When working on projects across multiple computers, while syncing using a remote repository managed by git. you might encounter scenarios where you need to run the same code but keep certain directories, like a results folder, unique to each machine:

  • Machine-Specific Data: The results folder may contain data that’s only relevant to a particular machine.
  • Large Files: Syncing large result files can slow down your version control system.
  • Privacy and Security: Results might contain sensitive information that shouldn’t be shared.

step 1: add results to .gitignore

The .gitignore file tells Git which files or directories to ignore in a repository. So first create a .gitignore file:

touch .gitignore

And, add one line to .gitignore telling Git to ignore the results directory and all of its contents.

results/

And that’s it! Theoretically one still needs to mkdir a results folder locally if it doesn’t exists, and remember to

step 2: (optional) remove previous results tracked by git

If the results folder was previously tracked by Git, you’ll need to remove it from the repository’s tracking while keeping the folder on your local filesystem.

(at the project’s root directory, in terminal, run)

git rm -r --cached results/
git commit -m "Stop tracking the results folder"
git push origin main

step 3: (optional)

Making the results folder to point to a specific location outside your project directory:

  • On Linux/macOS:

    ln -s /path/to/your/local/results results

  • On Windows (Command Prompt):

    mklink /D "results" "C:\path\to\your\local\results"