rehype plugin to add metadata to the <head>.
Contents
What is this?
This package is a unified (rehype) plugin to manage the metadata (Open Graph, Twitter Cards, SEO, etc.) that can be found in <head>. It focusses on reasonable and useful metadata that is supported by several and popular vendors instead of every possible field supported somewhere.
unified is a project that transforms content with abstract syntax trees (ASTs). rehype adds support for HTML to unified. hast is the HTML AST that rehype uses. This is a rehype plugin that adds metadata to the head in the tree.
When should I use this?
This plugin is particularly useful as a metadata manager when you’re working with articles that are supposed to be shared on the web, whether it’s on Twitter or Slack. You can define key/value pairs, either with frontmatter, with other plugins, or as options, and this plugin will generate the HTML used by different services.
This plugin works on complete documents (not fragments). A different plugin, rehype-document, wraps fragments in documents.
Install
This package is ESM only. In Node.js (version 16+), install with npm:
npm install rehype-meta
In Deno with esm.sh:
import rehypeMeta from 'https://esm.sh/rehype-meta@4'
In browsers with esm.sh:
<script type="module">
import rehypeMeta from 'https://esm.sh/rehype-meta@4?bundle'
</script>
Use
Say our module example.js looks as follows:
import {rehype} from 'rehype'
import rehypeMeta from 'rehype-meta'
const file = await rehype()
.data('settings', {fragment: true})
.use(rehypeMeta, {
author: 'Jane Doe',
authorFacebook: 'janedoe',
authorTwitter: '@jane',
copyright: true,
description:
'The city has changed drastically over the past 40 years, yet the M.T.A. map designed in 1979 has largely endured.',
image: {
alt: 'M.T.A. map designed in 1979',
height: '550',
url: 'https://static01.nyt.com/images/2019/12/02/autossell/mta-promo-image/mta-crop-facebookJumbo.jpg',
width: '1050'
},
modified: '2019-12-03T19:13:00.000Z',
name: 'The New York Times',
og: true,
origin: 'https://www.nytimes.com',
pathname: '/interactive/2019/12/02/nyregion/nyc-subway-map.html',
published: '2019-12-02T10:00:00.000Z',
readingTime: 11.1,
section: 'New York',
separator: ' | ',
siteAuthor: 'The New York Times',
siteTags: [
'US Politics',
'Impeachment',
'NATO',
'London',
'Food',
'Poverty',
'Climate Change',
'Global Warming'
],
siteTwitter: '@nytimes',
tags: [
'Subway',
'Map',
'Public Transit',
'Design',
'MTA',
'Massimo Vignelli',
'NYC'
],
title:
'The New York City Subway Map as You’ve Never Seen It Before',
twitter: true,
type: 'article'
})
.process('')
console.log(String(file))
…now running node example.js yields:
<head>
<title>The New York City Subway Map as You’ve Never Seen It Before | The New York Times</title>
<link rel="canonical" href="https://www.nytimes.com/interactive/2019/12/02/nyregion/nyc-subway-map.html">
<meta name="description" content="The city has changed drastically over the past 40 years, yet the M.T.A. map designed in 1979 has largely endured.">
<meta name="keywords" content="Subway, Map, Public Transit, Design, MTA, Massimo Vignelli, NYC, US Politics, Impeachment, NATO, London, Food, Poverty, Climate Change, Global Warming">
<meta name="author" content="Jane Doe">
<meta name="copyright" content="© 2019 Jane Doe">
<meta property="og:type" content="article">
<meta property="og:site_name" content="The New York Times">
<meta property="og:url" content="https://www.nytimes.com/interactive/2019/12/02/nyregion/nyc-subway-map.html">
<meta property="og:title" content="The New York City Subway Map as You’ve Never Seen It Before">
<meta property="og:description" content="The city has changed drastically over the past 40 years, yet the M.T.A. map designed in 1979 has largely endured.">
<meta property="og:image" content="https://static01.nyt.com/images/2019/12/02/autossell/mta-promo-image/mta-crop-facebookJumbo.jpg">
<meta property="og:image:alt" content="M.T.A. map designed in 1979">
<meta property="og:image:width" content="1050">
<meta property="og:image:height" content="550">
<meta property="article:published_time" content="2019-12-02T10:00:00.000Z">
<meta property="article:modified_time" content="2019-12-03T19:13:00.000Z">
<meta property="article:author" content="https://www.facebook.com/janedoe">
<meta property="article:section" content="New York">
<meta property="article:tag" content="Subway">
<meta property="article:tag" content="Map">
<meta property="article:tag" content="Public Transit">
<meta property="article:tag" content="Design">
<meta property="article:tag" content="MTA">
<meta property="article:tag" content="Massimo Vignelli">
<meta name="twitter:card" content="summary_large_image">
<meta name="twitter:image" content="https://static01.nyt.com/images/2019/12/02/autossell/mta-promo-image/mta-crop-facebookJumbo.jpg">
<meta name="twitter:image:alt" content="M.T.A. map designed in 1979">
<meta name="twitter:site" content="@nytimes">
<meta name="twitter:creator" content="@jane">
<meta name="twitter:label1" content="Posted in">
<meta name="twitter:data1" content="New York">
<meta name="twitter:label2" content="Reading time">
<meta name="twitter:data2" content="12 minutes">
</head>
API
This package exports no identifiers. The default export is rehypeMeta.
