fire_api_local 1.0.0

A pure Dart SQLite-backed fire_api implementation for local Firestore-style documents, queries, streams, storage, and vector search.

1.0.0

• Initial local Firestore-compatible fire_api adapter.
• Added pure Dart SQLite storage through package:sqlite3.
• Added document CRUD, collection queries, streams, atomic updates, field transforms, vector value persistence, and exact nearest-vector queries.
• Added local parity coverage for query operators, cursors, root prefixes, subcollections, batched deletes, stream change types, and vector thresholds.
• Added persisted scalar field index rows for equality-style query candidate narrowing.
• Added indexed candidate narrowing for numeric/string range filters and indexed candidate ordering for scalar orderBy queries.
• Split the local backend into focused storage, query, index, vector, stream, cursor, codec, and aggregation implementation files.
• Added exact indexed count() and sum(field) aggregation paths for simple query shapes, with separate scalar and array index rows to preserve Firestore query semantics.
• Added local collection stream removed change events with the last known document payload.
• Added composite index planning that intersects multiple equality, array, and range index candidates before loading and validating documents.
• Added persisted snapshot change-log rows for local document writes and deletes.
• Updated collection streams to suppress no-op query snapshot emissions when unrelated writes do not affect the result set.
• Added resumable local snapshot change replay through changesSince(...) and streamChangesSince(...).
• Added cross-handle snapshot notifications for file-backed databases using the persisted change log.
• Added local value encoding for DateTime, bytes, LocalGeoPoint, and DocumentReference, including nested map/list values.
• Fixed byte blob equality indexing so Uint8List fields can be queried without being mistaken for array fields.
• Added a persisted vector side table with dimension and magnitude metadata for exact vector query candidate acceleration.
• Added Firestore-like query validation for inequality/order and unsupported mixed operators.
• Added a local filesystem FireStorage adapter.
• Added benchmark harness coverage for large document and vector workloads.
• Split local adapter tests into focused core, query, index, stream, vector, and storage suites with additional edge-case coverage.