NullifyPDF - Privacy Edition

NullifyPDF is a local-first PDF privacy tool. It helps detect personal data, redact it irreversibly, or replace it with reversible placeholders before a PDF is shared with third parties or uploaded to AI systems.

Tip

First time using NullifyPDF? Start with our User Guide — takes 5 minutes.

📋 Quick Overview

NullifyPDF goes beyond simple text covering. It uses Natural Language Processing (NLP) engines to identify entities like names, addresses, email addresses, IBANs, and credit card numbers. At export, the user chooses irreversible anonymization or reversible pseudonymization with an encrypted restore map.

🎯 Feature	✨ Benefit
🧠 Auto-detection	Automatic detection tuned for Italian documents
🔐 Local-first	No cloud uploads during PDF processing
⚡ Responsive UI	Background scanning with live preview
🛡️ Privacy Export	Irreversible redaction or reversible placeholders

✨ Key Features

🧠 Automatic Redaction — Automatic detection of PII in Italian documents: names, locations, emails, phones, IBANs, credit cards, crypto addresses, Italian fiscal codes and IDs. A dialog lets you choose which data types to scan for, and matching is whole-word so ordinary text is not caught
🔄 Update Check — Optional, user-triggered check for new releases (stable builds see stable only, beta builds see both) showing the release notes and a one-click download into your Downloads folder; can be enabled at every start from the Info dialog (off by default)
🗄️ Fluid UI & Thread-Safe — PySide6 modern dark-mode interface with zero UI freezing. Text extraction in worker thread with QMutex serialization
📖 Persistent Dictionaries — Blocklist and Allowlist synchronized to disk (~/.nullifypdf) with O(1) fast-path matching
🛡️ Privacy Export Modes — Choose irreversible removal or reversible pseudonymization with a separately encrypted restore map; the exported file is sanitised (annotations, attachments, scripts, metadata, hidden text) and verified
📖 Built-in guide and tooltips — Press F1 for an Italian guide to every function; every control shows a description on hover
🔍 Review of uncertain hits — Detections without enough evidence are suggested for review instead of being applied or lost silently
🇮🇹 Italian name recognition on demand — the spaCy Italian model (CC BY-NC-SA 3.0, non-commercial) is not in the installer: the first Auto Redact shows its licence, asks for consent and downloads it (SHA-256 verified). Declining keeps rules and dictionaries only
🧠 Optional local models — Three reviewed ONNX PII models (MiniLMv2 EN/IT, PiiScan, OpenMed Italian FastClinical) can be downloaded by the user from pinned Hugging Face revisions; files are SHA-256 verified, the download progress is shown inside the button, and models only add review suggestions
📄 Word import — DOCX paragraphs and tables can be imported locally into the analysis workflow; legacy binary DOC files are not supported
🔎 OCR Support — Bundled EN/IT Tesseract language data for scanned PDFs
🖼️ Blindfold Mode — One-click image/logo censoring with professional placeholder: [ IMAGE REMOVED ]
📦 Native Cross-Platform — Automated build scripts generate Windows .exe, macOS ZIP bundles, and Linux .deb/.rpm packages
🎯 Drag & Drop Support — Native file drag-and-drop on main window
📊 Logging & Diagnostics — Local file-based logging (~/.nullifypdf/logs/) with debug mode for advanced troubleshooting

⚠️ Tool Limitations

To keep NullifyPDF lightweight, 100% offline, and secure, be aware of these technical limits:

❌ Limitation	💡 Workaround
Handwritten Text	NLP models cannot analyze non-digitized handwriting.
Password-Protected PDFs	Encrypted documents are blocked at load. Decrypt before importing.
Legacy Word DOC files	Only modern `.docx` files are supported; save legacy files as DOCX before importing.
Digital Signatures Invalidated	Privacy export changes the PDF; cryptographic signatures (PAdES, notarized) become invalid.

Warning

Digital signatures will be invalidated after redaction. Save unredacted originals separately for formal records.

🚀 Getting Started

📋 System Requirements

✅ Python 3.13 (required for PyMuPDF wheel compatibility)
✅ 2 GB disk space (dependencies + Italian model)
✅ 4 GB RAM minimum (8 GB recommended for large PDFs)

Operating System Support:

✅ Windows 10/11
✅ macOS 11+
✅ Linux: Ubuntu 24.04 / 26.04 and Debian 13 (.deb), RHEL / Rocky / Alma 9-10 and current Fedora (.rpm). The packages are built against glibc 2.34, so Ubuntu 22.04 and Debian 12 also run them, but these are no longer tested

⚙️ Installation

👤 End Users — Use Pre-Built Executable

Download the latest pre-compiled executable from Releases:

Windows: NullifyPDF_vX.Y.Z_Windows.exe
macOS: NullifyPDF_vX.Y.Z_macOS.zip
Ubuntu / Debian: NullifyPDF_vX.Y.Z_Ubuntu.deb
Fedora / RHEL: NullifyPDF_vX.Y.Z_Fedora.rpm

No installation needed on Windows/macOS - just run or unzip. Linux users can install the .deb or .rpm package.

