LUCERNA Archive

A Quiet Place for Your Reading Life

保存阅读人生的数字书房
Preserve the traces of who you were while reading.

English · 中文

---

Overview

LUCERNA Archive is a digital reading room that preserves your reading life. Connect your Kindle or WeRead, and your highlights and notes are quietly archived into fragments that carry time, mood, and a sense of space.

This is not a knowledge management tool. It's not a second brain. It's a room where, years later, you can meet the person you were while reading.

Not a dashboard. Not a second brain. A printed reading room.

概述

LUCERNA Archive 是一间保存阅读人生的数字书房。连接 Kindle 或微信读书后，应用将划线与笔记整理成有时间、情绪和空间感的阅读档案。

它关注的不是"知识管理效率"，而是"多年之后还能重新遇见当时读书的自己"。

---

Features

	Feature	Description
📖	Kindle Import	Parse My Clippings.txt with full Chinese & English format support, automatic deduplication
📱	WeRead Sync	One-click sync highlights, notes, and reviews via browser extension or cookie
🏠	The Room	Immersive dark reading room — random fragments slowly surface with breathing animations
📄	Fragments	Floating paper-slip stream of all highlights and notes, filterable by book
🌊	Timeline	Your reading life arranged by season and month, not by numbers
📚	Library	Private book archive — every book becomes its own room
🌐	Bilingual UI	Full Chinese / English interface toggle
🔒	Local-First	All data stays in your browser's IndexedDB — nothing is uploaded, ever
💾	Export & Backup	Export your entire archive as JSON; restore anytime

---

Design Language

The interface evokes the atmosphere of a late-night study: deep slate-blue backgrounds, warm wood-brown structures, soft amber lamplight, and paper textures. Animations are slow and breathing — like light shifting in a room, not flashy app feedback.

Visual keywords: Midnight · Wood · Warm lamp · Paper · Rain · Film grain · Floating paper slips · Breathing

Copy principles: Literary but not pretentious. Short, light, quiet.

✅ Do	❌ Don't
Listening to your Kindle.	Syncing...
43 fragments returned tonight.	Imported successfully
You marked this on a quiet night.	Total highlights

---

Tech Stack

Frontend    React 19 · TypeScript 6 · Vite 8
Database    sql.js (client-side SQLite via IndexedDB)
Auth        Supabase (optional cloud sync)
Extension   Chrome Extension (WeRead one-click import)

---

Getting Started

Prerequisites

• Node.js ≥ 18
• npm or pnpm

Install & Run

# Clone the repository
git clone https://github.com/huyan1349/Archive-web.git
cd Archive-web

# Install dependencies
npm install

# Start development server
npm run dev

# Build for production
npm run build

Import from Kindle

Drag your Kindle's My Clippings.txt into the app window, or click 「导入我的文件」 in the sidebar.

Sync from WeRead

1. Install the LUCERNA browser extension (see extension/ directory)
2. Open weread.qq.com and log in
3. The extension automatically captures your session — click sync

---

Project Structure

Archive-web/
├── src/
│   ├── core/
│   │   ├── analysis/        # Reading analysis & Midnight Recall
│   │   ├── auth/            # Supabase authentication
│   │   ├── database/        # IndexedDB data layer with repository pattern
│   │   ├── importer/        # Import pipeline & deduplication
│   │   ├── kindle-db/       # Kindle Clippings parser
│   │   ├── parser/          # Multi-format Clippings parser (zh/en)
│   │   └── weread/          # WeRead API client & sync
│   ├── data/                # React hooks & data providers
│   ├── App.tsx              # Main application shell
│   ├── AuthProvider.tsx     # Auth context
│   ├── LoginPage.tsx        # Supabase login
│   └── DevPanel.tsx         # Development utilities
├── extension/
│   ├── manifest.json        # Chrome Extension manifest
│   ├── background.js        # Service worker
│   ├── lucerna-content.js   # Main content script
│   ├── weread-content.js    # WeRead cookie extraction
│   └── popup.html/js        # Extension popup UI
└── LUCERNA Archive/         # Product documentation
    ├── 01-product-overview.md
    ├── 02-mvp-scope.md
    ├── 03-technical-architecture.md
    ├── 04-development-roadmap.md
    └── 05-design-language.md

---

Data Architecture

┌─────────────┐     ┌──────────────┐     ┌──────────────┐
│   books      │     │  fragments    │     │   imports     │
├─────────────┤     ├──────────────┤     ├──────────────┤
│ id           │◄────┤ book_id       │     │ id            │
│ title        │     │ type          │     │ source_path   │
│ author       │     │ content       │     │ imported_at   │
│ created_at   │     │ note          │     │ imported_count│
│ updated_at   │     │ location      │     │ skipped_count │
└─────────────┘     │ clipped_at    │     │ failed_count  │
                    │ source_hash   │     └──────────────┘
                    │ is_favorite   │
                    └──────────────┘
                           │
                    ┌──────┴───────┐
                    │ fragment_moods│
                    ├──────────────┤
                    │ fragment_id   │
                    │ mood_id       │
                    └──────────────┘

Deduplication strategy: Stable hash from book_title + author + type + location + clipped_at + content, with fallback to book_title + author + type + location + content when timestamp is missing.

Privacy principles:

• Default write to local IndexedDB only
• No account required
• No uploads
• No AI calls
• No user data in logs

---

Roadmap

Phase 1 — MVP ✅

• Kindle Clippings import & parsing
• WeRead sync via browser extension
• The Room / Fragments / Timeline / Library
• Bilingual UI (zh/en)
• Local-first data with export/backup

Phase 2 — Upcoming

• AI-powered mood analysis for fragments
• Dynamic Rooms with spatial layout
• Midnight Recall — random fragment notifications
• Menu bar quick-access on macOS
• Paper warmth adjustment (display settings)
• Annual reading climate report

Phase 3 — Exploring

• Local LLM / embedding for semantic search
• AirPods & Focus Mode integration
• iCloud Drive manual backup
• Vision Pro spatial reading room

---

License

MIT

---

Crafted with care by huyan