document_processor.py

import io
import docx
from docx.enum.text import WD_PARAGRAPH_ALIGNMENT
from docx.shared import Pt, Inches

def extract_content_from_docx(file_path):
    """
    Extract content from a Word document organized by headings.
    This function can handle both:
    1. Standard Word heading styles (Heading 1, Heading 2, etc.)
    2. Markdown-style headings (# Heading, ## Subheading, etc.)
    
    Args:
        file_path (str): Path to the Word document
        
    Returns:
        list: A list of dictionaries containing heading level, heading text, and content
    """
    try:
        doc = docx.Document(file_path)
        document_structure = []
        current_heading = {"level": 1, "heading": "Untitled", "content": ""}
        current_content = []
        
        for para in doc.paragraphs:
            text = para.text.strip()
            
            # Check if paragraph is a Word heading style
            if para.style and para.style.name and para.style.name.startswith('Heading'):
                # Save the previous section regardless of content
                if current_heading:
                    current_heading["content"] = "\n".join(current_content)
                    document_structure.append(current_heading)
                    current_content = []
                
                # Start a new section
                # Get the last character and convert to int, default to 1 if conversion fails
                try:
                    heading_level = int(para.style.name[-1])  # Extract the number from "Heading X"
                except (IndexError, ValueError):
                    heading_level = 1  # Default to heading level 1 if extraction fails
                current_heading = {
                    "level": heading_level,
                    "heading": text,
                    "content": ""
                }
            
            # Check if paragraph is markdown-style heading (# Heading)
            elif text.startswith('#'):
                # Count the number of # to determine heading level
                level = 0
                for char in text:
                    if char == '#':
                        level += 1
                    else:
                        break
                
                # Only process if it's a valid heading (has content after the #)
                if level > 0 and level <= 6 and len(text) > level and text[level] == ' ':
                    # Save the previous section regardless of content
                    if current_heading:
                        current_heading["content"] = "\n".join(current_content)
                        document_structure.append(current_heading)
                        current_content = []
                    
                    # Extract the heading text without the # symbols
                    heading_text = text[level:].strip()
                    
                    # Start a new section
                    current_heading = {
                        "level": level,
                        "heading": heading_text,
                        "content": ""
                    }
                else:
                    # This is not a heading, just regular content with # in it
                    if text:
                        current_content.append(text)
            else:
                # Add paragraph to current content if it's not empty
                if text:
                    current_content.append(text)
        
        # Add the last section regardless of content
        if current_heading:
            current_heading["content"] = "\n".join(current_content)
            document_structure.append(current_heading)
        
        return document_structure
    
    except Exception as e:
        print(f"Error extracting content from docx: {str(e)}")
        return []


def create_newsletter_docx(document_structure):
    """
    Create a newsletter Word document from processed content.
    
    Args:
        document_structure (list): A list of dictionaries containing heading level, 
                                  heading text, and processed content
        
    Returns:
        bytes: The Word document as bytes
    """
    doc = docx.Document()
    
    # Set page margins
    sections = doc.sections
    for section in sections:
        section.top_margin = Inches(1)
        section.bottom_margin = Inches(1)
        section.left_margin = Inches(1)
        section.right_margin = Inches(1)
    
    # Add a title
    title = doc.add_heading("Internal Newsletter", 0)
    title.alignment = WD_PARAGRAPH_ALIGNMENT.CENTER
    
    # Add a horizontal line
    doc.add_paragraph().add_run("_" * 50).bold = True
    
    # Process each section
    for item in document_structure:
        heading_level = item["level"]
        heading_text = item["heading"]
        content = item.get("processed_content", item["content"])  # Use processed content if available
        
        # Add the heading
        doc.add_heading(heading_text, heading_level)
        
        # Add the content
        paragraphs = content.split('\n')
        for para_text in paragraphs:
            if para_text.strip():
                p = doc.add_paragraph()
                p.add_run(para_text)
        
        # Add some space after each section
        doc.add_paragraph()
    
    # Save the document to a bytes buffer
    buffer = io.BytesIO()
    doc.save(buffer)
    buffer.seek(0)
    
    return buffer.getvalue()