Tom is right. Using a set is an easy way to prevent the returned collection
from containing duplicates (which can get out of hand in recursive tree-walking
searches if your script branches and re-joins several times). It also prevents
you from having to de-dupe your resulting list at the end or do membership
testing inside your search function.
However, to call myself out on some short-sightedness, there’s no reason to
create a new set every time; set.update makes things a little clearer (and less
verbose).
def recursiveFindReads(node=None, group=nuke.root(), reads=None):
if reads is None:
reads = set()
if node is None:
node = nuke.selectedNode() # Let this error on purpose if no node
selected
for n in node.dependencies(nuke.INPUTS | nuke.HIDDEN_INPUTS):
cls = n.Class()
if cls == 'Read':
reads.add(n)
elif cls == 'Group':
reads.update(recursiveFindReads(node=nuke.allNodes('Output',
group=n)[0], group=n, reads=reads))
else:
reads.update(recursiveFindReads(node=n, group=group, reads=reads))
return reads
-Nathan
From: Tom Ward
Sent: Friday, April 27, 2012 4:39 AM
To: Nuke Python discussion
Subject: Re: [Nuke-python] Re: selecting a node from its name
>From what I can see it's used to create a new set of unique items (correct me
>if I'm wrong!)
for example:
listA = set(["a", "b", "c"])
listB = set(["b", "c", "d"])
listC = listA.union(listB)
print listC
prints:
# Result:
set(['a', 'c', 'b', 'd'])
see http://docs.python.org/library/stdtypes.html#set
Tom
On 27/04/2012 12:08, Howard Jones wrote:
Out of interest ...
what does 'union' do in this script
Howard
----------------------------------------------------------------------------
From: Nathan Rusch mailto:[email protected]
To: [email protected]
Sent: Friday, 27 April 2012, 3:42
Subject: Re: [Nuke-python] Re: selecting a node from its name
Same function with one line cleaned up (sorry, it’s been a long day...)
def recursiveFindReads(node=None, group=nuke.root(), reads=None):
if reads is None:
reads = set()
if node is None:
node = nuke.selectedNode() # Let this error on purpose if no node
selected
for n in node.dependencies(nuke.INPUTS | nuke.HIDDEN_INPUTS):
cls = n.Class()
if cls == 'Read':
reads.add(n)
elif cls == 'Group':
reads =
reads.union(recursiveFindReads(node=nuke.allNodes('Output', group=n)[0],
group=n, reads=reads))
else:
reads = reads.union(recursiveFindReads(node=n, group=group,
reads=reads))
return reads
If you really DO need to select them, just run this:
nukescripts.clear_selection_recursive()
[n.setSelected(True) for n in recursiveFindReads()]
-Nathan
From: invisfx
Sent: Thursday, April 26, 2012 7:35 PM
To: [email protected]
Subject: [Nuke-python] Re: selecting a node from its name
what I meant by loop is
depNodes = selectedNode.dependencies()
only selects the immediate dependencies, not all the way up the graph. So I
have to re select those nodes and do it again.
or am I wrong?
----------------------------------------------------------------------------
_______________________________________________
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
_______________________________________________
Nuke-python mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
