Feature: literate language mode and support for literate coffee-script

[sjorek]

… what should I say, here is support for literate coffee-script. If one digs deeper into this mode, you will recognize that code and comments have always separate segments. This is necessary, because large blocks of literate comments will certainly get out of sync with the code they precede. This solution prevents this by not having literate comments side-by-side with code. But of course, non-literate comments in code are processed as usual.

A live-example:

• code: https://github.com/sjorek/groc/blob/master/lib/namespace.litcoffee
• renders to: http://sjorek.github.io/groc/namespace.html

Any feedback, especially to the side-by-side related stuff is always welcome !

Cheers,
Stephan

[leonardt]

+1

[leonardt]

It seems that this offsets code blocks with comments by one line

compared to http://nevir.github.io/groc/cli.html as an example, where code will be lined up with the comment, I think this may be a simple parsing issue.

[sjorek]

This is a deliberate design decision and obviously a bad one. Look at this literate-coffescript-file (and it's source) to get an idea of why I did it: … without this behavior code and comments get to much out of sync.