# Imageboard Grabber - Further Development & Roadmap

This document outlines a roadmap and future development plan for the **Imageboard Grabber** application on Hugging Face Spaces.

---

## 🚀 1. Feature Enhancements

### 1.1 Support for Authenticated APIs (API Keys & Username)

- **Goal:** Allow users to safely search and download from private or rate-limited sites (like Danbooru premium tiers, e621 personal accounts).
- **Implementation:**
  - Add credential fields (API Key / Password Hash, Username) to the **Add Site** dialog.
  - Securely transmit credentials to the server using HTTP headers or request bodies.
  - Store credentials locally in the browser's `localStorage` or session-only cookies so they are never stored globally on the server.

### 1.2 Persistent Site Configuration

- **Goal:** Avoid losing newly added schemas when refreshing or clearing session storage.
- **Implementation:**
  - Sync the list of added custom sites with IndexedDB or `localStorage`.
  - Provide an **Export/Import Sites** button to allow users to backup their custom configurations as JSON.

### 1.3 Advanced Query Builder

- **Goal:** Help non-technical users build complex booru search queries.
- **Implementation:**
  - Add UI controls for searching by rating (Safe, Questionable, Explicit), score range, upload date, or file extension.
  - Autocomplete / tag suggestion box by querying the active site's tag API endpoint (e.g., `/tags.json?search[name_matches]=*`).

### 1.4 Enhanced Lightbox & Media Viewer

- **Goal:** Support video playback and metadata exploration.
- **Implementation:**
  - Auto-detect video extensions (`.webm`, `.mp4`) in the Lightbox and render a native HTML `<video>` player with volume and speed controls.
  - Show side-panel info: Artist, Character, Copyright, and Meta tags. Click a tag to search.

---

## 🛠️ 2. Architectural & Performance Improvements

### 2.1 Concurrent / Streamed Server Downloads

- **Goal:** Speed up ZIP compilation and handle massive file downloads without causing OOM (Out Of Memory) crashes.
- **Implementation:**
  - Implement streaming ZIP using `fflate` on the server so bytes are sent back to the client immediately as they download, rather than buffering entire gigabytes in server memory.
  - Limit concurrent download workers per request to avoid IP bans from image hosters.

### 2.2 Client-Side Image Caching

- **Goal:** Reduce bandwidth usage and accelerate browsing.
- **Implementation:**
  - Register a Service Worker to cache thumbnails and UI assets.
  - Use `Cache-Control` headers for downloaded image URLs.

---

## 🧪 3. Quality Assurance & Automated Testing

### 3.1 End-to-End Tests with Playwright

- **Goal:** Ensure code modifications do not break key features like site additions or tag searches.
- **Implementation:**
  - Add E2E test suites inside a `tests/` directory.
  - Mock API responses from e621/Danbooru to test search, grid rendering, and selection UI without hitting actual boorus.

### 3.2 Unit and API Integration Tests

- **Goal:** Validate parsing logic, schema validation, and URL building helpers.
- **Implementation:**
  - Write Vitest suites to verify mapping functions in `src/lib/grabber/sites.ts` and zip streaming endpoints.

---

## 🌐 4. Deployment & Infrastructure Optimization

### 4.1 Hugging Face Storage Integration

- **Goal:** Retain user session data or cache assets on persistent storage.
- **Implementation:**
  - Connect a Hugging Face Dataset or persistent Space storage to backup logs or temporary configuration files if requested.