pdfembed 0.1.3

CLI/TUI tool to run OCR locally and overlay a searchable text layer on PDFs.

pdfembed

---

CLI/TUI tool to run OCR locally and overlay a searchable text layer on PDFs. By default, a Textual-based TUI launches; use --cli for the classic CLI.

License: BSD-3-Clause (see LICENSE).

Quickstart

• TUI (default):
  python -m pdfembed.cli or pdfembed

• CLI:
  python -m pdfembed.cli --cli --file sample.pdf --dpi 300

TUI Controls

• f: select PDF file(s) (opens a file dialog; multiple selection allowed)
• o: select output folder (opens a folder dialog; defaults to the first PDF's directory)
• v: toggle overlay visibility (debug)
• s: start OCR
• q: quit
• DPI is fixed to the default in TUI; change via CLI --dpi if needed.

While OCR is running, a "Processing... please wait" indicator is shown and other keys are ignored until completion.

CLI Options (key ones)

• --file <pdf1> [pdf2 ...] or --dir <folder>: input PDFs
• --output <dir>: output directory (default: input location)
• --dpi <int>: render DPI (default 300)
• --visible: make overlay text visible (debug)
• --font <path>: TTF font for overlay text
• --log-level <LEVEL>: logging level (INFO by default)

Dependencies

• Textual (TUI)
• tkinter (file dialogs, stdlib)
• onnxocr / pypdfium2 / pypdf / reportlab / opencv-python / numpy