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.

[1.0.1] - 2026-05-13¶

Critical: handle() now returns both successful and failed FileData results. Previously, successful uploads were silently dropped.
FileData.to_dict() now handles bytes (renders as <N bytes>) for JSON serialization.

Removed lru_cache from S3 client factory — clients are now created per-upload to respect credential changes.
Removed duplicate _detect_size from S3Engine — uses inherited detect_stream_size().
Removed unused FileField.file attribute.
Made Pydantic import lazy in FileModel() — no longer a hidden hard dependency.
Renamed join_s3_key → join_cloud_key (alias preserved for backward compatibility).
Renamed cloud engine factory methods to _create_client for consistency.

Store.total_size property — sum of sizes for successful uploads.
__repr__ methods on FileData, Store, and FileField.
pydantic optional extra in pyproject.toml.
S3Engine lazy export in top-level __init__.py.
Comprehensive docstrings across all public APIs.
Full test suite (175 tests) covering data structures, utilities, engines, orchestration, and integration.
MkDocs Material documentation site with GitHub Pages deployment.
Ruff formatting and linting configuration.
Coverage configuration with branch coverage reporting.
PyPI Trusted Publishing workflow for GitHub Releases.
PyPI package metadata, badges, and release validation checks.

Initial release with LocalStorage, MemoryStorage, S3Storage, GCSStorage, and AzureStorage.
File validation for size, extension, and content type.
Sync/async callbacks for filenames, destinations, filters, and metadata.
Multi-field upload support with per-field configuration.
Atomic local writes with collision handling.
FileModel() helper for generating Pydantic upload models.