A metadata parser for code fences in markdown
Many markdown processors can parse the language token associated with a code fence. fenceparser is meant for parsing other metadata besides language token. It supports
- ranges, (for example,
{1} {3, 7} ins{9..11, 88} del{90, 101..167}) and - key-value pairs (for example,
caption='Hello, World')
This package is ESM only.
In Node.js (version 12.20+, 14.14+, or 16.0+), install with npm:
npm install @microflash/fenceparserIn Deno, with esm.sh:
import parse from "https://esm.sh/@microflash/fenceparser";In browsers, with esm.sh:
<script type="module">
import parse from "https://esm.sh/@microflash/fenceparser?bundle";
</script>Say, you have the following code fence
```js {1} {3, 7} {9..11, 88} {90, 101..112} text-color='--text-default' syntax_theme="nord" css=`{ *: { display: none }}`
remark will provide the meta and lang for the above code fence.
{
"lang": "js",
"meta": "{1} {3, 7} {9..11, 88} {90, 101..104} text-color='--text-default' os=\"macOS 26 Tahoe\" syntax_theme=nord css=`{ *: { display: none }}`"
}Use the fenceparser to parse the meta as follows.
import parse from "@microflash/fenceparser";
const metadata = parse("{1} {3, 7} {9..11, 88} {90, 101..104} text-color='--text-default' os=\"macOS 26 Tahoe\" syntax_theme=nord css=`{ *: { display: none }}`");
console.log(metadata);Running the above example yields.
{
'*': [
1, 3, 7, 9, 10,
11, 88, 90, 101, 102,
103, 104
],
'text-color': '--text-default',
os: 'macOS 26 Tahoe',
syntax_theme: 'nord',
css: '{ *: { display: none }}'
}The default export is parse function.
The following options are available. All of them are optional.
rangeKey(default:*): specifies the range key to group ranges without annotations
- The key and value must be separated by equality sign
=. - The value can be unquoted if there are no spaces, else it must be wrapped in single-quotes
', double-quotes"or backticks`. - The key can contain letters, numbers, underscore, hyphen, and dollar sign.
- A range can be a single number, or a pair of numbers denoting the start and end values separated by double-dots
... - A range must be specified in curly braces.
- Multiple ranges can be specified in a single pair of curly braces. They should be separated by comma
,. - Ranges can also be annotated; the annotation should be prefixed before the starting curly brace. The default annotation is
*. You can customize this annotation by passingrangeKeyvalue in the options. - Ranges will be grouped in a single array by their annotations.
import parse from "@microflash/fenceparser";
console.log(parse("{100}"));Running the above example yields.
{
'*': [ 100 ]
}import parse from "@microflash/fenceparser";
console.log(parse("{3, 7} {9..11, 101..105}"));Running the above example yields.
{
'*': [
3, 7, 9, 10, 11,
101, 102, 103, 104, 105
]
}import parse from "@microflash/fenceparser";
console.log(parse("{3, 7} ins{9..11, 13} del{101..105}"));Running the above example yields.
{
'*': [ 3, 7 ],
ins: [ 9, 10, 11, 13 ],
del: [ 101, 102, 103, 104, 105 ]
}import parse from "@microflash/fenceparser";
console.log(parse("data-theme=synthwave callback=`(code) => copyToClipboard(code)`"));Running the above example yields.
{
'data-theme': 'synthwave',
callback: '(code) => copyToClipboard(code)'
}import parse from "@microflash/fenceparser";
console.log(parse("{100, 102}", { rangeKey: "highlight" }));Running the above example yields.
{
highlight: [ 100, 102 ]
}Check the fixtures for more examples on the syntax.
Any changes in the parser should have corresponding tests.
Run the tests with the following command.
pnpm test