saclay-format 0.2.17

python parser and usefull script related to the saclay format

Saclay_format (v0.2.17)

This repository defines a data standard to describe parts of the inputs/outputs of mean-field theory codes in the nuclear physics and astrophysics communities. It adheres to the results of the 05/2025 and 10/2025 ESNT workshops entitled "Simulating dense matter on three-dimensional coordinate space meshes (part I and II)".

It also defines and implements two storage formats for the data:

1. A format relying on a series of custom binary files containing the raw data along with a YAML header containing metadata. This format ensures compatibility with the wdata format, requires a YAML library and is not 100% portable.

2. A HDF5 format. This requires the HDF5 library and ensures portability.

Content

• src/saclay_format — Python3 package to parse and visualize the data standard
• docs — documentation on this package and specification of the data content and formats
• examples — nuclear mean-field outputs in this format
• tests — unit tests of the package

Requirements

• numpy — Python3 package for linear algebra
• pyyaml — Python3 package for YAML parsing
• h5py — Python3 package for HDF5 parsing

Installation via PyPI

The parser is available as a PyPI package: https://pypi.org/project/saclay-format/

Install it using pip:

pip install saclay-format

This also installs several command-line scripts:

• saclay_format_ev8_to_ev1
• saclay_format_ev1_to_cr1
• saclay_format_hdf5_yaml_converter
• saclay_format_compute_densities
• saclay_format_visualisation_rho
• saclay_format_visualisation_rho_TD
• saclay_format_visualisation_ascii

Running the parser

From a Python terminal, import the package and use its read() and write() methods. Example:

import saclay_format

data = saclay_format.read('file.yml')
# Modify or inspect data...
saclay_format.write(data, 'file_out.h5')

To run the unit tests:

cd tests/
python3 test_saclay_format.py

Examples

Example structure

examples/20Ne_bcs_hdf5/
├── state.h5
├── fields.h5
└── README.md

examples/alpha_alpha_3MeV/
├── reaction_alpha_alpha_3MeV_header.yml
├── reaction_alpha_alpha_3MeV_rho_n.bin
├── reaction_alpha_alpha_3MeV_rho_p.bin
├── README_visualisation_convert.md
└── script_test.convert.sh

examples/Ca40_ev8/
├── Ca40_HF_SLy4.yaml
├── Ca40_HF_SLy4_*.dat
└── README_test.md

Conversion test

python3 ../../src/saclay_format/hdf5_yaml_converter.py -h5 Ca40_HF_SLy4.yaml
python3 ../../src/saclay_format/ev8_to_ev1.py Ca40_HF_SLy4.h5 -o Ca40_HF_SLy4_ev1.h5
python3 ../../src/saclay_format/ev1_to_cr1.py Ca40_HF_SLy4_ev1.h5 -o Ca40_HF_SLy4_cr1.h5

Visualization test

Static:

python3 ../../src/saclay_format/visualisation_rho.py reaction_alpha_alpha_3MeV_header.yml

Dynamic:

python3 ../../src/saclay_format/visualisation_rho_TD.py reaction_alpha_alpha_3MeV_header.yml

Conversion and round-trip test:

python3 ../../src/saclay_format/hdf5_yaml_converter.py -h5 reaction_alpha_alpha_3MeV_header.yml
mv reaction_alpha_alpha_3MeV_header.h5 reaction_alpha_alpha_3MeV_converted.h5
python3 ../../src/saclay_format/visualisation_rho_TD.py reaction_alpha_alpha_3MeV_converted.h5
python3 ../../src/saclay_format/hdf5_yaml_converter.py -yaml reaction_alpha_alpha_3MeV_converted.h5
python3 ../../src/saclay_format/visualisation_rho_TD.py reaction_alpha_alpha_3MeV_converted.yaml
rm reaction_alpha_alpha_3MeV_converted*

Todo

• Should we stick to a constant rank for the numpy arrays returned by the read method ?
• Should we clarify the Python dictionary returned by read ?
• Change t0 and frame to tmin and nt
• Should we stick to [ix,iy,iz...] with column-major storage or go back to the numpy native storage and a [...,iz,iy,ix] order for the indices ?

Project layout

This project follows the structure recommended by the Hitchhiker’s Guide to Python: https://docs.python-guide.org/writing/structure