On Wed, Sep 28, 2011 at 13:36, Whit Armstrong <[email protected]>wrote:
> Chuck and Martin, thanks for your help. > > Here is the R binding I've been working on. Comments welcome. > https://github.com/armstrtw/rzmq > > It's a good start, but still needs more robust testing. > > Additionally, does anyone know of a good zmq_poll example in an > interpreted language? > I don't know if it's a 'good' example, but: pyzmq's poll/select wrapping (Cython): https://github.com/zeromq/pyzmq/blob/master/zmq/core/poll.pyx And replacing Tornado's ioloop with zmq_poll, for writing full applications: https://github.com/zeromq/pyzmq/blob/master/zmq/eventloop/ioloop.py -MinRK > I need to think about that feature a bit more before I implement it. > > -Whit > > > > On Wed, Sep 28, 2011 at 9:26 AM, Martin Sustrik <[email protected]> > wrote: > > On 09/28/2011 02:31 PM, Whit Armstrong wrote: > > > >> I modified the code to use HWM=1, but the sender still manages to send > >> more than 10 messages even if I only call zmq_recv once on the PULL > >> socket. > > > > There's just 1 message in the 0MQ queue, the rest of the messages you see > to > > be in flight are stored in TCP buffers. > > > > Martin > > > _______________________________________________ > zeromq-dev mailing list > [email protected] > http://lists.zeromq.org/mailman/listinfo/zeromq-dev >
_______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
