PyMuPDF

PyMuPDF is the Python binding of MuPDF. It opens MOBI files with the same Document API it uses for PDF and EPUB: the book is laid out into pages that you can read as text, render or convert to PDF.

MOBI support comes from MuPDF and has the same limits: no HUFF/CDIC-compressed books, no DRM, and the text records are read as one HTML document (MOBI 6 style), without KF8 reconstruction or EXTH metadata. For complete MOBI and AZW3 unpacking use mobi. The AGPL license requires you to publish your source code if you distribute your program or run it as a network service.

Installation#

pip install pymupdf

Basic usage: extract text from a MOBI file#

import pymupdf

try:
    doc = pymupdf.open("book.mobi")
except Exception as e:  # e.g. unknown compression method (HUFF/CDIC)
    raise SystemExit(f"cannot open: {e}")

doc.layout(width=420, height=595, fontsize=11)
print(doc.metadata["format"], doc.page_count, "pages")
text = "\n".join(page.get_text() for page in doc)
with open("book.txt", "w", encoding="utf-8") as f:
    f.write(text)

How to open .azw or .azw3 files#

MuPDF recognizes MOBI by the BOOKMOBI signature, and lists only .mobi, .prc and .pdb as MOBI extensions. To be safe with .azw and .azw3 files, pass the type explicitly:

doc = pymupdf.open("book.azw3", filetype="mobi")

How to convert MOBI to PDF#

import pymupdf

doc = pymupdf.open("book.mobi")
doc.layout(width=420, height=595, fontsize=11)
pdf = pymupdf.open("pdf", doc.convert_to_pdf())
pdf.save("book.pdf", garbage=3, deflate=True)

How to render the first page to PNG#

doc = pymupdf.open("book.mobi")
doc[0].get_pixmap(dpi=150).save("page1.png")

How to search for a word in a MOBI book#

for page in doc:
    hits = page.search_for("whale")
    if hits:
        print(f"page {page.number + 1}: {len(hits)} hits")

Tips#

• pymupdf.open(stream=data, filetype="mobi") opens a MOBI file from bytes.
• Page count depends on layout(); set the page size and font size before using page numbers.