Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[2.1.1] - 2025-05-28

exclude parameter for ConfigurationSpace.random_unique_configurations() to exclude previously generated configurations
exclude_file and output_file parameters for random_unique_structure_substitutions() to support batch generation workflows
Configuration.to_dict() and Configuration.from_dict() methods for serialisation
save_configurations() and load_configurations() utility functions for JSON file I/O
Documentation for batch generation workflow in random_sampling.ipynb

Varying composition enumeration via unique_configurations_by_composition method
generate_partitions utility for integer partitioning
compute_mapping_vectors for species permutation mapping
satisfies_bounds for occupancy constraint validation
Comprehensive progress tracking with nested progress bars
Verbose output options for composition enumeration
unique_structure_substitutions_by_composition in pymatgen interface
Multi-level disorder enumeration documentation and examples
Complete documentation restructure with separate Theory and User Guide sections
Type hints throughout the codebase with mypy static type checking in CI

BREAKING: Minimum Python version raised to 3.10
BREAKING: Removed ColourOperation class (moved to feature/colour-operations branch)
Optimised Configuration storage using np.int8 arrays
Implemented batched symmetry operations with cached index mappings
Migrated from individual loops to vectorised NumPy operations
Updated CI to GitHub Actions with Python 3.10-3.14 support
CI now uses pytest as test runner (unittest test suite maintained)
Converted narrative documentation from reStructuredText to Markdown
Species exchange symmetry optimisation reduces analyses by 40-50%

Substantially improved configuration enumeration performance
Example: 2×2×2 supercell of TiOF₂ shows 4.7× speedup (35.6s → 7.6s) with 13% memory overhead
Vectorised operations and optimised data types throughout