Saltar a contenido

🛠️ Core Infrastructure

Status: Implemented (Feb 2026)

The Core Infrastructure category contains the foundational classes, wrappers, and utilities used to build the rest of the Store Client System. These modules do not contain business data themselves; rather, they provide standard behaviors (CRUD, History, Draft editing) that other stores inherit or use.

🗂️ Stores

All these modules are located in src/lib/stores/core/.

Module Purpose
entityStore Base generic class factory (createEntityStore) for CRUD operations. Used to create all Reference State stores (brokerStore, assetStore, etc.). Handles normalization, caching, and optimistic merging.
EditBuffer Bidirectional edit buffer class for pending time-series modifications. Provides sync between chart click-to-edit, CSV textareas, and manual forms.
TimeSeriesStore A specialized generic cache for managing historical data points (keyed by ISO date). Handles gap detection (getMissingRanges) to minimize API calls and merges partial time-series datasets.

📐 Architecture & Flow (EditBuffer Pattern)

One of the most complex UX challenges is handling bulk edits on time-series charts (like Asset Prices or FX Rates): you want reactive inputs from multiple sources (chart clicks, CSV, forms) but you don't want to pollute the central store until the user clicks "Save".

The EditBuffer solves this by sitting between the UI interactions and the backend API.

---
config:
  layout: elk
---
graph TD
    Chart["📈 Chart (Click to Edit)"]
    CSV["📄 CSV Textarea"]
    Form["📝 Manual Form (+)"]

    Buffer["🛡️ EditBuffer (Pending Edits)"]
    SaveBtn["💾 Bulk Save"]
    API["🐍 Backend API"]

    Chart -. "adds/updates" .-> Buffer
    CSV -. "parses into" .-> Buffer
    Form -. "adds" .-> Buffer

    Buffer -->|"2-way sync"| CSV
    Buffer -->|"highlights on"| Chart

    Buffer -->|"computes"| SaveBtn
    SaveBtn -->|"Bulk POST"| API

🧠 How it Works

  1. Initialization: When the user enters "Edit Mode" on a chart (like FX rates or Asset prices), an empty EditBuffer is instantiated.
  2. Editing: The user can click on the chart to move a point, type directly into a CSV textarea, or use a form. All these actions add or update PendingEdit entries inside the EditBuffer.
  3. Synchronization: The EditBuffer maintains a 2-way sync. E.g., if a user drags a point on the chart, the CSV textarea automatically updates its text to reflect the new value.
  4. Bulk Save: When the user clicks Save, the EditBuffer aggregates all pending edits and sends a single bulk request to the backend API, updating the actual historical data.