I am pleased to announce the release of AmunPy 1.0.0, the first stable version of the Python interface for reading and analyzing AMUN Code simulation snapshots.
This release represents a complete rewrite and modernization of the package, providing a clean, tested, and well-defined API for both 2D and 3D data.
Highlights of version 1.0.0
- Fully redesigned AmunXML and AmunH5 readers with strict snapshot integrity verification
- Robust support for multiple compression formats (none, lz4, zstd, lzma)
- Unified and tested binary decoding pipeline (compression and byte-level encodings)
- Consistent handling of adaptive mesh refinement (AMR) metadata
- Cross-format validation ensuring identical results between XML and HDF5 snapshots
- Comprehensive automated test suite (all tests passing)
- Modern Python packaging based on pyproject.toml (PEP 517/621)
AmunPy 1.0.0 is intended as a stable foundation for future extensions, including derived fields, advanced analysis tools, and visualization workflows.
Project links
- Source code (GitLab):
https://gitlab.com/gkowal/amunpy - Python Package Index (PyPI):
https://pypi.org/project/amunpy

Recent Comments