Follow the instructions here to install Anaconda or Miniconda.
Assure that your terminal prompt is now prefixed by the conda environment that is currently activated as below
(base) $
If you don't see the (base)
text (indicating you are in the "base" conda environment), it is suggested that you follow this practice
If this is your first time setting up the repository you should use the following command to create the website's environment from file.
conda env create -f environment.yml
You should see a number of packages install using both conda and pip.
To activate, or enter, the newly created environment, you can use the following command.
NOTE: This will need to be each time you wish serve the documentation and open a new terminal.
conda activate mivp-agent-web
You should see this change in your environment in the conda's terminal prompt prefix.
If new dependencies are added to the environment.yml
you can update your local environment by the use of the following command.
IMPORTANT NOTE: This should be run ONLY when the (mivp-agent-web)
environment is activated.
conda env update -f environment.yml
Because the mivp-agent package is not currently on PyPI or other indexes, we can not directly include it in the environment.yml
file. To allow mkdocstrings to import the package and read the doc strings, we must install it in the mivp-agent-web
conda environment. Navigating to the moos-ivp-agent/src/python_module
directory (wherever that is on your system) and running the following command.
pip3 install -e .
Secondly, mkdocstrings needs a directory to watch for updates to know when to regenerate documentation. The plugin is configured to watch a mivp_agent
directory in the root of this repository. All you need to do is link that directory to the moos-ivp-agent/src/python_module/src/mivp_agent
in the main repository (wherever you put that). For example on my system I run the following command
ln -s /Users/carter/src/moos-ivp-agent/src/python_module/src/mivp_agent /Users/carter/src/mivp-agent.github.io/mivp_agent
NOTE: You will still need to refresh the browser to view updates to the python documentation
A local version of the documentation can be served through the following command in the root directory.
mkdocs serve
A static site can be build with the following command.
mkdocs build
- General
- Python API