entity-storage

Runtime storage layer for entity instances, designed to pair with entity-graph. While entity-graph describes what exists (entities, properties, relationships), entity-storage defines how to fetch, search, create, update, and delete the actual data.

What it provides

• A contribution-based registry for entity storage implementations.
• A single service (EntityStorageService) to route CRUD operations by entity name.
• DTOs for list, pagination, and single-item responses.
• Query key helpers to support cache invalidation (e.g., React Query).
• Optional mapping helpers for aggregate fetch results.

What it is (and how it differs from usual approaches)

Most applications hardcode a repository or API client per entity, tied to a single backend and a fixed schema. This module is different: it routes data access by entityName at runtime, using contributions that can come from multiple bounded contexts. That makes it a storage platform rather than a collection of static services.

The real power comes from its link to entity-graph: the graph tells the UI what entities exist and how they relate; the storage tells the UI where to fetch and mutate the data. Together, they turn a dynamic domain model into a working dynamic UI without hardcoding endpoints or per-entity services.

Concrete outcomes: fewer per-entity repositories, consistent data access across domains, and faster delivery of baseline screens (lists, detail, edit).

Why it matters

• Separates model discovery (entity-graph) from data access (entity-storage).
• Lets multiple backends or storage strategies coexist behind one API.
• Enables dynamic UI to load and mutate entity data without hardcoding endpoints.
• Provides consistent cache keys and invalidation hooks across entities.

How it works

• Extensions contribute a storage implementation per entity name.
• The storage exposes fetchItem, search, createItem, updateItemProperty, and deleteItem.
• The service resolves the right store by entity name and forwards calls.
• Additional invalidation keys can be contributed for cross-query consistency.