markdown-it-directive
中文指南
Directive extension for markdown-it markdown parser. Make your Markdown great again!
Basically follow Generic directives/plugins syntax spec on CommonMark Community and compatable with markdown-it-container.
Recommended for using with markdown-it-directive-webcomponents.
Install
npm i markdown-it-directive
Required dependencies
Please note that to start using the package, your project must also have the following installed: markdown-it
, @types/markdown-it
. Check out the peerDependencies
section of package.json
for accurate information.
API
const md = require('markdown-it')()
.use(require('markdown-it-directive'))
.use((md) => {
md.inlineDirectives['aaa'] = ({state, content, dests, attrs, contentStart, contentEnd, directiveStart, directiveEnd}) => {
};
md.blockDirectives['aaa'] = ({
state, content, contentTitle, inlineContent, dests, attrs,
contentStartLine, contentEndLine,
contentTitleStart, contentTitleEnd,
inlineContentStart, inlineContentEnd,
directiveStartLine, directiveEndLine
}) => {
};
});
markdown-it-directive plugin only extracts valid directive from the markdown document, and passes it to the corresponding registered handler
for further processing.
There are two types of directive and handler
, inline level and block level.
After loading this plugin, two arrays inlineDirectives
and blockDirectives
will be added to the md
instance for registering inline level and block level handler
, and it will add inline_directive
and block_directive
rules to block ruler
and inline ruler
to extract directive from the document.
Here are three directive formats that can be recognized:
text before :directive-name[content](/link "destination" /another "one"){.class #id name=value name="string!"} text after
:: directive-name [inline content] (/link "destination" /another "one") {.class #id name=value name="string!"} content title ::
::: directive-name [inline content] (/link "destination" /another "one") {.class #id name=value name="string!"} content title ::
content
:::
The name of the handler
, which is the key of the array, only allows lowercase letters and -
, the directive name in the document is case-insensitive, and all _
will be converted into -
to find the corresponding handler
.
handler
will receive the parsed link destinations (ie those in ()
) (ie dests
), attributes (ie attrs
), not unescaped content
and the position of each component of the directive, and needs to store processed things in the MarkdownIt Token stream. Full details can be found in [markdown-it-directive-webcomponents] (https://github.com/hilookas/markdown-it-directive-webcomponents).
Example
const md = require('markdown-it')()
.use(require('markdown-it-directive'))
.use((md) => {
md.inlineDirectives['directive-name'] = ({state, content, dests, attrs, contentStart, contentEnd, directiveStart, directiveEnd}) => {
const token = state.push('html_inline', '', 0);
token.content = JSON.stringify({ directive: 'directive-name', content, dests, attrs }) + '\n';
};
md.blockDirectives['directive-name'] = ({
state, content, contentTitle, inlineContent, dests, attrs,
contentStartLine, contentEndLine,
contentTitleStart, contentTitleEnd,
inlineContentStart, inlineContentEnd,
directiveStartLine, directiveEndLine
}) => {
const token = state.push('html_block', '', 0);
token.map = [ directiveStartLine, directiveEndLine ];
token.content = JSON.stringify({
directive: 'directive-name (block)', content, contentTitle, inlineContent, dests, attrs,
}) + '\n';
};
});
console.dir(md.render(`text before :directive-name[content](/link "destination" /another "one"){.class #id name=value name="string!"} text after
:: directive-name [inline content] (/link "destination" /another "one") {.class #id name=value name="string!"} content title ::
::: directive-name [inline content] (/link "destination" /another "one") {.class #id name=value name="string!"} content title ::
content
:::`));
More examples can be found in test.js
.
License
MIT
Copyright (c) 2020, lookas