Re: [Nuke-python] more views than actual views

Thu, 31 Mar 2011 00:24:14 -0700

Thanks again. 

In this case animated only knobs are all I need but thanks for both methods. Puts my python to shame!

Howard

On 31 Mar 2011, at 07:34, Nathan Rusch <[email protected]> wrote:

True, that works if your knob is animated...
 
What happens when it’s split but just set to static values? Or a mixture of static and animated components?
 
-Nathan

 
Sent: Wednesday, March 30, 2011 3:27 PM
Subject: Re: [Nuke-python] more views than actual views
 
No need to parse Knob::toScript.  Just iterator over the knobs Animation objects.
 
for v in nuke.views():
    print theKnob.animations(view=v)[0].view()
 
On Wed, Mar 30, 2011 at 3:00 PM, Nathan Rusch <[email protected]> wrote:
I recently posted a little info in this vein on VFXNation (http://www.vfxnation.com/showthread.php/252-Multiview-setExpression%28%29?p=2196#post2196), as I found myself dealing with a similar case recently.
 
Basically the only reliable way to deal with split knobs (to my knowledge) for the time being is to parse the knob’s .toScript() string. The code I wrote is for seeing if a given node name is present in any _expression_ links on a knob, and if so, at which indices and views. Needless to say, it gets pretty indented

Attachment: Open-mouthed smile
Description: Binary data

.
 
The rough code to check how many view curves exist for a knob is something like this:
 
------------------------------------------
# Assumes the knob in question is assigned to ‘knob’
import re
fullViewsString = r’default\s\{|’ + ‘|’.join([r‘%s\s\{’ % v for v in nuke.views()])
viewsRE = re.compile(‘(%s)’ % fullViewsString)
knobViews = [match.rstrip(‘ {‘) for match in viewsRE.findall(knob.toScript())]
------------------------------------------
 
‘knobViews’ will be a nice list of all the view curves that exist on that knob. Just ignore the ‘default’ and you’ve got all the split views.
 
Hope this helps.
 
-Nathan

 
Sent: Wednesday, March 30, 2011 2:33 PM
Subject: Re: [Nuke-python] more views than actual views
 
Thanks Nathan (and other Nathan)

One of the other things that is odd is that if you unsplit right it changes to default rather than take the right's values which is counter intuitive.
I guess the question is if any body bothers splitting to this level. I'm coding for it but it means things running more often to catch it.

Is there a way of checking if a knob has been split and to how far?

H
 
From: Nathan Rusch <[email protected]>
To: Nuke Python discussion <[email protected]>
Sent: Wed, 30 March, 2011 16:42:41
Subject: Re: [Nuke-python] more views than actual views

Nuke always keeps a ‘default’ knob curve around when you start splitting knobs. So if you split off your ‘left’ view, you’re creating curves for ‘left’ and ‘default,’ which, as you can imagine, everything else falls through to. Once you split off ‘right,’ you’re up to 3 curves.
 
Why they let you split a knob into more components than there are views in your script is a great question... I imagine there was a reason for it when they designed it that way, but I have no idea if that reason is still valid or applicable.
 
-Nathan

 
Sent: Wednesday, March 30, 2011 12:56 AM
Subject: [Nuke-python] more views than actual views
 
Dear All.

Does anybody know why it is possible to end up with more curves per view than there are views?

That is create a stereo project

split off a knob to left right,
you get 2 curves, knobs etc

knob.le
knob

split it again and you get
knob.le
knob.re
knob.*

the last one is an extra knob beyond the amount of views you have.

I can understand from a code pov why when you split knobs in the first place you get
knob.le
knob
  and not
knob.le
knob.re

but I cant understand the need for the extra view.

Is this a bug or clever behaviour.

It means btw if you want to know how many views are in a project you have to add 1 to the result.
ie len(nuke.views()) is potentially 1 short unless its safe to ignore the extra one.

Howard

_______________________________________________
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
_______________________________________________
Nuke-python mailing list
[email protected]
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python

Reply via email to