mdast is speedy Markdown parser for multipurpose analysis (a syntax tree) in JavaScript. NodeJS, and the browser. Lots of tests. 100% coverage.
It's close, but not yet there, to being able to compile the AST back to markdown again: it still has some slight issues with loose lists.
npm:
$ npm install mdast
Component.js:
$ component install wooorm/mdast
Bower:
$ bower install mdast
See Nodes for information about he returned nodes.
var mdast = require('mdast');
var ast = mdast.parse('Some *emphasis*, **strongness**, and `code`.');
Yields:
{
"type" : "root",
"children" : [
{
"type": "paragraph",
"children": [
{
"type": "text",
"value": "Some "
},
{
"type": "emphasis",
"children": [{
"type": "text",
"value": "emphasis"
}]
},
{
"type": "text",
"value": ", "
},
{
"type": "strong",
"children": [{
"type": "text",
"value": "strongness"
}]
},
{
"type": "text",
"value": ", and "
},
{
"type": "inlineCode",
"value": "code"
},
{
"type": "text",
"value": "."
}
]
}
]
}
And passing that document into mdast.stringify
mdast.stringify(ast);
Yields:
Some *emphasis*, **strongness**, and `code`\.
Yeah, the escaped period is nasty, but it works :)
defaults to true.
mdast.parse('hello ~~hi~~ world', {
'gfm' : true
});
Yields:
{
"type": "root",
"children": [
{
"type": "paragraph",
"children": [
{
"type": "text",
"value": "hello "
},
{
"type": "delete",
"children": [{
"type": "text",
"value": "hi"
}]
},
{
"type": "text",
"value": " world"
}
]
}
]
}
defaults to true.
var source =
'Header 1 | Header 2\n' +
':------- | -------:\n' +
'Cell 1 | Cell 2\n' +
'Cell 3 | Cell 4\n';
mdast.parse(source, {
'tables' : true
});
Yields:
{
"type": "root",
"children": [
{
"type": "table",
"align": [
"left",
"right"
],
"children": [
{
"type": "tableHeader",
"children": [
{
"type": "tableCell",
"children": [{
"type": "text",
"value": "Header 1"
}]
},
{
"type": "tableCell",
"children": [{
"type": "text",
"value": "Header 2"
}]
}
]
},
{
"type": "tableRow",
"children": [
{
"type": "tableCell",
"children": [{
"type": "text",
"value": "Cell 1"
}]
},
{
"type": "tableCell",
"children": [{
"type": "text",
"value": "Cell 2"
}]
}
]
},
{
"type": "tableRow",
"children": [
{
"type": "tableCell",
"children": [{
"type": "text",
"value": "Cell 3"
}]
},
{
"type": "tableCell",
"children": [{
"type": "text",
"value": "Cell 4"
}]
}
]
}
]
}
]
}
“Pedantic”, used by Gruber's Markdown, matches emphasis-marks inside words. It's mostly not what you want.
Defaults to false.
mdast.parse('some_file_name', {
'pedantic' : true
});
Yields:
{
"type": "root",
"children": [
{
"type": "paragraph",
"children": [
{
"type": "text",
"value": "some"
},
{
"type": "emphasis",
"children": [{
"type": "text",
"value": "file"
}]
},
{
"type": "text",
"value": "name"
}
]
}
]
}
“Breaks” prettifies line breaks inside a paragraph.
Defaults to false.
mdast.parse('A\nparagraph', {
'gfm' : true,
'breaks' : true
});
Yields:
{
"type": "root",
"children": [
{
"type": "paragraph",
"children": [
{
"type": "text",
"value": "A"
},
{
"type": "break"
},
{
"type": "text",
"value": "paragraph"
}
]
}
]
}
Whereas with breaks false (and GFM true), mdast.parse would yield:
{
"type": "root",
"children": [
{
"type": "paragraph",
"children": [{
"type": "text",
"value": "A\nparagraph"
}]
}
]
}
“Footnotes” enables use of inline- and reference-style footnotes. Its also possible to reference other footnotes inside footnotes.
Defaults to false.
var source =
'Something something[^or something?]\n' +
'And something else[^1]\n\n' +
'[^1]: This content here can contains paragraphs,\n' +
' - and lists\n';
mdast.parse(source, {
'footnotes' : true
});
Yields:
{
"type": "root",
"children": [
{
"type": "paragraph",
"children": [
{
"type": "text",
"value": "Something something"
},
{
"type": "footnote",
"id": "footnote-1"
},
{
"type": "text",
"value": "\nAnd something else"
},
{
"type": "footnote",
"id": "1"
}
]
}
],
"footnotes": {
"1": [
{
"type": "paragraph",
"children": [{
"type": "text",
"value": "This content here can contains paragraphs,"
}]
},
{
"type": "list",
"ordered": false,
"children": [
{
"type": "listItem",
"children": [{
"type": "text",
"value": "and lists"
}]
}
]
}
],
"footnote-1": [
{
"type": "paragraph",
"children": [{
"type": "text",
"value": "or something?"
