DictSQLite

English

日本語

English

日本語

Appearance

Changelog ​

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

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

v2.1.3 - 2026-07-04 ​

Fixed ​

  • Hardened write-buffer flushing so pending writes are restored if storage flush fails.
  • Fixed delete/clear paths that could allow pending buffered writes to reappear after removal.
  • Fixed table-name prefix handling in async deletion paths.
  • Fixed separate-table persistence so durable modes write table data directly instead of losing lazy-mode table updates.
  • Fixed compression/decompression consistency for separate-table storage.
  • Updated anyhow lockfile entry to 1.0.103 to resolve RUSTSEC-2026-0190.

Improved ​

  • Added bulk storage reads for batch cache misses to reduce SQLite round trips.
  • Reworked warm-cache byte accounting to avoid O(n) cache scans on hot read promotion.
  • Improved benchmark coverage across data sizes, batch sizes, record counts, persistence modes, storage modes, table modes, cold reads, mutation operations, and threaded access.
  • Improved benchmark graph generation with compact labels, aggregation, p95 latency views, and less label overlap.
  • Added GitHub Actions benchmark CSV/graph artifacts, benchmark-history export, and Markdown run summaries.
  • Configured GitHub benchmark comparison as informational only to avoid false alerts from runner variance.

Changed ​

  • Bumped package version to 2.1.3.

v2.1.1 - 2026-03-03 ​

Security ​

  • Updated pyo3 to v0.28.2 (Rust-Python bindings security and compatibility fixes)
  • Updated pythonize to 0.28 (aligned with pyo3 0.28)
  • Updated papaya to 0.2 (concurrent hash map security fixes)
  • Updated base64 to 0.22

Changed ​

  • Updated actions/cache to v5
  • Updated actions/checkout to v6
  • Updated actions/setup-python to v6
  • Updated google/osv-scanner-action to v2.3.3
  • Updated criterion to 0.8 (benchmarking dependency)

v2.1.0 - 2026-02-11 ​

Security ​

  • Fixed vulnerabilities in used libraries

Fixed ​

  • Fixed internal bugs

v2.0.9 - 2026-01-08 ​

Changed ​

  • Internal code organization and cleanup for improved code quality and maintainability

Note ​

  • No functional changes or improvements
  • No performance impact
  • API remains identical to v2.0.8
  • Full backward compatibility with v2.0.8

v2.0.8 - 2026-01-03 ​

Added ​

  • ARM Package Distribution: Pre-built packages for ARM64 (aarch64) and ARM32 (armv7l)
  • Windows on ARM (ARM64) package support
  • Multi-architecture CI/CD pipeline with GitHub Actions
  • Automated PyPI deployment for all architectures

Changed ​

  • PyPI package distribution now includes ARM architecture wheels
  • GitHub Actions automated build and release workflow

Improved ​

  • Installation on ARM platforms (no compilation needed)
  • Cross-platform deployment consistency

Note ​

  • Functionality and API remain identical across all architectures
  • Performance characteristics maintained from v2.0.7
  • ARM benchmark testing under consideration for future releases

v2.0.7 - 2025-12-29 ​

Added ​

  • Updated dependencies to latest versions (pyo3 0.27.2, dashmap 6.1, tokio 1.42)
  • Enhanced documentation with migration guide and examples

Changed ​

  • Improved Python-Rust data conversion performance
  • Enhanced async runtime stability

Fixed ​

  • Fixed test stability issues with pytest-v6
  • Resolved import errors in benchmark tests

v2.0.6 - 2025-12-06 ​

Added ​

  • Final optimization achieving 1.31x better overall performance than fastest version
  • LRU tracking optimization for Memory/Lazy modes

Changed ​

  • Write performance improved to 154K ops/sec (4.4x improvement)
  • Read performance: 8.42x faster than fastest version
  • Bulk insert optimization with single transaction

Performance ​

  • Read: 510,256 ops/sec (WriteThrough mode)
  • Mixed operations: 285,657 ops/sec (Lazy mode)

v2.0.5 - 2025-11-15 ​

Added ​

  • Python 3.13 compatibility preparation

Changed ​

  • Updated pyo3 to 0.26 for improved Python-Rust data conversion
  • Updated dashmap to 6.0 for better concurrent access performance
  • Updated tokio to 1.40 for enhanced async processing

