Extending PhysioMotion4D

PhysioMotion4D is an early-alpha toolkit. Prefer small, explicit extensions that match the current class boundaries over large compatibility layers.

Where to Start

Workflows: add or modify complete pipelines.
CLI wrappers: expose workflow classes for repeatable command-line use.
Utilities: add focused image, contour, transform, or USD helpers.
Experiments: keep exploratory research in experiments/. Experiments document prior and ongoing research used to define the toolkit, but they are not examples for users or developers.

Runtime Class Pattern

import logging

from physiomotion4d import PhysioMotion4DBase

class MyWorkflow(PhysioMotion4DBase):
    def __init__(self, input_file: str, log_level: int | str = logging.INFO):
        super().__init__(class_name="MyWorkflow", log_level=log_level)
        self.input_file = input_file

    def process(self) -> str:
        self.log_info("Processing %s", self.input_file)
        return self.input_file

CLI Wrapper Pattern

import argparse
import sys

from my_module import MyWorkflow

def main() -> int:
    parser = argparse.ArgumentParser(description="Run my workflow")
    parser.add_argument("--input-file", required=True)
    args = parser.parse_args()

    workflow = MyWorkflow(input_file=args.input_file)
    workflow.process()
    return 0

if __name__ == "__main__":
    sys.exit(main())

Documentation Requirements

Update docstrings for changed public methods.
Update or add examples that use the actual current API.
Regenerate docs/API_MAP.md after public API changes.
Avoid documenting planned APIs as if they are installed.

Testing Requirements

Use synthetic ITK images and small PyVista meshes where possible. Mark tests that require downloaded or manually prepared data with requires_data.

py -m pytest tests/ -m "not slow and not requires_data" -v