Changelog¶

All notable changes to ParquetFrame are documented here.

For the complete changelog with all releases, please see our CHANGELOG.md file in the repository root.

Latest Release¶

[0.3.2] - 2025-10-14¶

🛠️ Windows Compatibility & Bug Fixes¶

This release resolves Windows CI/CD pipeline issues and enhances cross-platform compatibility.

🔧 Bug Fixes¶

Windows CI/CD Pipeline: Fixed ArrowException timezone database errors on Windows
File Permission Errors: Resolved Windows file locking issues during test cleanup
Test Suite Compatibility: 382/383 tests now pass across all platforms (99.7% success rate)
ORC Format Handling: Graceful ORC test skipping on Windows due to Arrow limitations
Temporary File Management: Robust cleanup with retry logic and garbage collection

⚡ Improvements¶

Enhanced Windows-specific error handling with platform detection
Improved file handle management with proper cleanup on Windows
Added retry logic with 3-attempt cleanup and progressive permission fixes
Windows-compatible temporary directory fixtures in test infrastructure
Better cross-platform testing support

📊 Test Coverage¶

57% overall code coverage (exceeds 40% requirement)
382/383 tests passing (99.7% success rate)
Cross-platform compatibility: Windows, macOS, and Linux

[0.3.1] - 2025-10-14¶

🔧 Test Infrastructure & Bug Fixes¶

Fixed CLI test expectations for multi-format output
Enhanced SQL test compatibility with Arrow warning messages
Improved directory handling in multi-format path resolution
Updated test assertions for new multi-format behavior

[0.3.0] - 2025-10-14¶

🎉 Major Feature Release: Multi-Format Support¶

This release transforms ParquetFrame into a comprehensive multi-format data processing framework while maintaining full backward compatibility.

✨ New Features¶

Multi-Format Support: CSV (.csv, .tsv), JSON (.json, .jsonl, .ndjson), ORC (.orc), and Parquet (.parquet, .pqt)
Automatic Format Detection: Smart detection from file extensions with manual override capability
Handler-Based Architecture: Extensible design with format-specific handlers
Enhanced CLI: Multi-format support in all CLI commands (pframe info, pframe run)
Intelligent Backend Selection: Improved file size threshold (100MB) for pandas/Dask selection
Format-Specific Features: TSV delimiter detection, JSON Lines support, ORC big data integration

🔧 Improvements¶

Enhanced error handling with graceful dependency management
Comprehensive path resolution for all file formats
Updated documentation with multi-format examples and guides
CLI format override with --format option
Improved help text and examples throughout

🧪 Testing¶

37 comprehensive tests (23 core + 14 CLI) covering all formats
47% code coverage on core modules with robust edge case handling
End-to-end multi-format workflow testing

📚 Documentation¶

Complete multi-format guide (docs/formats.md)
Updated README with comprehensive examples
Enhanced CLI documentation
Format-specific usage patterns and best practices

Previous Releases¶

[0.2.3.2] - 2025-10-13¶

Added¶

Automatic pandas/Dask backend selection based on file size
Smart file extension handling for parquet files (.parquet, .pqt)
Seamless conversion between pandas and Dask DataFrames
Full API compatibility with pandas and Dask operations
Comprehensive CLI interface
SQL query support with DuckDB
AI-powered natural language queries
Performance benchmarking tools

View Full Changelog¶

For detailed release notes, breaking changes, and migration guides, see the complete CHANGELOG.md.