In Python, a module lets you organize functions, classes, and constants into a common namespace. For example, os is a module and os.getcwd is a function inside that module, and you access the contents of a module by looking up Python attributes on it.

An HDAModule is a Python module that is associated with a particular digital asset type. It lets you store a library of Python code in one location in your asset, and you can invoke that code from parameters, event handlers, and callbacks inside that asset.

The module’s source code is stored in the Python Module section of the Scripts tab in the Type Properties dialog. For example, suppose the digit asset is an object named gear and the Python Module section contains the following:

def position():
    return (hou.frame() * 1.2, 0.0, 3.2)

def onButtonPress():
    print "you pressed the button"

def onLoaded():
    print "onLoaded section running"

Unlike regular Python modules, which you access by name, you access a digital asset’s Python module by calling hou.NodeType.hdaModule on its node type. For example, suppose you created an object-level digital asset named gear and put the above code in its Python Module section. You could then access the contents of the Python module as follows:

>>> node_type = hou.nodeType(hou.objNodeTypeCategory(), "gear")
>>> node_type.hdaModule().position()
(1.2, 0.0, 3.2)
>>> node_type.hdaModule().onButtonPress()
you pressed the button

One use for the Python module is drive parameter expressions on nodes inside the digital asset. For example, suppose /obj/gear1 is an instance of the digital asset and /obj/gear1/geo1 is a node inside the asset. You could put the following inside geo1's tx parameter expression:

hou.node("..").type().hdaModule().position()[0]

For convenience, you can also access the module from a node instance of the digital asset using hou.OpNode.hdaModule. So, you could simplify the above expression to:

hou.node("..").hdaModule().position()[0]

And since you don’t need to use the hou. prefix inside expressions, you could further simplify it to:

node("..").hdaModule().position()[0]

The following example shows how you might run code in the module from the Callback Script field of a button parameter:

hou.pwd().hdaModule().onButtonPress()

In an event handler script, such as On Loaded, you can use the kwargs dict to access the node type:

kwargs["type"].hdaModule().onLoaded()

Note that Houdini creates a local kwargs dict that’s accessible from the Python Module, too. It contains one entry with the key "type", to give you access to the hou.NodeType defined by the digital asset.

If you find that a digital asset has too much Python code to store in one module, it’s possible to create submodules. For example, if you want to create a submodule named bar, put its source code in a new digital asset section (say, "bar_PythonModule"). Then, from the Python Module section, you can write the following:

import toolutils
bar = toolutils.createModuleFromSection("bar", kwargs["type"], "bar_PythonModule")

print("Hello world!")

print "Hello world!"

bar now appears as a submodule of the main module. If, for example, the bar_PythonModule section contains:

def foo():
    return 3.2

then you could write the following from a parameter on the digital asset node:

pwd().hdaModule().bar.foo()

Note that the Python Module code is stored in a section of the digital asset named "PythonModule". For example, you can get a string containing that source code using node_type.definition().sections()["PythonModule"].contents().