Skip to content

dev_setup_without_docker_with_conda.md

Latest commit

 

History

History
executable file
·
64 lines (46 loc) · 2.11 KB

dev_setup_without_docker_with_conda.md

File metadata and controls

executable file
·
64 lines (46 loc) · 2.11 KB

Development with conda

If you do not have docker, you can setup your local development environment with conda. However, the Python version will be 3.9 and versions of some Python packages will differ from those defined for Docker and poetry installations. This can cause issues where code that runs on Docker does not run in the conda environment or produces different results. Any such issues should be posted on GitHub.

Prerequisites

A recent version of conda must be installed. See:

The environment.yml dependency specification is tested in eis_toolkit with the libmamba solver instead of the default. If you encounter installation issues following this guide further, especially on Windows, you can enable the libmamba solver globally(!) as follows:

conda install -n base conda-libmamba-solver
conda config --set solver libmamba

See https://conda.github.io/conda-libmamba-solver/getting-started/ for further info.

Set up a local conda environment

Run all commands in the repository root unless instructed otherwise

  1. Install dependencies and create a new conda environment using the provided environment.yml file. The environment name is defined in environment.yml (eis_toolkit).
conda env create -f environment.yml
# You can overwrite an existing environment named eis_toolkit with the --force flag
conda env create -f environment.yml --force
  1. Activate the environment.
conda activate eis_toolkit

  1. With the environment active, the package and all its dependencies should be available for execution.

  2. E.g. run pytest to verify that the test suite works with your installation. If not, you should first verify your installation and secondly make a GitHub issue if you cannot figure out the problem.

Further info

You can add your own packages to the environment as needed. E.g. jupyterlab:

# -c conda-forge specifies the conda-forge channel, which is recommended
conda install -n eis_toolkit -c conda-forge jupyterlab