Mobu Look - Motionbuilder Marker Look Serializer

Save and load Motionbuilder FBModelMarker look properties to json.

How to use

The save function dump a list of FBModelMarker to a json file:

import mobulook
markers = []  # <- List of FBModelMarker to save
mobulook.save("C:/look.json", markers)

The load function apply the content of a saved json file to markers in scene with matching names.

mobulook.load("C:/look.json")

Only the FBModelMarker.Name are saved in the json, namespaces are ommited. When loading the file, an optional prefix namespace can be applied find the correct markers in scene.

# Load "look.json" to matching markers under "Character_Ctrl" namespace
mobulook.load("C:/look.json", namespace="Character_Ctrl")

Examples

Save all FBModelMarker in hierarchy

import pyfbsdk
import mobulook

def hierarchy(root): # (pyfbsdk.FBModel) -> Generator[pyfbsdk.FBModel, None, None]
    """Iter through models in hierarchy starting from 'root'."""
    root = root or pyfbsdk.FBSystem().SceneRootModel
    stack = [root]
    while stack:
        model = stack.pop()
        yield model
        stack.extend(model.Children)

# Save all FBModelMarkers in the hierarchy of "MyRootModel"
root = pyfbsdk.FBFindModelByName("MyRootModel")
mobulook.save("C:/output.json", hierarchy(root))

Save all FBModelMarker of current character

import pyfbsdk
import mobulook

character = pyfbsdk.FBApplication().CurrentCharacter
markers = []

# Body Nodes (FK)
for node_id in pyfbsdk.FBBodyNodeId.values.values():
    model = character.GetCtrlRigModel(node_id)
    if model:
        markers.append(model)

# Effectors
for effector in pyfbsdk.FBEffectorId.values.values():
    for effector_set in pyfbsdk.FBEffectorSetID.values.values():
        model = character.GetEffectorModel(effector, effector_set)
        if model:
            markers.append(model)

# Extensions
for extension in character.CharacterExtensions
    markers.extend(extension.Components)

mobulook.save("C:/output.json", markers)

Contributing

Dev environment setup using pyenv and poetry.

Running the tests in cli

Install python 2.7 with pyenv and set it as our local env. Then we tell poetry to use this version to create a venv with the necessary dependencies (mainly pytest).

pyenv install 2.7.18
pyenv local 2.7.18
for /f "tokens=*" %a in ('pyenv exec python -c "import sys; print(sys.executable)"') do poetry env use %a
poetry install

Now that we have a venv compatible with Motionbuilder 2020 (python 2.7.11), we need to add the site-packages directory of our poetry venv to PYTHONPATH (to find pytest) and the mobulook/src directory (to find our module).

set PYTHONPATH=<PATH_TO_VENV>\Lib\site-packages;%PYTHONPATH%
set PYTHONPATH=<PATH_TO_MOBULOOK>\src;%PYTHONPATH%

Motionbuilder 2022 and above

We can now run the tests using the mobupy interpreter.

mobupy -m pytest

Motionbuilder 2020 and below

We can't use the mobupy interpreter of Motionbuilder 2020, it crashes as soon as you use pyfbsdk.

Instead we can use the main application to run our tests.

Create a python runner script. Note that the path to our test file has to be an absolute path, the __file__ variable will be undefined in our case.

# runner.py
import pytest
pytest.main(["<PATH_TO_MOBULOOK>/test_mobulook.py", "--capture=sys"])

And execute this runner script from Motionbuilder CLI

motionbuilder.exe <PATH_TO_RUNNER_SCRIPT>

Check the Python Editor Window of Motionbuilder for the results.

Running the tests in VSCode

This only work with Motionbuilder >= 2022.

Follow the first step of the section above to install a python venv. But instead of adding to PYTHONPATH in a shell, we can create an .env file at the root of our package directory and reference it in the project settings:

mobulook/.env:

PYTHONPATH=<PATH_TO_VENV>\Lib\site-packages;${PYTHONPATH}
PYTHONPATH=<PATH_TO_MOBULOOK>\src;${PYTHONPATH}

mobulook/.vscode/settings.json:

{
    "python.defaultInterpreterPath": "<PATH_TO_MOBUPY>",
    "python.testing.unittestEnabled": false,
    "python.testing.pytestEnabled": true,
    "python.envFile": "${workspaceFolder}/.env",
}

Now you can run the tests in the Testing tab of VSCode.

tahv/mobulook