Update app.py

app.py

@@ -1,174 +1,119 @@
 import os
-import shutil
-import streamlit as st
-from io import BytesIO
+from getpass import getpass
+
+openai_api_key = os.getenv('OPENAI_API_KEY')
+openai_api_key = openai_api_key
+
 
-# Importing LlamaIndex components
 from llama_index.llms.openai import OpenAI
 from llama_index.embeddings.openai import OpenAIEmbedding
-from llama_index.core import Settings, SimpleDirectoryReader, VectorStoreIndex, StorageContext
-from llama_index.vector_stores.qdrant import QdrantVectorStore
-from llama_index.core.memory import ChatMemoryBuffer
-import qdrant_client
+from llama_index.core import Settings
 
-# =============================================================================
-# Configuration and Global Initialization
-# =============================================================================
+Settings.llm = OpenAI(model="gpt-3.5-turbo",temperature=0.4)
+Settings.embed_model = OpenAIEmbedding(model="text-embedding-ada-002")
 
-# Ensure that the OpenAI API key is available
-openai_api_key = os.getenv("OPENAI_API_KEY")
-if not openai_api_key:
-    raise ValueError("Please set your OPENAI_API_KEY environment variable.")
+from llama_index.core import SimpleDirectoryReader
 
-# System prompt for the chat engine
-SYSTEM_PROMPT = (
-    "You are an AI assistant who answers the user questions, "
-    "use the schema fields to generate appropriate and valid json queries"
-)
+documents = SimpleDirectoryReader("new_file").load_data()
 
-# Configure the LLM and embedding models
-Settings.llm = OpenAI(model="gpt-3.5-turbo", temperature=0.4)
-Settings.embed_model = OpenAIEmbedding(model="text-embedding-ada-002")
+from llama_index.core import VectorStoreIndex, StorageContext
+from llama_index.vector_stores.qdrant import QdrantVectorStore
+import qdrant_client
 
-# Load initial documents from a directory called "new_file"
-if os.path.exists("new_file"):
-    documents = SimpleDirectoryReader("new_file").load_data()
-else:
-    documents = []
+client = qdrant_client.QdrantClient(
+    location=":memory:",
+)
 
-# Set up the Qdrant vector store (using an in-memory collection for simplicity)
-client = qdrant_client.QdrantClient(location=":memory:")
 vector_store = QdrantVectorStore(
-    collection_name="paper",
+    collection_name = "paper",
     client=client,
     enable_hybrid=True,
     batch_size=20,
 )
+
 storage_context = StorageContext.from_defaults(vector_store=vector_store)
 
-# Build the initial index and chat engine
-index = VectorStoreIndex.from_documents(documents, storage_context=storage_context)
-chat_memory = ChatMemoryBuffer.from_defaults(token_limit=3000)
+index = VectorStoreIndex.from_documents(
+    documents,
+    storage_context=storage_context,
+)
+
+query_engine = index.as_query_engine(
+    vector_store_query_mode="hybrid"
+)
+
+from llama_index.core.memory import ChatMemoryBuffer
+
+memory = ChatMemoryBuffer.from_defaults(token_limit=3000)
+
 chat_engine = index.as_chat_engine(
     chat_mode="context",
-    memory=chat_memory,
-    system_prompt=SYSTEM_PROMPT,
+    memory=memory,
+    system_prompt=(
+        """You are an AI assistant who answers the user questions,
+           use the schema fields to generate appriopriate and valid json queries"""
+    ),
 )
 
