Sim-Database

[SeeChen]

Goal

Design and implement a front-end simulated backend data engine for SeeChen Website.

Planned direction

This issue is intentionally planned after Version A - Amber. Amber will keep the data layer simple and focus on static JSON resources, pagination, and page-level integration.

The future Sim-Database should demonstrate a backend-like data access chain in a pure front-end environment:

1. API facade: pages call an API-like interface instead of reading files directly.
2. Service / merge layer: combines multiple logical data sources for page needs.
3. Logical query layer: describes queries without knowing actual JSON / CSV file paths.
4. Compiler / planner layer: maps logical queries such as select / filter / cursor into static shard access plans.
5. Storage adapter layer: reads JSON / CSV shards and returns normalized results.

Initial scope for a future version

• Support JSON first.
• Support one showcase collection before expanding.
• Support filter, sort, limit, cursor, and projection.
• Support static shard routing through manifest / index files.
• Keep the same repository-facing API so it can later be replaced by a real backend.

Out of scope for Version A - Amber

• Full SQL parser.
• Full MongoDB query compatibility.
• CSV + JSON synchronization.
• Complex indexing engine.
• Using Sim-Database as a required runtime dependency for normal pages.