}]
}
]
}
}
mdast.parse returns node objects---just plain vanilla JS objects. Every node implements the following "Node" interface.
interface Node {
type: string;
}
Most nodes implement the "Parent" interface (block/inline nodes which accept other nodes as children)...
interface Parent <: Node {
children: [Node];
}
...all others, with the exception of Table, HorizontalRule, Break, and Footnote, implement "Text" (nodes which accept a value.)
interface Parent <: Text {
value: string;
}
A root element houses all other nodes. In addition, it hold a footnote property, housing the content of Footnotes by their keys (if "footnotes" is true in options, that is).
interface Root <: Parent {
type: "root";
footnotes: { [nodes] } | null;
}
interface Paragraph <: Parent {
type: "paragraph";
}
interface Blockquote <: Parent {
type: "blockquote";
}
A heading, just like with HMTL, greater-than-or-equal-to 0, lower-than-or-equal-to 6.
interface Heading <: Parent {
type: "heading";
depth: 1 <= uint32 <= 6;
}
Occurs at block level (see inlineCode for code spans). Code sports a language tag (when using Github Flavoured Markdown fences, null otherwise).
interface Code <: Text {
type: "code";
lang: string | null;
}
Occurs at inline level (see code for code blocks). Inline code does not sport a lang tag.
interface InlineCode <: Text {
type: "inlineCode";
}
A string of raw HTML.
interface HTML <: Text {
type: "html";
}
A list.
interface List <: Parent {
type: "list";
ordered: true | false;
}
An item in a list (always occurs inside a list).
interface ListItem <: Parent {
type: "listItem";
}
Tabular data, with alignment. Its children are either tableHeader (the first child), or tableRow (all other children).
table.align is a list of align types.interface Table <: Parent {
type: "table";
align: [alignType];
}
enum alignType {
"left" | "right" | "center";
}
A table header. Its children are always tableCell.
interface TableHeader <: Parent {
type: "tableHeader";
}
A table row. Its children are always tableCell.
interface TableRow <: Parent {
type: "tableRow";
}
Tabular data cell.
interface TableCell <: Parent {
type: "tableCell";
}
Just a horizontal rule.
interface HorizontalRule <: Node {
type: "horizontalRule";
}
If you want, you can use "breaks": true, and instead of newlines, break nodes will show up.
interface Break <: Node {
type: "break";
}
Slightly important text.
interface Emphasis <: Parent {
type: "emphasis";
}
Super important text.
interface Strong <: Parent {
type: "strong";
}
Content ready for removal.
interface Delete <: Parent {
type: "delete";
}
The humble hyperlink.
interface Link <: Parent {
type: "link";
title: string | null;
href: string;
}
The figurative figure.
interface Image <: Parent {
type: "image";
title: string | null;
alt: string | null;
src: string;
}
A footnote. These occur as inline nodes.
interface Footnote <: Node {
type: "footnote";
id : string
}
Everything thats just text, is wrapped in a text node (d'oh):
interface Text <: Text {
type: "text";
}
Run the benchmark yourself:
$ npm run benchmark
On a MacBook Air, it parser about 5 megabytes of markdown per second, depending on how much markup v.s. plain text the document contains, and which language the document is in, that's the entire works of Shakespeare, in a second.
benchmarks * 56 fixtures (total: 47Kb markdown)
110 op/s » mdast.parse -- this module
This project was initially a fork of marked.
Copyright (c) 2011-2014, Christopher Jeffrey. (MIT License)
MIT © 2014 Titus Wormer
FAQs
Markdown Abstract Syntax Tree format
The npm package mdast receives a total of 47,195 weekly downloads. As such, mdast popularity was classified as popular.
We found that mdast demonstrated a not healthy version release cadence and project activity because the last version was released a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Ransomware payment rates hit an all-time low in 2024 as law enforcement crackdowns, stronger defenses, and shifting policies make attacks riskier and less profitable.
Security News
require(esm) backported to Node.js 20, easing the transition to ESM-only packages and reducing complexity for developers as Node 18 nears end-of-life.
Security News
PyPI now supports iOS and Android wheels, making it easier for Python developers to distribute mobile packages.