ndcube Installation Instructions

Development Version

This page outlines how to install the development version of NDCube. To install NDCube we will use a combination of conda, conda environments, pip and git. We will assume these are all installed on your current system. NDCube also requires Python 3. Python 2 is not supported.

1. Create Conda Environment: ndcube-dev

From command line in any directory, do:

$ conda config --append channels conda-forge
$ conda create -n ndcube-dev sunpy hypothesis pytest-mock pip sphinx coverage ipython jupyter
$ source activate ndcube-dev on windows this is $ activate ndcube-dev

The first line opens a conda channel so that SunPy and its dependencies can be installed. The second line creates the ndcube-dev conda environment with a list of dependencies. The final line activates, or puts us into the new enviroment.

2. Clone NDCube Repository

First we will create a directory to hold the repo.

$ cd
$ mkdir github_repos
$ cd github_repos
$ git clone https://github.com/sunpy/ndcube.git ndcube_dev

If you already have this repo cloned, make sure they are up-to-date but by pulling the latest version of the master branches:

$ cd ~/github_repos/ndcube_dev
$ git pull origin master

assuming that origin is the remote pointing to the main ndcube repo, i.e. https://github.com/sunpy/ndcube.git. To determine what the correct remote name is, cd into the repo's directory and do

$ git remote -v

3. Install the Development Versions of the Repo

$ cd ~/github_repos/ndcube_dev
$ pip install -e .

You should now be ready to use NDCube. To check it's installed, open an Python, IPython, or Jupyter Notebook session from any directory and try
>>> import ndcube