v1.3.0

This release introduces pattern matching functionality, a renamed package structure, and significantly improved documentation.

• Added Pattern Matching Module: new conllu_tools.matching module for finding tokens and sentences matching complex linguistic criteria.
• Codebase Reorganization: restructured modules for better maintainability and separation of concerns
• Bug Fixes: various stability improvements and bug fixes
• Comprehensive documentation overhaul with new User Guide sections
• Expanded API Reference with detailed docstrings
• New examples section with practical use cases
• Pattern Matching guide with syntax reference

Breaking Changes:

• Package has been renamed for clarity and consistency.

Full Changelog: v1.2.0...v1.3.0

v.1.2.0

• Add nlp_utilities.validators with validate_features and validate_xpos.
• Add converters/xpos/format_xpos to normalize various XPOS formats to Perseus.
• Refactor nlp_utilities.normalizers: replace old APIs with normalize_morphology(upos, xpos, feats, feature_set, ref_features=None) (breaking change).
• Update nlp_utilities/data/dalme_features.json (expanded VerbType and Case).
• Add docs/source/api_reference/validators.rst.
• Other improvements to documentation.
• Misc. bug fixes.
• Test-suite expansion: new/updated tests covering validators, formatters, converters, and normalizers.

Note: callers of the old normalize_features/normalize_xpos must be updated to the new API.

Full Changelog: v1.1.0...v1.2.0

v1.1.0

This version:

• Adds support for LLCT XPOS formatting and conversion.
• Fixes an issue with package data files, ensuring all resources are now properly included at installation.
• Improves and clarified documentation throughout.
• Enhances normalization routines with case validation, ensuring stricter consistency across forms.

Full Changelog: v1.0.1...v1.1.0

v1.0.1

Erratum ad Editionem Primam or the “Oops, Wrong URL” Edition. Just a quick fix: the homepage link was pointing to the wrong place. The code was perfect; the metadata was… less so. No code, functionality, or documentation content has changed.

Full Changelog: v1.0.0...v1.0.1

v1.0.0

First stable release of this toolkit for working with Latin treebanks and annotated corpora. This complete re-factoring of the earlier code provides robust utilities for converting between brat standoff and CoNLL-U formats, validating CoNLL-U files against Universal Dependencies guidelines, evaluating system outputs with metrics including UAS, LAS, MLAS, and enhanced dependencies, and converting between different morphological tagsets (Perseus, ITTB, PROIEL, LLCT, DALME). The package includes extensive test coverage (153 tests, 100% passing) and complete documentation with user guides, API reference, and developer documentation now available at https://gpizzorno.github.io/latin-nlp-utilities/

Full Changelog: https://github.com/gpizzorno/latin-nlp-utilities/commits/v1.0.0