Automatic Versioning Guide

About   |   Requirements   |   Configuration   |   Author

About

This is a quick guide on how to implement automatic semantic versioning using husky.js, commitlint, semantic release and github actions.

Requirements

• git
• node.js
• yarn or npm*

NOTE: If you choose to use npm, replace yarn with npm run in the command to install dependencies and when executing the release script in the workflow.

Configuration

HuskyJS

1. Install husky

yarn add husky --dev

2. Enable Git hooks

yarn husky install

3. To automatically have Git hooks enabled after install, edit package.json

// package.json

{
  "private": true, // <- your package is private, you only need postinstall
  ...
  "scripts": {
    ...
    "postinstall": "husky install"
  }
  ...
}

Commitlint

1. Install commitlint 🔗

yarn add @commitlint/cli @commitlint/config-conventional --dev

2. Configure

// commitlint.config.js

module.exports = {
  extends: ["@commitlint/config-conventional"],
};

3. Add hook

yarn husky add .husky/commit-msg 'yarn commitlint --edit $1'

Semantic Release

1. Install semantic-release

yarn add semantic-release --dev

2. Install plugins 🔗

• Default plugins

  These four plugins are already part of semantic-release and are listed in order of execution. They do not have to be installed separately:

  • @semantic-release/commit-analyzer
  • @semantic-release/release-notes-generator
  • @semantic-release/npm
  • @semantic-release/github

• Additional plugins

yarn add @semantic-release/git @semantic-release/changelog --dev

3. semantic-release configuration 🔗

• Create a .releaserc file, written in YAML, with optional extensions: .yaml / .yml / .json / .js
• 📝 config file example
• edit package.json and add release script

// package.json

{
  ...
  "scripts": {
    ...
    "release": "semantic-release"
  }
  ...
}

4. Set up Continuous Integration 🔗

Create a workflow with Continuous Integration processes to be executed whenever a new change is sent to the main branch, see an example here.

Optional

Run linters against staged git files and don't let errors slip into your code base

1. Install lint-staged

yarn add lint-staged --dev

2. Set up the pre-commit git hook to run lint-staged

yarn husky add .husky/pre-commit 'npx lint-staged'

3. Install some linters, like ESLint and Prettier

4. Configure lint-staged to run linters and other tasks:

• see an example file