doesn't already exist) and set permissions
RUN mkdir -p /nonexistent && chmod 777 /nonexistent
RUN mkdir -p /nonexistent/.EasyOCR && chmod 777 /nonexistent/.EasyOCR

RUN apt-get update && apt-get install -y --no-install-recommends libicu-dev
RUN apt install ttf-mscorefonts-installer -y

RUN apt-get update && apt-get install -y \
      libfontconfig1 \
  && rm -rf /var/lib/apt/lists/*


COPY .build/startup.sh /app/

ENTRYPOINT ["/app/.build/startup.sh"]
# Default to running the API, change to "streamlit" to run frontend
CMD [""]