Try out 6.2v3.  One of these bugs sounds like this one which was fixed:
BUGID 15967-An erroneous expression link was created if you renamed a
node whose name was in use within a group or a gizmo.

-deke

On Sun, Apr 3, 2011 at 09:43, John RA Benson
<[email protected]> wrote:
> 6.2v1 linux and osx
> also turns out it's not the updateUI callback. Arg.
>
> Also dealing with a nasty bug where a gizmo that's getting internal stuff 
> built will break connections when it's copied. Any hints on that one 
> appreciated. I'm hoping I can do something where I don't use the onCreate 
> callback but call the function that builds the gizmo after a paste:
>
> s = nuke.selectedNodes(filter='myGizmo')
>        for n in s:
>                buildMyGizmo(node=n)
>
> I see something called nukescripts.drop.dropData and in the docs:
>        # This function is called whenever data is dropped onto the DAG. 
> Override it to perform other actions.
>        # If you handle the drop, return True, otherwise return None.
>
> But I'm nut sure exactly if this applies to what I want to do or how to do it.
>
> any help appreciated - thanks!
> jrab
>
>
> On Apr 3, 2011, at 5:56 PM, Deke Kincaid wrote:
>
>> What version of nuke?  This sounds like a bug about erroneous expressions 
>> linking from inside gizmos to external nodes.  It was fixed a few v releases 
>> ago.
>>
>> -deke
>>
>> On Apr 3, 2011, at 7:52, John RA Benson <[email protected]> 
>> wrote:
>>
>>> Hey there -
>>>
>>> I have a gizmo that is doing a funny thing. Sometimes when I create it, a 
>>> green expression link will appear to another gizmo of the same class. If I 
>>> click on the gizmo it's pointing to, the expression goes away. Yeah, I'm 
>>> doing some funny stuff with updateUI, but it's nothing (although I could be 
>>> totally wrong) that would create that link.
>>>
>>> I can do this:
>>> s = nuke.selectedNode()
>>> d = s.dependencies(nuke.EXPRESSIONS)[0]
>>> d.dependent(nuke.EXPRESSIONS)
>>>
>>> and I get 's' showing me that it has an expression pointing to 'd', and 
>>> d.dependent() points back to 's'. Ok, good.
>>>
>>> if I deselect s and then select d (the green arrow then goes away) and do 
>>> d.dependent() again, I get an empty list.
>>>
>>> None of the knobs on s or d have an expression, so maybe it's from 
>>> something inside the gizmo that I'm pythonically setting?
>>>
>>> What would be swell is to find out where this mystery expression is coming 
>>> from in the first place. is there a 
>>> 'tellMeWhichKnobHasTheExpressionSinceNuke.EXPRESSIONS_tellsMeThereIsOne' 
>>> function?
>>>
>>> hmmn - it appears to be due to the updateUI callback, but it would really 
>>> be good to know which expression is being mysteriously generated by it.
>>>
>>> Cheers!
>>> JRAB_______________________________________________
>>> Nuke-python mailing list
>>> [email protected]
>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>> _______________________________________________
>> Nuke-python mailing list
>> [email protected]
>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>
> _______________________________________________
> Nuke-python mailing list
> [email protected]
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>
_______________________________________________
Nuke-python mailing list
[email protected]
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python

Reply via email to