[rabbitmq-discuss] .net version of the subscriber

Tue Jan 22 23:46:10 GMT 2013

On 22/01/13 18:43, jdepp wrote:
> I am trying to interprate the following and convert into .net code to
> implement creating multiple consumers and have them subscribe to a queue.
>
> def callback(ch, method, properties, body):
>      print " [x] Received %r" % (body,)
>      time.sleep( body.count('.') )
>      print " [x] Done"
>      ch.basic_ack(delivery_tag = method.delivery_tag)
>
> channel.basic_qos(prefetch_count=1)
> channel.basic_consume(callback,
>                        queue='task_queue')
>
> channel.start_consuming()
>
>
> what is the callback method? Is that the function that is called

yes.

> and how is this faster than using a basicget within your own loop.

I answered that in the other thread.

> It is still one consumer to a queue

Yes. Depending on your definition of 'consumer'. It's certainly just one 
client connected to rabbit. And it's certainly just one AMQP consumer 
(since there is only one basic.consume method). But it *could* 
potentially spawn a new thread for every message received, or use a 
thread pool.

> right or does rabbit implement its own threading somehow.

Depends on the client library. There are certainly some that maintain 
thread pools for invoking consumer callbacks - the RabbitMQ Java client 
is an example - but usually even then only a single thread is used for 
all messages arriving for a single AMQP consumer, or even across an 
entire channel, since that way messages can be guaranteed to get 
processed on order.

Regards,

Matthias.