med-notes/data_structure/docx_to_md.py
2025-06-29 13:04:14 +00:00

334 lines
10 KiB
Python

#!/usr/bin/env python3
"""
Convert a directory of DOCX documents to Markdown format with proper formatting.
"""
import os
import re
import argparse
from pathlib import Path
from typing import Optional, List
# Check each library individually
MAMMOTH_AVAILABLE = False
PANDOC_AVAILABLE = False
DOCX_AVAILABLE = False
try:
import mammoth
MAMMOTH_AVAILABLE = True
print("✓ mammoth library available")
except ImportError:
print("✗ mammoth not available")
try:
import pypandoc
PANDOC_AVAILABLE = True
print("✓ pypandoc library available")
except ImportError:
print("✗ pypandoc not available")
try:
from docx import Document
from markdownify import markdownify as md
DOCX_AVAILABLE = True
print("✓ python-docx library available")
except ImportError:
print("✗ python-docx not available")
# Show available methods
available_methods = []
if MAMMOTH_AVAILABLE:
available_methods.append("mammoth")
if PANDOC_AVAILABLE:
available_methods.append("pandoc")
if DOCX_AVAILABLE:
available_methods.append("python-docx")
if available_methods:
print(f"Available conversion methods: {', '.join(available_methods)}")
else:
print("No conversion libraries available!")
print("Install with: pip install mammoth pypandoc python-docx markdownify")
def clean_markdown_text(text: str) -> str:
"""
Clean and format markdown text using better approaches than manual regex.
This function addresses common formatting issues:
- Multiple whitespaces
- Inconsistent list formatting
- Paragraph spacing
- Line break normalization
"""
# 1. Normalize line endings
text = text.replace("\r\n", "\n").replace("\r", "\n")
# 2. Fix multiple consecutive spaces (but preserve code blocks)
lines = text.split("\n")
cleaned_lines = []
in_code_block = False
for line in lines:
# Check if we're entering/exiting a code block
if line.strip().startswith("```"):
in_code_block = not in_code_block
cleaned_lines.append(line)
continue
if not in_code_block:
# Replace multiple spaces with single space, but preserve indentation
leading_spaces = len(line) - len(line.lstrip())
content = line.lstrip()
# Collapse multiple spaces in content
content = re.sub(r" {2,}", " ", content)
line = " " * leading_spaces + content
cleaned_lines.append(line)
text = "\n".join(cleaned_lines)
# 3. Fix list formatting
# Ensure proper spacing for unordered lists
text = re.sub(r"^(\s*)[-*+](\S)", r"\1- \2", text, flags=re.MULTILINE)
# Ensure proper spacing for ordered lists
text = re.sub(r"^(\s*)(\d+\.)(\S)", r"\1\2 \3", text, flags=re.MULTILINE)
# 4. Fix paragraph spacing - remove excessive blank lines
text = re.sub(r"\n{3,}", "\n\n", text)
# 5. Clean up common punctuation issues (more targeted than your original)
# Fix missing space after punctuation at end of sentences
text = re.sub(r"([.!?])([A-Z])", r"\1 \2", text)
# Fix missing space after commas (but not in numbers)
text = re.sub(r",(?=[a-zA-Z])", ", ", text)
# Fix missing space after colons and semicolons
text = re.sub(r"([;:])(?=[a-zA-Z])", r"\1 ", text)
# Remove spaces before punctuation
text = re.sub(r" +([,.;:!?])", r"\1", text)
# 6. Clean up markdown formatting issues
# Fix bold/italic markers with spaces
text = re.sub(r"\*\* +", "**", text)
text = re.sub(r" +\*\*", "**", text)
text = re.sub(r"\* +", "*", text)
text = re.sub(r" +\*", "*", text)
# 7. Final cleanup
text = text.strip()
return text
def convert_docx_with_mammoth(docx_path: Path) -> str:
"""Convert DOCX to Markdown using mammoth (recommended approach)."""
try:
with open(docx_path, "rb") as docx_file:
result = mammoth.convert_to_markdown(docx_file)
if result.messages:
print(f"Mammoth warnings: {result.messages}")
return result.value
except Exception as e:
print(f"Mammoth conversion error: {e}")
raise
def convert_docx_with_pandoc(docx_path: Path) -> str:
"""Convert DOCX to Markdown using pandoc (alternative approach)."""
return pypandoc.convert_file(str(docx_path), "md", format="docx")
def convert_docx_with_python_docx(docx_path: Path) -> str:
"""Convert DOCX to Markdown using python-docx (basic fallback)."""
