Developer Setup

Developer Environment Setup

For the installation of qp for the purpose of development, we recommend that you use a separate Anaconda virtual environment with qp installed in “editable” mode with all dev optional dependencies added.

In this guide, we will name this developer environment qp_dev and we will assume that an Anaconda with a minimum Python version of 3.10 has been previously installed.

To install the developer environment:

bash

# Clone the repo and enter it
git clone https://github.com/LSSTDESC/qp.git
cd qp

# Creating the environment from the YAML
conda env create -n qp_dev -f environment.yml

# Activate the environment
conda activate qp_dev

# Install qp in editable mode with dev dependencies
pip install -e .[dev]

zsh

# Clone the repo and enter it
git clone https://github.com/LSSTDESC/qp.git
cd qp

# Creating the environment from the YAML
conda env create -n qp_dev -f environment.yml

# Activate the environment
conda activate qp_dev

# Install qp in editable mode with dev dependencies
pip install -e '.[dev]'

Running tests

To run coverage tests, run the following on the command line from the base qp directory.

./do_cover.sh

The output is in the cover/ folder, which will output HTML files that provide an overview of the coverage status. For a pull request, the goal is to have 100% coverage.

Building Documentation

To build the documentation locally, start by making sure that you have the appropriate documentation packages installed:

bash

pip install -e .[docs]

zsh

pip install -e .[docs]

Once you have the appropriate packages, run the following lines of code to make the documentation:

cd docs/
make html

The HTML files will be generated in the _build/ folder inside the docs/ folder.

Where to go from here

Now that you’ve got a development environment set up, take a look at Code Structure to get a sense of how qp is structured. Make sure to read over the Contribution Guidelines before getting started, as this covers the intended workflow for developers to follow, in addition to expected naming conventions and so on.

If you are looking to work on a new parameterization type, take a look at Creating new parameterizations for detailed instructions and a code template to get you started. Or if you’re looking for ideas of where to start, take a look at Development Priorities or the open issues.