[rabbitmq-discuss] Experimental STOMP adapter available

Tony Garnock-Jones tonyg at lshift.net
Sun Feb 3 15:17:15 GMT 2008


Hi again,

Tony Garnock-Jones wrote:
> How were you testing it? Our Erlang STOMP serialiser always supplies a 
> content-length header, and from my reading of the Net::Stomp code, it 
> looks like it should use the content-length, when there is one, to do a 
> single block read.

I've just checked, by sending

$stomp->send({destination=>'foo',
	      bytes_message=>1,
	      body=>($ARGV[0] or "test\0message")});

that the receiver correctly handles embedded NULs - which means it's in 
the branch that uses the content-length to do a block read. (Of course, 
it's probably still not doing any buffering!)

I'd be interested to hear more about your testing, and how 
RabbitMQ+STOMP performs for your use cases.

Regards,
   Tony




More information about the rabbitmq-discuss mailing list