Skip to content

nanomsg wrapper for python with multiple backends (CPython and ctypes) should support 2/3 and Pypy

License

Notifications You must be signed in to change notification settings

freepn/nanomsg-python

 
 

Repository files navigation

nanomsg-python

GitHub CI Workflow Status

This fork of nanomsg-python is currently tested on Python 2.7 and Python 3.6+ on Linux, Darwin, and Windows (the latter two somewhat theoretical).

Python library for nanomsg which does not compromise on usability or performance.

Like nanomsg this library is still experimental, the API is fairly stable but if you plan to use it at this time be prepared to get your hands dirty, fixes and enhancements are very welcome.

The following versions of Python are supported CPython 2.6+, 3.2+ and Pypy 2.1.0+

Bugs and change requests can be made here.

nanommsg library in /usr/local

If you're nanomsg is in /usr/local and your machine is not configured to find it there you can rename the usr_local_setup.cfg to setup.cfg to fix the problem.

Example

from __future__ import print_function
from nanomsg import Socket, PAIR, PUB
s1 = Socket(PAIR)
s2 = Socket(PAIR)
s1.bind('inproc://bob')
s2.connect('inproc://bob')
s1.send(b'hello nanomsg')
print(s2.recv())
s1.close()
s2.close()

Or if you don't mind nesting you can use Socket as a context manager

with Socket(PUB) as pub_socket:
    .... do something with pub_socket
# socket is closed

The lower level API is also available if you need the additional control or performance, but it is harder to use. Error checking left out for brevity.

from nanomsg import wrapper as nn_wrapper
from nanomsg import PAIR, AF_SP

s1 = nn_wrapper.nn_socket(AF_SP, PAIR)
s2 = nn_wrapper.nn_socket(AF_SP, PAIR)
nn_wrapper.nn_bind(s1, 'inproc://bob')
nn_wrapper.nn_connect(s2, 'inproc://bob')
nn_wrapper.nn_send(s1, b'hello nanomsg', 0)
result, buffer = nn_wrapper.nn_recv(s2, 0)
print(bytes(buffer))
nn_wrapper.nn_term()

License

MIT

Authors

Tony Simpson

About

nanomsg wrapper for python with multiple backends (CPython and ctypes) should support 2/3 and Pypy

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 68.5%
  • C 29.9%
  • Batchfile 1.6%