doc = Document(docx_path)
content = []
for paragraph in doc.paragraphs:
# Simple paragraph to markdown conversion
text = paragraph.text.strip()
if text:
# Basic heading detection (you might want to improve this)
if paragraph.style.name.startswith("Heading"):
level = (
int(paragraph.style.name.split()[-1])
if paragraph.style.name.split()[-1].isdigit()
else 1
)
content.append("#" * level + " " + text)
else:
content.append(text)
content.append("") # Add blank line after each paragraph
# Handle tables
for table in doc.tables:
content.append("") # Add blank line before table
for i, row in enumerate(table.rows):
row_data = [cell.text.strip() for cell in row.cells]
content.append("| " + " | ".join(row_data) + " |")
if i == 0: # Add separator after header
content.append("|" + "---|" * len(row_data))
content.append("") # Add blank line after table
return "\n".join(content)
def convert_single_docx(
input_path: Path, output_dir: Path, method: str = "mammoth"
) -> bool:
"""
Convert a single DOCX file to Markdown.
Args:
input_path: Path to the DOCX file
output_dir: Directory to save the Markdown file
method: Conversion method ('mammoth', 'pandoc', or 'python-docx')
Returns:
True if successful, False otherwise
"""
try:
print(f"Converting: {input_path.name}")
# Choose conversion method
if method == "mammoth" and MAMMOTH_AVAILABLE:
markdown_content = convert_docx_with_mammoth(input_path)
elif method == "pandoc" and PANDOC_AVAILABLE:
markdown_content = convert_docx_with_pandoc(input_path)
elif method == "python-docx" and DOCX_AVAILABLE:
markdown_content = convert_docx_with_python_docx(input_path)
else:
print(f"Method '{method}' not available. Trying alternatives...")
if MAMMOTH_AVAILABLE:
markdown_content = convert_docx_with_mammoth(input_path)
elif PANDOC_AVAILABLE:
markdown_content = convert_docx_with_pandoc(input_path)
elif DOCX_AVAILABLE:
markdown_content = convert_docx_with_python_docx(input_path)
else:
print("No conversion libraries available!")
return False
# Clean and format the markdown
cleaned_content = clean_markdown_text(markdown_content)
# Create output file path
output_path = output_dir / f"{input_path.stem}.md"
# Write the markdown file
with open(output_path, "w", encoding="utf-8") as f:
f.write(cleaned_content)
print(f"✓ Saved: {output_path}")
return True
except Exception as e:
print(f"✗ Error converting {input_path.name}: {str(e)}")
return False
def convert_directory(
input_dir: str, output_dir: str, method: str = "mammoth", recursive: bool = False
):
"""
Convert all DOCX files in a directory to Markdown.
Args:
input_dir: Directory containing DOCX files
output_dir: Directory to save Markdown files
method: Conversion method to use
recursive: Whether to search subdirectories
"""
input_path = Path(input_dir)
output_path = Path(output_dir)
# Validate input directory
if not input_path.exists():
print(f"Error: Input directory '{input_dir}' does not exist.")
return
# Create output directory if it doesn't exist
output_path.mkdir(parents=True, exist_ok=True)
# Find all DOCX files
if recursive:
docx_files = list(input_path.rglob("*.docx"))
else:
docx_files = list(input_path.glob("*.docx"))
# Filter out temporary files (start with ~$)
docx_files = [f for f in docx_files if not f.name.startswith("~$")]
if not docx_files:
print("No DOCX files found in the specified directory.")
return
print(f"Found {len(docx_files)} DOCX files to convert...")
# Convert each file
successful = 0
failed = 0
for docx_file in docx_files:
if convert_single_docx(docx_file, output_path, method):
successful += 1
else:
failed += 1
print(f"\nConversion complete:")
print(f"✓ Successful: {successful}")
print(f"✗ Failed: {failed}")
def main():
parser = argparse.ArgumentParser(
description="Convert DOCX documents to Markdown format",
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog="""
Examples:
python docx_to_md.py /path/to/docx/files /path/to/output
python docx_to_md.py ./documents ./markdown --method pandoc --recursive
python docx_to_md.py ~/docs ~/markdown -r
""",
)
parser.add_argument("input_dir", help="Directory containing DOCX files")
parser.add_argument("output_dir", help="Directory to save Markdown files")
parser.add_argument(
"--method",
"-m",
choices=["mammoth", "pandoc", "python-docx"],
default="mammoth",
help="Conversion method (default: mammoth)",
)
parser.add_argument(
"--recursive",
"-r",
action="store_true",
help="Search subdirectories recursively",
)
args = parser.parse_args()
print("DOCX to Markdown Converter")
print("=" * 30)
print(f"Input directory: {args.input_dir}")
print(f"Output directory: {args.output_dir}")
print(f"Method: {args.method}")
print(f"Recursive: {args.recursive}")
print()
convert_directory(args.input_dir, args.output_dir, args.method, args.recursive)
if __name__ == "__main__":
main()