Dabscm Library Reference

(scm markdown)

Markdown parser and HTML renderer (CommonMark subset)

Overview

(scm markdown) is a small, dependency-free Markdown parser and HTML renderer covering a practical CommonMark subset. It has two entry points: parse-markdown turns Markdown text into an s-expression AST you can walk, and markdown->html renders that AST to an HTML fragment.

It is the engine behind this documentation site's library preambles, but it is written to be reusable anywhere you need to turn Markdown into HTML or into your own output format.

Rendering Markdown to HTML

The common case is one call:

(import (scm markdown))

(markdown->html "# Title\n\nHello **world** and `code`.")
;; => "<h1>Title</h1>\n<p>Hello <strong>world</strong> and <code>code</code>.</p>"

Text and link URLs are HTML-escaped for you, so user-supplied content is safe to render.

Walking the AST

When you want to target something other than HTML (a Word document, a terminal, LaTeX, …), parse to the AST and walk it yourself:

(parse-markdown "- one\n- two")
;; => ((bullet-list (item "one") (item "two")))

(parse-markdown "See [docs](http://example.com).")
;; => ((paragraph "See " (link ("docs") "http://example.com") "."))

Block nodes are heading, paragraph, code-block, blockquote, bullet-list, ordered-list and thematic-break. Inline nodes are plain strings plus strong, emph, code and link. See the library source header for the exact shapes.

Supported subset

  • ATX headings (# … ######), with trailing #s stripped
  • Fenced code blocks (``` or ~~~) with an optional info/language string
  • Bullet lists (-, *, +) and ordered lists (1. / 1))
  • Blockquotes (>), parsed recursively
  • Thematic breaks (---, ***, ___)
  • Inline code, **strong**, *emphasis*, [links](url) and backslash escapes

By design it does not handle setext headings, reference links, images, raw HTML, tables, or nested lists — keep the input simple and predictable.

2 bindings

markdown->html
Syntax: (markdown->html s)
Library: (scm markdown)
Description: Renders the Markdown text in string s to an HTML fragment (no
  surrounding <html>/<body>). Text and link URLs are HTML-escaped. See the
  library header for the supported Markdown subset.
Example:
  (markdown->html "# Hi\n\nA *para*.")
    => "<h1>Hi</h1>\n<p>A <em>para</em>.</p>"
parse-markdown
Syntax: (parse-markdown s)
Library: (scm markdown)
Description: Parses the Markdown text in string s into a list of block nodes
  (an s-expression AST). See the library header for the supported subset and
  the node shapes. Use markdown->html to render the result, or walk the AST
  yourself to target another backend.
Example:
  (parse-markdown "# Title\n\nHello **world**")
    => ((heading 1 "Title") (paragraph "Hello " (strong "world")))