Skip to content

Latest commit

 

History

History
77 lines (51 loc) · 3.26 KB

README.md

File metadata and controls

77 lines (51 loc) · 3.26 KB
Raw

This project is work in progress. if you are interested in contributing or otherwise have input please touch base via github

about

q is a queueing toolkit. the idea is to provide a universal application programming interface that can be used throughout the entire application development lifecycle without the need to commit to a specific queueing technology or to set up complex queueing environments where such are not required. you can think of it as an ORM for queueing.

q runs on multiple back-ends and has bindings to many programing languages. and so, while during development you will most likely run it in-memory and let it clear when the process dies, you may choose a redis back-end on your test environment and running dedicated servers backed by rabbitMQ, amazon SQS or some other enterprise queueing system on production.

see more about the core library at https://github.com/tomerd/q

q bindings for node.js

q bindings for node.js uses the ffi module to bind to q's native API. q is represented by the node-q module which exposes a simple API:

  • version: returns the version of q

  • connect(config): connects to the library and initialized a connection to the backend specified by the config param. see further documentation on backends at the core library.

  • disconnect: disconnect from the library. no further calls can be made after this.

  • post(queue, job):/ posts a job to a named queue (aka channel). a job is represented by the simple json.

    the job's json consists of 3 fields:

    • uid (string, optional): a unique identifier for the job, useful if you plan on updating or rescheduling the job.
    • data (string): the payload, will be used by the consumer to actually perform the job.
    • run_at (date, optional): if you want the job to be scheduled in the future, use this field to specify the target timestamp.

  • reschedule(uid, run_at): reschedule the job identified by uid to a new target date.

  • cancel(uid): cancel a scheduled jobs.

  • worker(queue, &delegate): register a worker [function] for a named queue (aka channel). the worker will start receiving jobs immediately.

  • observer(queue, &delegate): register an observer [function] for a named queue (aka channel). the observer will start receiving jobs immediately. the difference between an observer and a worker is that the observer is passive in nature and as such is notified only after a worker has completed the job successfully.

  • drop: careful, drops all queues! useful in development scenarios when you need to flush the entire queue system.

usage example
var q = require('./node-q');

q.connect(	{ driver: 'redis',  host: '127.0.0.1'  });

var total = 100;
for (var index=0; index < total; index++)
{
	var uid = q.post("channel1", { data: "node " + index });
	console.info("posted %s", uid);
}

var received = 0;
q.worker("channel1", function(data)
{
	received++;
	console.info("node worker received [%s]", data);
});

setInterval(function()
{
	console.info(received + "/" +  total);
	
	if (received == total)
	{
		console.info("done");
		q.disconnect();
		process.exit(0);
	}
	
}, 1000);