By John Gruber
Atoms: We are mostly sold out... but there is more!
Version 1.0b4 is out. No new features, just bug fixes and improved integration with Blosxom. Thanks to everyone who submitted bug reports. I strongly recommend this version to anyone using earlier betas.
A handful of people have asked if there’s a way to translate Markdown in reverse — to turn HTML back into Markdown-formatted plain text. The short answer is yes, by using Aaron Swartz’s new version of html2text:
html2text is a Python script that convers a page of HTML into clean, easy-to-read plain ASCII. Better yet, that ASCII also happens to be valid Markdown (a text-to-HTML format).
html2text works so well that I’m planning to use it to convert most of my old Daring Fireball articles (the ones I wrote in raw HTML). It’s worth noting that if you start with a Markdown document, translate it to HTML, then use html2text to go back to Markdown, it won’t give you the exact same document you started with. That sort of complete round-trip fidelity simply is not possible, but html2text comes pretty close.
Also, much like Markdown and SmartyPants, html2text works as a BBEdit text filter. Simply save a copy in the Unix Filters folder in your BBEdit Support folder.
Markdown and html2text are now available from the Mac OS X Services menu, thanks to Gust’s HumaneText, a free utility for Mac OS X 10.3 or later. This means you can use Markdown from Services-aware apps, including SubEthaEdit and TextEdit.