Convert DOC to DOCX and Extract Tables with Python & LibreOffice

Purpose

Legacy .doc files need programmatic reading or modification. Modern Python libraries handle .docx but not .doc.

Why Convert to DOCX

.docx

is XML‑based, avoiding encoding problems.

The python-docx library can manipulate .docx files but cannot work with the older .doc format.

Conversion Options

If Microsoft Word is installed on Windows, pywin32 can use Word’s COM interface. For a free, cross‑platform solution, LibreOffice provides a command‑line converter usable on Windows, Linux, and macOS.

Environment Setup

Install LibreOffice

Install via the official installer on Windows or with a package manager on Linux:

sudo apt update && sudo apt install libreoffice

sudo dnf install libreoffice

sudo zypper install libreoffice

Verify the installation:

soffice --version

Install python-docx

pip install python-docx

Implementation

1. Convert .doc to .docx

Use LibreOffice in headless mode from Python:

import os
import subprocess

def convert_doc_to_docx(doc_path, output_dir, soffice_path=None):
    """Convert a .doc file to .docx using LibreOffice.
    Args:
        doc_path (str): Path to the .doc file.
        output_dir (str): Directory for the converted file.
        soffice_path (str, optional): Full path to the soffice executable.
    Returns:
        str: Path to the generated .docx file.
    """
    if not doc_path.lower().endswith('.doc'):
        raise ValueError("Input file must be .doc format")
    os.makedirs(output_dir, exist_ok=True)
    base_name = os.path.splitext(os.path.basename(doc_path))[0]
    docx_path = os.path.join(output_dir, f"{base_name}.docx")

    if soffice_path is None:
        try:
            subprocess.run(["soffice", "--version"], check=True,
                             stdout=subprocess.PIPE, stderr=subprocess.PIPE)
            soffice_cmd = ["soffice"]
        except FileNotFoundError:
            raise RuntimeError("LibreOffice is not installed or not in PATH.")
    else:
        soffice_cmd = [soffice_path]

    try:
        subprocess.run(
            soffice_cmd + ["--headless", "--nodefault", "--nologo",
                           "--convert-to", "docx", "--outdir", output_dir, doc_path],
            check=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
        if not os.path.exists(docx_path):
            raise RuntimeError(f"Conversion failed, output not found: {docx_path}")
        return docx_path
    except subprocess.CalledProcessError as e:
        raise RuntimeError(f"Conversion failed: {e.stderr.decode()}")

2. Read and optionally modify tables in the converted DOCX

Use python-docx to extract table data:

from docx import Document

def read_docx_tables(docx_path):
    """Extract tables from a .docx file.
    Args:
        docx_path (str): Path to the .docx file.
    Returns:
        list: List of tables, each table is a list of row lists.
    """
    try:
        doc = Document(docx_path)
    except Exception as e:
        raise RuntimeError(f"Unable to read .docx file: {e}")

    tables_data = []
    for i, table in enumerate(doc.tables):
        print(f"
Table {i + 1}:")
        table_data = []
        for row in table.rows:
            row_data = [cell.text.strip() for cell in row.cells]
            # Optional: clean special characters
            row_data = [clean_text(cell) for cell in row_data]
            table_data.append(row_data)
            print(row_data)
        tables_data.append(table_data)
    return tables_data

The conversion function checks for LibreOffice, runs the conversion, and returns the path of the generated .docx file. The table‑reading function returns a nested list that can be processed further or modified using the same library.

Summary

By installing LibreOffice and python-docx, legacy .doc files can be converted to .docx on any major OS and their tables can be programmatically accessed and edited with Python.