Changelog

v1.2.0 - January 26, 2026

🐍 Python Version Update

Breaking Changes:

Improvements:

Fixed unclosed file handles in test suite
Added .venv and venv to .gitignore and black exclude patterns
Pinned black to 25.x for consistent formatting across all supported Python versions
Updated download count to 150,000+

📈 Documentation Enhancement

Documentation Improvements:

🚀 Major Performance and Bug Fix Release

New Features:

Optimized insert_multiple Performance: Added configurable batched processing with batch_size parameter (default: 1,000)
- Reduces fsync operations from O(n) to O(n/batch_size) for significant performance improvements
- Memory-efficient iterator-based processing supports large datasets and generators
- Maintains correct index state management during batch operations
- Addresses issue #54: “Optimize db.insert_multiple(points) Execution Time”
Enhanced Data Validation: Added comprehensive batch_size validation with proper error handling
- batch_size must be >= 1, raises ValueError with descriptive message for invalid values
- Consistent validation across both TinyFlux.insert_multiple() and Measurement.insert_multiple()

Bug Fixes:

Fixed Critical Measurement Filtering Bug: Resolved issue where get_field_values() returned incorrect results
- Index path was returning field values from all measurements instead of filtering by specified measurement
- Now ensures perfect parity between storage path and index path behaviors across all get_* methods
- Fixes issue #58: “get_field_values not working correctly”

Testing & Quality:

Achieved 100% test coverage with 175 passing tests
Added comprehensive test suites for batching, validation, and measurement filtering
Verified consistency across all storage types (Memory, CSV) and access patterns
Enhanced test fixtures to reduce code duplication and improve maintainability

Documentation:

Updated Sphinx documentation with detailed batch_size parameter usage and performance considerations
Added examples demonstrating optimal batch sizes for different use cases

Updated Python support: Removed Python 3.7, added Python 3.13 support
Updated CI/CD workflows to test against Python 3.9-3.13
Updated GitHub Actions to latest versions for better security and performance
Updated documentation and package metadata for new Python version requirements
Fixed decorator bug in temp_storage_op that was introduced in decorator refactoring
Updated download count to reflect 150,000+ downloads

Tags and Fields can be removed from individual points. See the documentation for more (resolves issue #27).

Fixed bug that allowed user to delete key/field tags with .update() and .update_all(). (resolves issue #36).

Tag and field keys can be compacted when using CSVStorage, saving potentially many bytes per Point (resolves issue #32).
Fixed bug that causes tag values of ‘’ to be serialized as “_none” (resolves issue #33).

Fix bug that caused values of 0.0 to be serialized as None/null rather than “0.0”.