Add metadata to the <head>.
Parameters
options (Options, optional) — configuration
Returns
Transform (Transformer).
Notes
Algorithm
- adds a
<head> if one doesn’t already exist - overwrites existing metadata in
<head> (for example, when a <title> already exists, it’s updated)
Config
There are three ways to configure the metadata of a document.
- pass an object as
options when using rehypeMeta - define it in YAML front matter (by integrating with
vfile-matter) - define an object at
file.data.meta
Configuration is created by extending the defaults, with these objects, in the above order (so file.data.meta is preferred over options). Only options is enough if every file has the same metadata. If your workflow enables front matter, that’s a good way to keep data in files. Alternatively, do it yourself by adding data at file.data.meta, which can also be done by plugins:
Image
Image metadata (TypeScript type).
Fields
alt (string, optional, example: 'M.T.A. map designed in 1979') — alt text of imageheight (number | string, optional, example: '550') — height of imageurl (string, required, example: 'https://static01.nyt.com/images/…/mta-crop-jumbo.jpg') — URL of imagewidth (number | string, optional, example: '1050') — width of image
Options
Configuration (TypeScript type).
Fields
author
Name of the author of the document (string, optional, example: 'Jane Doe').
Affects: meta[name=author], meta[name=copyright].
authorFacebook
Facebook username of the author of the document (string, optional, example: 'example').
Affects: meta[property=article:author].
Twitter username of the author of the document (string, optional, example: '@janedoe').
Affects: meta[name=twitter:creator].
color
Hexadecimal theme color of document or site (string, optional, example: '#bada55').
Affects: meta[name=theme-color].
copyright
Whether to add copyright metadata (boolean, default: false).
Affects: meta[name=copyright].
description
Description of the document (string, optional, example: 'The city has changed drastically over the past 40 years, yet the M.T.A. map designed in 1979 has largely endured.').
Affects: meta[name=description], meta[property=og:description].
image
One or more images associated with the document (Array<Image | string> | Image | string, optional); if strings are passed, they are seen as Image objects with a url field set to that value.
Affects: meta[property=og:image], meta[name=twitter:card], meta[name=twitter:image].
modified
Date the document was last modified (Date or string, optional, example: '2019-12-03T19:13:00.000Z').
👉 Note: parsing a string is inconsistent, prefer dates.
Affects: meta[property=article:modified_time].
name
Name of the whole site (string, optional, example: 'The New York Times').
Affects: title, meta[property=og:site_name], meta[property=og:title].
og
Whether to add Open Graph metadata (boolean, default: false).
Affects: meta[property=og:site_name], meta[property=og:url], meta[property=og:title], meta[property=og:description], meta[property=og:image], meta[property=article:published_time], meta[property=article:modified_time], meta[property=article:author], meta[property=article:section], meta[property=article:tag], meta[name=twitter:card].
ogNameInTitle
Whether to add the site name name to the og:title (boolean, default: false).
