#!/usr/bin/env python3 """ Advanced markdown to PDF converter that renders Mermaid diagrams using Playwright. Creates actual visual diagrams in the PDF output. """ import markdown import re import sys import asyncio import tempfile import base64 from pathlib import Path from playwright.async_api import async_playwright def extract_mermaid_diagrams(md_content): """Extract Mermaid diagrams from markdown content.""" # Pattern to match Mermaid code blocks mermaid_pattern = r'```mermaid\n(.*?)\n```' diagrams = [] matches = re.finditer(mermaid_pattern, md_content, flags=re.DOTALL) for i, match in enumerate(matches): diagram_content = match.group(1).strip() diagrams.append({ 'index': i, 'content': diagram_content, 'full_match': match.group(0) }) return diagrams async def render_mermaid_to_image(diagram_content, output_path): """Render a Mermaid diagram to PNG using Playwright.""" html_template = f"""
{diagram_content}
""" try: async with async_playwright() as p: browser = await p.chromium.launch() page = await browser.new_page() # Set content and wait for Mermaid to render await page.set_content(html_template) await page.wait_for_timeout(2000) # Wait for Mermaid to render # Find the SVG element and take screenshot svg_element = await page.query_selector('.mermaid svg') if svg_element: await svg_element.screenshot(path=str(output_path)) await browser.close() return True else: print(f"Could not find rendered SVG for diagram") await browser.close() return False except Exception as e: print(f"Error rendering Mermaid diagram: {e}") return False async def process_mermaid_diagrams(md_content, temp_dir): """Process all Mermaid diagrams and replace with image references.""" diagrams = extract_mermaid_diagrams(md_content) processed_content = md_content # Process diagrams in reverse order to maintain string positions for diagram in reversed(diagrams): png_filename = f"diagram_{diagram['index']}.png" png_path = temp_dir / png_filename print(f"Rendering diagram {diagram['index']}...") # Try to render the diagram if await render_mermaid_to_image(diagram['content'], png_path): # Convert image to base64 for embedding with open(png_path, 'rb') as img_file: img_data = base64.b64encode(img_file.read()).decode() # Replace with embedded image img_tag = f'''
Mermaid Diagram {diagram['index']}
''' processed_content = processed_content.replace(diagram['full_match'], img_tag) else: # Fallback to styled code block if rendering fails styled_fallback = f'''
Diagram {diagram['index']} (Rendering Failed)
{diagram['content']}
''' processed_content = processed_content.replace(diagram['full_match'], styled_fallback) return processed_content def process_manual_breaks(html_content): """Convert manual page break markers to proper CSS page breaks.""" # Pattern to match our manual page break markers page_break_pattern = r'[\s\n]*

PAGE BREAK

[\s\n]*' # Replace with proper CSS page break page_break_replacement = '
' processed_html = re.sub(page_break_pattern, page_break_replacement, html_content, flags=re.IGNORECASE | re.MULTILINE) return processed_html def create_html_template(content): """Create a complete HTML document with PDF-optimized CSS.""" html_template = f""" A Four-Tiered Cognitive Architecture for Advanced AI Reasoning {content} """ return html_template async def convert_md_to_pdf_with_rendered_mermaid(input_file, output_file=None): """Convert markdown to PDF with rendered Mermaid diagrams.""" input_path = Path(input_file) if not input_path.exists(): print(f"Error: Input file {input_file} not found") return False if output_file is None: output_file = input_path.with_suffix('.pdf') output_path = Path(output_file) print(f"Converting {input_file} to {output_file}") try: # Create temporary directory for images with tempfile.TemporaryDirectory() as temp_dir: temp_path = Path(temp_dir) # Read markdown content with open(input_path, 'r', encoding='utf-8') as f: md_content = f.read() print("Processing Mermaid diagrams...") # Process Mermaid diagrams first md_content = await process_mermaid_diagrams(md_content, temp_path) # Convert markdown to HTML md = markdown.Markdown(extensions=['codehilite', 'fenced_code', 'tables']) html_content = md.convert(md_content) # Process manual page breaks html_content = process_manual_breaks(html_content) # Create complete HTML document full_html = create_html_template(html_content) # Save HTML file for inspection html_file = input_path.stem + '_rendered.html' with open(html_file, 'w', encoding='utf-8') as f: f.write(full_html) print(f"HTML file saved as: {html_file}") # Convert HTML to PDF using WeasyPrint try: import weasyprint print("Converting HTML to PDF using WeasyPrint...") weasyprint.HTML(filename=str(html_file)).write_pdf(str(output_path)) print(f"PDF successfully created: {output_path}") return True except ImportError: print("WeasyPrint not available. Install with: pip install weasyprint") return False except Exception as e: print(f"WeasyPrint conversion failed: {e}") return False except Exception as e: print(f"Error during conversion: {e}") return False def main(): if len(sys.argv) < 2: print("Usage: python md_to_pdf_with_rendered_mermaid.py [output.pdf]") print("\nRequirements:") print(" pip install weasyprint markdown playwright") print(" playwright install chromium") sys.exit(1) input_file = sys.argv[1] output_file = sys.argv[2] if len(sys.argv) > 2 else None success = asyncio.run(convert_md_to_pdf_with_rendered_mermaid(input_file, output_file)) sys.exit(0 if success else 1) if __name__ == "__main__": main()