Conduit & Mobile Devices - Week 5

["John Stowers" <john stowers gmail com>]

Hey Team,

Apologies for a few days late, I have been preparing for GUADEC and my
talk, etc. This week I was able to successfully run the first
libsyncml test from python (i.e. a port of one of the C libsyncml test
cases to python). This is significant because It showed that the
marshaling of callbacks between libsyncml and python now seems to be
working - at least to some extent. It also obviously implies that most
of the rest of the autogenerated binding is doing the right thing, and
I have sufficient API coverage (>>90%) to actually do the useful
stuff.

The depressingly short piece of code is

import syncml

def manager_cb(eventType, session):
	print "CALLBACK ", eventType

c = syncml.Transport(syncml.SML_TRANSPORT_HTTP_CLIENT)
s = syncml.Transport(syncml.SML_TRANSPORT_HTTP_SERVER)

c.SetConfigOption("URL", "http://127.0.0.1:12020";)
s.SetConfigOption("PORT","12020")

cm = syncml.Manager(c)
cm.SetEventCallback(manager_cb)

sm = syncml.Manager(s)
sm.SetEventCallback(manager_cb)

c.Initialize()
s.Initialize()

cm.Start()
sm.Start()

datastr = "<SyncML></SyncML>"
data = syncml.TransportData(datastr, len(datastr),
syncml.SML_MIMETYPE_XML, False)

c.Send(data)
while 1:
	sm.Dispatch()

Next week I am not sure what I will do. I will be at GUADEC.

Wahoo!

John