Introduction to PyMuPDF: Installation, Features, and Usage Guide

PyMuPDF is the Python interface to MuPDF, a lightweight viewer and library for PDF, XPS, EPUB, CBZ, and other document formats, offering high‑quality anti‑aliased rendering and fast performance.

The library supports a wide range of functions including decryption, metadata access, raster (PNG) and vector (SVG) rendering, text search, extraction of text and images, conversion to formats such as HTML, XML, JSON, and full PDF manipulation like creating, merging, splitting, rotating, and annotating pages.

Installation is straightforward via pip install PyMuPDF ; optional dependencies such as Pillow, fontTools, and pymupdf‑fonts enhance functionality for image saving and font handling.

Basic usage starts with importing the library ( import fitz ) and checking the version ( print(fitz.__doc__) ). A document is opened with doc = fitz.open(filename) , after which Document methods like doc.page_count , doc.metadata , doc.get_toc() , and doc.load_page(pno) provide access to pages and their properties.

Pages can be accessed directly ( page = doc[pno] ) or loaded ( page = doc.load_page(pno) ). Iterating over pages is supported ( for page in doc: ), and each page offers methods to retrieve links ( links = page.get_links() ), annotations ( for annot in page.annots(): ), and widgets ( for field in page.widgets(): ).

Rendering a page to an image is done with pix = page.get_pixmap() , and the resulting pixmap can be saved ( pix.save("page-%i.png" % page.number) ) or converted to SVG via page.get_svg_image() . The pixmap object also provides size and color‑space information.

Text and image extraction uses page.get_text(opt) where opt can be "text", "blocks", "words", "html", "json", "xml", etc., allowing fine‑grained control over the output format. Searching for a string on a page is performed with areas = page.search_for("mupdf") , returning rectangles that locate each occurrence.

PDF‑specific operations include deleting, copying, moving, and inserting pages via methods such as Document.delete_page() , Document.copy_page() , Document.move_page() , Document.insert_page() , and Document.new_page() . Documents can be merged or split using Document.insert_pdf() and by creating new empty PDFs with fitz.open() . Saving changes is done with Document.save() , optionally using incremental=True for fast incremental updates.

Finally, resources are released by calling Document.close() , which closes the underlying file and frees associated buffers.