Version control is the practice of tracking and managing changes to files.
Does this look familiar?
├── stats-publication
│ ├── publication-analysis-code.R
│ ├── publication-analysis-code-v2.R
│ ├── publication-analysis-code-v2 NEW METHODOLOGY.R
│ ├── publication-analysis-code-v2 NEW METHODOLOGY AB-changes.R
│ ├── publication-analysis-code-final.R
│ ├── publication-analysis-code-final April 2023.R
Git
Git is a free and open source software for version control.
It allows you to:
Record any changes you make to files (these records are called ‘commits’)
Undo changes and revert to previous version of files (if required)
Collaborate with others on the same project using ‘branches’
Git Commits
Commits contain information on:
what change was made,
when the change was made,
why the change was made, and
who made the change.
Git tracks changes to the content of files, not just the file as a whole. This means the information above can be recorded for changes as small as one character on one line of code.
What is in a code repository?
A version controlled code repository will usually contain files for one project and can contain:
Code (e.g. R scripts)
Documentation (e.g. README)
Configuration files
…but NOT DATA!
Ignoring files
A .gitignore file tells Git which files in the code repository it shouldn’t track changes to.
Can include specific files, folders or file extensions
GitHub is often used as the main copy of a code repository (or ‘remote’). Analysts or developers can take a copy (or ‘clone’) of the repository from GitHub to work on locally.
Use Git locally to track changes and regularly ‘push’ to GitHub
Use GitHub to facilitate code review and merging of branches
Why use Git and GitHub?
Preferable to lots of copies of the same file with various names!
