Bernardyn

Bernardyn is a PySide6/PyQtGraph workbench for publication-oriented plotting of small-angle scattering and diffraction curves. It keeps scientific arrays out of widgets and stores complete, source-independent graphs in portable HDF5 packages.

Current capabilities

• Browse all discoverable NXcanSAS/simple-HDF5 entries and 2–4 column text data.
• Keep a workspace-wide dataset catalog and independent graph tabs.
• Plot General I(Q), Guinier, rod/sheet Guinier, Kratky, dimensionless Kratky, Porod, modified Porod, Zimm, and Debye–Bueche views.
• Style each curve, errors, labels, axes, legends, typography, and annotations.
• Render 3D line waterfalls and common-grid surfaces with an isolated, lazily-loaded PyQtGraph OpenGL backend and 2D fallback.
• Save one graph or a deduplicated multi-graph workspace as *.bernardyn.h5, including canonical data, resolved snapshots, configuration, checksums, previews, and 3D renderer data.
• Export PNG/JPEG/SVG, CSV, Igor ITX, and canonical data through PyIrena's H5XP writer. H5XP export is data interoperability, not Bernardyn graph persistence.

The native schema is documented in bernardyn/schemas/graph-package-v2.md. User-facing workflow and file-format documentation is in docs/. Scripts and automation can use the typed Python facade or bernardyn-plot CLI; see headless recipes. Local MCP-capable agents can use the five-tool bernardyn-mcp server for the same recipe workflow.

User-visible changes are recorded in changelog.md.

Install and run

Bernardyn targets Python 3.10–3.13 and shares its Qt plotting dependencies with PyIrena. For development, keep the Bernardyn and pyirena repositories next to each other and create the pinned Conda environment from the Bernardyn root:

conda env create -f environment.yml
conda activate bernardyn
bernardyn-doctor
bernardyn

The environment file installs both repositories in editable mode. If an older bernardyn environment already exists with Python 3.14, remove and recreate it:

conda deactivate
conda env remove -n bernardyn
conda env create -f environment.yml
conda activate bernardyn

For an already compatible Python 3.10–3.13 environment, installation can also be performed manually:

python -m pip install -e "../pyirena[qtplot]"
python -m pip install -e ".[dev]"
bernardyn

To diagnose an installation:

bernardyn-doctor
bernardyn-doctor --json

If an existing environment reports that the PyIrena shared API is missing with No module named 'six', repair that older environment with python -m pip install six. New Bernardyn installations install this compatibility dependency automatically.

The OpenGL renderer needs a usable graphics context. Before enabling a 3D renderer, Bernardyn checks that Qt can create the required OpenGL/GLX widget in a separate process. If that check fails (for example on a remote or headless X session), Bernardyn stays running, displays a 2D offset-waterfall fallback, and retains the 3D document configuration. bernardyn-doctor also reports whether the actual 3D context is ready.

Development

pytest
ruff check bernardyn tests
python -m build
twine check dist/*

Extension entry points are bernardyn.sources, bernardyn.transforms, and bernardyn.renderers; see bernardyn/EXTENDING.md.

Scope

Version 1.0 is a plotting and presentation application. Model fitting, reduction, destructive source modification, movies, and calibrated 2D detector workflows are intentionally outside 1.0. Existing experimental 0.1 containers are not treated as compatible graph packages.

Bernardyn is distributed under the MIT License.