Update query.py

query.py

@@ -70,10 +70,12 @@ class VectaraQuery():
                 print(f"Received data chunk: {json.dumps(data, indent=2)}")  # Debugging line
 
                 if 'result' not in data:
+                    print("No 'result' in data")
                     continue
 
                 res = data['result']
                 if 'responseSet' not in res:
+                    print("No 'responseSet' in result")
                     continue
                 
                 response_set = res['responseSet']
@@ -81,15 +83,24 @@ class VectaraQuery():
                 if response_set:
                     for result in response_set['response']:
                         if 'text' not in result:
+                            print("No 'text' in result")
                             continue
                         text = result['text']
                         print(f"Processing text: {text}")  # Debugging line
 
                         # Adjusting regex patterns to be more flexible
-                        reason_match = re.search(r"Reason Why it Can't be Used:\s*(.*?)(?=\n|$)", text, re.DOTALL)
-                        alternative_match = re.search(r"Alternative:\s*(.*?)(?=\n|$)", text, re.DOTALL)
-                        notes_match = re.search(r"Notes:\s*(.*?)(?=\n|$)", text, re.DOTALL)
+                        reason_match = re.search(r"Reason Why it Can't be Used:\s*(.*?)(?:\n|$)", text, re.DOTALL)
+                        alternative_match = re.search(r"Alternative:\s*(.*?)(?:\n|$)", text, re.DOTALL)
+                        notes_match = re.search(r"Notes:\s*(.*?)(?:\n|$)", text, re.DOTALL)
                         
+                        # Improved regex to capture multiline fields
+                        if not reason_match:
+                            reason_match = re.search(r"DISCUSSION\s*-\s*(.*?)(?=\n|\r\n)", text, re.DOTALL)
+                        if not alternative_match:
+                            alternative_match = re.search(r"Alternative\s*:\s*(.*?)(?=\n|\r\n)", text, re.DOTALL)
+                        if not notes_match:
+                            notes_match = re.search(r"Notes\s*:\s*(.*?)(?=\n|\r\n)", text, re.DOTALL)
+
                         reason = reason_match.group(1).strip() if reason_match else "Not available"
                         alternative = alternative_match.group(1).strip() if alternative_match else "Not available"
                         notes = notes_match.group(1).strip() if notes_match else "Not available"