10.3 Git Internals - Git References

The question now is, when you run git branch (branchname), how does Git know the SHA-1 of the last commit? The answer is the HEAD file.

The HEAD file is a symbolic reference to the branch you’re currently on. By symbolic reference, we mean that unlike a normal reference, it doesn’t generally contain a SHA-1 value but rather a pointer to another reference. If you look at the file, you’ll normally see something like this:

If you run git checkout test, Git updates the file to look like this:

When you run git commit, it creates the commit object, specifying the parent of that commit object to be whatever SHA-1 value the reference in HEAD points to.

You can also manually edit this file, but again a safer command exists to do so: symbolic-ref. You can read the value of your HEAD via this command:

You can also set the value of HEAD:

You can’t set a symbolic reference outside of the refs style:

We just finished discussing Git’s three main object types, but there is a fourth. The tag object is very much like a commit object – it contains a tagger, a date, a message, and a pointer. The main difference is that a tag object generally points to a commit rather than a tree. It’s like a branch reference, but it never moves – it always points to the same commit but gives it a friendlier name.

As discussed in Git асослари, there are two types of tags: annotated and lightweight. You can make a lightweight tag by running something like this:

That is all a lightweight tag is – a reference that never moves. An annotated tag is more complex, however. If you create an annotated tag, Git creates a tag object and then writes a reference to point to it rather than directly to the commit. You can see this by creating an annotated tag (-a specifies that it’s an annotated tag):

Here’s the object SHA-1 value it created:

Now, run the cat-file command on that SHA-1 value:

Notice that the object entry points to the commit SHA-1 value that you tagged. Also notice that it doesn’t need to point to a commit; you can tag any Git object. In the Git source code, for example, the maintainer has added their GPG public key as a blob object and then tagged it. You can view the public key by running this in a clone of the Git repository:

The Linux kernel repository also has a non-commit-pointing tag object – the first tag created points to the initial tree of the import of the source code.

The third type of reference that you’ll see is a remote reference. If you add a remote and push to it, Git stores the value you last pushed to that remote for each branch in the refs/remotes directory. For instance, you can add a remote called origin and push your master branch to it:

Then, you can see what the master branch on the origin remote was the last time you communicated with the server, by checking the refs/remotes/origin/master file:

Remote references differ from branches (refs/heads references) mainly in that they’re considered read-only. You can git checkout to one, but Git won’t point HEAD at one, so you’ll never update it with a commit command. Git manages them as bookmarks to the last known state of where those branches were on those servers.