inv-pdf 0.1.0

Invert PDF colors page-by-page with parallel processing — dark-mode PDFs from any source

inv-pdf

Invert PDF colors (dark-mode) for a file or entire directory — with parallel processing across CPU cores.

---

The Problem

Reading white-background PDFs at night or on OLED screens is painful. inv-pdf inverts every page of a PDF to produce a dark-mode version, with near-black pixels clamped to pure black to avoid muddy greys.

---

Installation

pip install inv-pdf

Dependencies: pymupdf and Pillow are installed automatically.

---

Usage

CLI

# Invert all PDFs in current directory (creates file_inverted.pdf for each)
inv-pdf .

# Specify output directory
inv-pdf ./pdfs -o ./dark-pdfs

# Custom DPI, suffix, and parallelism
inv-pdf ./pdfs --dpi 150 --suffix _dark --jobs 4

# Full help
inv-pdf --help

Python API

from inv_pdf import invert_pdf

# Invert a single PDF
invert_pdf("document.pdf", "document_inverted.pdf", dpi=300)

---

Options

Flag	Default	Description
DIR	.	Directory of PDFs to process
-o / --output	Same as input	Output directory
-s / --suffix	_inverted	Suffix added to output filenames
-d / --dpi	300	Render resolution (higher = better quality, slower)
-j / --jobs	CPU count	Number of parallel worker processes
--ext	.pdf	File extension to match

---

How It Works

1. Opens each PDF with PyMuPDF (pymupdf)
2. Rasterises each page at the specified DPI
3. Inverts pixel values: 255 − x for each RGB channel via Pillow
4. Clamps near-black pixels (all channels < 100) to pure (0, 0, 0)
5. Re-encodes each page as PNG and embeds it back into a new PDF
6. Preserves the original TOC and metadata
7. Saves with deflate=True, garbage=4, clean=True for compact output

---

Requirements

• Python 3.8+
• pymupdf >= 1.23.0
• Pillow >= 9.0.0

---

License

MIT © Santt997