Affects: meta[property=og:title].
origin
Origin the file will be hosted on (string, optional, example: https://www.nytimes.com).
Affects: link[rel=canonical], meta[property=og:url].
pathname
Absolute pathname of where the file will be hosted (string, default: /, example: /interactive/2019/12/02/nyregion/nyc-subway-map.html).
Affects: link[rel=canonical], meta[property=og:url].
published
Date the document (or site) was first published (Date or string, optional, example: '2019-12-02T10:00:00.000Z').
👉 Note: parsing a string is inconsistent, prefer dates.
Affects: meta[name=copyright], meta[property=article:published_time].
readingTime
Estimated reading time in minutes for the document ([number, number] or number, optional, example: 1.219403). If two numbers are given, they represent a range of two estimates.
Affects: meta[name=twitter:label1], meta[name=twitter:data1], meta[name=twitter:label2], meta[name=twitter:data2].
section
Section associated with the document (string, optional, example: 'New York').
Affects: meta[property=article:section], meta[name=twitter:label1], meta[name=twitter:data1].
separator
Value to use to join the title and name together (string, default: ' - ').
Affects: title, meta[property=og:title].
siteAuthor
Name of the author of the whole site (string, optional, example: 'The New York Times').
Affects: meta[name=author], meta[name=copyright].
Tags associated with the whole site (Array<string>, optional, example: ['US Politics', 'Impeachment', 'NATO', 'London', 'Food', 'Poverty', 'Climate Change', 'Global Warming']).
Affects: meta[name=keywords].
Twitter username of the whole site (string, optional, example: '@nytimes').
Affects: meta[name=twitter:site].
Tags associated with the document (Array<string>, optional, example: ['Subway', 'Map', 'Public Transit', 'Design', 'MTA', 'Massimo Vignelli', 'NYC']).
Affects: meta[name=keywords], meta[property=article:tag].
title
Title of the document (string, optional, example: 'The New York City Subway Map as You’ve Never Seen It Before').
Affects: title, meta[property=og:title].
Whether to add Twitter metadata (boolean, default: false).
Affects: meta[name=twitter:card], meta[name=twitter:image], meta[name=twitter:site], meta[name=twitter:creator], meta[name=twitter:label1], meta[name=twitter:data1], meta[name=twitter:label2], meta[name=twitter:data2].
type
What the document refers to ('article' | 'website', default: 'website').
Affects: meta[property=og:type], meta[property=article:published_time], meta[property=article:modified_time], meta[property=article:author], meta[property=article:section], meta[property=article:tag].
The following metadata can be added by rehype-meta.
link[rel=canonical]
Affected by: origin, pathname.
If origin is 'https://example.com' and path is '/path/':
<link rel="canonical" href="https://example.com/path/">
If origin is 'https://example.com' and path is not set:
<link rel="canonical" href="https://example.com/">
Affected by: author, siteAuthor.
If author is 'Jane':
<meta name="author" content="Jane">
If siteAuthor is 'John':
<meta name="author" content="John">
If author is 'Jane' and siteAuthor is 'John':
<meta name="author" content="Jane">
Affected by: copyright, author, siteAuthor, published.
The below examples depend on the current date, so for example purposes, say it was the year 2030.
If copyright is not true, meta[name=copyright] is not added.
If copyright is true and author is 'Jane':
<meta name="copyright" content="© 2030 Jane">
If copyright is true and siteAuthor is 'John':
<meta name="copyright" content="© 2030 John">
If copyright is true, author is 'Jane', and siteAuthor is 'John':
<meta name="copyright" content="© 2030 Jane">
If copyright is true, author is 'Jane', and published is '2015':
<meta name="copyright" content="© 2015 Jane">
Affected by: description.
If description is 'Lorem ipsum':
<meta name="description" content="Lorem ipsum">
Affected by: tags, siteTags.
If tags is ['a', 'b']:
<meta name="keywords" content="a, b">
If siteTags is ['b', 'c']:
<meta name="keywords" content="b, c">
If tags is ['a', 'b'] and siteTags is ['b', 'c']:
<meta name="keywords" content="a, b, c">
Affected by: color.
