New package commonmark: yet another markdown parser?

June 3, 2015

Last week the commonmark package was released on CRAN. The package implements some very thin R bindings to John Macfarlane’s (author of pandoc) cmark library. From the cmark readme:

cmark is the C reference implementation of CommonMark, a rationalized version of Markdown syntax with a spec. It provides a shared library (libcmark) with functions for parsing CommonMark documents to an abstract syntax tree (AST), manipulating the AST, and rendering the document to HTML, groff man, CommonMark, or an XML representation of the AST.

Each of the R wrapping functions parses markdown and renders it to one of the output formats:

md <- "
## Test
My list:
  - foo
  - bar"

The markdown_html function converts markdown to HTML:

<p>My list:</p>

Obviously the dynamic content rendered from markdown is not a full HTML document in itself. To create a full HTML page you would insert one or more of these snippets in an HTML template with static header and footer content and possibly some css/js to make the page more exciting.

The markdown_xml function gives the parse tree in xml format:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE CommonMark SYSTEM "CommonMark.dtd">
  <header level="2">
    <text>My list:</text>
  <list type="bullet" tight="true">

Most of the value in commonmark and is probably in the latter. There already exist a few nice markdown converters for R including the popular rmarkdown package, which uses pandoc to convert markdown to several presentation formats.

The formal commonmark spec makes markdown suitable for more strict documentation purposes, where we might currently be inclined to use json or xml. For example we could use it to parse files from R packages in a way that allows for archiving and indexing individual news items, without ambiguity over indentation rules and such.