pdfcpu is a PDF processing library written in Go.

It is compatible with all PDF versions. Support for PDF 2.0 (ISO-32000-2) is evolving and continuously improving.

👉 CLI Installation instructions

👉 API Installation instructions

Validate against PDF 2.0 (ISO-32000-2):

pdfcpu validate input.pdf

Merge two PDFs:

pdfcpu merge out.pdf in1.pdf in2.pdf

See API documentation for usage examples.

• Validate, optimize, split, trim, and merge PDFs
• Extract and manipulate images, fonts, and metadata
• Encrypt and decrypt PDFs
• Resize and rotate pages
• Add and remove stamps and watermarks
• Manage digital signatures (ongoing work)
• Manage attachments and more...

Common operations and examples:

Complete list of supported commands:

pdfcpu aims to provide comprehensive PDF processing capabilities implemented in Go.

It focuses on correctness, robustness and independence from external dependencies.

• comprehensive PDF processing functionality
• minimal external dependencies
• predictable and stable behavior

• Project documentation: https://pdfcpu.io

• Command help: pdfcpu [command] --help

• Package documentation: https://pkg.go.dev/github.com/pdfcpu/pdfcpu

• API documentation: https://pkg.go.dev/github.com/pdfcpu/pdfcpu/pkg/api

• Examples:

  • https://github.com/pdfcpu/pdfcpu/tree/master/pkg/api/test
  • https://github.com/pdfcpu/pdfcpu/tree/master/pkg/samples

Contributions are welcome.

• Report bugs or propose changes via issues
• Discuss ideas on the discussion board
• For PRs, please open an issue or discussion first v

• Base your work on the latest commit
• Include verbose output (pdfcpu cmd -vv ...) and a sample PDF when reporting issues
• Please sign your commits

Crashes may occur due to the wide variety of PDF producers and formats in use, including older or non-compliant files. In many cases this is related to validation issues or edge cases in the parser.

Even with relaxed validation, some files cannot be processed. These cases are essential for improving pdfcpu by extending validation and handling additional real-world PDFs.

If you encounter a crash, please report it.

Start by validating the file using the CLI:

pdfcpu validate -vv <file.pdf>

Include in your report:

• the command used
• verbose output (-vv)
• a sanitized sample PDF

If validation crashes for a PDF that opens in Adobe Reader or macOS Preview, it is likely we can extend relaxed validation and provide a fix.

If the file cannot be opened by both Adobe Reader and macOS Preview, we cannot support it.

Please include a sample PDF to reproduce the issue whenever possible.

Thanks 💚 to all contributors:

Horst Rutter	haldyr	Vyacheslav	Erik Unger	Richard Wilkes	minenok-tutu	Mateusz Burniak
Dmitry Harnitski	ryarnyah	Sam Giffney	Carlos Eduardo Witte	minusworld	Witold Konior	joonas.fi
Henrik Reinstädtler	VMorozov-wh	Benoit KUGLER	Adam Greenhall	moritamori	JanBaryla	TheDiscordian
Rafael Garcia Argente	truyet	Christian Nicola	Benjamin Krill	Peter Wyatt	Kroum Tzanev	Stefan Huber
Juan Iscar	Eng Zer Jun	Dmitry Ivanov	Rene Kaufmann	Christian Heusel	Chris	Lukasz Czaplinski
Joel Silva Schutz	semvis123	guangwu	Yoshiki Nakagawa	Steve van Loben Sels	Yaofu	vsenko
Alexis Hildebrandt	Sivukhin Nikita	Joachim Bauch	kalimit	Andreas Erhard	Matsumoto Toshi	Carl Wilson
LNAhri	vishal	Andreas Deininger	Robert Raines	Frank Anderson	Sven Lilienthal	Florian Kinder
mdmcconnell	Bradley Erickson	doronbehar	joeyave	Zhenbang Wei

This project is released with a Contributor Code of Conduct. By participating, you agree to abide by its terms.

Use of pdfcpu assumes compliance with all applicable copyrights for any processed PDF content, including embedded resources such as fonts and images.

Gopher artwork by Renee French

Apache License 2.0