PyMultiWFN

PyMultiWFN is a modernization of the Multiwfn wavefunction analysis program, rewritten in Python for easier extension and packaging. The MVP keeps the core data structures, FCHK parsing, and density evaluation while avoiding any compilation step so it can ship as a pure-Python wheel to TestPyPI.

What’s in the MVP

• Pure-Python package, pip install-able without compilation.
• FchkLoader to read Gaussian .fchk files into a Wavefunction object.
• Vectorized Gaussian basis evaluation and electron-density calculator.
• Minimal CLI (pymultiwfn path/to/file.fchk) for quick inspection.
• Optional Fortran wrappers documented in pymultiwfn/math/fortran (not required for install).

Install (TestPyPI)

Publish a build to TestPyPI, then install:

python -m pip install --upgrade pip build twine
python -m build
python -m twine upload --repository testpypi dist/*

# install from TestPyPI (replace VERSION)
python -m pip install -U --extra-index-url https://test.pypi.org/simple pymultiwfn==0.1.1

Quick start

import numpy as np
from pymultiwfn.io.loader import load_wavefunction
from pymultiwfn.math.density import calc_density

wfn = load_wavefunction("molecule.fchk")
points = np.array([[0.0, 0.0, 0.0]])  # Bohr
density = calc_density(wfn, points)
print("ρ(0,0,0) =", density[0])

Command line:

pymultiwfn molecule.fchk   # prints header info and a density sanity check

Package layout

pymultiwfn/
├── core/         # Wavefunction, Atom/Shell containers, definitions & constants
├── io/           # Parsers (currently .fchk)
├── math/         # Basis and density evaluation; optional f2py stubs
├── analysis/     # Stubs for bonding/density/orbital analyses (extensible)
├── vis/          # GUI/plotting placeholders
├── utils/        # Helpers
└── config.py     # Runtime configuration singleton

Development tips

• Keep changes additive; use NumPy vectorization for heavy math.
• Use pymultiwfn/math/fortran/lebedev.pyf if you need compiled grids, but it is optional for wheels.
• consistency_verifier/ can be used to compare outputs with the original Multiwfn Fortran code.

License & citation

• Code: MIT License (see LICENSE).
• Please cite the original Multiwfn papers when results rely on its algorithms:
  Tian Lu, Feiwu Chen, J. Comput. Chem. 33, 580–592 (2012)
  Tian Lu, J. Chem. Phys. 161, 082503 (2024)