GVU

Python automation for GaussView 6 on macOS. Load and compare structures, inspect native menus and enabled actions, align models, control animations, map density/ESP surfaces and arrange windows through one shared API.

GVU requires a separately installed GaussView application. The tested version is GaussView 6.0.16. Accessibility and Screen Recording permissions are required for the application invoking Python.

Install

uv add gvu

For a standalone viewer script, let uv install the published package automatically:

# /// script
# requires-python = ">=3.11"
# dependencies = ["gvu>=0.2.0,<0.3"]
# ///
from pathlib import Path
from gvu import Alignment, Workflow, view

if __name__ == "__main__":
    view(Workflow(
        source=Path(__file__).resolve().parent / "molecule.gjf",
        alignment=Alignment("X", "Y"),
    ))

Run it with uv run viewer.py. Use --check for data preflight, --new to request fresh views and --verbose for diagnostic JSON Lines events.

Documentation

Start with the documentation or browse its source index:

• Getting started for installation and viewer scripts.
• Viewing workflows for comparisons, alignment, playback and surfaces.
• Menus and controls and the menu reference for native actions and their state.
• API reference for public classes and signatures.
• Coverage and troubleshooting for tested behavior and current limitations.
• Development for checks, documentation builds and releases.

Menu discovery does not establish complete automation of every dialog. Unavailable or ambiguous actions must be reported explicitly; preference fields, arbitrary context menus and some specialist controls remain incomplete or unverified.

Develop

git clone https://github.com/ghovax/gvu.git
cd gvu
uv sync --extra docs
uv run python -m unittest discover -s tests
uv run sphinx-build -W --keep-going -b html docs docs/_build/html

The documentation uses Sphinx, MyST and the Read the Docs theme. Open docs/_build/html/index.html after building. GitHub Pages publishes the site, and .readthedocs.yaml also supports importing this repository into Read the Docs.

Released wheels and source distributions contain the Python library, not the GaussView application or Gaussian calculation data. GVU is distributed under the MIT license.