Update app.py

app.py

@@ -157,10 +157,17 @@ def identify_headers_with_openrouter(pdf_path, model, LLM_prompt, pages_to_check
                 y1 = spans[0]['bbox'][3]
                 # if y0 < top_margin or y1 > (page_height - bottom_margin):
                 #     continue
-                text = " ".join(s.get('text','') for s in spans).strip()
-                if text:
+                for s in spans:
+                    # text,font,size,flags,color
+                    ArrayofTextWithFormat={'Font':s.get('font')},{'Size':s.get('size')},{'Flags':s.get('flags')},{'Color':s.get('color')},{'Text':s.get('text')}
+                
                     # prefix with page for easier mapping back
-                    lines_for_prompt.append(f"PAGE {pno+1}: {text}")
+                    lines_for_prompt.append(f"PAGE {pno+1}: {ArrayofTextWithFormat}")
+
+                # text = " ".join(s.get('text','') for s in spans).strip()
+                # if text:
+                #     # prefix with page for easier mapping back
+                #     lines_for_prompt.append(f"PAGE {pno+1}: {text}")
                     lines_on_page += 1
         
         if lines_on_page > 0: