Roger Steve Ruiz is a software engineer.
Reading time: 4 mins

Check for changes on git pull

Written on 01 Oct 2021 (Link to this post)
automation devex git

As projects grow, there are dependencies that fall outside of the project’s dependency graph. This means that there are some actions you need to perform on your code base when things change. Using the post-merge git-hook can help automate this for your team.

Table of contents

Working on large projects with people means that you’re going to be collaborating asynchronously a lot. Even with the best intentions, it can be difficult to track what’s changed after pulling code down from your Git repository. Things like keeping dependencies up to date and running local database migrations are examples of possible slippage that can lead to developer confusion or friction. But other times, you may just need a reminder that some internal documentation has changed and that folks show read it.

It’s really helpful that whenever you do a git pull, you are given a list of files that change. But if you’re not paying attention or that list of changes is too long, you can get lost and forget to run certain things on your local machine.

With a post-merge Git hook, you can run an arbitrary number of commands whenever certain files or directories change. These examples only run echo commands as it could be intrusive to run code on other’s machines, but it’s easy enough to run a command rather than only suggest running a command if you’d like. After all, the post-merge hook is a local Bash script.

Anatomy of a post-merge hook

Below is an example of a Git hook that is saved as an executable file in your project’s .git/hooks/post-merge file. This one is written using Bash, but feel free to tweak things

Viewing the contents of the file
cat .git/hooks/post-merge
#!/usr/bin/env bash

set -eo pipefail

function changed {
  local chg_str=$1
  git diff --name-only 'HEAD@{1}' HEAD | grep "^${chg_str}" > /dev/null 2>&1

if changed 'migrations/'; then
  echo "🗄️ The migrations/ directory has changed. You may want to run \`make db_dev_migrate\`."

The first thing you’ll notice is that we have a function called changed which takes a single argument and uses that to search using grep through the output of git diff with some flags and options. Let’s break those down to make sure we understand what’s going on.

Running a diff between the last pull and HEAD

For a more up-to-date and thorough explanation of the following git diff command, please read the official Git documentation on git-diff and git-rev-parse . Those docs will be more current and may be better at explaining things than my summary.

When you run git diff with the --name-only flag, you’re asking for just the filenames for any files that were changed. After --name-only, we are passing the two <refname> that we want to compare. In this case, it’s 'HEAD@{1}' and HEAD respectively. The @{1} means the previous value of HEAD that existed before the git pull took place. That particular <refname> is also quoted in single-quotes to prevent issues with the @ at-sign and {} braces in the Bash script.

Checking for changes

With a changed function, we’re able to write if-statements with simple strings to check for the existence of a file in the output of the git-diff command. This is useful for checking both exact files and directories. But make sure you include the full path to a file if it’s nested in a directory.

Anatomy of a changed if-statement

Like I said above, the string that’s passed in as an argument for changed is grepping for a value that starts the string. Notice the ^ in the changed function above. This means that checking for files or paths means that we will need to include the full path and not just any part of a file or directory.

The reason for this is important, but I won’t go into it here. Take a look at what the git-diff command outputs after you’ve done a git-pull to better understand this. I will say that it’s always better to be specific rather than general when checking for things. If that’s enough reason, then just keep that in mind.

Now the example above shows that we’re checking for a mirgations/ directory to have changes in it. But we can easily add other if-statements to check for other common files or directories that have changed.

Check for changes in the docs/adrs directory.

if changed 'docs/adrs/'; then
  echo "📄 The docs/adrs directory has changed. Please review any ADR changes."

Check for changes of the package.json file.

if changed 'package.json'; then
  echo "🧰 The contents of package.json has changed. Run \`npm install\` or \`yarn\` to update your local dependencies."

The options here are limited only by the files or directories that are changing in your repository. And as I mentioned earlier, this script exists locally on your machine. So even if your team leverages the post-merge hook for everyone, you can customize this script yourself and check for explicit changes to files and folders.

This post was written by a human & not by artificial intelligence (AI) tools. I don't have anything against AI but I am interested in differenciating content created by people versus machines. To find out more about the Not by AI badge, please click it.

If you enjoyed this post, please explore other posts by the topics listed below.

automation devex git