improved devcontainer for langchain

dev.Dockerfile

# This is a Dockerfile for running unit tests

# Use the Python base image
FROM python:3.11.2-bullseye


# install bash completion - to make make work
RUN apt-get update && apt-get install -y bash-completion

# Install nodejs & yarn (for docs)
# inspired by https://github.com/nikolaik/docker-python-nodejs/blob/main/Dockerfile
RUN \
  echo "deb https://deb.nodesource.com/node_20.x bullseye main" > /etc/apt/sources.list.d/nodesource.list && \
  wget -qO- https://deb.nodesource.com/gpgkey/nodesource.gpg.key | apt-key add - && \
  echo "deb https://dl.yarnpkg.com/debian/ stable main" > /etc/apt/sources.list.d/yarn.list && \
  wget -qO- https://dl.yarnpkg.com/debian/pubkey.gpg | apt-key add - && \
  apt-get update && \
  apt-get upgrade -yqq && \
  apt-get install -yqq nodejs yarn 

# Add user python - this also creates the homedir
RUN useradd -ms /bin/bash python

# Create some dirs 
# used for poetry env
RUN mkdir -p /opt/poetry
# used for poetry virtual envs
RUN mkdir -p /opt/virtual-envs

# set the correct ownership
RUN chown python /opt/virtual-envs
RUN chown python /opt/poetry

# Switch to python user
USER python


# Create a Python virtual environment for Poetry and install it
# Define the version of Poetry to install (default is 1.4.2)

ARG POETRY_VERSION=1.5.1
ARG POETRY_HOME=/opt/poetry

RUN python3 -m venv ${POETRY_HOME} && \
  $POETRY_HOME/bin/pip install --upgrade pip && \
  $POETRY_HOME/bin/pip install poetry==${POETRY_VERSION}

# Test if Poetry is installed in the expected path
RUN echo "Poetry version:" && $POETRY_HOME/bin/poetry --version

# add poetry to paths
RUN echo $PATH
RUN echo "export PATH=$PATH:/opt/poetry/bin" >> ~/.bashrc
RUN cat ~/.bashrc
RUN echo $PATH


# Set the working directory for the app
# This is the path where the code gets mounted
# This is crucial as poetry will calculate the name of the poetry env based on the path
WORKDIR /workspaces/

# force the settings to not store things in code-dir/.venv
ENV POETRY_VIRTUALENVS_PATH=/opt/virtual-envs
ENV POETRY_VIRTUALENVS_IN_PROJECT=false

# Copy only the dependency files for installation
# We put them on the right place so poetry can find them again
COPY libs/langchain/pyproject.toml libs/langchain/poetry.lock libs/langchain/poetry.toml libs/langchain/

# Install the Poetry dependencies (this layer will be cached as long as the dependencies don't change)
RUN cd libs/langchain/ &&  $POETRY_HOME/bin/poetry install --sync --no-interaction --no-ansi --no-root --with dev,test

#  Unable to find installation candidates for nvidia-cudnn-cu11 (8.5.0.96)
#RUN $POETRY_HOME/bin/poetry install --no-interaction --no-ansi --with dev,test,test_integration
# adding extended-testing makes it fail
# CRITICAL:root:A GDAL API version must be specified. Provide a path to gdal-config using a GDAL_CONFIG environment variable or use a GDAL_VERSION environment variable.
#RUN $POETRY_HOME/bin/poetry install --no-interaction --no-ansi --with dev,test -E extended-testing

# This is more to calculate a hash for bumping
# We copy this (though it's overridden by the mount)
# This allows us to detect if something changed in libs/langchain
# and if needed re-install local
COPY libs/langchain libs/langchain
# If there is a need to rebuild , the previous poetry packages will be cached
RUN cd libs/langchain/ && $POETRY_HOME/bin/poetry install --sync --with dev,test 

# note this also depends on langchain 
# todo: figure out how to also cache this based in the files , so we don't need to reinstall
COPY pyproject.toml poetry.lock poetry.toml ./
RUN $POETRY_HOME/bin/poetry install --sync --no-interaction --no-ansi  --with docs