twitchdown

Dead simple Markdown parser for react-like libraries

Kudos

Kudos goes to Jason Miller and the contributors to Snarkdown as this is based off their hard work.

Features

• Still fast
• Still small (9.5kB, 3kB gzipped)
• Still simply - pass a Markdown string, get an array of components (created with the given createElement function)
• Option to use custom {@ } tags and handlers
• Option to add <p> tags around text
• Option to replace --, --- and ... with an endash, emdash and ellipsis respectively
• Integrate with code highlighters like react-syntax-highlighter
• Minified version included (require('twitchdown/index.min'))

Limitations

As twitchdown uses regular expressions for parsing, it is limited in the formatting that it can handle. Although most good practices should be handled correctly, here are some formatting issues that aren't

• lists with paragraphs

  - item
  - item
  
    with a paragraph break
  - item

• items where line continuations aren't indented

  - item
  - item
  continued item
  - item

• items without spacing between the point and the text

  - item
  -item
  - item

Example

For a more "real life" example with lazy loading, see the Markdown component of MARSS

var twitchdown = require('twitchdown');
// var React = require('react');
// var SyntaxHighlighter = require('react-syntax-highlighter');
// var docco = require('react-syntax-highligher/styles/hljs').docco;

// Valid if options.parseArguments is falsey
var customTag = (attributes) => {
  return `First is '${attributes[0]}', the rest is '${attributes.splice(1).join(',')}`
};

// Valid if options.parseArguments is truthy
var superTag = (attributes) => {
  return `You are super '${attributes.name}' because ${attributes.arguments.join(',')}`
};

// Use a code highlighter to highlight code in ``` tags. The function should
// return an element as created by the used createElement function
// var highlight = (code, language) => {
//   return React.createElement(SyntaxHighlighter, {
//     showLineNumbers: true,
//     style: defaultStyle,
//     language
//   }, [ code ]);
// }

const markdown = `#Test

<script> evilFunction() </script>

This is some <em>test</em> markdown
- good [me](me)
- one --- {@custom first second "third"}
- list item
  over multiple lines
  - sub list
    over multiple lines
  - woot

{@super name=bob twitch}

\`\`\`javascript
function hello() {
  console.debug('hello');
}
\`\`\`
`;

var elements = twitchdown(markdown, {
  // Function to use for creating elements
  // createElement: React.createElement,
  // Highlighter function for code blocks
  //highlight: highlight,
  // These HTML tags and their contents will be completely removed (defaults to <script> tags)
  removeTags: [ 'script' ],
  // These HTML tags will be removed, but their contents will be kept
  stripTags: [ 'em' ],
  // Custom tag handlers
  customTags: {
    custom: {
      handler: customTag,
      // If set, this will override the global `tagsInParagraph` option
      inParagraph: true,
      // If set, this will override the global `parseArguments` option
      parseArguments: false
    },
    super: superTag
  },
  // Whether or not to parse attributes passed to custom tags. Note that when
  // enabled ALL custom tag arguments will be parsed into an object, rather
  // than left as an array
  parseArguments: true,
  // If true and `paragraphs` is true , custom tags will be placed p tags
  tagsInParagraph: false,
  // Opens external links in a new window
  openExternalInNewWindow: true,
  // Reference links
  referenceLinks: {
    me: 'https://example.com/'
  },
  // Add id tags to any headings
  headingIds: true,
  // Add an offset number to all h tags (e.g. h1 becomes h2)
  headingOffset: 1,
  // Wrap text in p tags
  paragraphs: true,
  /// Replace --, --- and ... with an endash, emdash and ellipsis respectively
  replacePunctuation: true
});

console.log(elements);

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

Unreleased

1.6.1 - 2022-10-19

Fixed

• Typo in CHANGELOG

1.6.0 - 2022-10-19

Added

• replacePunctuation option for replacement of --, ---, ... with an endash (–), emdash (—) and ellipsis (…) respectively

1.5.0 - 2022-10-11

Added

• headingOffset option allowing the offsetting of h tags. For example if headingOffset is set to 1 in the options, an h1 heading will be formatted as as h2 heading

Fixed

• Handling of reference links

1.4.2 - 2020-08-02

Added

• Added voidTags (an array of HTML element tags that don't have a closing tag)

Fixed

• Fix example around highlighter

1.4.1 - 2020-06-14

Fixed

• Adding of spaces after non-block elements such as links if they are directly after block elements

1.4.0 - 2020-02-18

Added

• openExternalInNewWindow option to open external links in a new window

Fixed

• Handling of sub-lists and multi-line items in lists

1.3.3 - 2020-02-15

Changed

• Fixed handling of p tags when using paragraphs option

1.3.2 - 2019-05-14

Changed

• Fixed handling of void HTML elements
• Updated dependencies

1.3.1 - 2019-02-22

Added

• key values to each element as required by React
• Render tests for both React and Preact
• Reference links to CHANGELOG

Changed

• Fix paragraph and inline formatting issue that caused inline formatted text to be left outside of a paragraph
• Fixed parsing of attributes in html tags

Removed

• href attribute if href not given and not reference link available

1.3.0 - 2018-08-08

Added

• Add parseArguments option to parse custom tag attributes into an object
• Add tagsInParagraph option to set if custom tags should be placed in p tags
• Allow customTags to be given as Objects with their own values for the parseArguments and tagsInParagraph (as inParagraph)

Changed

• Set main file as non-minified version

1.2.0 - 2018-07-24

Added

• Added title attribute to images

Changed

• Fix so attributes are passed to custom tag function as an array for custom tags inside of urls for links and images as they are for tags outside of urls

Removed

• Empty alt attributes from images with no title

1.1.0 - 2018-07-16

Added

• Allow custom tags in image and link urls
• Added headingIds option to add id attributes to heading based on the text of the heading (replacing spaces with - and removing all symbols etc
• Add tests to linting and ensure all development packages are up-to-date

Changed

• Fix placement of components in p tags
• Set main file as minified version

1.0.1 - 2018-07-11

Added

• Linting of code

Changed

• Change code to ES3

1.0.0 - 2018-07-10

Initial Release