If color is '#bada55':
<meta name="theme-color" content="#bada55">
Affected by: og, twitter, image.
If twitter is not true, meta[name=twitter:card] is not added. If twitter is true, og is true, and there is no valid image, no meta[name=twitter:card] is added either, because Twitter assumes a summary in this case.
If twitter is true and there is a valid image:
<meta name="twitter:card" content="summary_large_image">
If twitter is true and there is no valid image:
<meta name="twitter:card" content="summary">
Affected by: twitter, authorTwitter.
If twitter is not true, meta[name=twitter:creator] is not added.
If twitter is true and authorTwitter is '@example':
<meta name="twitter:creator" content="@example">
Affected by: twitter, section, readingTime.
👉 Note: this data is used by Slack, not by Twitter.
If twitter is not true, meta[name=twitter:label1] and meta[name=twitter:data1] are not added.
If twitter is true and section is 'Food':
<meta name="twitter:label1" content="Posted in">
<meta name="twitter:data1" content="Food">
If twitter is true, section is not defined, and readingTime is 3.083:
<meta name="twitter:label1" content="Reading time">
<meta name="twitter:data1" content="4 minutes">
Affected by: twitter, section, readingTime.
👉 Note: this data is used by Slack, not by Twitter.
If twitter is not true, section is not defined, or readingTime is not defined, meta[name=twitter:label2] and meta[name=twitter:data2] are not added.
If twitter is true, section is defined, and readingTime is 0.8:
<meta name="twitter:label2" content="Reading time">
<meta name="twitter:data2" content="1 minute">
If twitter is true, section is defined, and readingTime is [8, 12]:
<meta name="twitter:label2" content="Reading time">
<meta name="twitter:data2" content="8-12 minutes">
Affected by: twitter, image.
If twitter is not true, meta[name=twitter:image] and meta[name=twitter:image:alt] are not added.
👉 Note: only one image is added.
If twitter is true and image is 'https://example.com/image.png':
<meta name="twitter:image" content="https://example.com/image.png">
If twitter is true and image is ['https://example.com/a.png', 'https://example.com/b.png']:
<meta name="twitter:image" content="https://example.com/a.png">
If twitter is true and image is {url: 'https://example.com/a.png', alt: 'A', width: '670', height: '1012'}:
<meta name="twitter:image" content="https://example.com/a.png">
<meta name="twitter:image:alt" content="A">
Affected by: twitter, siteTwitter.
If twitter is not true, meta[name=twitter:site] is not added.
If twitter is true and siteTwitter is '@example':
<meta name="twitter:site" content="@example">
meta[property=article:author]
Affected by: og, type, authorFacebook.
If og is not true or type is not 'article', meta[property=article:author] is not added.
If og is true, type is 'article', and authorFacebook is 'jane':
<meta property="article:author" content="https://www.facebook.com/jane">
meta[property=article:modified_time]
Affected by: og, type, modified.
If og is not true or type is not 'article', meta[property=article:modified_time] is not added.
If og is true, type is 'article', and modified is '2017-04-26T22:37:10-05:00':
<meta property="article:modified_time" content="2017-04-27T03:37:10.000Z">
meta[property=article:published_time]
Affected by: og, type, published.
If og is not true or type is not 'article', meta[property=article:published_time] is not added.
If og is true, type is 'article', and published is '2014-06-30T15:01:35-05:00':
<meta property="article:published_time" content="2014-06-30T20:01:35.000Z">
meta[property=article:section]
Affected by: og, type, section.
If og is not true or type is not 'article', meta[property=article:section] is not added.
If og is true, type is 'article', and section is 'Politics':
<meta property="article:section" content="Politics">
meta[property=article:tag]
Affected by: og, type, tag.
If og is not true or type is not 'article', meta[property=article:tag] are not added.
👉 Note: up to 6 tags are added.
If og is true, type is 'article', and tags is ['US Politics', 'Impeachment', 'NATO', 'London', 'Food', 'Poverty', 'Climate Change']:
<meta property="article:tag" content="US Politics">
<meta property="article:tag" content="Impeachment">
<meta property="article:tag" content="NATO">
<meta property="article:tag" content="London">
<meta property="article:tag" content="Food">
<meta property="article:tag" content="Poverty">
Affected by: og, description.
