Architecture

flow_props follows a narrow pipeline:

1. cfd-io reads a CFD dataset
2. flow_props.profiles.extract_profiles slices wall-normal profiles at chosen stations
3. reduction modules compute derived quantities from those profiles

Module Layout

• profiles.py: wall-normal profile extraction
• bl.py: boundary-layer edge detection and integral thicknesses
• entropy_layer.py: entropy-layer thickness
• wall.py: wall shear and thermal quantities
• config.py: TOML section loading for the CLI
• cli.py: Typer-based command-line interface

Design Intent

The package is deliberately lightweight.

• it does not own CFD file I/O
• it assumes cfd-io has already constructed a dataset
• it focuses on profile-based property extraction rather than general field operations

That separation keeps flow_props easy to reason about and consistent with the rest of the workspace.