I am currently reading Smooth CoffeeScript which is a wonderful introduction to CoffeeScript.

Two PDF versions are available: with and without solutions to exercices, and there is also an interactive version which can be browsed directly. Other resources are available, e.g. JavaScript Basics.

I found this book thanks to a post by Karl Broman (who also has a very nice website full of interesting resources for those interested in statistics, genetics, and programming language).

CoffeeScript is a little language that compiles into JavaScript.

Hopefully, the syntax is cleaner than Javascript itself, and we can use JS libraries. It comes with a REPL and it has an Emacs mode too. Something weird about Emacs mode for Node.js or CoffeeScript: when running node or coffee as inf-mode, strange escape characters will be printed instead of the nice looking REPL prompt we get in classic Terminal. It looks like it results from Node.js using Readline. As suggested on Stack Overflow, it is possible to use something like export NODE_NO_READLINE=1 (in your .bashrc or .profile file); however, if you want Raedline support when running node or coffee in a Terminal, it might just easier to change the command that Emacs call (for example, create two sh scripts, say node2 and coffee2, that set the above environment variable before running either node or coffee). For node, we can use (setq inferior-js-program-command "node2"). It appears that an even easier solution is proposed on Node.js help page for the REPL: using a combination of NODE_NO_READLINE and rlwrap.

Literate programming is possible thanks to docco. Incidentally, I used a similar tool for a statistical introduction to health measurement with R: rocco which plays nicely with R code. Finally, it relies on Node.js, which is another of those nice projects I discovered last year. But see this cool “Introduction to Node.js with Ryan Dahl:”

Now the question I’d like to ask is: Would it be possible to play with d3.js with CoffeeScript directly?¹ It looks like Harlan T Wood asked the same question some months ago and provided an answer, see his gist (also on http://bl.ocks.org/1091420).