Performance ​

  • Write operations: ~5% faster
  • Concurrent operations: ~12% faster

v2.0.4 - 2025-11-01 ​

Security ​

  • CRITICAL: Fixed RUSTSEC-2025-0020 (buffer overflow in pyo3 0.20.3)
  • Upgraded pyo3 to 0.24.1 with Bound API support
  • Fixed sensitive data logging in examples
  • Replaced unsafe tempfile.mktemp() with mkstemp()

Security Validation ​

  • Cargo Audit: 0 vulnerabilities (was 1)
  • CodeQL (Python): 0 alerts (was 5)
  • CodeQL (Rust): 0 alerts

v2.0.3 - 2025-10-25 ​

Fixed ​

  • CRITICAL: Resolved database lock issues in async operations
  • Implemented queue-based operation serialization
  • Fixed concurrent access conflicts

Added ​

  • Persistent ThreadPoolExecutor for stable async operations
  • NORMAL locking mode for better concurrency

Performance ​

  • 50 concurrent writes: 10,027 ops/s
  • 50 concurrent reads: 145,345 ops/s
  • Cache hit rate: 80%

v2.0.2 - 2025-10-20 ​

Added ​

  • Dictionary-compatible API methods
    • items() - Returns iterator of (key, value) tuples
    • values() - Returns iterator of all values
    • update(dict) - Bulk update from dictionary
    • pop(key, default) - Remove key and return value
    • setdefault(key, default) - Set default if key doesn't exist

Changed ​

  • Enhanced compatibility with v1.8.8 codebases
  • All persistence modes (Memory/Lazy/WriteThrough) support new methods

v2.0.1 - 2025-10-15 ​

Added ​

  • AsyncDictSQLite persistence implementation
  • LRU cache eviction for automatic memory management
  • Storage fallback functionality

Changed ​

  • Improved async processing stability
  • Enhanced cache miss handling

Fixed ​

  • Memory leak prevention with LRU eviction

v2.0.0 - 2025-10-09 ​

Added ​

  • MAJOR: Complete rewrite in Rust using PyO3 bindings
  • True Python asyncio support with awaitable methods
  • AES-256-GCM encryption (upgraded from AES-CBC)
  • Safe Pickle validation with whitelist-based verification
  • Multiple persistence modes (Memory/Lazy/WriteThrough)
  • LRU cache for efficient memory management

Performance ​

  • Async writes: 300x faster (30s → 0.1s for 1000 items)
  • Sync writes: 43x faster (29.79K → 1.30M ops/sec)
  • Batch reads: 5-10x faster

Changed ​

  • Mutex lock count reduced 100x in async operations
  • SQL transaction count reduced 100x
  • PBKDF2 key derivation with 100,000 iterations

Security ​

  • AES-256-GCM with authenticated encryption
  • SQL injection protection with parameterized queries
  • Integrity verification with GCM tags

v1.8.8 - 2025-09-20 ​

Added ​

  • Experimental JSON mode implementation
  • AI-generated documentation (Japanese and English)

Fixed ​

  • Fixed schema parameter type annotation (bool → str | None)
  • Fixed multiple Pylint warnings

v1.8.7 - 2025-09-20 ​

Fixed ​

  • PyPI package fixes
  • License file updates

v1.8.6 - 2025-09-18 ​

Security ​

  • CRITICAL: Fixed pickle deserialization vulnerability

v1.8.5 - 2025-09-18 ​

Added ​

  • AI-generated comprehensive documentation
  • Extended test coverage

v1.8.4 - 2025-09-18 ​

Changed ​

  • Code quality improvements
  • Stability enhancements

v1.8.3 - 2025-09-18 ​

Added ​

  • Experimental security updates

v1.8.2 - 2025-09-15 ​

Fixed ​

  • Python version support corrections

v1.8.1 - 2025-09-15 ​

Changed ​

  • File organization improvements

Fixed ​

  • Various bug fixes

v1.8.0 - 2025-09-13 ​

Added ​

  • MAJOR: Initial stable release

v1.7.3 - 2025-09-10 ​

Added ​

  • Pydantic integration
  • WAL mode support
  • Preparation for v2.0

v1.3.7 - 2025-08-15 ​

Added ​

  • Performance improvements
  • Feature enhancements

v1.3.3 - 2025-08-01 ​

Added ​

  • Initial practical release - Core functionality established

Legend ​

  • Added: New features
  • Changed: Changes in existing functionality
  • Deprecated: Soon-to-be removed features
  • Removed: Removed features
  • Fixed: Bug fixes
  • Security: Security-related changes
  • Performance: Performance improvements