Update app.py

app.py

@@ -10,17 +10,16 @@ import tempfile
 
 os.environ["OPENAI_API_KEY"] = "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"
 
-@st.cache_resource
-def load_vectorstore(_file_path):
+def load_vectorstore(file_path):
     if os.path.exists("faiss_index"):
         embeddings = OpenAIEmbeddings()
         return FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
     
-    file_ext = _file_path.split('.')[-1].lower()
+    file_ext = file_path.split('.')[-1].lower()
     if file_ext == 'pdf':
-        loader = PyPDFLoader(_file_path)
+        loader = PyPDFLoader(file_path)
     else:
-        loader = TextLoader(_file_path)
+        loader = TextLoader(file_path)
     
     docs = loader.load()
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
@@ -29,21 +28,13 @@ def load_vectorstore(_file_path):
     embeddings = OpenAIEmbeddings()
     vectorstore = FAISS.from_documents(splits, embeddings)
     vectorstore.save_local("faiss_index")
-    st.sidebar.success("✅ Document indexed!")
     return vectorstore
 
 def get_rag_chain(vectorstore):
     retriever = vectorstore.as_retriever(search_kwargs={"k": 4})
     llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
     
-    # FIXED: Clean single-line prompt
-    prompt_template = """You are a helpful assistant. Answer using ONLY this context: {context}
-
-Question: {question}
-
-Answer concisely. If not in context: "I don't have that information.""""
-    
-    prompt = ChatPromptTemplate.from_template(prompt_template)
+    prompt = ChatPromptTemplate.from_template("Context: {context}. Question: {question}. Answer using context only.")
     
     chain = (
         {"context": retriever, "question": lambda x: x} 
@@ -54,7 +45,7 @@ Answer concisely. If not in context: "I don't have that information.""""
     return chain
 
 st.title("🧠 Dynamic RAG Chatbot")
-st.markdown("**Upload PDF/TXT → Ask ANY question → Perfect answers!**")
+st.markdown("Upload PDF/TXT → Ask ANY question!")
 
 uploaded_file = st.file_uploader("📤 Upload PDF or TXT", type=['pdf', 'txt'])
 
@@ -65,19 +56,15 @@ if uploaded_file is not None:
     
     st.success(f"✅ Loaded: {uploaded_file.name}")
     
-    with st.spinner("🔄 Indexing document..."):
-        try:
-            vectorstore = load_vectorstore(file_path)
-            chain = get_rag_chain(vectorstore)
-            st.session_state.chain = chain
-            st.session_state.ready = True
-            st.session_state.doc_name = uploaded_file.name
-        except Exception as e:
-            st.error(f"Error: {str(e)}")
-            st.session_state.ready = False
+    with st.spinner("🔄 Indexing..."):
+        vectorstore = load_vectorstore(file_path)
+        chain = get_rag_chain(vectorstore)
+        st.session_state.chain = chain
+        st.session_state.ready = True
+        st.session_state.doc_name = uploaded_file.name
 
 if 'ready' in st.session_state and st.session_state.ready:
-    st.success(f"🚀 Ready! Document: **{st.session_state.doc_name}**")
+    st.success(f"🚀 Ready! Document: {st.session_state.doc_name}")
     
     if "messages" not in st.session_state:
         st.session_state.messages = []
@@ -92,7 +79,7 @@ if 'ready' in st.session_state and st.session_state.ready:
             st.markdown(query)
         
         with st.chat_message("assistant"):
-            with st.spinner("🔍 Searching..."):
+            with st.spinner("Searching..."):
                 response = st.session_state.chain.invoke(query)
                 st.markdown(response)
         
@@ -103,12 +90,4 @@ if 'ready' in st.session_state and st.session_state.ready:
         st.rerun()
 
 else:
-    st.info("👆 **Upload PDF/TXT to start!**")
-    st.markdown("**Test with:** Resume, notes, research papers...")
-
-with st.sidebar:
-    st.markdown("### ✅ Medium-Level RAG")
-    st.markdown("- Dynamic upload")
-    st.markdown("- Vector search") 
-    st.markdown("- FAISS indexing")
-    st.markdown("- LCEL pipeline")
+    st.info("👆 Upload PDF/TXT to start chatting!")