Update main.py

main.py

@@ -543,6 +543,26 @@ def write_markdown_document(pdf_path: Path, out_dir: Path) -> Optional[Path]:
         logger.warning(f"  No textual content extracted from {pdf_path.name}")
         return None
 
+    # --- FORMATTING POST-PROCESSING ---
+    import re
+    # 0. Normalize Windows CRLF to Unix LF so regex works correctly
+    markdown_content = markdown_content.replace('\r\n', '\n').replace('\r', '\n')
+    
+    # 1. Ensure headers have a blank line before and after them
+    markdown_content = re.sub(r'^(#{1,6}\s+.*)$', r'\n\n\1\n\n', markdown_content, flags=re.MULTILINE)
+    
+    # 2. Fix arbitrary hard line breaks in paragraphs inside sentences
+    markdown_content = re.sub(r'([a-z,:\-])\n([a-zA-Z])', r'\1 \2', markdown_content)
+    
+    # 3. Ensure double newlines between sentences across lines
+    markdown_content = re.sub(r'([.?!])\n([A-Z])', r'\1\n\n\2', markdown_content)
+    
+    # 4. Clean up excessive newlines
+    markdown_content = re.sub(r'\n{3,}', '\n\n', markdown_content)
+    
+    markdown_content = markdown_content.strip()
+    # ----------------------------------
+
     if not markdown_content.endswith("\n"):
         markdown_content += "\n"