What is git?

27 November 2022 · Filed in Platform Infrastructure

Git is a very well known open source control tool. But what is it and how is it used?

Source Control Management

Source code for a product, program or Infrastructure can and does change. Keeping track of which version is current is the job of source control. There is normally a default branch or trunk, from which all other branches stem from.

Code can be checked out by a user. In the older versions of source control software this would lock the file to that user, until they checked in back in (with changes usually). The other users would only get read access.

Newer software was written to overcome these limitation like CVS, Subversion.

Where did git come from?

Git was created by Linus Torvalds after BitKeeper withdrew the free (gratis) version. The Mercurial project was also started at the same time (April 2005). These projects created distributed source control systems.

The Source code can be hosted by a Git server. Additional functionality exists from Git as a Service sites such as Github, Gitlab, and Bitbucket. Using a git client with the Git server, the code can easily be distributed amongst users, with updates published in the form of Diffs. The git hosting sites extend this by giving an easier route of merging back changes into the default branch by way of Pull Requests or PRs.

Having the code repository distributed amongst multiple developers means that a group can work on the code at the same time, and to aid merging, can create local copies of the code, called branches. These branches track changes to the code enabling efficient disk usage and easy merging.

Working with git

The basic workflow begins with cloning the git repo to you local machine. If you already have the repo locally, you can fetch changes to the repo, and merge them into your branch. This can also be done in a single pull operation, which combines the fetch and merge based on the current branch.

If desired, a new branch can be created locally and marked as the current branch with the checkout command.

Changes to the code and tests are then performed at this point. The developer would then add the changes to the index followed by a commit to store those changes in the repo. The developer would then push the changes to the remote origin repository. Once pushed, the hosting service will then enable merging of those changes by way of creating a Pull Request.

Pull requests are then usually reviewed by the maintainers of the repository, and merged into the default branch, unless some comments are added.

Want to know more? Come ask me questions. You can find me on Slack.

If you found this useful, subscribe to updates and look out for my YouTube channel content.

Previous Post: System Reliability and Canaries Next Post: What fish teach us about asynchronous operations

Tags: developer · devops · git · source control