jash.js allows you to more easily use node as a shell scripting facility. It will collate all output in an easy to consume way - avoiding pitfalls that can cause you to lose data.

You can call any commandline utility like this: (Why '$'? It's a shell prompt!)

var $ = require('jash');
$.ls('-l', '/tmp/foo', function(statusCode, stdout, stderr) {
	console.log('files are', stdout);
}

var getStreamConsumer = function() {
		return function(data) {
			console.log('I got some data!', data);
		};
};

or with your own callbacks for stdout and stderr. This is nice for long running processes and the like

$.tail('-f', '/var/log/foo.log', getStreamConsumer(), getStreamConsumer());

Since the child process is returned, you can also do whatever you want:

var child = $.ls('-l', '/tmp/foo');
child.stdout.on ...
child.stderr.on ...
child.on('exit') ...

The full API:

$.cmd(arg1, arg2, arg3, [options],[cb [, stdoutHandler, stdErrHandler]]);

where options is the options argument passed to child_process.spawn: http://nodejs.org/api/child_process.html#child_process_child_process_spawn_command_args_options

arguments can also be passed in an array as the first argument:

$.cmd([arg1, arg2, arg3], ...

When only one callback is passed, cb is a function that takes 3 arguments:

function(statusCode, stdout, stderr) { ...

You can also pass 3 callbacks:

cb = function(statusCode) ...
stdout = function(data) ...
stderr = function(data) ...

jash is written by Tom Fairfield <www.github.com/fairfieldt>