Python SimpleXMLRPCServer返回值

Question

I just started using a XMLRPC server and clients to connect my raspberry pi to a computer.

My server looks like this:

from SimpleXMLRPCServer import SimpleXMLRPCServer
from SimpleXMLRPCServer import SimpleXMLRPCRequestHandler
import numpy as np

allow_reuse_address = True   
ip = '...'
port = 8000  

class RequestHandler(SimpleXMLRPCRequestHandler):
rpc_paths = ('/RPC2',)  

server = SimpleXMLRPCServer((ip, port), requestHandler=RequestHandler)
server.register_introspection_functions() 

def Spectrum():
    data = ... # it's a numpy array
    return data

server.register_function(Spectrum, 'Spectrum')  
server.serve_forever()

My client looks like this:

#!/usr/bin/env python

import xmlrpclib
import numpy as np

[...]

def getSpectrum():
try:
    s = xmlrpclib.ServerProxy(server)
    v = s.Spectrum()
    print v         

except:
    print "no data"

My server is running and my test function shows that it works. But my function getSpectrum() always throws an exception. I figured out that it works fine if my return value is a float instead of a numpy array:

def Spectrum():
    data = ... # it's a numpy array
    return float(data[0][0])

I don't have any idea what's wrong but I think it should be possible to return a numpy array. Do you know how to fix that?

Answer 1

The xmlrpclib only supports marshalling of standard python types. numpy arrays are an extension type and therefor cannot be serialized out of the box. A simple solution would be to return a list representation of the numpy array and when you receive the data you turn it into a numpy array again:

def Spectrum():
    data = ... # it's a numpy array
    return data.tolist()
...
def getSpectrum():
    try:
        s = xmlrpclib.ServerProxy(server)
        v = numpy.asarray(s.Spectrum())
        ...