5 Best Ways to Read Microsoft Word Documents with Python

💡 Problem Formulation: Developers often need to read and process the content of Microsoft Word documents programmatically. Consider the scenario where you have a DOCX file and you want to extract the text within it to analyze the document, search for certain keywords, or migrate content to another format. The input is a Microsoft Word (.docx) file, and the desired output is a string representation of its contents.

Method 1: Using python-docx Library

The python-docx library allows users to create, modify, and extract information from Word documents. This method is ideal for structured data extraction, as it provides functionality to access document properties, text, and even style information.

Here’s an example:

from docx import Document

def read_docx(file_path):
    doc = Document(file_path)
    fullText = []
    for para in doc.paragraphs:
        fullText.append(para.text)
    return '\n'.join(fullText)
  
print(read_docx('example.docx'))
  

Output:

This is a text example from
the document's first paragraph.

This is from the second paragraph.
  

This script opens a .docx file, loops through each paragraph, and appends the paragraph text to a list. The joined list elements form a string that represents the document text.

Method 2: Using PyWin32 (Windows Only)

PyWin32 allows Python to interact with COM objects, enabling the automation of Microsoft Office applications on Windows. This method leverages Microsoft Word itself, so it provides high fidelity reading of documents, preserving formatting and other properties.

Here’s an example:

import win32com.client

def read_docx_win32(file_path):
    word = win32com.client.Dispatch("Word.Application")
    doc = word.Documents.Open(file_path)
    doc_content = doc.Content.Text
    doc.Close()
    word.Quit()
    return doc_content
  
print(read_docx_win32('example.docx'))
  

Output:

This is a text example extracted by Microsoft Word via PyWin32,
including style and formatting data.
  

This script uses COM automation to open the Word document with Microsoft Word, extract the text contents, and then closes the document and quit Word.

Method 3: Using textract Library

textract is a Python library that extracts text out of any document, including Word files. It calls command line utilities or Python libraries behind the scenes, saving you the hassle of handling multiple document formats.

Here’s an example:

import textract

text = textract.process('example.docx')
print(text.decode('utf-8'))
  

Output:

This is some text that has been extracted from a .docx file by the textract library.
  

textract abstracts away the complexity of reading different text file formats, returning the extracted text as a byte string that can be decoded to UTF-8.

Method 4: Using Unoconv Tool

Unoconv is a command line utility that can convert between any document format supported by LibreOffice/OpenOffice. It’s a cross-platform tool that can also be used to extract text from documents.

Here’s an example:

import subprocess

def extract_text_with_unoconv(file_path):
    subprocess.call(['unoconv', '--stdout', '-f', 'txt', file_path])

text = extract_text_with_unoconv('example.docx')
print(text)
  

Output:

This is example text converted from .docx to plain text using unoconv.
  

The function calls unoconv to output the text of the Word document to standard output, which can then be captured or redirected as needed.

Bonus One-Liner Method 5: Using the command line

For a quick and dirty one-liner, the Linux command line provides tools to convert and extract text from a Word document using utilities like antiword or catdoc.

Here’s an example:

antiword example.docx
  

This command outputs the content of a Word document to the terminal using the antiword utility, which can be handy for quick checks or piping into other commands.

Summary/Discussion

• Method 1: python-docx Library. The python-docx library is Pythonic and great for reading docx files. Strengths: Good for structured and styled content. Weaknesses: Does not handle older .doc files.
• Method 2: PyWin32 (Windows Only). PyWin32 provides a Windows-native way of reading Word documents. Strengths: High fidelity reading. Weaknesses: Windows-only and requires Microsoft Word installed.
• Method 3: textract Library. Textract supports many file formats, reducing the need for multiple libraries. Strengths: Versatile and easy. Weaknesses: External dependencies may need to be installed.
• Method 4: Unoconv Tool. Unoconv provides flexibility and works across platforms. Strengths: Compatible with many document formats. Weaknesses: Requires LibreOffice/OpenOffice.
• Method 5: Command Line Tools. Using command line tools like antiword is quick and straightforward. Strengths: Fast and simple for plain text. Weaknesses: Not a Python solution, and may have limited availability.