Setup VOLTTRON for development

Prerequisites

This document assumes you already have

  1. A Linux environment - This can be a Linux machine, a virtual machine or a docker container.

  2. VOLTTRON Pre-requisites software - Refer: VOLTTRON Pre-requisites

  3. Forked the volttron-* repository you are editing. See forking volttron-core for development

Running VOLTTRON with source code

  1. Create a new directory, say volttron_instance1

  2. Create a new virtual environment in that directory and activate it

  3. Install volttron-core, volttron-lib-auth and volttron-lib-zmq.

    You can point to your clone fork directory for the library you are editing and use pypi packages for rest. You can do this using pip or poetry

    1. Using pip

      • pip install -e <path to cloned volttron-core dir> or pip install volttron-core

      • pip install -e <path to cloned volttron-lib-auth dir> or pip install volttron-lib-auth

      • pip install -e <path to cloned volttron-lib-zmq dir> or pip install volttron-lib-zmq

    or

    1. Using poetry and volttron-zmq wrapper

      • Clone the volttron-zmq library that is a wrapper that is dependent on volttron-core, volttron-lib-auth, and volttron-lib-zmq

      • Update pyproject.toml file of volttron-zmq to point to cloned source directory of volttron-core, and/or volttron-lib-auth, and/or volttron-lib-zmq and mark these source package as editable so that you don’t have to rebuild the wheel

      • run poetry install

    If you updating one of volttron agents or volttron-lib-* packages, you can simply install volttron using pip intall volttron

  4. Optional: export VOLTTRON_HOME=/path/to/volttron-home` Defaults to $HOME/.volttron. Note: VOLTTRON>= 10 cannot work with volttron home directory used for previous versions of VOLTTRON

  5. Start volttron volttron -l volttron.log & This creates a new poetry project in $VOLTTRON_HOME and use that.

  6. You can check the status by running the command vctl status

  7. To install agents use vctl install <agent package name in pypi> or vctl install <path to forked agent source dir>

  8. To install any optional libraries use the vctl install-lib command. Example: vctl install-lib volttron-lib-bacnet-driver or vctl install-lib <path to forked library directory>

  9. Using vctl install and vctl install-lib ensures that versions of python packages used by VOLTTRON are all compatible with each other. Using pip install <library> might work if there is no dependency version mismatch but pip doesn’t do dependency management as well as poetry.

Note:

To debug your source code during development, you could run/debug volttron and vctl commands from within an IDE. Refer: PyCharm development environment

After executing the above commands your environment is setup for starting volttron. The next step is to begin developing an agent(see agent development, or continue on with setting up pycharm for use with this environment(see Pycharm development environment)

Note: To exit the poetry shell use the exit command. this deactivates the environment and exits the spawned shell.