pycogk

Python package for using PicoGK and related LEAP 71 geometry libraries from Python.

This is an unofficial community library.

• Official PicoGK project (C++ / C#): https://github.com/leap71/PicoGK
• This Python package repository: https://github.com/ghedo44/PycoGK

The goal is to keep a PicoGK-like API where practical while adapting the workflow to Python, including a vedo-based viewer layer.

Library Summary

This repository currently ships six Python libraries.

Library	Primary role	Inspired by C# repo	Typical use
picogk	Runtime bridge and core geometry API	PicoGK	Lattice/voxel/mesh/field creation, IO, viewer orchestration
shape_kernel	High-level modeling toolkit	LEAP71_ShapeKernel	Base shapes, frames/splines, utility functions, composition workflows
lattice_library	Beam-lattice workflow framework	LEAP71_LatticeLibrary	Cell arrays + lattice logic + beam thickness strategy
implicit_library	TPMS/implicit construction framework	LEAP71_LatticeLibrary (Implicit Library part)	Modular implicits, coordinate transforms, split/wall logic
penrose_pattern	2D Penrose tiling primitives	LEAP71_QuasiCrystals	Aperiodic 2D tilings and subdivision logic
quasi_crystal	3D quasi-crystal building blocks	LEAP71_QuasiCrystals	Icosahedral face/tile inflation and quasi-crystal assembly

Architecture At A Glance

Practical dependency picture:

1. picogk is the runtime-facing foundation.
2. shape_kernel builds on picogk.
3. lattice_library and implicit_library build on picogk (and interoperate with each other).
4. penrose_pattern and quasi_crystal provide aperiodic tiling/quasi-crystal logic and can be combined with picogk visualization/output workflows.

Install

From PyPI:

pip install pycogk

From this repository:

pip install "git+https://github.com/ghedo44/PycoGK.git"

Import Name

The distribution name is pycogk, while imports are per-library:

import picogk
import shape_kernel
import lattice_library
import implicit_library
import penrose_pattern
import quasi_crystal

Runtime Notes

pycogk loads the PicoGK native runtime using ctypes.

Bundled runtime targets currently include:

1. win-x64
2. osx-arm64

If your platform is not bundled, set PICOGK_RUNTIME_PATH to a compatible binary.

PowerShell example:

$env:PICOGK_RUNTIME_PATH = "C:\path\to\picogk.1.7.dll"
python your_script.py

Quick Start (picogk)

from picogk import Lattice, Mesh, VedoViewer, Voxels, go

viewer = VedoViewer(title="pycogk Quickstart")

def task() -> None:
    with Lattice() as lat:
        lat.AddSphere((0.0, 0.0, 0.0), 10.0)
        with Voxels.from_lattice(lat) as vox:
            with Mesh.from_voxels(vox) as msh:
                viewer.Add(msh, nGroupID=1)
                viewer.SetViewAngles(30.0, 20.0)
                viewer.RequestUpdate()
                print("triangles:", msh.nTriangleCount())

go(0.5, task, end_on_task_completion=False, viewer=viewer)

Close the viewer window to end the program.

In-Depth: picogk

picogk is the low-level runtime API. It wraps the PicoGK native library and exposes core primitives needed by the rest of the stack.

Main capabilities include:

1. Runtime lifecycle and viewer orchestration via Library, go, and viewer interfaces.
2. Geometry primitives such as Lattice, Voxels, Mesh, and PolyLine.
3. Field workflows with ScalarField, VectorField, metadata, and field utilities.
4. IO and processing helpers for mesh, image, slice, CSV, CLI, OpenVDB, and utility workflows.

Use picogk when you need direct, explicit control over runtime objects and conversion steps.

In-Depth: shape_kernel

shape_kernel is a higher-level modeling toolkit inspired by LEAP 71 ShapeKernel concepts. In the original C# framing, ShapeKernel bridges low-level kernel operations and higher-level engineering design intent.

Main capabilities include:

1. Parametric base shapes (BaseBox, BaseCylinder, BaseSphere, BasePipe, BaseLens, and more).
2. Frame and spline infrastructure (LocalFrame, Frames, control splines, tangential splines).
3. Modulations and geometric composition helpers (LineModulation, SurfaceModulation, Sh).
4. Utility classes for measurement, mesh operations, cylindrical utilities, and color/visualization helpers.

Use shape_kernel when you want expressive modeling code with less low-level plumbing.

In-Depth: lattice_library

lattice_library ports the flexible lattice workflow from the C# Lattice Library. The core idea is to keep three concerns independent so they can be recombined:

1. Cell array generation (ICellArray and concrete arrays).
2. Lattice connection logic (ILatticeType and concrete lattice types).
3. Beam thickness strategy (IBeamThickness and concrete thickness models).

Representative classes include:

1. Cell arrays: RegularUnitCell, RegularCellArray, ConformalCellArray.
2. Lattice types: BodyCentreLattice, OctahedronLattice, RandomSplineLattice.
3. Thickness models: ConstantBeamThickness, CellBasedBeamThickness, GlobalFuncBeamThickness, BoundaryBeamThickness.

