Update app.py

app.py

@@ -2,12 +2,30 @@ from typing import List
 
 import pytesseract
 from PIL import Image
+import re
 
 import gradio as gr
 
 def tesseract_ocr(filepath: str, languages: List[str]):
     image = Image.open(filepath)
-    return pytesseract.image_to_string(image=image, lang=', '.join(languages))
+    extracted_text =  pytesseract.image_to_string(image=image, lang=', '.join(languages))
+    return extracted_text
+
+def search_and_highlight(text, keyword):
+    highlighted_text = re.sub(f"({keyword})", r"<mark>\1</mark>", text, flags=re.IGNORECASE)
+    return highlighted_text
+
+def ocr_and_search(str, keyword):
+    if image is None:
+        return "Please upload an image."
+
+    extracted_text = tesseract_ocr(str)
+
+    if keyword:
+        highlighted_text = search_and_highlight(extracted_text, keyword)
+        return highlighted_text
+    else:
+        return extracted_text
 
 title = "Tesseract OCR"
 description = "Gradio demo for Tesseract."
@@ -18,7 +36,7 @@ demo = gr.Interface(
     fn=tesseract_ocr, 
     inputs=[
         gr.Image(type="filepath", label="Input"), 
-        gr.CheckboxGroup(language_choices, type="value", value=['eng'], label='language')
+        gr.Textbox(label="Keyword to Highlight", placeholder="Enter a keyword...")  # Keyword input
         ],
     outputs='text',
     title=title,