Update app.py

app.py

@@ -6,35 +6,41 @@ from langchain_community.vectorstores import FAISS
 from langchain_google_genai import ChatGoogleGenerativeAI
 import re
 
-
 genai.configure(api_key="AIzaSyD2o8vjePJb6z8vT_PVe82lVWMD3_cBL0g")
 
-
 def format_gemini_response(text):
+    bold_pattern = r"\*\*(.*?)\*\*"
+    italic_pattern = r"\*(.*?)\*"
+    code_pattern = r"```(.*?)```"
+    text = text.replace('\n', '<br>')
+    formatted_text = re.sub(code_pattern, "<pre><code>\\1</code></pre>", text)
+    formatted_text = re.sub(bold_pattern, "<b>\\1</b>", formatted_text)
+    formatted_text = re.sub(italic_pattern, "<i>\\1</i>", formatted_text)
+    return formatted_text
+
+def predict(message: str, chat_his: List[List[str]], d: dict) -> str:
+    if not message.strip():
+        return "Error: Message cannot be empty.", chat_his, d
 
-  bold_pattern = r"\*\*(.*?)\*\*"
-  italic_pattern = r"\*(.*?)\*"
-  code_pattern = r"```(.*?)```"
-  text = text.replace('\n', '<br>')
-  formatted_text = re.sub(code_pattern,"<pre><code>\\1</code></pre>",text)
-  formatted_text = re.sub(bold_pattern, "<b>\\1</b>", formatted_text)
-  formatted_text = re.sub(italic_pattern, "<i>\\1</i>", formatted_text)
-
-  return formatted_text
-def predict(message :str , chat_his ,d ) -> str:
     model = genai.GenerativeModel("gemini-pro")
     his = []
-    # for i,j in history:
-    #     his.extend([
-    #         {"role": "user", "parts": i},
-    #         {"role": "model", "parts": j},
-    #     ])
-    chat = model.start_chat(
-        history=his
-    )
-    response = chat.send_message(message)
-    return format_gemini_response(response.text),chat_his, d
-iface = gr.Interface(fn = predict,inputs = ["text","list","json"],outputs = "text")
-iface.launch()
-
+    for i, j in chat_his:
+        his.extend([
+            {"role": "user", "parts": i},
+            {"role": "model", "parts": j},
+        ])
 
+    chat = model.start_chat(history=his)
+    response = chat.send_message(message)
+    
+    # Update chat history
+    chat_his.append((message, response.text))
+    
+    return format_gemini_response(response.text), chat_his, d
+
+iface = gr.Interface(
+    fn=predict,
+    inputs=["text", "list", "json"],
+    outputs="html"  # Change to HTML for proper rendering
+)
+iface.launch(share=True)