Use lattice_library when designing beam-based infill/meta-material workflows where topology and thickness rules are customized independently.

In-Depth: implicit_library

implicit_library ports the Implicit Library concepts from the C# Lattice Library project. It focuses on TPMS-style implicits and modular composition logic.

Main capabilities include:

1. Coordinate transformations (ScaleTrafo, RadialTrafo, FunctionalScaleTrafo, CombinedTrafo).
2. Splitting logic and signed-region control (FullWallLogic, FullVoidLogic, half-wall and void variants).
3. Raw TPMS patterns (RawGyroidTPMSPattern, RawLidinoidTPMSPattern, Schwarz variants, transitions).
4. Modular implicit assembly (ImplicitModular) and ready presets in TPMSPresets.

Use implicit_library when constructing procedural implicit materials with tunable wall/void behavior and spatial transforms.

In-Depth: penrose_pattern

penrose_pattern provides 2D aperiodic tiling primitives inspired by the Penrose workflow from the C# Quasi Crystals repository.

Main capabilities include:

1. Penrose pattern generation via tile subdivision logic.
2. Tile primitives such as RhombicTile, LargeRhombicTile, and SmallRhombicTile.
3. Supporting geometric entities like RobinsonTriangle.

Use penrose_pattern when you need non-periodic 2D tilings with local order and no translational symmetry.

In-Depth: quasi_crystal

quasi_crystal provides 3D quasi-crystal building blocks, aligned with the C# quasi-crystal concepts (icosahedral faces, quasi-tiles, and inflation/substitution rules).

Main capabilities include:

1. Rhombic-face representation (IcosehedralFace).
2. Quasi-tile families (QuasiTile, QuasiTile_01 to QuasiTile_04).
3. Inflation/subdivision support (QuasiTileInflation).
4. Crystal-level assembly utilities (QuasiCrystal).

Use quasi_crystal when exploring aperiodic 3D structures and rule-driven quasi-tile expansion.

Documentation

Current long-form docs are available for:

1. picogk
2. shape_kernel

picogk docs

1. Book Home
2. Chapter 1: Orientation
3. Chapter 2: Installation and Runtime
4. Chapter 3: First Project
5. Chapter 4: Core Concepts
6. Chapter 5: Modeling Workflow
7. Chapter 6: Viewer and Interaction
8. Chapter 7: Data IO
9. Chapter 8: Troubleshooting
10. Chapter 9: Production and Publishing

shape_kernel docs

1. Book Home
2. Chapter 1: Orientation
3. Chapter 2: Concept Mapping
4. Chapter 3: Frames and Splines
5. Chapter 4: Base Shapes
6. Chapter 5: Modulations and Utility Functions
7. Chapter 6: Visualization and Preview
8. Chapter 7: Measurements and Runtime Helpers
9. Chapter 8: Migration Guide

For lattice_library, implicit_library, penrose_pattern, and quasi_crystal, use source and tests as the current reference points.

Examples

Runnable examples are organized by library focus.

picogk examples

1. examples/picogk/basic_usage.py
2. examples/picogk/01_lattice_to_mesh.py
3. examples/picogk/02_voxel_boolean_filters.py
4. examples/picogk/03_scalar_vector_fields.py
5. examples/picogk/04_openvdb_io.py
6. examples/picogk/05_stl_io_and_mesh_math.py
7. examples/picogk/06_slicing_and_cli.py
8. examples/picogk/07_images_and_tga.py
9. examples/picogk/08_animation_and_csv.py
10. examples/picogk/09_polyline_and_viewer.py
11. examples/picogk/10_utils_tempfolder_log.py
12. examples/picogk/11_viewer_controls_and_timelapse.py

shape_kernel examples

1. examples/shape_kernel/example_spline.py
2. examples/shape_kernel/ex_base_box_showcase.py
3. examples/shape_kernel/ex_base_cylinder_showcase.py
4. examples/shape_kernel/ex_base_lens_showcase.py
5. examples/shape_kernel/ex_base_pipe_segment_showcase.py
6. examples/shape_kernel/ex_base_pipe_showcase.py
7. examples/shape_kernel/ex_base_ring_showcase.py
8. examples/shape_kernel/ex_base_sphere_showcase.py
9. examples/shape_kernel/ex_basic_lattices.py
10. examples/shape_kernel/ex_lattice_manifold_showcase.py
11. examples/shape_kernel/ex_lattice_pipe_showcase.py
12. examples/shape_kernel/migration_sh_preview_equivalents.py
13. examples/shape_kernel/migration_visualization_equivalents.py

Additional library references

Integration and behavior checks for additional libraries are currently covered in tests:

1. tests/lattice_library
2. tests/implicit_library

Status

pycogk is actively developed and has production-readiness gates, but it is still an unofficial project and not a drop-in official replacement.

For readiness details, see:

1. Production Readiness
2. Parity Matrix
3. Publishing Guide