I recently realised that every day I do the same set of commands over and over:

yarn eslint src/some/folders/**/*.js --fix
git add .
git commit -m "commit message"
git push

What if I could increase my productivity by merging this all into one command? 🤔

The magic command

Now when I run:

magic "commit message"

My code will be linted, committed and pushed! 🎉

To accomplish this, I’m storing the following aliases in my ~/.bashrc file:

alias lint="git add . && git diff HEAD --name-only --diff-filter=d | grep '.*.js$' | xargs yarn eslint --fix"
alias magic='f() {branch=$(git rev-parse --abbrev-ref HEAD); lint && git add . && git commit -m "$1" && git push -u origin $branch;};f'

Read on if you want a bit more of an explanation of what’s happening.

Linting your changed files

git add . && git diff HEAD --name-only --diff-filter=d |
grep '.*.js$' |
xargs yarn eslint --fix

• I run git add . first to make sure any new files that I’ve added are being tracked by git
• git diff --staged --name-only will grab the file names of all staged files
• --diff-filter=d ignores the files that have been deleted (we don’t need to lint those!)
• the grep will find only files ending in .js
• Finally we run yarn eslint --fix on each of those files

You may have to modify this depending on your linting setup, and what file extensions you use. Thanks to Masato Ohba, who provided the initial snippet.

Adding, committing, and pushing your code

f() {
    branch=$(git rev-parse --abbrev-ref HEAD);
    lint && git add . && git commit -m "$1" && git push -u origin $branch;
}

• We are storing the current branch name in $branch
• $1 refers to the first argument that we pass in - the commit message
• -u is the shortened form of --set-upstream (which you only need to do the first time you push after creating a branch)

Thanks for reading!

Let me know if this works for you, or if you have any cool aliases you’d like to share!