Gerrit Voss wrote:
> Hi,
> 
> On Tue, 2006-05-09 at 08:37 -0500, Patrick Hartling wrote:
>>> interesting, one quick test, could you add a call to
>>>
>>> beginEditCP(node, VolMask); 
>>> node->invalidateVolume(); 
>>> endEditCP(node,VolMask);
>>>
>>> after you activated a node.  
>> Doing this fixes things for me.
> 
> ah good, I hope I spotted problem in Node::updateVolume (in combination
> with Node::changed). Once the cvs is back online I'll hopt to commit
> something to fix it.
> 
>>> >From a quick glance over the code involved in travmask and volume
>>> handling it might be a little bit optimistic at some points ;-)
>>>
>>> Currently my feeling that is it more OpenSG bug exposed by OSG::Group
>>> change?
>>>
>>> I'll see if I can easily recreate the problem.
>> Let me know if there is anything I can do to help you with that. The data I
>> am using is confidential, but I might be able to make up something if you
>> need a test case.
> 
> So far I would wait until we can try what I have in mind ;-). If you
> need it earlier (before the cvs is back online) the thing you could try
> is modifying Node::updateVolume to
> 
> void Node::updateVolume(void)
> {
>     if(_sfVolume.getValue().getInstance().isValid() == true ||
>        getTravMask() == 0)
>     {
>         return; 
>     }
> 
> 
> you also can remove the later getTravMask if needed. 

The above change fixes the problems I was seeing. Thanks very much! I am
glad to be able to get my OpenSG source up to date again.

 -Patrick

> regards,
>   gerrit



-- 
Patrick L. Hartling                    | VP Engineering, Infiscape Corp.
PGP: http://tinyurl.com/2msw3          | http://www.infiscape.com/

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to