Update app.py

app.py

@@ -4,14 +4,22 @@ import mimetypes
 import os
 import threading
 import uvicorn
+import re
 
-# Start FastAPI server in a background thread
+# === Fix surrogate character issue ===
+def sanitize(text):
+    if isinstance(text, str):
+        text = re.sub(r'[\uD800-\uDFFF]', '', text)
+        return text.encode("utf-8", "ignore").decode("utf-8")
+    return text
+
+# Start FastAPI in a background thread
 def run_fastapi():
     uvicorn.run("main:app", host="0.0.0.0", port=8000)
 
 threading.Thread(target=run_fastapi, daemon=True).start()
 
-# Backend API URLs
+# Backend API endpoints
 FILE_API_URL = "http://127.0.0.1:8000/chat-with-file"
 URL_API_URL = "http://127.0.0.1:8000/chat-with-url"
 IMAGE_API_URL = "http://127.0.0.1:8000/extract-text-from-image"
@@ -19,6 +27,9 @@ AUDIO_API_URL = "http://127.0.0.1:8000/transcribe-audio"
 
 chat_sessions = {"file": [], "url": []}
 
+def format_chat(history):
+    return "\n\n".join([f"👤 {sanitize(q)}\n🤖 {sanitize(a)}" for q, a in history])
+
 def ask_from_file(file_obj, question, session_id):
     if not file_obj or not question.strip():
         return "⚠️ Please upload a file and enter a question.", chat_sessions["file"]
@@ -34,7 +45,7 @@ def ask_from_file(file_obj, question, session_id):
             context = "\n".join([f"Q: {q}\nA: {a}" for q, a in history]) + f"\nQ: {question}"
             data = {"question": context}
             response = requests.post(FILE_API_URL, files=files, data=data)
-            answer = response.json().get("answer", "⚠️ No answer returned.")
+            answer = sanitize(response.json().get("answer", "⚠️ No answer returned."))
             chat_sessions["file"].append((question, answer))
             return format_chat(chat_sessions["file"]), chat_sessions["file"]
     except Exception as e:
@@ -47,7 +58,7 @@ def ask_from_url(url, question, session_id):
         history = chat_sessions["url"]
         context = "\n".join([f"Q: {q}\nA: {a}" for q, a in history]) + f"\nQ: {question}"
         response = requests.post(URL_API_URL, json={"url": url, "question": context})
-        answer = response.json().get("answer", "⚠️ No answer returned.")
+        answer = sanitize(response.json().get("answer", "⚠️ No answer returned."))
         chat_sessions["url"].append((question, answer))
         return format_chat(chat_sessions["url"]), chat_sessions["url"]
     except Exception as e:
@@ -58,7 +69,7 @@ def extract_text_from_image(image_path):
         with open(image_path, "rb") as f:
             files = {"file": ("image.png", f, "image/png")}
             response = requests.post(IMAGE_API_URL, files=files)
-            return response.json().get("answer", "⚠️ No text extracted.")
+            return sanitize(response.json().get("answer", "⚠️ No text extracted."))
     except Exception as e:
         return f"❌ Error: {str(e)}"
 
@@ -67,13 +78,10 @@ def transcribe_audio(audio_path):
         with open(audio_path, "rb") as f:
             files = {"file": ("audio.wav", f, "audio/wav")}
             response = requests.post(AUDIO_API_URL, files=files)
-            return response.json().get("answer", "⚠️ No transcript returned.")
+            return sanitize(response.json().get("answer", "⚠️ No transcript returned."))
     except Exception as e:
         return f"❌ Error: {str(e)}"
 
-def format_chat(history):
-    return "\n\n".join([f"\ud83d\udc64 {q}\n\ud83e\udd16 {a}" for q, a in history])
-
 def clear_file_chat():
     chat_sessions["file"] = []
     return "", chat_sessions["file"]
@@ -122,9 +130,9 @@ h1 {
 """
 
 with gr.Blocks(css=custom_css) as demo:
-    gr.Markdown("# \ud83e\udd16 AI Chatbot with File, Web, Image, Audio & Chat History")
+    gr.Markdown("# 🤖 AI Chatbot with File, Web, Image, Audio & Chat History")
 
-    with gr.Tab("\ud83d\udcc2 Chat with File"):
+    with gr.Tab("📂 Chat with File"):
         file_input = gr.File(label="Upload File (.txt, .csv, .docx, .pdf)", file_types=[".txt", ".csv", ".docx", ".pdf"])
         file_question = gr.Textbox(label="Your Question", placeholder="Ask something based on file")
         file_button = gr.Button("Ask from File")
@@ -134,7 +142,7 @@ with gr.Blocks(css=custom_css) as demo:
         file_button.click(fn=ask_from_file, inputs=[file_input, file_question, gr.State("file")], outputs=[file_output, gr.State("file")])
         clear_file.click(fn=clear_file_chat, inputs=[], outputs=[file_output, gr.State("file")])
 
-    with gr.Tab("\ud83c\udf10 Chat with Website"):
+    with gr.Tab("🌐 Chat with Website"):
         url_input = gr.Textbox(label="Website URL", placeholder="https://example.com")
         url_question = gr.Textbox(label="Your Question", placeholder="Ask something based on webpage")
         url_button = gr.Button("Ask from URL")
@@ -144,16 +152,16 @@ with gr.Blocks(css=custom_css) as demo:
         url_button.click(fn=ask_from_url, inputs=[url_input, url_question, gr.State("url")], outputs=[url_output, gr.State("url")])
         clear_url.click(fn=clear_url_chat, inputs=[], outputs=[url_output, gr.State("url")])
 
-    with gr.Tab("\ud83d\uddbc Extract Text from Image"):
+    with gr.Tab("🖼️ Extract Text from Image"):
         image_input = gr.Image(label="Upload Image", type="filepath")
         image_button = gr.Button("Extract Text")
         image_output = gr.Textbox(label="Extracted Text", lines=8)
         image_button.click(fn=extract_text_from_image, inputs=image_input, outputs=image_output)
 
-    with gr.Tab("\ud83c\udfa4 Transcribe Audio"):
+    with gr.Tab("🎤 Transcribe Audio"):
         audio_input = gr.Audio(label="Upload Audio", type="filepath")
         audio_button = gr.Button("Transcribe Audio")
         audio_output = gr.Textbox(label="Transcript", lines=8)
         audio_button.click(fn=transcribe_audio, inputs=audio_input, outputs=audio_output)
 
-demo.launch(share=True)
+demo.launch(share=False)