This is off of memory so I apologize if I don't get all of the details right.
The base SimpleXMLRPCServer uses TCPServer as it's server component and SimpleXMLXMLRPCRequestHandler as it's handler. The handler is a subclass of BaseHTTPRequestHandler, which itself doesn't handle any multithreading. You need to use one of the *MixIn classes if you want to handle concurrent requests. -Jeff On 3/28/07, Erik Johnson <[EMAIL PROTECTED]> wrote: > > "Achim Domma" <[EMAIL PROTECTED]> wrote in message > news:[EMAIL PROTECTED] > > Hi, > > > > is SimpleXMLRPCServer multithreaded or how does it handle multiple > > clients? I want to implement a simple server which will be queried by > > multiple processes for work to be done. The server will simply hold a > > queue with files to process. The clients will ask for the next file. > > > > Do I have to sync access to the queue or is the server not threaded at > all? > > > I don't claim to know much about the internals of the module, but it > imports BaseHTTPServer which I am sure is multithreaded. I did not test > this with functions, but when you register an object in the server, > different clients are accessing the same object: > (The following example is modified from that given in the module > documentation: http://docs.python.org/lib/module-SimpleXMLRPCServer.html) > > > $ cat server.py > #!/usr/bin/env python > > from SimpleXMLRPCServer import SimpleXMLRPCServer > > # Create server > server = SimpleXMLRPCServer(("localhost", 8000)) > server.register_introspection_functions() > > # Register pow() function; this will use the value of > # pow.__name__ as the name, which is just 'pow'. > server.register_function(pow) > > # Register a function under a different name > def adder_function(x,y): > return x + y > server.register_function(adder_function, 'add') > > > # Register an instance; all the methods of the instance are > # published as XML-RPC methods (in this case, just 'div'). > class MyObj: > SEQ = None > > def __init__(self): > self.SEQ = range(10000) > self.index = 0 > > def next_seq(self): > n = self.SEQ[self.index] > self.index += 1 > return n > > server.register_instance(MyObj()) > > # Run the server's main loop > server.serve_forever() > > > # First client calls: > >>> import xmlrpclib > >>> url = 'http://localhost:8000' > >>> s = xmlrpclib.Server(url) > >>> s.next_seq() > 0 > >>> s.next_seq() > 1 > >>> s.next_seq() > 2 > > > # second client calls > >>> import xmlrpclib > >>> url = 'http://localhost:8000' > >>> s = xmlrpclib.Server(url) > >>> s.next_seq() > 3 > > > I did not programmatically drive these calls at the same time to try to > force simultaneous access from separate requests, but my intuition is that > it's a pretty good bet that SimpleXMLRPCServer has taken care of re-entrant > issues. It may be instructive for you to run that test yourself and > explicitly verify requests do not interact. > > Alternatively, if you really want to know for sure... Python is an > open-source project. You can dig around in SimpleXMLRPCServer.py yourself > and figure out exactly what it is doing: > > >>> import SimpleXMLRPCServer > >>> SimpleXMLRPCServer > <module 'SimpleXMLRPCServer' from > '/usr/lib/python2.4/SimpleXMLRPCServer.pyc'> > >>> > > [EMAIL PROTECTED] ~ > $ vi /usr/lib/python2.4/SimpleXMLRPCServer.py > > > > see also: > http://docs.python.org/lib/module-BaseHTTPServer.html > http://docs.python.org/lib/module-SocketServer.html > > > > > -- > http://mail.python.org/mailman/listinfo/python-list > -- http://mail.python.org/mailman/listinfo/python-list
