Update app.py

app.py

@@ -9,11 +9,39 @@ from qdrant_client import QdrantClient, models
 from langchain_openai import ChatOpenAI
 import gradio as gr
 import logging
+from typing import List, Tuple
+from dataclasses import dataclass
+from datetime import datetime
 
 # Configure logging
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
+@dataclass
+class Message:
+    role: str
+    content: str
+    timestamp: str
+
+class ChatHistory:
+    def __init__(self):
+        self.messages: List[Message] = []
+    
+    def add_message(self, role: str, content: str):
+        timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+        self.messages.append(Message(role=role, content=content, timestamp=timestamp))
+    
+    def get_formatted_history(self, max_messages: int = 5) -> str:
+        """Returns the most recent conversation history formatted as a string"""
+        recent_messages = self.messages[-max_messages:] if len(self.messages) > max_messages else self.messages
+        formatted_history = "\n".join([
+            f"{msg.role}: {msg.content}" for msg in recent_messages
+        ])
+        return formatted_history
+    
+    def clear(self):
+        self.messages = []
+
 # Load environment variables
 load_dotenv()
 
@@ -78,45 +106,104 @@ llm = ChatOpenAI(
     model="meta-llama/Llama-3.3-70B-Instruct"
 )
 
-# Create prompt template
+# Create prompt template with chat history
 template = """
-You are an expert assistant specializing in the Mawared HR System. Your task is to provide accurate and contextually relevant answers strictly based on the provided context. If the context lacks sufficient information, ask targeted clarifying questions to gather specific details required for a precise response. Always ensure clarity, brevity, and relevance in your answers.
-Context:
+You are an expert assistant specializing in the Mawared HR System. Your task is to provide accurate and contextually relevant answers based on the provided context and chat history. If you need more information, ask targeted clarifying questions.
+
+Previous Conversation:
+{chat_history}
+
+Current Context:
 {context}
-Question:
+
+Current Question:
 {question}
+
+Please provide a response that takes into account both the current context and the previous conversation history. If you refer to information from the chat history, make it clear where that information came from.
+
 Answer:
 """
 
 prompt = ChatPromptTemplate.from_template(template)
 
-# Create the RAG chain
-rag_chain = (
-    {"context": retriever, "question": RunnablePassthrough()}
-    | prompt
-    | llm
-    | StrOutputParser()
-)
+# Create the RAG chain with chat history
+def create_rag_chain(chat_history: str):
+    chain = (
+        {
+            "context": retriever,
+            "question": RunnablePassthrough(),
+            "chat_history": lambda x: chat_history
+        }
+        | prompt
+        | llm
+        | StrOutputParser()
+    )
+    return chain
+
+# Initialize chat history
+chat_history = ChatHistory()
 
 # Gradio Function
-def ask_question_gradio(question):
+def ask_question_gradio(question, history: gr.Chatbot) -> Tuple[str, gr.Chatbot]:
     try:
+        # Add user question to chat history
+        chat_history.add_message("User", question)
+        
+        # Get formatted history
+        formatted_history = chat_history.get_formatted_history()
+        
+        # Create chain with current chat history
+        rag_chain = create_rag_chain(formatted_history)
+        
+        # Generate response
         response = ""
         for chunk in rag_chain.stream(question):
             response += chunk
-        return response
+        
+        # Add assistant response to chat history
+        chat_history.add_message("Assistant", response)
+        
+        # Update Gradio chat history
+        history.append((question, response))
+        
+        return "", history
     except Exception as e:
         logger.error(f"Error during question processing: {e}")
-        return "An error occurred. Please try again later."
+        return "", history + [("Error", "An error occurred. Please try again later.")]
+
+def clear_chat():
+    chat_history.clear()
+    return None
 
 # Gradio Interface
-iface = gr.Interface(
-    fn=ask_question_gradio,
-    inputs=gr.Textbox(label="Ask a question about Mawared HR System:"),
-    outputs=gr.Textbox(label="Answer:"),
-    title="Mawared HR Assistant",
-    description="Ask questions about the Mawared HR system, and this assistant will provide answers based on the available context."
-)
+with gr.Blocks() as iface:
+    gr.Markdown("# Mawared HR Assistant")
+    gr.Markdown("Ask questions about the Mawared HR system, and this assistant will provide answers based on the available context and conversation history.")
+    
+    chatbot = gr.Chatbot(
+        height=400,
+        show_label=False,
+    )
+    
+    with gr.Row():
+        question_input = gr.Textbox(
+            label="Ask a question:",
+            placeholder="Type your question here...",
+            scale=9
+        )
+        clear_button = gr.Button("Clear Chat", scale=1)
+    
+    question_input.submit(
+        ask_question_gradio,
+        inputs=[question_input, chatbot],
+        outputs=[question_input, chatbot]
+    )
+    
+    clear_button.click(
+        lambda: (None, None),
+        outputs=[chatbot, question_input],
+        _js="() => { window.location.reload(); }"
+    )
 
 # Launch the Gradio App
 if __name__ == "__main__":