# https://hub.docker.com/_/python/tags?name=3.11.13-slim-bookworm
ARG BASE_IMAGE=docker.io/python:3.13.9-slim-bookworm@sha256:ee6c0814498e24cc54424475b9611edb99072250b2b30f9fb83d9256a1469b7a
FROM ${BASE_IMAGE}

# Set the timezone to UTC
RUN ln -snf /usr/share/zoneinfo/Etc/UTC /etc/localtime

# Set-up groups, users, and directories
RUN adduser --uid 1000 --system --group --no-create-home airbyte

# Create the cache airbyte directories and set the right permissions
RUN mkdir --mode 755 /airbyte && \
    mkdir --mode 755 /custom_cache && \
    mkdir /secrets && \
    mkdir /config && \
    mkdir /nonexistent && \
    chown airbyte:airbyte /airbyte && \
    chown -R airbyte:airbyte /custom_cache && \
    chown -R airbyte:airbyte /secrets && \
    chown -R airbyte:airbyte /config && \
    chown -R airbyte:airbyte /nonexistent && \
    chown -R airbyte:airbyte /tmp


ENV POETRY_VIRTUALENVS_CREATE=false
ENV POETRY_VIRTUALENVS_IN_PROJECT=false
ENV POETRY_NO_INTERACTION=1

# Create and set up pip cache directory
# TODO: Remove this block if not needed.
ENV PIP_CACHE_DIR=/pip_cache
RUN mkdir -p ${PIP_CACHE_DIR} && chown -R airbyte:airbyte ${PIP_CACHE_DIR}

# Install pip and poetry
RUN pip install --upgrade \
      pip==24.0 \
      setuptools==78.1.1 \
      poetry==1.8.4

# Install system dependencies 
RUN apt-get update && \
    apt-get upgrade -y && \
    apt-get dist-upgrade -y && \
    apt-get clean
RUN apt-get install -y socat=1.7.4.4-2

# Install system dependencies related to OCR and PDF processing
# TODO: move these out of the base image since they are only needed for a few connectors
RUN apt-get update && \
    apt-get install -y \
      tesseract-ocr=5.3.0-2 \
      poppler-utils=22.12.0-2+deb12u1

# NLTK data directory (this may not be needed)
# ENV NLTK_DATA_DIR="/usr/share/nltk_data"
# RUN mkdir -p 755 ${NLTK_DATA_DIR}
# TODO: remove this block if not needed. If it is needed, finish copying over from Airbyte CI.

# Set a whole bunch of permissions for the non-root user (delete this block if not needed)
# RUN mkdir --mode 755 /airbyte && \
#     mkdir --mode 755 /custom_cache && \
#     mkdir -p 755 /usr/share/nltk_data && \
#     chown -R airbyte:airbyte /airbyte && \
#     chown -R airbyte:airbyte /custom_cache && \
#     chown -R airbyte:airbyte /secrets && \
#     chown -R airbyte:airbyte /config && \
#     chown -R airbyte:airbyte /usr/share/pki/ca-trust-source && \
#     chown -R airbyte:airbyte /usr/share/pki/nltk_data && \
#     chown -R airbyte:airbyte /etc/pki/ca-trust && \
#     chown -R airbyte:airbyte /tmp