Update app.py

app.py

@@ -5,7 +5,6 @@ from byaldi import RAGMultiModalModel
 from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 from qwen_vl_utils import process_vision_info
 import torch
-import re
 
 # Load models
 def load_models():
@@ -54,39 +53,19 @@ def ocr_and_search(image, keyword):
     
     # Save extracted text to JSON
     output_json = {"query": text_query, "extracted_text": extracted_text}
-    # json_output = json.dumps(output_json, ensure_ascii=False, indent=4)
+    json_output = json.dumps(output_json, ensure_ascii=False, indent=4)
     
     # Perform keyword search
     keyword_lower = keyword.lower()
     sentences = extracted_text.split('. ')
     matched_sentences = [sentence for sentence in sentences if keyword_lower in sentence.lower()]
-
-
-    #########################################################################################################################################################
     
-    highlighted_text = matched_sentences
-    if search_queries:
-        keywords = [kw.strip() for kw in search_queries.split(",")]
-        for keyword in keywords:
-            highlighted_text = re.sub(
-                re.escape(keyword),
-                f"<span style='background-color: #f1c40f; color: #000; font-weight: bold;'>{keyword}</span>",
-                highlighted_text,
-                flags=re.IGNORECASE
-            )
-
-
-    ####################################################################################################################################################
-
-
-    return extracted_text, highlighted_text #, json_output
+    return extracted_text, matched_sentences #, json_output
 
 
 # Gradio App 
 def app(image, keyword):
     
-    # extracted_text, search_results, json_output = ocr_and_search(image, keyword)
-
     extracted_text, search_results = ocr_and_search(image, keyword)
     
     search_results_str = "\n".join(search_results) if search_results else "No matches found."