Marko

A markdown parser with high extensibility.

Marko is a markdown parser written in pure Python that complies CommonMark's spec v0.28. It is designed to be highly extensible, see Extend Marko for details.

Marko requires Python 3.4 or higher, Python 2.7 support is still in plan but I guess it is not needed.

Why Marko

Among all implementations of Python's markdown parser, it is a common issue that user can't easily extend it to add his own features. Furthermore, Python-Markdown and mistune don't comply CommonMark's spec. It is a good reason for me to develop a new markdown parser and use it.

Respecting that Marko complies CommonMark's spec at the same time, which is a super complicated spec, Marko's performance will be affected. A benchmark result shows that Marko is 3 times slower than Python-Markdown, but a bit faster than Commonmark-py, much slower than mistune. If performance is a bigger concern to you than spec compliance, you's better choose another parser.

Use Marko

The installation is very simple:

$ pip install marko

And to use it:

from marko import Markdown
markdown = Markdown()
print(markdown(text))

Marko also provides a simple CLI, for example, to render a document and output to a html file:

$ cat my_article.md | marko > my_article.html

Extend Marko

Please refer to Document

License

Marko is released under MIT License

Change Log

• v0.3.4: Fix bugs about extensions.
• v0.3.1: Pangu extension.
• v0.3.0: Change the entry function to a class, add TOC and footnotes extensions.
• v0.2.0: Github flavored markdown and docs.
• v0.1.0: Commonmark spec tests.