Serialize (encode) HTML character references.

• What is this?
• When should I use this?
• Install
• Use
• API
  • stringifyEntities(value[, options])
• Algorithm
• Types
• Compatibility
• Security
• Related
• Contribute
• License

This is a small and powerful encoder of HTML character references (often called entities). This one has either all the options you need for a minifier/formatter, or a tiny size when using stringifyEntitiesLight.

You can use this for spec-compliant encoding of character references. It’s small and fast enough to do that well. You can also use this when making an HTML formatter or minifier, because there are different ways to produce pretty or tiny output. This package is reliable: '`' characters are encoded to ensure no scripts run in Internet Explorer 6 to 8. Additionally, only named references recognized by HTML 4 are encoded, meaning the infamous ' (which people think is a virus) won’t show up.

This package is ESM only. In Node.js (version 14.14+, 16.0+), install with npm:

npm install stringify-entities

In Deno with esm.sh:

import {stringifyEntities} from 'https://esm.sh/stringify-entities@4'

In browsers with esm.sh:

<script type="module">
  import {stringifyEntities} from 'https://esm.sh/stringify-entities@4?bundle'
</script>

import {stringifyEntities} from 'stringify-entities'

stringifyEntities('alpha © bravo ≠ charlie 𝌆 delta')
// => 'alpha © bravo ≠ charlie 𝌆 delta'

stringifyEntities('alpha © bravo ≠ charlie 𝌆 delta', {useNamedReferences: true})
// => 'alpha © bravo ≠ charlie 𝌆 delta'

This package exports the identifiers stringifyEntities and stringifyEntitiesLight. There is no default export.

Encode special characters in value.

Whether to only escape possibly dangerous characters (boolean, default: false). Those characters are ", &, ', <, >, and `.

Whether to only escape the given subset of characters (Array<string>). Note that only BMP characters are supported here (so no emoji).

If you do not care about the following options, use stringifyEntitiesLight, which always outputs hexadecimal character references.

Prefer named character references (&) where possible (boolean?, default: false).

Prefer the shortest possible reference, if that results in less bytes (boolean?, default: false).

⚠️ Note: useNamedReferences can be omitted when using useShortestReferences.

Whether to omit semicolons when possible (boolean?, default: false).

⚠️ Note: This creates what HTML calls “parse errors” but is otherwise still valid HTML — don’t use this except when building a minifier. Omitting semicolons is possible for certain named and numeric references in some cases.

Create character references which don’t fail in attributes (boolean?, default: false).

⚠️ Note: attribute only applies when operating dangerously with omitOptionalSemicolons: true.

Encoded value (string).

By default, all dangerous, non-ASCII, and non-printable ASCII characters are encoded. A subset of characters can be given to encode just those characters. Alternatively, pass escapeOnly to escape just the dangerous characters (", ', <, >, &, `). By default, hexadecimal character references are used. Pass useNamedReferences to use named character references when possible, or useShortestReferences to use whichever is shortest: decimal, hexadecimal, or named. There is also a stringifyEntitiesLight export, which works just like stringifyEntities but without the formatting options: it’s much smaller but always outputs hexadecimal character references.

This package is fully typed with TypeScript. It exports the additional types Options and LightOptions types.

This package is at least compatible with all maintained versions of Node.js. As of now, that is Node.js 14.14+ and 16.0+. It also works in Deno and modern browsers.

This package is safe.

• parse-entities — parse (decode) HTML character references
• wooorm/character-entities — info on character references
• wooorm/character-entities-html4 — info on HTML 4 character references
• wooorm/character-entities-legacy — info on legacy character references
• wooorm/character-reference-invalid — info on invalid numeric character references

Yes please! See How to Contribute to Open Source.

MIT © Titus Wormer