👨‍💻 Developers — Install from Source

Clone the repository

git clone https://github.com/overwrite00/NullifyPDF.git
cd NullifyPDF

Verify Python 3.13

# Windows
py -3.13 --version

# macOS/Linux
python3.13 --version

Run automated setup (recommended)
```
python setup_env.py
```
This script automatically:
- Creates isolated virtual environment (.venv)
- Installs all dependencies
- Installs the Italian spaCy model for development (the installer downloads it on demand)

Activate environment & launch

# Windows (PowerShell)
.\.venv\Scripts\Activate.ps1
python NullifyPDF.py

# macOS/Linux (Bash)
source .venv/bin/activate
python3.13 NullifyPDF.py

🤖 Automation Scripts

The repository includes cross-platform Python scripts for developers:

🔧 setup_env.py — Environment Setup

Configures development environment with Python 3.13, virtual environment, and NLP models.

python setup_env.py

What it does:

Detects OS (Windows/macOS/Linux)
Creates .venv with Python 3.13
Installs requirements.txt dependencies
Installs the Italian spaCy model for development (the installer does not bundle it)

Automatic OS detection:

Windows: Uses py -3.13 launcher
macOS/Linux: Uses python3.13 directly

🏗️ build_local.py — Build Executable

Compiles standalone executable with PyInstaller.

python build_local.py

Features:

Cleans temporary directories
Auto-detects your OS
Reads version dynamically from code
Generates named executables such as NullifyPDF_vX.Y.Z_Windows.exe
Bundles EN/IT OCR data (downloaded automatically if missing)

Linux bonus: On Ubuntu/Fedora, automatically builds .deb and .rpm packages in dist/

✓ Running Tests

Verify core behavior with smoke tests:

# Activate venv first
source .venv/bin/activate  # or .venv\Scripts\Activate.ps1 on Windows

pytest tests/ -v

Test coverage:

Detection pipeline: word index, rectangle geometry, scoring, partial-match policy, Italian lexicon
Redaction engine, export sanitiser and export verification (survivor and leak checks)
Privacy placeholders, encrypted restore maps (Argon2id envelope, recovery key) and reconstruction
Encrypted dictionaries with expiry and the secure key store
Offscreen GUI flows (review, export, reconstruction, guide and tooltips, model dialog, Info window)
Local model catalog, on-demand Italian model download, DOCX import, update check
Licence consistency and build configuration

📚 Documentation

📄 Document	📖 Purpose
USER_GUIDE.md	Step-by-step usage instructions
CONTRIBUTING.md	How to contribute code & report issues
ARCHITECTURE.md	System design & technical overview
DEVELOPMENT.md	Local dev setup, testing & builds
OCR_SETUP.md	OCR data setup and build notes
TROUBLESHOOTING.md	Common issues & solutions
CHANGELOG.md	Release history & updates
SECURITY.md	Security model & responsible disclosure
THIRD_PARTY_NOTICES.md	Licences of data and components
LICENSE	AGPL-3.0-or-later licence text

🔒 Security & Privacy

✅ 100% Local Processing — All analysis happens on your machine
✅ No Cloud Uploads — PDF content never leaves your computer; the only network access is the optional update check and the model downloads you approve
✅ Open Source — Full code transparency
✅ No Telemetry — Zero user tracking

Important

Irreversible removal cannot be undone from the exported PDF, and it does not by itself guarantee anonymisation: indirect identifiers can remain. A pseudonymized PDF and its restore map are personal data. Always keep backups of original documents. Pseudonymization requires the encrypted restore map and its password.

See SECURITY.md for responsible disclosure and privacy details.

🛠️ Tech Stack

Technology	Purpose
Python 3.13	Core language (required for PyMuPDF compatibility)
PySide6 (Qt6)	Modern dark-mode GUI with multi-threading
PyMuPDF (fitz)	High-performance PDF manipulation
Microsoft Presidio	PII (Personally Identifiable Information) detection
spaCy	NLP for entity recognition (Italian model, downloaded on demand)
ONNX Runtime	CPU inference for the optional local PII models
cryptography	Encrypted pseudonymization restore maps

📝 License

GNU Affero General Public License v3.0 or later (AGPL-3.0-or-later)

NullifyPDF is free of charge and is not sold by its author. It is AGPL-licensed because it uses PyMuPDF, which is available under the AGPL-3.0 or a commercial licence from Artifex. Anyone who distributes the program, modified or not, must give the recipients the source code under the same licence. Releases up to v2.4.0-beta.4 were published under the MIT licence.

The Italian name-recognition model (spaCy it_core_news_md, CC BY-NC-SA 3.0, non-commercial) is not part of the installer: the program offers to download it, shows its licence and asks for consent first. Licences of all third-party components are listed in THIRD_PARTY_NOTICES.md.

