Python-Markdown 2.2 Release Notes

We are pleased to release Python-Markdown 2.2 which makes improvements on 2.1. While 2.2 is primarily a bug fix release, some internal improvements were made to the parser, and a few security issues were resolved.

Python-Markdown supports Python versions 2.5, 2.6, 2.7, 3.1, and 3.2 out of the box.

Backwards-incompatible Changes

While Python-Markdown has received only minor internal changes since the last release, there are a few backward-incompatible changes to note:

  • Support had been dropped for Python 2.4. No guarantees are made that the library will work in any version of Python lower than 2.5. Additionally, while the library had been tested with Python 2.5, consider Python 2.5 support to be depreciated. It is not likely that any future versions will continue to support any version of Python less than 2.6.

  • For many years Python-Markdown has identified <ins> and <del> tags in raw HTML input as block level tags. As they are actually inline level tags, this behavior has been changed. This may result in slightly different output. While in most cases, the new output is more correct, there may be a few edge cases where a document author has relied on the previous incorrect behavior. It is likely that a few adjustments may need to be made to those documents.

  • The behavior of the enable_attributes keyword has been slightly altered. If authors have been using attributes in documents with safe_mode on, those attributes will no longer be parsed unless enable_attributes is explicitly set to True. This change was made to prevent untrusted authors from injecting potentially harmful JavaScript in documents. This change had no effect when not in safe_mode.

What’s New in Python-Markdown 2.2

The docs were refactored and can now be found at http://packages.python.org/Markdown/. The docs are now maintained in the Repository and are generated by the setup.py build_docs command.

The Sane_Lists extension was added. The Sane Lists Extension alters the behavior of the Markdown List syntax to be less surprising by not allowing the mixing of list types. In other words, an ordered list will not continue when an unordered list item is encountered and vice versa.

Markdown now excepts a full path to an extension module. In other words, your extensions no longer need to be in the primary namespace (and start with mdx_) for Markdown to find them. Just do Markdown(extension=['path.to.some.module']). As long as the provided module contains a compatible extension, the extension will be loaded.

The BlockParser API was slightly altered to allow blockprocessor.run to return True or False which provides more control to the block processor loop from within any Blockprocessor instance.

Various bug fixes have been made. See the commit log for a complete history of the changes.