If og is not true, meta[property=og:description] is not added.
If og is true and description is 'Lorem ipsum':
<meta property="og:description" content="Lorem ipsum">
Affected by: og, image.
If og is not true, meta[property=og:image], meta[property=og:image:alt], meta[property=og:image:width], and meta[property=og:image:height] are not added.
👉 Note: up to 6 images are added.
If og is true and image is 'https://example.com/image.png':
<meta property="og:image" content="https://example.com/image.png">
If og is true and image is ['https://example.com/a.png', 'https://example.com/b.png']:
<meta property="og:image" content="https://example.com/a.png">
<meta property="og:image" content="https://example.com/b.png">
If og is true and image is {url: 'https://example.com/a.png', alt: 'A', width: '670', height: '1012'}:
<meta property="og:image" content="https://example.com/a.png">
<meta property="og:image:alt" content="A">
<meta property="og:image:width" content="670">
<meta property="og:image:height" content="1012">
Affected by: og, name.
If og is not true, meta[property=og:site_name] is not added.
If og is true and name is 'Example':
<meta property="og:site_name" content="Example">
Affected by: og, ogNameInTitle, title, name, separator.
If og is not true, meta[property=og:title] is not added.
If og is true and title is 'About':
<meta property="og:title" content="About">
If og is true, ogNameInTitle is true, title is 'About', and name is 'Site':
<meta property="og:title" content="About - Site">
If og is true, ogNameInTitle is true, title is 'About', name is 'Site', and separator is ' | ':
<meta property="og:title" content="About | Site">
Affected by: og, type.
If og is not true, meta[property=og:type] is not added.
If og is true and type is 'website':
<meta property="og:type" content="website">
If og is true and type is 'article':
<meta property="og:type" content="article">
Affected by: og, origin, pathname.
If og is not true, meta[property=og:url] is not added.
If og is true, origin is 'https://example.com', and path is '/path/':
<meta property="og:url" content="https://example.com/path/">
If origin is 'https://example.com' and path is not set:
<meta property="og:url" content="https://example.com/">
title
Affected by: title, name, separator.
If title is 'About':
<title>About</title>
If name is 'Example':
<title>Example</title>
If title is 'About' and name is 'Example':
<title>About - Example</title>
If title is 'About', name is 'Example', and separator to ' | ':
<title>About | Example</title>
Examples
Example: frontmatter in markdown
This example shows how it’s possible to combine the different data sources to pass site wide info as options and define more specific data from within markdown files with frontmatter.
Say we have the following file example.md:
---
title: Neptune
author: U. Le Verrier
authorTwitter: '@leverrier123'
description: Neptune is blue.
tags:
- neptune
- blue
---
# Neptune
To do: write some stuff about why neptune is cool.
…and a module example.js:
import remarkFrontmatter from 'remark-frontmatter'
import remarkParse from 'remark-parse'
import remarkRehype from 'remark-rehype'
import rehypeDocument from 'rehype-document'
import rehypeMeta from 'rehype-meta'
import rehypeStringify from 'rehype-stringify'
import {read} from 'to-vfile'
import {unified} from 'unified'
import {matter} from 'vfile-matter'
const file = await read('example.md')
// Define where the generated file will be available.
file.data.meta = {
origin: 'https://planets.com',
pathname: '/neptune/'
}
await unified()
.use(remarkParse)
.use(remarkFrontmatter)
.use(function () {
return function (_, file) {
matter(file)
}
})
.use(remarkRehype)
// `rehype-document` manages non-metadata things in `<head>`.