Notice. NullifyPDF is free software, distributed under the AGPL-3.0-or-later. Every component and model it uses remains under its own licence, listed below with a link to the official text; using the program means respecting those conditions. The Italian spaCy model is for non-commercial use only and is downloaded only after your consent.

Component	Licence	Used for
NullifyPDF	AGPL-3.0-or-later	this program
PyMuPDF / MuPDF	AGPL-3.0 or commercial (Artifex)	PDF reading and redaction
PySide6 / Qt	LGPL-3.0	graphical interface
Presidio	MIT	personal-data detection
spaCy	MIT	language processing
spaCy it_core_news_md	CC BY-NC-SA 3.0 (non-commercial use only)	Italian model, downloaded after consent
MiniLMv2 EN/IT NER (optional download)	MIT	optional local model by Alberto Schiabel (jkomyno)
PiiScan (optional download)	Apache-2.0 (training data: Ai4Privacy OpenPII, CC BY 4.0)	optional local model
OpenMed PII Italian FastClinical Small (optional download)	Apache-2.0	optional local model
MiniLMv2 EN/IT NER (optional download)	MIT	optional local model by Alberto Schiabel (jkomyno)
PiiScan (optional download)	Apache-2.0 (training data: Ai4Privacy OpenPII, CC BY 4.0)	optional local model
OpenMed PII Italian FastClinical Small (optional download)	Apache-2.0	optional local model
onnxruntime	MIT	running local models
tokenizers	Apache-2.0	tokenizing for local models
cryptography	Apache-2.0 or BSD-3-Clause	encryption
keyring	MIT	key storage (Linux/macOS)
tldextract	BSD-3-Clause	URL recognition
NumPy	BSD-3-Clause	numerical computing
Tesseract OCR	Apache-2.0	OCR of scanned documents
tessdata_fast (eng, ita)	Apache-2.0	OCR language data
ISTAT, elenco dei comuni	CC BY 4.0	place names
Wikidata	CC0 1.0	given names and surnames

See LICENSE for the full text.

🤝 Contributing

Want to help improve NullifyPDF? See CONTRIBUTING.md for guidelines.

Last updated: 2026-11-11
User Guide →

Name		Name	Last commit message	Last commit date
.githooks		.githooks
.github		.github
images		images
scripts		scripts
tests		tests
.gitignore		.gitignore
ARCHITECTURE.md		ARCHITECTURE.md
CHANGELOG.md		CHANGELOG.md
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
CONTRIBUTING.md		CONTRIBUTING.md
DEVELOPMENT.md		DEVELOPMENT.md
LICENSE		LICENSE
NullifyPDF.py		NullifyPDF.py
OCR_SETUP.md		OCR_SETUP.md
PDF_Checker.py		PDF_Checker.py
README.md		README.md
SECURITY.md		SECURITY.md
THIRD_PARTY_NOTICES.md		THIRD_PARTY_NOTICES.md
TROUBLESHOOTING.md		TROUBLESHOOTING.md
USER_GUIDE.md		USER_GUIDE.md
build_local.py		build_local.py
dictionary_dialog.py		dictionary_dialog.py
export_engine.py		export_engine.py
help_guide.py		help_guide.py
lexicon_data_it.py		lexicon_data_it.py
lexicon_it.py		lexicon_it.py
licences_info.py		licences_info.py
model_dialog.py		model_dialog.py
model_manager.py		model_manager.py
onnx_ner.py		onnx_ner.py
pdf_sanitize.py		pdf_sanitize.py
pii_detection.py		pii_detection.py
privacy_core.py		privacy_core.py
requirements.txt		requirements.txt
review_dialog.py		review_dialog.py
secure_store.py		secure_store.py
setup_env.py		setup_env.py
spacy_models.py		spacy_models.py
update_check.py		update_check.py
word_import.py		word_import.py
word_lists.py		word_lists.py

Navigation Menu

Repository navigation

About

Repository files navigation

NullifyPDF - Privacy Edition

📋 Quick Overview

✨ Key Features

⚠️ Tool Limitations

🚀 Getting Started

📋 System Requirements

⚙️ Installation

🤖 Automation Scripts

📚 Documentation

🔒 Security & Privacy

🛠️ Tech Stack

📝 License

🤝 Contributing

About

Releases

Used by

Contributors

Languages

Latest commit

History

Folders and files

Repository files navigation

NullifyPDF - Privacy Edition

📋 Quick Overview

✨ Key Features

⚠️ Tool Limitations

🚀 Getting Started

📋 System Requirements

⚙️ Installation

🤖 Automation Scripts

📚 Documentation

🔒 Security & Privacy

🛠️ Tech Stack

📝 License

🤝 Contributing

About

Topics

Resources

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Releases

Used by

Contributors

Languages