Heres what I was putzing around with today. Wanted to make a more robust
context-aware node creation function. I'm no python expert so suggestions
and criticisms are welcome. I would like to find a better detection method
for 3d nodes, since the 'render_mode' knob is not in all 3d nodes, like
Lights.
## context aware node function
def contextAwareCreateNode(node2d=None, node3d=None, nodeDeep=None):
nodes = nuke.selectedNodes()
try:
if nodeDeep and selectedNodesDeep(nodes):
return nuke.createNode(nodeDeep)
elif node3d and selectedNodes3d(nodes):
return nuke.createNode(node3d)
else:
raise ValueError
except:
return nuke.createNode(node2d)
## Returns true if all selected nodes are Deep nodes
def selectedNodesDeep(nodes):
if not nodes:
nodes = nuke.selectedNodes()
for n in nodes:
try:
n.deepSampleCount(0,0)
ret = True
except(ValueError):
ret = False
break
return ret
## Returns true if all selected nodes are 3D nodes
def selectedNodes3d(nodes):
if not nodes:
nodes = nuke.selectedNodes()
for n in nodes:
try:
## Need to find better detection method for 3D nodes, this
method doesnt include Lights
n['render_mode'].getValue()
ret = True
except(ValueError):
ret = False
break
return ret
nuke.menu('Nodes').addCommand('Transform/Transform',
'contextAwareCreateNode(node2d=\'Transform\', node3d=\'TransformGeo\',
nodeDeep=\'DeepTransform\')', 't', icon="2D.png")
nuke.menu('Nodes').addCommand('Merge/Merge','contextAwareCreateNode(node2d=\'Merge\',
node3d=\'Scene\', nodeDeep=\'DeepMerge\')', 'm', icon="Merge.png")
nuke.menu('Nodes').addCommand('Color/ColorCorrect',
'contextAwareCreateNode(node2d=\'ColorCorrect\',
nodeDeep=\'DeepColorCorrect\')', 'c', icon="ColorCorrect.png")
nuke.menu('Nodes').addCommand('Image/Write',
'contextAwareCreateNode(node2d=\'Write\', nodeDeep=\'DeepWrite\')', 'w',
icon="Write.png")
On Fri, Jun 13, 2014 at 3:00 PM, Anthony Kramer <[email protected]>
wrote:
> Good point...
>
> For my particular purposes, I do want to know if the output of the
> selected node has deep data, which DeepToImage and DeepToPoints do not. The
> deepSampleCount method actually returns False for those nodes as well.
>
>
> On Fri, Jun 13, 2014 at 2:54 PM, Nathan Rusch <[email protected]>
> wrote:
>
>> Be careful with with the deepSampleCount approach, as it will require
>> computation of image data if it isn’t already available.
>>
>> The canSetInput test isn’t a bad idea and would work most of the time,
>> but what about DeepToImage or DeepToPoints?
>>
>> -Nathan
>>
>>
>> *From:* Anthony Kramer <[email protected]>
>> *Sent:* Friday, June 13, 2014 2:47 PM
>> *To:* Nuke Python discussion <[email protected]>
>> *Subject:* Re: [Nuke-python] Identify deep node with python
>>
>> Nice, that seems to work, thanks!
>>
>>
>> On Fri, Jun 13, 2014 at 2:42 PM, Jose Fernandez de Castro <
>> [email protected]> wrote:
>>
>>> Another hacky way (not thoroughly tested though) could be to do the
>>> following:
>>> try:
>>> nuke.selectedNode().deepSampleCount(0,0)
>>> print("Deep")
>>> except(ValueError):
>>> print("Not deep")
>>>
>>> On Fri, Jun 13, 2014 at 2:26 PM, Anthony Kramer
>>> <[email protected]> wrote:
>>> > Its hacky but heres what I came up with. It tests whether all the
>>> node(s)
>>> > you have selected can connect to a DeepTransform node, which presumably
>>> > means they are deep nodes:
>>> >
>>> >
>>> > nodes = nuke.selectedNodes()
>>> >
>>> > nukescripts.clear_selection_recursive()
>>> >
>>> > test = nuke.createNode('DeepTransform')
>>> >
>>> > for n in nodes:
>>> >
>>> > if not test.canSetInput(0, n):
>>> >
>>> > ret = False
>>> >
>>> > break
>>> >
>>> > else:
>>> >
>>> > ret = True
>>> >
>>> > nuke.delete(test)
>>> >
>>> > print ret
>>> >
>>> >
>>> >
>>> > If anyones got a better idea, I'd love to hear it.
>>> >
>>> >
>>> >
>>> >
>>> > On Fri, Jun 13, 2014 at 12:24 PM, Nathan Rusch <
>>> [email protected]>
>>> > wrote:
>>> >>
>>> >> Off the top of my head, I don’t think there’s anything incriminating
>>> about
>>> >> deep nodes other than the fact that their class names start with
>>> "Deep" (for
>>> >> now, at least).
>>> >>
>>> >> -Nathan
>>> >>
>>> >>
>>> >> From: Anthony Kramer
>>> >> Sent: Friday, June 13, 2014 11:36 AM
>>> >> To: Nuke Python discussion
>>> >> Subject: [Nuke-python] Identify deep node with python
>>> >>
>>> >> Looking for a way with python to identify if the selected node(s) as
>>> deep
>>> >> nodes.
>>> >>
>>> >> -ak
>>> >>
>>> >> ________________________________
>>> >> _______________________________________________
>>> >> Nuke-python mailing list
>>> >> [email protected], http://forums.thefoundry.co.uk/
>>> >> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>> >>
>>> >>
>>> >> _______________________________________________
>>> >> Nuke-python mailing list
>>> >> [email protected], http://forums.thefoundry.co.uk/
>>> >> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>> >>
>>> >
>>> >
>>> > _______________________________________________
>>> > Nuke-python mailing list
>>> > [email protected], http://forums.thefoundry.co.uk/
>>> > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>> >
>>>
>>>
>>>
>>> --
>>> Jose Fernandez de Castro
>>> _______________________________________________
>>> Nuke-python mailing list
>>> [email protected], http://forums.thefoundry.co.uk/
>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>>
>>
>>
>> ------------------------------
>> _______________________________________________
>> Nuke-python mailing list
>> [email protected], http://forums.thefoundry.co.uk/
>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>
>>
>> _______________________________________________
>> Nuke-python mailing list
>> [email protected], http://forums.thefoundry.co.uk/
>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>
>>
>
_______________________________________________
Nuke-python mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
