How to use named functions with Async waterfall in node.js

Lets start with a real world example where you want to create a user in CouchDB. This is a two step process that involves asking for a unique uuid and then creating a user based on that.

• The following code (+/-) abstracts out each individual step:

  	function getUUID() {
  	nano.request(
  	{
  	method : 'GET',
  	path : '_uuids?count=1'
  	},
  	function (err, body, headers) {
  	if (err) {
  	console.error(err);
  	} else {
  	console.log(body.uuids[0]);
  	}
  	}
  	);
  	}
  	function createUser(uuid) {
  	var usersDB = nano.db.use('_users');
  	usersDB.insert(
  	{
  	'name' : uuid,
  	'password' : 'pwdUser1',
  	'roles' : [],
  	'type' : 'user'
  	},
  	'org.couchdb.user:'+ uuid,
  	function(err, body, header) {
  	if (err) {
  	console.error('[_users.insert] ', err);
  	} else {
  	console.log('you have inserted the uuid.')
  	console.log(body);
  	}
  	}
  	);
  	}

  view raw abstracted_methods.js hosted with ❤ by GitHub
• But once you have the abstracted methods, how do you chain their results? This is where the waterfall approach from the Async NPM package comes in.
• Since you've already abstracted your methods, you may try something like this (+/-) to make them a little more waterfall friendly:
• And then you may try the simplest way you can think of asking waterfall to run:
• But that will fail to run because the syntax of the sentence has ended up being written in a manner where callback must be a variable that is already defined and ready to be used. Making an anonymous method here and bringing up the entire code bodies of the abstracted named-methods would defeat the readability and reuseability that we have been trying to achieve. Here's a tweak that will make it happen: