Using the development branch

A new version of PCOT is released every few months, but PCOT is continuously being updated and refined. If you want to try the latest development branch, this page is for you.

List of changes

A list of changes for the dev build from the previous release can be found as the most recent entry in the dev branch's releases file.

Installing the dev branch

You'll need to delete the old PCOT conda environment and recreate it with the new dependencies. I am assuming you already have a working PCOT installation and know how to open an Anaconda shell.

The process is very similar to a full install, because dependencies often change. From this shell:

  • Go into the directory into which you installed PCOT. It should contain the pyproject.toml file.
  • conda deactivate if you are already in the PCOT environment.
  • conda env remove -n pcot to delete the old environment.
  • git checkout dev to move to the dev branch.
  • git pull to get the latest changes.
  • conda create -n pcot python=3.11 poetry to create a new environment with the correct Python and the Poetry dependency manager.
  • conda activate pcot to work in the new environment.
  • poetry install to install the dependencies.

Some modern Mac users previously had to do some Weird Stuff with an osx-64 environment to get things working; hopefully this will no longer be necessary.

Problems with pre-existing Python packages

You may have pre-existing packages in your system level Python installation which conflict with those in Anaconda. Typically, this manifests as lots of errors when you run poetry install. To fix this, tell Anaconda not to look at your system install at all:

conda env config vars set PYTHONNOUSERSITE=1 -n pcot
conda deactivate
conda activate pcot
poetry install

Updating the dev branch

From time to time, it's a good idea to git pull on this branch to download the most recent changes. If you get errors running PCOT after this, it may be that the dependencies have changed. Try running poetry install to refresh them.

Going back to the old version

If you have problems, you can deactivate and delete the old environment, check out the master branch and recreate and reinstall the environment - as if you were doing a standard install.