[rabbitmq-discuss] py-amqplib being reworked

Barry Pederson bp at barryp.org
Mon Dec 1 21:36:08 GMT 2008

I've had some time to work on py-amqplib over the holiday, and
have done some serious rearranging of code.  The biggest
thing has been to breakup the monstrosity that was there before to 
better reflect how the protocol is layered.

I've also started working on features people have been having problems 
with or needs for, like non-blocking waits, flow-control, and handling 
of returned messages.

It's still got a ways to go, needs lots of testing, and I still may make 
some large changes, but for anyone interested I've started a new 
Mercurial repository at:


The API is essentially unchanged so old users of amqplib
*should* still work.

The Connection() initializer has a "use_threading" option
that defaults to False, giving the old single-threaded behavior,
but with True it creates a helper thread that allows for
timeouts and flow control.

I'm not happy with some of the details about waiting and handling
flow-control and returned messages, and error handling needs work, so 
it's definitely a work-in-progress (actually, it always has been, but 
it's even more so for the near future).


More information about the rabbitmq-discuss mailing list