Update app.py

app.py

@@ -672,6 +672,25 @@ Grading JSON:
         print(f"❌ Failed to parse Gemini JSON mapping: {e}")
         return [] 
 
+def normalize_question_id(qid):
+    """
+    Normalize question ID to a standard format for matching.
+    Converts formats like:
+    - "1(a)" -> "1.a"
+    - "2(c).i" -> "2.c.i"
+    - "3.d.ii" -> "3.d.ii" (already normalized)
+    """
+    if not qid:
+        return qid
+    
+    # Replace parentheses format: 1(a) -> 1.a
+    qid = re.sub(r'(\d+)\(([a-zA-Z])\)', r'\1.\2', qid)
+    
+    # Replace format like 2(c).i -> 2.c.i
+    qid = re.sub(r'(\d+)\(([a-zA-Z]+)\)\.', r'\1.\2.', qid)
+    
+    return qid
+
 def imprint_marks_using_mapping(pdf_path, grading_json, output_pdf, expected_ids=None, rows=GRID_ROWS, cols=GRID_COLS): 
     """ 
     Convert PDF to images, create grid-numbered images for batch sending to Gemini, 
@@ -737,12 +756,27 @@ def imprint_marks_using_mapping(pdf_path, grading_json, output_pdf, expected_ids
             if qid is None or cell_number is None: 
                 continue 
 
-            marks_list = next((g["marks_awarded"] for g in grading_json.get("grading", []) if g["question"] == qid), []) 
-            if not marks_list: 
+            # Normalize the question ID from Gemini mapping
+            normalized_qid = normalize_question_id(qid)
+            
+            # Try exact match first with normalized ID
+            marks_list = next((g["marks_awarded"] for g in grading_json.get("grading", []) 
+                              if g["question"] == normalized_qid), [])
+            
+            # If no match, try case-insensitive match
+            if not marks_list:
                 marks_list = next((g["marks_awarded"] for g in grading_json.get("grading", []) 
-                                   if g["question"].lower() == (qid or "").lower()), []) 
+                                   if g["question"].lower() == normalized_qid.lower()), [])
+            
+            # If still no match, try with original qid
+            if not marks_list:
+                marks_list = next((g["marks_awarded"] for g in grading_json.get("grading", []) 
+                                  if g["question"] == qid), [])
 
-            marks_text = ",".join(marks_list) if marks_list else "?" 
+            marks_text = ",".join(marks_list) if marks_list else "?"
+            
+            if marks_text == "?":
+                print(f"⚠️ No marks found for question '{qid}' (normalized: '{normalized_qid}') on page {page_num}") 
 
             row = (cell_number - 1) // cols 
             col = (cell_number - 1) % cols