Kafka

Fast, vim-like language learning.

Features

• Vim-style navigation: Navigate through text using vim keybinds
• Vocabulary tracking: Track word familiarity across 6 stages (unseen to known)
• Multiple import formats: PDF, EPUB, plain text
• AI import and cleanup: OCR on non-text sources, AI text cleanup for OCR errors
• Fast and responsive: Built with Svelte

Quick Start

Prerequisites

• Python 3.10+
• Node.js 18+
• (Optional) Tesseract OCR for PDF text extraction: sudo apt install tesseract-ocr tesseract-ocr-chi-sim
• (Optional) ANTHROPIC_API_KEY environment variable for AI text cleanup

Running the Application

# Make start script executable
chmod +x start.sh

# Start both backend and frontend
./start.sh

Or run them separately:

# Terminal 1: Backend
cd backend
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python run.py

# Terminal 2: Frontend
cd frontend
npm install
npm run dev

Then open http://localhost:5173 in your browser.

Vim Keybindings

Navigation

Key	Action
h / l	Previous / next word
j / k	Down / up (by line)
w / b	Next / previous word
W / B	Next / previous sentence
gg	Go to beginning
G	Go to end
Ctrl+f / Ctrl+b	Page down / up

Vocabulary

Key	Action
0	Mark as unseen
1	Mark as unknown (light yellow)
2-4	Learning stages (darker yellow)
5	Mark as known (blue)
x	Mark as irrelevant

General

Key	Action
i	Import file (in library)
q	Close reader
?	Toggle help
Esc	Close dialogs

Vocabulary Stages

Stage	Color	Description
0	None	Unseen
1	Light yellow	Seen but unknown
2-4	Darker yellow	Learning stages
5	Blue	Known
-1	None	Irrelevant (names, etc.)

Project Structure

lingq_copy/
├── backend/           # FastAPI Python backend
│   ├── app/
│   │   ├── main.py   # API entry point
│   │   ├── models.py # Database models
│   │   ├── routers/  # API endpoints
│   │   └── services/ # Business logic
│   └── data/         # SQLite DB and content files
├── frontend/          # Svelte TypeScript frontend
│   └── src/
│       ├── components/
│       └── lib/      # API client, stores, vim handler
└── context/          # Project documentation

API Endpoints

• GET /vocabulary/ - List vocabulary words
• PUT /vocabulary/{word}/stage/{stage} - Set word stage
• GET /content/ - List imported content
• GET /content/{id} - Get content with text and word positions
• POST /import/upload - Upload PDF/EPUB/TXT file
• POST /import/text - Import raw text

Full API documentation at http://localhost:8000/docs

Technology Stack

• Frontend: Svelte + TypeScript + Vite
• Backend: Python + FastAPI
• Database: SQLite
• Chinese tokenization: jieba
• PDF parsing: PyMuPDF + pytesseract (OCR)
• EPUB parsing: ebooklib