Skip to content

Latest commit

 

History

History
171 lines (147 loc) · 8.91 KB

CHANGELOG.md

File metadata and controls

171 lines (147 loc) · 8.91 KB

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.

[0.4.4] - 2026-01-07

Fixed

  • AKN4EU Parser Article ID Extraction: Fixed bug where AKN4EUParser was not correctly extracting xml:id attributes for articles. The AKNArticleExtractor now properly receives the id_attr='{http://www.w3.org/XML/1998/namespace}id' parameter for AKN4EU documents.

Added

  • Chained Intro Extraction for AKN4EU: New extract_content_with_chained_intro method in AKNArticleExtractor that properly combines list intro text (subparagraph) with all its points into a single paragraph entry, preserving the logical structure of legal provisions.

[0.4.3] - 2026-01-05

Fixed

  • Formex Article eID Parsing: Fixed bug in FormexArticleStrategy where articles with IDs starting with 3 (e.g., Article 300, 333) were incorrectly parsed. The lstrip('3') method was stripping all leading 3 characters instead of just the legacy prefix. Now correctly handles both normal 3-digit article numbers (300, 333, etc.) and legacy 4-digit prefixed IDs (3001 -> 001).

[0.4.2] - 2026-01-04

Changed

  • Released version 0.4.2 with bug fixes and improvements for cellar standard parsing and proposal parsing.

[0.4.1] - 2024-12-22

Added

  • End-to-End Test Suite: Comprehensive E2E tests for all clients and parsers:
    • EU Cellar client tests with multiple formats and error handling
    • Member state client tests (Finland, France, Germany, Italy, Luxembourg, Portugal)
    • Parser E2E tests for Cellar HTML, Veneto, Legifrance JSON, AKN, Finlex XML, Formex, German legislation, Italy Normattiva, and Luxembourg
  • GitHub Actions CI/CD: Automated testing workflow with coverage reporting:
    • Unit test execution with Poetry
    • Coverage badge generation and automatic updates
    • Codecov integration for detailed coverage reports
  • Client Module Enhancements:
    • New regional client: VenetoClient for Italian regional legislation
    • Enhanced Germany client with comprehensive RIS integration
    • New Portugal ELI portal client
    • Unit tests for Legilux, Malta, Normattiva, and other state clients
  • Coverage Tracking: Automated coverage badge generation from coverage.xml

Changed

  • Package Restructuring: Major reorganization for clarity and maintainability:
    • Renamed tulit.parsers → tulit.parser (singular for consistency)
    • Reorganized client modules by jurisdiction:
      • tulit.client.eu/ for EU-level clients (Cellar)
      • tulit.client.state/ for national clients (Finlex, Legifrance, etc.)
      • tulit.client.regional/ for regional clients (Veneto)
    • Reorganized test structure:
      • tests/unit/ for unit tests
      • tests/e2e/ for end-to-end tests
      • Organized by module type (client, parser)
  • Test Organization: Comprehensive test restructuring:
    • Split tests into unit and E2E categories
    • Organized client tests by jurisdiction (eu, state, regional)
    • Enhanced test fixtures with shared conftest configurations
    • Improved file path handling using locate_data_dir for portability

Improved

  • Test Coverage: Significantly expanded test suite:
    • Added unit tests for all major parser classes
    • Enhanced coverage for Akoma Ntoso parsers (AKN4EU, German LegalDocML, Luxembourg)
    • Comprehensive tests for HTML parsers (Cellar variants, Veneto)
    • Formex4Parser tests with additional assertions
    • XML parser helper tests
  • Code Quality: Enhanced maintainability and structure:
    • Better separation of concerns in client modules
    • Improved error handling in parser implementations
    • Enhanced documentation and inline comments
    • Consistent import patterns across modules

Fixed

  • Parser Implementations:
    • Removed base_dir parameter from xml.py parser
    • Enhanced Akoma Ntoso parser for article and section validation
    • Improved Luxembourg AKN parser functionality
    • Fixed RelaxNG schema structure in XML validation tests

Removed

  • Deprecated Scripts: Cleaned up obsolete test and migration scripts:
    • Removed scripts/run_all_clients.py and scripts/run_all_parsers.py
    • Removed old migration and structure analysis reports
    • Removed obsolete HTML parsing scripts
    • Cleaned up deprecated test files for better codebase clarity

