Skip to content

CHANGELOG.md

Latest commit

 

History

History
42 lines (27 loc) · 4.49 KB

CHANGELOG.md

File metadata and controls

42 lines (27 loc) · 4.49 KB
Raw

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.

Unreleased

Added

  • GitHub Actions CI (.github/workflows/ci.yml) — automated test pipeline triggered on every push and pull request. Runs a matrix across Python 3.11, 3.12, and 3.13; installs dev dependencies, lints with ruff, type-checks with mypy, and executes the full pytest suite with --cov coverage reporting. fail-fast: false ensures all matrix legs run to completion.

Fixed

  • Ruff linting violations — resolved all issues reported by ruff check on first CI run: sorted __all__ alphabetically in __init__.py (RUF022); replaced timezone.utc with the UTC alias throughout source and tests (UP017); replaced Decimal("0") string literals with Decimal(0) in bank_account.py (FURB157); updated SQLiteEventStore.__enter__ return annotation to Self (PYI034).

0.1.0 - 2026-08-23

Added

  • EventStore ABC (store.py) — backend-agnostic interface declaring append, load, save_snapshot, and load_snapshot with full PEP 484 type annotations.
  • SQLiteEventStore (sqlite_store.py) — fully ACID-compliant concrete implementation using the standard-library sqlite3 module. Enforces per-aggregate monotonically increasing version numbers via a UNIQUE(aggregate_id, version) constraint; translates IntegrityError to OptimisticConcurrencyError.
  • AggregateRoot (aggregate.py) — base class with integer version tracking, pending_events collection, record() / apply() / clear_pending_events() lifecycle methods, on_<EventType> handler dispatch convention, and snapshot_state() / _restore_from_snapshot() hooks.
  • Repository (repository.py) — orchestration layer that loads the latest snapshot, fetches subsequent events, calls AggregateRoot.rehydrate(), auto-triggers snapshot saves when the post-snapshot event count reaches the configurable snapshot_threshold (default 50), and persists pending events via save().
  • StoredEvent and SnapshotRecord (domain.py) — immutable frozen=True dataclasses carrying aggregate_id, version, event_type/state, payload, and a UTC occurred_at/taken_at timestamp.
  • Exception hierarchy (exceptions.py) — EventSourcingError base; OptimisticConcurrencyError, AggregateNotFoundError, UnknownEventTypeError, InsufficientFundsError, AccountClosedError concrete exceptions, all with human-readable messages identifying the aggregate_id and, where applicable, conflicting version numbers.
  • BankAccount worked example (examples/bank_account.py) — concrete aggregate demonstrating AccountOpened, MoneyDeposited, MoneyWithdrawn, and AccountClosed events; enforces InsufficientFundsError and AccountClosedError guard conditions; implements snapshot_state() / _restore_from_snapshot() for snapshot-aware rehydration.
  • Full pytest test suite — 69 tests across test_sqlite_store.py, test_aggregate.py, test_rehydration.py, test_domain.py, and test_bank_account.py; 100% line coverage of the event_sourcing package.
  • pyproject.toml packaging — PEP 517/518 build configuration via Hatchling; dev extras for pytest and pytest-cov; src/ layout with event_sourcing as the sole wheel package; requires-python = ">=3.11".

Architecture

  • SQLite (stdlib sqlite3) was chosen as the sole backend for this release to provide zero-infrastructure, fully ACID, CI/CD-friendly event storage. The EventStore interface is designed so a PostgreSQL or other server-backed implementation can be substituted without changing application code. See docs/requirements/SRS.md §4 for the full trade-off analysis.

Documentation

  • docs/requirements/SRS.md — Software Requirements Specification covering functional requirements (FR-01 – FR-25), non-functional requirements, backend trade-off analysis, out-of-scope items, glossary, and acceptance criteria.
  • docs/design/DESIGN.md — Architecture and design document covering the three-layer architecture, module responsibilities, key interfaces (with full type-hinted pseudocode), SQLite schema, rehydration algorithm, snapshot strategy, and concurrency model.