22.5. Merging in Git

22.5.1. How to Merge

A merge in Git occurs when two branches are combined in the repository.

Example

Let's say a programmer wants to merge a branch called test into main. To accomplish this:

  1. The programmer switches to the branch they want to merge into. In this case:

    $ git checkout main

  2. The programmer then runs the merge command:

    $ git merge test

When successful, the code in the test branch is pulled into main.

22.5.2. Merge Conflicts

This process is often seamless. In the example in the previous section, a programmer created a branch to change the HTML and the other programmer did the same to change the CSS. Because the two programmers changed different files, the merge of the updated HTML and updated CSS won't create a conflict.

A merge conflict occurs when a change was made to the same line of code on both branches. Git doesn't know which change to accept, so it is up to the programmers to resolve it. Merge conflicts are minor on small applications, but can cause issues with large enterprise applications.

Even though the thought of ruining software can be scary, every programmer deals with a merge conflict during their career. The best way to deal with a merge conflict is to face it head on and rely on teammates for support!

22.5.2.1. Ways to Avoid Merge Conflicts

Even though merge conflicts are normal in Git, it is also normal for programmers to want to do everything they can to avoid them. Here are some tips on how to avoid a merge conflict:

  1. Git has a dry-run option for many commands. When a programmer uses that option, Git outputs what WILL happen, but doesn't DO it. With merging in Git, the command to perform a dry-run and make sure there aren't any conflicts is git merge --no-commit --no-ff <branch>. The --no-commit and --no-ff syntax tells Git to run the merge without committing the result to the repository.

  2. Before merging in a branch, any uncommitted work that would cause a conflict needs to be dealt with. A programmer can opt to not commit that work and instead stash it. By using the git stash command, the uncommitted work is saved in the stash, and the repository is returned to the state at the last commit. If the programmer wants to retrieve stashed work later, they can do so with the command git stash pop.

22.5.3. Check Your Understanding

Question

If a programmer is on the branch test and wants to merge a branch called feature into main, what steps should they take?