HTML template strings for the Browser with support for Server Side Rendering in Node.
$ npm install nanohtml
var html = require('nanohtml')
var el = html`
<body>
<h1>Hello planet</h1>
</body>
`
document.body.appendChild(el)
Node doesn't have a DOM available. So in order to render HTML we use string concatenation instead. This has the fun benefit of being quite efficient, which in turn means it's great for server rendering!
var html = require('nanohtml')
var el = html`
<body>
<h1>Hello planet</h1>
</body>
`
console.log(el.toString())
Modules like jsdom
implement (parts of)
the DOM in pure JavaScript. If you don't really need the performance of
string concatenation, or use nanohtml components that modify the raw DOM, use
nanohtml/dom
to give nanohtml a custom Document.
var JSDOM = require('jsdom').JSDOM
var nanohtml = require('nanohtml/dom')
var jsdom = new JSDOM()
var html = nanohtml(jsdom.window.document)
var el = html`
<body>
<h1>Hello planet</h1>
</body>
`
el.appendChild(html`<p>A paragraph</p>`)
el.outerHTML === '<body><h1>Hello planet</h1><p>A paragraph</p></body>'
By default all content inside template strings is escaped. This is great for
strings, but not ideal if you want to insert HTML that's been returned from
another function (for example: a markdown renderer). Use nanohtml/raw
for
to interpolate HTML directly.
var raw = require('nanohtml/raw')
var html = require('nanohtml')
var string = '<h1>This a regular string.</h1>'
var el = html`
<body>
${raw(string)}
</body>
`
document.body.appendChild(el)
var html = require('nanohtml')
var el = html`
<body>
<button onclick=${onclick}>
Click Me
</button>
</body>
`
document.body.appendChild(el)
function onclick (e) {
console.log(`${e.target} was clicked`)
}
If you have more than one root element they will be combined with a DocumentFragment.
var html = require('nanohtml')
var el = html`
<li>Chashu</li>
<li>Nori</li>
`
document.querySelector('ul').appendChild(el)
Use a javascript object to conditionally add HTML attributes.
var html = require('nanohtml')
var customAttr = isFuzzy ? { 'data-hand-feel': 'super-fuzzy' } : {}
var el = html`
<div ${ customAttr }></div>
`
Parsing HTML has significant overhead. Being able to parse HTML statically, ahead of time can speed up rendering to be about twice as fast.
$ browserify -t nanohtml index.js > bundle.js
var browserify = require('browserify')
var nanohtml = require('nanohtml')
var path = require('path')
var b = browserify(path.join(__dirname, 'index.js'))
.transform(nanohtml)
b.bundle().pipe(process.stdout)
{
"name": "my-app",
"private": true,
"browserify": {
"transform": [
"nanohtml"
]
},
"dependencies": {
"nanohtml": "^1.0.0"
}
}
At the time of writing there's no Webpack loader yet. We'd love a contribution!
Add nanohtml to your .babelrc
config.
Without options:
{
"plugins": [
"nanohtml"
]
}
With options:
{
"plugins": [
["nanohtml", {
"useImport": true
}]
]
}
useImport
- Set to true to use import
statements for injected modules.
By default, require
is used.appendChildModule
- Import path to a module that contains an appendChild
function. Defaults to "nanohtml/lib/append-child"
.Use the @rollup/plugin-commonjs plugin with @rollup/plugin-node-resolve. Explicitly import the browser or server entrypoint in your application. E.g.:
import html from 'nanohtml/lib/browser';
Shout out to Shama and Shuhei for their contributions to Bel, yo-yoify and pelo. This module is based on their work, and wouldn't have been possible otherwise!