Technical Details

  • CI/CD Pipeline: Comprehensive GitHub Actions workflow for automated testing
  • Test Framework: pytest-based test suite with enhanced fixtures
  • Coverage Reporting: Integrated coverage tracking with badge generation
  • Module Structure: Clear separation between unit and E2E tests
  • Backward Compatibility: Import aliases maintained for smooth migration

Migration Notes

  • Update imports from tulit.parsers to tulit.parser (backward compatible)
  • Client imports now organized by jurisdiction:
    • from tulit.client.eu.cellar import CellarClient
    • from tulit.client.state.finlex import FinlexClient
    • from tulit.client.regional.veneto import VenetoClient

0.4.0 - 2025-12-10

Added

  • Parser Registry System: New ParserRegistry class for dynamic parser selection and management
  • Domain Models: Type-safe domain objects (Article, Citation, Recital, Chapter, ArticleChild) using dataclasses
  • Text Normalization Strategies: Comprehensive normalization system with composable strategies:
    • WhitespaceNormalizer for whitespace handling
    • UnicodeNormalizer for character normalization
    • PatternReplacementNormalizer for custom replacements
    • CompositeNormalizer for combining strategies
  • Article Extraction Strategies: Strategy pattern implementation for parser-specific article extraction
  • Custom Exception Hierarchy: Granular error handling with ParserError, ParseError, ValidationError, ExtractionError, and FileLoadError
  • XML Utilities Package: Centralized XMLNodeExtractor and XMLValidator classes
  • Akoma Ntoso Package: Modular structure for Akoma Ntoso parsers:
    • AkomaNtosoParser base class
    • AKN4EUParser for EU documents
    • GermanLegalDocMLParser for German documents
    • LuxembourgAKNParser for Luxembourg documents
    • Factory functions for automatic format detection
  • Cellar HTML Parsers Package: Organized EU Cellar parsers into dedicated package:
    • CellarHTMLParser for semantic XHTML
    • CellarStandardHTMLParser for simple HTML structure
    • ProposalHTMLParser for EU legislative proposals
  • Comprehensive Documentation: New architecture guide and updated API documentation
  • Sphinx Documentation: Integrated documentation build system

Changed

  • Major Refactoring: Reorganized codebase following SOLID principles and design patterns
  • Module Organization: Split large monolithic files into focused modules:
    • parser.py: Reduced from 825 to 315 lines
    • xml.py: Reduced from 907 to 663 lines
    • akomantoso.py: Split into 7 focused modules
    • HTML parsers organized into cellar/ package
  • Import Paths: Updated structure with backward-compatible re-exports:
    • tulit.parsers.html.html_parser replaces tulit.parsers.html.xhtml
    • tulit.parsers.html.cellar package for Cellar parsers
    • tulit.parsers.xml.akomantoso package for Akoma Ntoso variants
  • HTMLParser Base Class: Moved from xhtml.py to html_parser.py for clarity
  • Package Documentation: Complete rewrite focusing on current architecture rather than migration

Improved

  • Code Quality: Better separation of concerns and single responsibility principle
  • Maintainability: Smaller, more focused modules (average 200-300 lines)
  • Extensibility: Easy to add new parsers through registry and strategy patterns
  • Type Safety: Domain models provide IDE autocompletion and type checking
  • Error Handling: Specific exception types for better debugging
  • Testing: Maintained 100% test coverage (126 tests passing)

Technical Details

  • Design Patterns: Registry, Strategy, Factory, Template Method
  • Architecture: Clean separation between parsing logic, data models, and utilities
  • Backward Compatibility: All existing code works without changes through re-exports
  • Documentation: Comprehensive architecture guide with examples

Fixed

  • Module import paths updated throughout test suite
  • Consistent namespace handling in XML parsers
  • Improved error messages in validation failures

Deprecated

  • tulit.parsers.html.xhtml: Use tulit.parsers.html.html_parser instead (backward compatible)

0.3.2 - Previous Release

Previous Features

  • Basic parser implementations for Formex, Akoma Ntoso, and HTML formats
  • Client implementations for Cellar, Normattiva, and other legal databases
  • Initial SPARQL query support
  • Basic JSON export functionality