I just committed a bunch of updates to the WS-Discovery stuff that adds a 
setVersion10() method onto the WSDiscoveryClient that should allow it to flip 
to the older WS-Discovery spec (and SOAP 1.1 and the older WS-Addressing 
version).   It also fixes a bunch of things with the wsa Actions and other 
bugs.    I'm hoping that will allow this to work.   Is there any way you could 
grab tomorrows snapshots (or build from the latest branches) and give it a spin 
and see?

On the server side, CXF services will now response to a 1.0 Probe as well as 
the 1.1 probe.  However, I do still have a bit of mapping to do to map the 
various scopes.  Right now, they only look at the 1.1 scope names.


Dan



On Apr 4, 2013, at 1:27 PM, Daniel Kulp <dk...@apache.org> wrote:

> 
> Hit send too soon….
> 
> On Apr 4, 2013, at 1:25 PM, Daniel Kulp <dk...@apache.org> wrote:
> 
>> 
>> On Apr 4, 2013, at 10:06 AM, Pampolini Matteo 
>> <matteo.pampol...@selex-es.com> wrote:
>>> Hello there,
>>> 
>>> I'm trying to write a simple application to discover ONVIF devices in my 
>>> network.
>> 
>> Very interesting…  Wasn't even aware of them.   If anyone would like to buy 
>> me one, I'd be more that happy to experiment more to get it working….   :-)  
>>  
>> 
>> 
>>> Code is mainly based on ws_discovery sample with the following simple 
>>> difference:
>>> 
>>> List<EndpointReference> references = client.probe(new 
>>> QName("http://www.onvif.org/ver10/network/wsdl";<http://www.onvif.org/ver10/network/wsdl>,
>>>  "NetworkVideoTransmitter"));
>>> 
>>> I'm writing because it does not work, of course. I do not even see (through 
>>> WireShark) multicast packets being sent over my network interface,
>>> can anyone please help me? Please note that ONVIF is based on WS-Discovery 
>>> version 2005/04.
>> 
>> There's two parts to this:  1) getting the UDP packets out    2) 
>> WS-Discovery 2005/04.  
>> 
>> No idea what to do about (1).   
> 
> 
> Does the ws-discovery sample work?    If you run the sample with wireshark, 
> does that get the packets?   I guess that would be the starting point for 
> looking at that.
> 
> (2) is definitely a big issue.  I'd likely try and use the CXF transformation 
> feature to map the namespaces and see if that works, but it obviously won't 
> work if (1) cannot be resolved.
> 
> 
> 
> -- 
> Daniel Kulp
> dk...@apache.org - http://dankulp.com/blog
> Talend Community Coder - http://coders.talend.com
> 

-- 
Daniel Kulp
dk...@apache.org - http://dankulp.com/blog
Talend Community Coder - http://coders.talend.com

Reply via email to