-# =============================================================================
-# Helper Functions
-# =============================================================================
-
-def process_uploaded_file(uploaded_file: BytesIO) -> str:
-    """
-    Process the uploaded file:
-      1. Save the file to an "uploads" folder.
-      2. Copy it to a temporary folder ("temp_upload") for reading.
-      3. Update the global documents list and rebuild the index and chat engine.
-    """
-    if uploaded_file is None:
-        return "No file uploaded."
-
-    # Ensure the uploads directory exists
-    uploads_dir = "uploads"
-    os.makedirs(uploads_dir, exist_ok=True)
-
-    # Save the uploaded file locally
-    file_name = uploaded_file.name
-    dest_path = os.path.join(uploads_dir, file_name)
-    with open(dest_path, "wb") as f:
-        f.write(uploaded_file.getbuffer())
-
-    # Prepare a temporary directory for processing the file
-    temp_dir = "temp_upload"
-    os.makedirs(temp_dir, exist_ok=True)
-    # Clear any existing file in temp_upload directory
-    for f_name in os.listdir(temp_dir):
-        os.remove(os.path.join(temp_dir, f_name))
-    shutil.copy(dest_path, temp_dir)
-
-    # Load new document(s) from the temporary folder using SimpleDirectoryReader
-    new_docs = SimpleDirectoryReader(temp_dir).load_data()
-
-    # Update global documents and rebuild the index and chat engine
-    global documents, index, chat_engine
-    documents.extend(new_docs)
-    index = VectorStoreIndex.from_documents(documents, storage_context=storage_context)
-    chat_engine = index.as_chat_engine(
-        chat_mode="context",
-        memory=chat_memory,
-        system_prompt=SYSTEM_PROMPT,
-    )
-
-    return f"File '{file_name}' processed and added to the index."
-
-def chat_with_ai(user_input: str) -> str:
-    """
-    Send user input to the chat engine and return the response.
-    """
+# def is_greeting(user_input):
+
+#     greetings = ["hello", "hi", "hey", "good morning", "good afternoon", "good evening", "greetings"]
+#     user_input_lower = user_input.lower().strip()
+#     return any(greet in user_input_lower for greet in greetings)
+# def is_bye(user_input):
+
+#     greetings = ["thanks", "thanks you", "thanks a lot", "good answer", "good bye", "bye bye"]
+#     user_input_lower = user_input.lower().strip()
+#     return any(greet in user_input_lower for greet in greetings)
+import gradio as gr
+def chat_with_ai(user_input, chat_history):
+    # if is_greeting(user_input):
+    #     response = 'hi, how can i help you?'
+    #     chat_history.append((user_input, response))
+    #     return chat_history, ""
+    # elif is_bye(user_input):
+    #     response = "you're wlocome"
+    #     chat_history.append((user_input, response))
+    #     return chat_history, ""
     response = chat_engine.chat(user_input)
-    # Extract references from the response (if any)
     references = response.source_nodes
-    ref = []
-    for node in references:
-        if "file_name" in node.metadata and node.metadata["file_name"] not in ref:
-            ref.append(node.metadata["file_name"])
-    complete_response = str(response)
-    if ref:
-        complete_response += "\n\nReferences: " + ", ".join(ref)
-    return complete_response
-
-# =============================================================================
-# Streamlit App Layout
-# =============================================================================
-
-st.set_page_config(page_title="LlamaIndex Chat & File Upload", layout="wide")
-st.title("Chat Interface for LlamaIndex with File Upload")
-
-# Use Streamlit tabs for separate Chat and Upload functionalities
-tab1, tab2 = st.tabs(["Chat", "Upload"])
-
-# -----------------------------------------------------------------------------
-# Chat Tab
-# -----------------------------------------------------------------------------
-with tab1:
-    st.header("Chat with the AI")
-    # Initialize chat history in session state if it does not exist
-    if "chat_history" not in st.session_state:
-        st.session_state["chat_history"] = []
-
-    # Display conversation history
-    for chat in st.session_state["chat_history"]:
-        st.markdown(f"**User:** {chat[0]}")
-        st.markdown(f"**AI:** {chat[1]}")
-        st.markdown("---")
-
-    # Input text for user query
-    user_input = st.text_input("Enter your question:")
-
-    # When the "Send" button is clicked, process the chat
-    if st.button("Send") and user_input:
-        with st.spinner("Processing..."):
-            response = chat_with_ai(user_input)
-        st.session_state["chat_history"].append((user_input, response))
-        st.experimental_rerun()  # Refresh the page to show updated history
-
-    # Button to clear the conversation history
-    if st.button("Clear History"):
-        st.session_state["chat_history"] = []
-        st.experimental_rerun()
-
-# -----------------------------------------------------------------------------
-# Upload Tab
-# -----------------------------------------------------------------------------
-with tab2:
-    st.header("Upload a File")
-    uploaded_file = st.file_uploader("Choose a file to upload", type=["txt", "pdf", "doc", "docx", "csv", "xlsx"])
-    if st.button("Upload and Process"):
-        if uploaded_file is not None:
-            with st.spinner("Uploading and processing file..."):
-                status = process_uploaded_file(uploaded_file)
-            st.success(status)
-        else:
-            st.error("No file uploaded.")
+    ref,pages = [],[]
+    for i in range(len(references)):
+      if references[i].metadata['file_name'] not in ref:
+        ref.append(references[i].metadata['file_name'])
+      # pages.append(references[i].metadata['page_label'])
+    complete_response = str(response) + "\n\n" 
+    if ref !=[] or pages!=[]:
+      chat_history.append((user_input, complete_response))
+      ref = []
+    elif ref==[] or pages==[]:
+      chat_history.append((user_input,str(response)))
+        
+    return chat_history, ""
+
+def clear_history():
+    return [], ""
+
+def gradio_chatbot():
+    with gr.Blocks() as demo:
+        gr.Markdown("# Chat Interface for LlamaIndex")
+
+        chatbot = gr.Chatbot(label="LlamaIndex Chatbot")
+        user_input = gr.Textbox(
+            placeholder="Ask a question...", label="Enter your question"
+        )
+
+        submit_button = gr.Button("Send")
+        btn_clear = gr.Button("Delete Context") 
+
+
+        chat_history = gr.State([])
+
+        submit_button.click(chat_with_ai, inputs=[user_input, chat_history], outputs=[chatbot, user_input])
+
+        user_input.submit(chat_with_ai, inputs=[user_input, chat_history], outputs=[chatbot, user_input])
+        btn_clear.click(fn=clear_history, outputs=[chatbot, user_input])  
+
+    return demo
 
+gradio_chatbot().launch(debug=True)