The use of Git in Statisical Programming

🚧 This repository is under construction yet. 🚧

It is intended to support the working group for the use of Git in Statistical Programming.

If you are interested in joining, please go here to find out more!

Take a look to the website (made with Quarto).

1. Objectives and Deliverables

1. Principles for using Git and example workflow(s) in context of the current QC process

2. PHUSE white paper on Git for statistical programming, to cover:

• The case for Git
• Challenges in using Git
• Current QC process
• Using Git for audit trails
• How Git can work with the current QC process
• How Git is typically used outside of statistical programming and why
• What the QC process could look like when using the benefits of a typical Git framework
• Regulators’ point of view

3. Core education on Git for statistical programmers
4. Soft proposal for an alternative QC process using Git features

2. How to Contribute

You can get involved in one of the following ways

2.1 - If you wish to join the working group please go here for the leads to reach out to them

2.2 - You can get involved in discussion via the issues

2.3 - You can propose a merge request via a fork if you see something that needs fixing

2.3. Using Git, to propose changes!

2.3.1 About cloning, branching and pushing

Basic steps to get into this "Git world"

0- Ensure you have installed Git, and you use the terminal to go to your favorite directory.

1- Clone the repo from your terminal git clone https://github.com/phuse-org/git-in-statistical-programming.git

2- Create a branch with your username git checkout -b yourname_dev All the changes and contributions will be done here

3- once this is done, add changes to stage git add .

4- once this is done, commit the changes with a message `git commit -m " i did this change (your summary of the changes)"

5- finally push the changes! git push origin yourname_dev

6- if you go now to the repo https://github.com/phuse-org/git-in-statistical-programming.git you will see at the top a message saying your changes has been pushed into your named branch.

7- if you want your changes being in the "main" branch (default page), you have to create a merge request. (green button top right saying "compare & pull request")

💡NOTE: If you like to use your Chat GPT to teach you more about these steps, use it. No fear to break anything! 💡NOTE2: There is a wonderful tutorial in [Coursera] (https://www.coursera.org/learn/making-data-science-work-for-clinical-reporting#modules) to practise on your own how to do these steps. Very easy and well explained (and you get a certification 🎓 at the end!)

2.3.2 About R

This repository will be using quarto. To support this repository, there is an renv lock file which is currently bare bones. (This means we can keep traceability of the R packages we are using within this repo)

# reinstall environment packages
renv::restore() # to install packages required

# if you made changes to the environment (installing new packages) you can save them by:
renv::snapshot()

For the poster it would be nice to have some stats in a jupyter or Rmarkdown about issues around Git (or some charts explaining key points). To start with,

0 - after you have clones the repo in your laptop

1 - you are in your branch

# in terminal, within your project directory
#create your own git branch:
git checkout -b yourname_dev
# or switch to your branch
git switch yourname_dev

2- A folder called "script_ideas" will be the SANDBOX with your scripts/ collaborative scripts/pictures to get some charts/numbers around this topic (for the paper or poster, or to discuss in our meetings).