pypdf

pypdf is a pure-Python library for working with PDF documents, providing capabilities for reading, splitting, merging, cropping, transforming, encrypting, and decrypting PDF files without requiring external dependencies such as Ghostscript, wkhtmltopdf, or LibreOffice. Originally created as PyPDF2 by Mathieu Fenniak in 2006, the project underwent a major rename and rebranding in 2022 to pypdf, consolidating the previously fragmented PyPDF2, PyPDF3, and PyPDF4 forks into a single, actively maintained codebase under the py-pdf GitHub organization. The library has accumulated over 8,000 GitHub stars and over 100 million downloads on PyPI, making it the most popular Python PDF manipulation library. pypdf supports a comprehensive range of PDF operations including page extraction, splitting and merging documents, page rotation (90, 180, 270 degrees), page cropping with mediabox and cropbox manipulation, overlay and watermark application, PDF encryption and decryption with 40-bit, 128-bit, and 256-bit AES and RC4 algorithms, metadata extraction and modification (title, author, subject, keywords, creation date), bookmark creation and navigation, form field reading and filling (AcroForm), image extraction from PDF pages, text extraction (including layout-preserving extraction mode), and hyperlink preservation. The library provides both a high-level API for common operations and a low-level API for direct manipulation of PDF objects (dictionaries, arrays, streams, names, strings) in the internal PDF structure. pypdf is pure Python with no compiled extensions, runs on CPython 3.8+, PyPy, and works cross-platform. Available under the BSD-3-Clause license.