This is my self-curated collection of python tools for Nuke.

I take no author credit for any of these whether I've written them or not. They've come from many different people and morphed over the years into what they are now. They've all been cleaned up to match my own PEP 8 preference for styling python code.

In general when I'm writing a tool that operates on a node, nodes, or all nodes I like to keep it open to what input it takes so alot of the rewrites involve tailoring functions in this manner. An example below:

def eat_my_nodes(nodes=[]):
  if not nodes:
    nodes=nuke.selectedNodes()
  if not nodes:
    nodes=nuke.allNodes()
  if nodes:
    for node in nodes:
      eat_str = 'Eating Node: {0}'.format(node.name())
      nuke.tprint(eat_str)
    return
  else:
    return

This way I can use this function programatically somewhere else, I can pass in a list of nodes using nuke.SelectedNodes() or some other way, or I can let it run on it's own and it will work as well.

I have another tool called MyNk that I use inside my dotnuke folder to automagically add this or other structures of python scripts to a namespaced python object. As part of that process the tool looks for the menus attribute in each function and uses that to generate my custom menus for me. The syntax is pretty straight forward. Here's an example for the above function.

__menus__ = {
  'Tools/Info/Eat Selected Nodes':  {
    'cmd': 'eat_my_nodes(nuke.selectedNodes())',
    'hotkey': '',
    'icon': ''
  },
  'Tools/Info/Eat All Read Nodes':  {
    'cmd': 'eat_my_nodes(nuke.allNodes('Read'))',
    'hotkey': '',
    'icon': ''
  }
}

So that should make sense, although unless you're using the other tool or code like it you can ignore this.