a small starter for building a component library with styled-components 💅

This component lib is set up to use styled-components, polished, and styled-components-modifiers 💅. This starter is designed to help you create your own components and easily publish them to npm.

To install dependencies with Yarn, run:

$ yarn

or to install with npm, run:

$ npm install

This component library borrows its structure from BEM and is set up to use Blocks, Elements, and Modifiers. Below is a description of each.

Blocks are the highest level of abstraction in the Blocks, Elements, Modifiers concept. They are responsible for providing the context for Elements, handling UI logic, and rendering the Elements within the Block. They are not connected to application state, nor do they handle any business logic.

Elements are the smallest, indivisible parts of UI. They are responsible for actually rendering the UI. They do not handle application logic or UI logic, but they do handle their own modifiers which modify the element’s style. Elements generally exist within the context of a Block (as their own file in the Block’s directory) allowing the reuse of that set of Elements, but they are not exclusively bound to blocks. An example of a stand-alone Element would be an A, Link, H3, or P. These common elements live in lib/elements.

This library utilizes styled-components-modifiers to build modifiers. Modifiers are small functions that allow us to alter the properties of an Element. They primarily live in the Element's file and are solely responsible for modifying styles. Some modifiers are common to multiple Elements. An example would be fontWeights. These common modifiers live in lib/modifiers

├ lib/
├── blocks/
|   ├── Card
|   |   ├── Body.js     // <- Element
|   |   ├── Footer.js   // <- Element
|   |   ├── Header.js   // <- Element
|   |   └── index.js    // <- Block
|   └── index.js        // <- export for all Blocks
├── elements/
|   ├── A
|   |   ├── __tests__
|   |   |   ├── __snapshots__
|   |   |   |   └── index.js.snap   // <- Snapshot Test
|   |   |   └── index.js            // <- Test
|   |   └── index.js                // <- Element
|   ├── Link
|   |   └── index.js                // <- Element
|   ├── H3
|   |   └── index.js                // <- Element
|   ├── P
|   |   └── index.js                // <- Element
|   └── etc.
|   └── index.js                    // <- export for all Blocks
├── modifiers/
|   ├── fontWeights
|   └── etc.
└── index.js                        // <- main export for the library

Helpful information on development workflow in this library lives here.

NOTE: The linter will run against everything in the lib directory. I've added an initial .eslintrc file for some basic configuration. Feel free to edit or replace it as needed. The intent is to help give you a guide for syntax as you build your application. However, if the linters are too distracting and / or confusing, feel free to ignore them.

This assumes you have eslint and eslint-watch installed. If you don't, run the following:

$ npm i -g eslint eslint-watch

or if you need permissions:

$ sudo npm i -g eslint eslint-watch

To run the linter once:

$ yarn lint:js

To run the watch task:

$ yarn lint:js:watch

I've also added a style linter for Sass / SCSS.

To run the style linter:

$ yarn lint:style

To run both linters:

$ yarn lint

An initial test suite has been setup with two tests (one passing and one intentionally failing). We're using Jest Snapshots for our initial test setup, though Enzym and Expect are also available. The basic test setup lives in ./__tests__. The main configuration for Jest lives at the bottom of package.json. I've also added a few handy scripts, which I've listed below. Jest also gives us a test coverage tool for free, so I've added that too. The setup is at the bottom of package.json. Everything is set to 90% coverage, but your welcome to update that to whatever you'd like.

To run the tests once:

$ npm test

To run the watch script (for only relevant test files)

$ npm run test:watch

To run the watch script (for all test files)

$ npm run test:watchAll

To view the coverage report:

$ npm run test:coverage:report

If you'd like to run the linters and tests at once (this is a nice check before pushing to Github or deploys), you can run:

$ npm run review

NOTE: When you run build, Babel will create a build directory. This is what your users will interact with when they use your library. Nothing in lib gets shipped with your published module.

Run once:

$ npm run build

Run the watch script:

$ npm run build:watch

NOTE: the build script runs in the prepublish script just before you publish to npm.

If you already have an account with npm, you can simply run:

$ npm login
$ npm publish

If you don't have an account with npm:

NOTE: Your email address is public

$ npm set init.author.name "Your Name"
$ npm set init.author.email "[email protected]"
$ npm set init.author.url "http://yourblog.com"
$ npm adduser
$ npm publish

I am thankful for any contributions made by the community. By contributing you agree to abide by the Code of Conduct in the Contributing Guidelines.

MIT