Markdown-it-include NPM

markdown-it-include

Markdown-it plugin which adds the ability to include markdown fragment files.

Install

node.js, browser:

npm install markdown-it-include --save
bower install markdown-it-include --save

Use

Let's create a markdown which uses a header and a footer from two separate files:

File: 'header.md'

# This is my header for all my markdowns

File: 'footer.md'

Follow me on twitter!

Let's assume that header.md and footer.md are located in /in/this/directory.

Now it's your turn to play markdown-it!

var md = require('markdown-it')()
            .use(require('markdown-it-include'), '/in/this/directory');

md.render('!!!include(header.md)!!!\n\n*your content*\n\n!!!include(footer.md)!!!');

It will produce this

<h1>This is my header for all my markdowns</h1>
<p><em>your content</em></p>
<p>Follow me on twitter!</p>

Options

var md = require('markdown-it')()
            .use(require('markdown-it-include'), options);

Options Type: String|Object

If it's a string, it's the same as options.root.

Otherwise, it must be an options object (or null / undefined when you only wish to use the default options).

The options object may contain any of these:

root

Type: String
Default: .

root is the base directory of all the markdown files.

get RootDir

Type: Function
Default: (options, state, startLine, endLine) => options.root

getRootDir() can be used to customize the root directory used for the includes in a very flexible way.

Here's an example where the root directory for the relative paths in the MarkDown is set to the directory the currently parsed MarkDown file is located in. Note that by setting this callback in the state.env state environment, we can have multiple different functions for multiple parse actions in parallel -- ahandy thing to have when we process the MarkDown content asynchronously, for example!

// md instance setup:
const options = {
  root: '/bogus/',
  // show the
  getRootDir: (options, state, startLine, endLine) =>
      state.env.getIncludeRootDir(options, state, startLine, endLine),
  bracesAreOptional: true
};

let md = require('markdown-it')()
           .use(require('markdown-it-include'), options);

...

// now in some async code rendering multiple MD files, we can do this:

// (`mdPath` is an absolute path pointing to the MD file being processed)
let mdPath = ...;

let env = {};
env.getIncludeRootDir = function (options, state, startLine, endLine) {
  return path.dirname(mdPath);
};

// Use the 'unwrapped' version of the md.render / md.parse process:
//
// let content = md.render(data); --> .parse + .renderer.render
//
// .parse --> new state + process: return tokens
// let tokens = md.parse(data, env)
let state = new md.core.State(data, md, env);   // <-- here our env is injected into state!
md.core.process(state);
let tokens = state.tokens;
// now call md.render():
let htmlContent = md.renderer.render(tokens, md.options, env);
// presto!              (End of `env` lifetime, BTW.)

includeRe

Type: RegExp
Default: /\!{3}\s*include(.+?)\!{3}/i

By default the !!!include(path)!!! statement is used to include markdown fragment files. This option allows to change the regular expression and then customize this statement.

bracesAreOptional

Type: Boolean
Default: false

When set to TRUE, this option allows users to write include statements like !!!include path !!! instead of !!!include(path)!!!`. Note that when this option is set, both these forms are acceptable.

This allows you to configure the plugin to use C/C++ like #include statements in your MarkDown if you're so inclined -- I know I am! :)

Example usage:

const options = {
  includeRe: /#include(.+)/,
  bracesAreOptional: true
};

let md = require('markdown-it')()
           .use(require('markdown-it-include'), options);
let html = md.render('#include(xyz.md)\n');

throwError

Type: Boolean
Default: true

When set to false, instead of throwing an error message, the error message will be written into the output. For references to possible error messages as well as how to change it, see options 'notFoundMessage' and 'circularMessage'.

Error messages will always be prefixed with the text "INCLUDE ERROR:" and will be rendered as <h1> headings wheenever possible so they will be pretty obvious in the generated output when they occur. After all, you don't want errors to silently hide under the rug.

notFoundMessage

Type: String
Default: File '{{FILE}}' not found.

With notFoundMessage the default error message when the to be included file cannot be found can be changed. The marker {{FILE}} in the message string will be replaced with the full file path.

circularMessage

Type: String
Default: Circular reference between '{{FILE}}' and '{{PARENT}}'.

With circularMessage the default error message when there is a circular reference between files can be changed. The markers {{FILE}} and {{FILE}} in the message string will be replaced with the respective full file paths.