.use(rehypeDocument, {
css: 'https://planets.com/index.css',
js: 'https://planets.com/index.js'
})
// Site wide metadata:
.use(rehypeMeta, {
copyright: true,
name: 'Planets',
og: true,
siteAuthor: 'J. Galle',
siteTags: ['planet', 'solar', 'galaxy'],
siteTwitter: '@the_planets',
twitter: true,
type: 'article'
})
.use(rehypeStringify)
.process(file)
console.log(String(file))
…now running node example.js yields:
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Neptune - Planets</title>
<meta content="width=device-width, initial-scale=1" name="viewport">
<link href="https://planets.com/index.css" rel="stylesheet">
<link rel="canonical" href="https://planets.com/neptune/">
<meta name="description" content="Neptune is blue.">
<meta name="keywords" content="neptune, blue, planet, solar, galaxy">
<meta name="author" content="U. Le Verrier">
<meta name="copyright" content="© 2023 U. Le Verrier">
<meta property="og:type" content="article">
<meta property="og:site_name" content="Planets">
<meta property="og:url" content="https://planets.com/neptune/">
<meta property="og:title" content="Neptune">
<meta property="og:description" content="Neptune is blue.">
<meta property="article:tag" content="neptune">
<meta property="article:tag" content="blue">
<meta name="twitter:site" content="@the_planets">
<meta name="twitter:creator" content="@leverrier123">
</head>
<body>
<h1>Neptune</h1>
<p>To do: write some stuff about why neptune is cool.</p>
<script src="https://planets.com/index.js"></script>
</body>
</html>
Some metadata can be automatically gathered, either by extracting it from the document, or by accessing the file system or Git. This is done by other plugins (see config) which “infer” that metadata and store their results on file.data.meta, which this plugin then looks at.
Taking this readme as an example and running the following code within this repo:
import remarkParse from 'remark-parse'
import remarkRehype from 'remark-rehype'
import rehypeDocument from 'rehype-document'
import rehypeInferTitleMeta from 'rehype-infer-title-meta'
import rehypeInferDescriptionMeta from 'rehype-infer-description-meta'
import rehypeInferReadingTimeMeta from 'rehype-infer-reading-time-meta'
import rehypeMeta from 'rehype-meta'
import rehypeStringify from 'rehype-stringify'
import {read} from 'to-vfile'
import {unified} from 'unified'
import unifiedInferGitMeta from 'unified-infer-git-meta'
const file = await unified()
.use(remarkParse)
.use(unifiedInferGitMeta) // Find published, modified, and authors in Git.
.use(remarkRehype)
.use(rehypeDocument)
.use(rehypeInferTitleMeta) // Find the main title.
.use(rehypeInferDescriptionMeta, {truncateSize: 64}) // Find the description.
.use(rehypeInferReadingTimeMeta) // Estimate reading time.
.use(rehypeMeta, {og: true, twitter: true, copyright: true})
.use(rehypeStringify)
.process(await read('readme.md'))
console.log(String(file))
Yields:
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>rehype-meta</title>
<meta content="width=device-width, initial-scale=1" name="viewport">
<meta name="description" content="rehype plugin to add metadata to the <head…">
<meta name="author" content="Titus Wormer">
<meta name="copyright" content="© 2019 Titus Wormer">
<meta property="og:type" content="website">
<meta property="og:title" content="rehype-meta">
<meta property="og:description" content="rehype plugin to add metadata to the <head…">
<meta name="twitter:label1" content="Reading time">
<meta name="twitter:data1" content="16-25 minutes">
</head>
<body>
<h1>rehype-meta</h1>
…
Types
This package is fully typed with TypeScript. It exports the additional types Image and Options.
It also registers expected fields on file.data.meta and file.data.matter with vfile. If you’re working with the file, make sure to import this plugin somewhere in your types, as that registers the new fields on the file.
/**
* @import {} from 'rehype-meta'
*/
import {VFile} from 'vfile'
const file = new VFile()
console.log(file.data.meta.title) //=> TS now knows that this is a `string?`.
Compatibility
Projects maintained by the unified collective are compatible with maintained versions of Node.js.
When we cut a new major release, we drop support for unmaintained versions of Node. This means we try to keep the current release line, rehype-meta@^4, compatible with Node.js 16.
This plugin works with rehype-parse version 3+, rehype-stringify version 3+, rehype version 4+, and unified version 6+.
Security
Use of rehype-meta is relatively safe, however, it is possible for an attacker to define metadata from within a because of the matter integration.
Contribute
See contributing.md in rehypejs/.github for ways to get started. See support.md for ways to get help.
This project has a code of conduct. By interacting with this repository, organization, or community you agree to abide by its terms.
License
MIT © Titus Wormer