Welcome to the home of PharmaDevOps

This PHUSE working group is focussed on ensuring R package authors new to concepts like CICD can can easily add these useful automations to pharmaverse R packages and related biostatistical packages.

We hope to both convince package developers of the value of CICD, provide a minimum bar of what is expected, and provide a set of tools to make it easy to implement.

Getting started with CICD

In this guidance, you’ll see a number of gh-actions that you can use to implement CICD on your project. We have categorised these into three groups:

  • Minimum CICD: These are the minimum CICD tools we recommend for all packages, including those in the pharmaverse and openstatsware. These CICD tools are focussed on ensuring your package works, and the documentation users see matches the latest release of your code.
  • Recommended CICD: These are tools we feel make your life developing and releasing robust code easier.
  • Enhanced CICD: These are tools are option tools for power users, or those with specific needs.

Each CICD action is described via a common pattern, with an emphasis on why we recommend it, and how to use it.

A recomended plan for success in implenting CICD on your project is to:

  • Read this post on the general anatomy of a gh-action
  • Browse our recomended gh-action’s below
  • Implement CICD on your git repo!

We’ve also collected togther this list of external learning resources.

Minimum CICD

The following CICD is what we consider a minimum bar for a package on the pharmaverse. These CICD tools are focussed on ensuring your package works, and documentation matches the code.

There is no need to use the specific code we suggest, if you are happy getting the same functionality from another source, like the ones built into usethis.

The following CICD are tools we feel make your life developing and releasing robust code easier.

Enhanced CICD

The following CICD are tools are optional tools for power users.