pdf2docx: Python Library for Converting PDF to DOCX with Features, Limitations, Installation, and Example

pdf2docx is a Python library that converts PDF files to DOCX by extracting layout, paragraphs, images, tables, and other elements using PyMuPDF and python-docx.

Features

<code>- Parse and create page layout
  - margins
  - sections and columns (up to two)
  - header and footer [TODO]

- Parse and create paragraphs
  - OCR text [TODO]
  - horizontal or vertical text
  - font style (font, size, bold/italic, color)
  - text style (highlight, underline, strikethrough)
  - list style [TODO]
  - external hyperlinks
  - paragraph alignment and spacing

- Parse and create images
  - inline images (grayscale/RGB/CMYK)
  - images with transparency
  - floating images

- Parse and create tables
  - border style (width, color)
  - cell background color
  - merged cells
  - vertical text in cells
  - hidden borders
  - nested tables

- Multiprocessing support</code>

The library also extracts table content and style, making it useful as a table extraction tool.

Limitations

<code>- No OCR for scanned PDFs
- Supports only left‑to‑right languages (no Arabic)
- No rotated text support
- Rule‑based parsing cannot guarantee 100% style fidelity</code>

Installation

<code>pip install pdf2docx</code>

Example

<code>from pdf2docx import parse

pdf_file = '/path/to/sample.pdf'
docx_file = 'path/to/sample.docx'

# convert pdf to docx
parse(pdf_file, docx_file)</code>

Run the script to generate the DOCX file.