Update app.py

app.py

@@ -1,17 +1,19 @@
 import streamlit as st
 import os
-from langchain.document_loaders import TextLoader
+from langchain_community.document_loaders import TextLoader  # ✅ FIXED
 from langchain.text_splitter import CharacterTextSplitter
 from langchain_openai import OpenAIEmbeddings, ChatOpenAI
 from langchain.vectorstores import FAISS
 from langchain.prompts import ChatPromptTemplate
 from langchain_core.output_parsers import StrOutputParser
 
-os.environ["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"] = st.secrets["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"]  # HF secrets
+# HF Spaces secrets
+os.environ["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"] = st.secrets["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"]
 
 @st.cache_resource
 def load_retriever():
     if not os.path.exists("faiss_index"):
+        st.info("Creating vector index... (first run)")
         loader = TextLoader("knowledge.txt")
         docs = loader.load()
         splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
@@ -19,28 +21,45 @@ def load_retriever():
         embeddings = OpenAIEmbeddings()
         vectorstore = FAISS.from_documents(chunks, embeddings)
         vectorstore.save_local("faiss_index")
+    
     embeddings = OpenAIEmbeddings()
     vectorstore = FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
     return vectorstore.as_retriever(search_kwargs={"k": 3})
 
+# Load components
 retriever = load_retriever()
 llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
-prompt = ChatPromptTemplate.from_template("Context: {context}\nQuestion: {question}\nAnswer:")
+prompt = ChatPromptTemplate.from_template(
+    "Use this context only: {context}\n\nQuestion: {question}\nAnswer:"
+)
 
 def get_response(query):
     context_docs = retriever.invoke(query)
     context = "\n".join([doc.page_content for doc in context_docs])
-    chain = {"context": lambda _: context, "question": lambda _: query} | prompt | llm | StrOutputParser()
+    chain = (
+        {"context": lambda _: context, "question": lambda _: query} 
+        | prompt 
+        | llm 
+        | StrOutputParser()
+    )
     return chain.invoke(query)
 
-st.title("🧠 RAG Chatbot on HF")
-if "messages" not in st.session_state: st.session_state.messages = []
+# UI
+st.title("🧠 RAG Chatbot")
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
 for msg in st.session_state.messages:
-    with st.chat_message(msg["role"]): st.markdown(msg["content"])
-if prompt := st.chat_input():
+    with st.chat_message(msg["role"]):
+        st.markdown(msg["content"])
+
+if prompt := st.chat_input("Ask about your documents..."):
     st.session_state.messages.append({"role": "user", "content": prompt})
-    with st.chat_message("user"): st.markdown(prompt)
+    with st.chat_message("user"):
+        st.markdown(prompt)
+    
     with st.chat_message("assistant"):
-        response = get_response(prompt)
-        st.markdown(response)
+        with st.spinner("Thinking..."):
+            response = get_response(prompt)
+            st.markdown(response)
     st.session_state.messages.append({"role": "assistant", "content": response})