Optical Character Recognition (OCR)

Extract plain text directly from any screen region into your Wayland clipboard.

Usage

1. Trigger OCR mode via the floating dock or CLI:

   shotdock -t
   

2. Drag a rectangular selection over any on-screen text, terminal buffer, error dialog, or image.
3. The extracted text is immediately written to your Wayland clipboard (wl-clipboard) and an interactive notification displays a preview of the recognized text.

Configuration

Configure the language model in ~/.config/shotdock/config.json:

{
  "ocr_lang": "eng"
}

The language code matches Tesseract dataset codes (e.g., eng, deu, fra, spa, jpn).

Dependencies

• tesseract: The OCR engine
• tesseract-data-eng: English language trained data (or language pack of your choice)
• wl-clipboard: Wayland clipboard tool (wl-copy)