High-performance 3D coordinate system library with unified differential geometry, frame algebra, and topological-physics application APIs
Project description
Coordinate System Library
High-performance 3D coordinate system and differential geometry library for Python.
Authors: Pan Guojun
Version: 8.1.0
License: MIT
DOI: https://doi.org/10.5281/zenodo.14435613
Highlights
- C++ math core: vec3, quat, coord3
- Intrinsic gradient curvature (default)
- Classical finite-difference curvature (reference)
- Analytical fast path for Sphere/Torus and surfaces that provide derivatives
- Caching for repeated samples
- Spectral geometry and U(3) frame utilities
Installation
pip install coordinate-system
Quick Start
Vectors and Frames
from coordinate_system import vec3, quat, coord3
v1 = vec3(1, 2, 3)
v2 = vec3(4, 5, 6)
dot = v1.dot(v2)
cross = v1.cross(v2)
q = quat(1.5708, vec3(0, 0, 1)) # 90 degrees around Z
rotated = q * v1
frame = coord3.from_angle(1.57, vec3(0, 0, 1))
world_pos = v1 * frame
local_pos = world_pos / frame
Curvature (Differential Geometry)
from coordinate_system import Sphere, compute_gaussian_curvature, compute_mean_curvature
sphere = Sphere(radius=1.0)
K = compute_gaussian_curvature(sphere, u=0.5, v=0.5)
H = compute_mean_curvature(sphere, u=0.5, v=0.5)
print(K, H)
Notes:
- Intrinsic method returns signed mean curvature.
- Classical method returns absolute mean curvature.
Topological Physics APIs
from coordinate_system import (
predict_dynamic_stall,
estimate_non_newtonian,
mass_from_winding,
infer_winding_from_mass,
nearest_dm_shell,
)
stall = predict_dynamic_stall(cl_max_static=1.35, delta_alpha_deg=15.0)
rheo = estimate_non_newtonian(mu_reference_pa_s=0.0035, mu_observed_pa_s=0.0052)
proton = mass_from_winding(53861)
n_e = infer_winding_from_mass(0.51099895e6)
dm = nearest_dm_shell(6.2e3)
print(stall, rheo, proton.mass_MeV, n_e, dm)
Notes:
topological_physicsnow focuses on application-oriented interfaces.- Low-level field-equation internals and raw coupling parameters are not part of the public API.
Curvature APIs
Intrinsic (default):
- compute_gaussian_curvature(surface, u, v, step_size=1e-3)
- compute_mean_curvature(surface, u, v, step_size=1e-3)
- compute_riemann_curvature(surface, u, v, step_size=1e-3)
- compute_curvature_tensor(surface, u, v, step_size=1e-3)
Classical (reference):
- gaussian_curvature_classical(surface, u, v, step_size=1e-3)
- mean_curvature_classical(surface, u, v, step_size=1e-3)
Methods
Intrinsic Gradient (default)
Computes curvature from the intrinsic frame and the gradient of the normal field. This path is usually faster and stable on smooth surfaces.
Classical Finite Differences
Uses 5-point stencils to compute first and second derivatives and then builds the fundamental forms. This is useful as a numerical reference.
Project Layout
coordinate_system/
coordinate_system.pyd/.so # C++ core (vec3, quat, coord3)
topological_physics.py # Application-level topological physics APIs
spectral_geometry.py # FourierFrame, spectral analysis
u3_frame.py # U3Frame, gauge field theory
differential_geometry.py # Surface curvature
visualization.py # 3D visualization
curve_interpolation.py # C2-continuous interpolation
topological_physics.py # Public-safe topological physics formulas
Performance Notes
Performance depends on hardware and step size. For local benchmarks:
- vec3 microbenchmarks:
bench/compare_perf.py - curvature examples:
examples/curvature_computation.py
Changelog
v8.1.0 (2026-03-04)
- Unified public physics interfaces into
topological_physics.py. - Switched to application-level APIs for dynamic-stall, non-Newtonian flow, particle mass, and DM shell scans.
- Removed low-level field-equation and raw coupling symbols from top-level public exports.
v8.0.0 (2026-03-04)
- Added
topological_physicsmodule with public-safe CFUT formula interfaces. - Added shared-parameter mass/winding APIs (
mass_from_winding_eV,winding_from_mass). - Added dynamic-stall and TME helper interfaces (
nmax_dynamic_stall,tme_conductance). - Restricted top-level export surface to avoid exposing secret field-equation entry points.
v7.1.2 (2026-02-09)
- Intrinsic curvature sampling reuse for numeric surfaces (13 position calls per point)
- Documentation updates and benchmark clarity
v7.1.1 (2026-02-05)
- Analytical curvature fast path (Sphere/Torus and surfaces with derivatives)
- Caching for curvature calculators and last-call results
- AVX2 / fast-math enabled in native build config
License
MIT License - Copyright (c) 2024-2026 Pan Guojun
Links
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file coordinate_system-8.1.0.tar.gz.
File metadata
- Download URL: coordinate_system-8.1.0.tar.gz
- Upload date:
- Size: 82.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.13.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
de91a17e597c9bc01a680564f8eb4fa80d2bf077ee3bd1c038c79ae1637dba24
|
|
| MD5 |
5e32827d50e76bb02f78d0ea730873b2
|
|
| BLAKE2b-256 |
1986b79073b1b530657fa9d65242daeab97f13a9682c3d18159501df9635ff9e
|
File details
Details for the file coordinate_system-8.1.0-cp313-cp313-win_amd64.whl.
File metadata
- Download URL: coordinate_system-8.1.0-cp313-cp313-win_amd64.whl
- Upload date:
- Size: 236.2 kB
- Tags: CPython 3.13, Windows x86-64
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.13.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
fbec1fae601b30b18534bcb0ae52f91fc1538e9fd2241066293d55317ac53b3a
|
|
| MD5 |
faf35a68143b5300be60176247d8c52a
|
|
| BLAKE2b-256 |
e014542350e25688b2b7884f63efbe64430ff63af4803771baa14be5114d45e9
|