fixed ner output again

app.py

@@ -24,6 +24,13 @@ DEFAULT_NER_LABELS = "person, organization, location, date, event"
 #     "Qwen/Qwen2-VL-7B-Instruct": AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", trust_remote_code=True, torch_dtype="auto", _attn_implementation="flash_attention_2").cuda().eval()
 
 # }
+
+class TextWithMetadata(list):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args)
+        self.original_text = kwargs.get('original_text', '')
+        self.entities = kwargs.get('entities', [])
+
 def array_to_image_path(image_array):
     # Convert numpy array to PIL Image
     img = Image.fromarray(np.uint8(image_array))
@@ -140,16 +147,12 @@ def run_example(image, model_id="Qwen/Qwen2-VL-7B-Instruct", run_ner=False, ner_
         if last_end < len(ocr_text):
             highlighted_text.append((ocr_text[last_end:], None))
         
-        # Store the original text and entities as attributes of the highlighted_text list
-        highlighted_text.original_text = ocr_text
-        highlighted_text.entities = ner_results
-        
-        return highlighted_text
+        # Create TextWithMetadata instance with the highlighted text and metadata
+        result = TextWithMetadata(highlighted_text, original_text=ocr_text, entities=ner_results)
+        return result
     
     # If NER is disabled, return the text without highlighting
-    result = [(ocr_text, None)]
-    result.original_text = ocr_text
-    result.entities = []
+    result = TextWithMetadata([(ocr_text, None)], original_text=ocr_text, entities=[])
     return result
 
 css = """