Python data persistent object for communication between threads

Question

As shown here , I set up a python Django application served by cherrypy wsgi server. The app is basically another IRC client. Here's the deal : it is very likely that I have to create several separate instances of my bot for every new server connection I need to establish. I need to be able to communicate with each bot. One suggested answer was to use multiprocessing.Queue . This object is data persistent and does allow me to communicate with a bot. However, I need all my bots to listen to the same signal simultaneously, to stop for instance. Every bot has to control whether the stop signal was for him or not. Therefore, I need an object, or some other method, that allows for each bot, running in a separate daemonic thread, to listen to a bunch of signals. The Bus system used by cherrypy for server wide messages is great but seems like an overkill here and I would have no idea how to implement it. Any suggestions ?

Answer 1

So, turns out I just mixed up variables and objects. Variables cannot be defined in one module and imported to another one (a = foo, then modify a from another module and import a from third module). This is possible with lists, dicts and stuff. Solution : at each bot launch (and attempt to connect to a given server), a queue is created and assigned to this server in a dict with a key : the nickname (that has to be unique on IRC). Each bot has a deamonic listener that reads the queue. Other modules import the dict, the specific queue, put a signal there, that is used by the listener. The system works perfectly. Thumbs up. I'm having difficulties with one aspect though : each bot can have multiples DCC connections and it's hard to assign data to each connection and not mix them up.