"""FoodHub Customer Service Chatbot — Streamlit frontend."""

import streamlit as st


# @st.cache_resource is Streamlit's standard pattern for initialising shared
# resources (models, DB connections, agents) that are expensive to create.
# It runs the function body exactly once per container lifetime and caches
# the result — all sessions and reruns reuse the same initialised agent.
@st.cache_resource
def get_chat_agent():
    try:
        from agent import run_chat_agent_query
        return run_chat_agent_query, None
    except Exception as e:
        return None, str(e)

# Assign at module level so the function is available as a normal callable.
# Streamlit's cache ensures the initialisation only runs once, not on every rerun.
run_chat_agent_query, _init_error = get_chat_agent()

# ---------------------------------------------------------------------------
# Page configuration
# ---------------------------------------------------------------------------
st.set_page_config(
    page_title="FoodHub Customer Service",
    page_icon="🍔",
    layout="centered",
)

st.title("🍔 FoodHub Customer Service")
st.caption(
    "Ask about your order status, delivery ETA, items, or payment. "
    "Please have your Order ID ready (e.g. O12486)."
)
st.divider()

if _init_error:
    st.error(f"Agent initialisation failed:\n\n```\n{_init_error}\n```")
    st.stop()

# ---------------------------------------------------------------------------
# Session state — persist conversation history across reruns
# ---------------------------------------------------------------------------
if "messages" not in st.session_state:
    st.session_state.messages = []

# ---------------------------------------------------------------------------
# Render existing conversation history
# ---------------------------------------------------------------------------
for message in st.session_state.messages:
    with st.chat_message(message["role"]):
        st.markdown(message["content"])

# ---------------------------------------------------------------------------
# Chat input
# ---------------------------------------------------------------------------
if prompt := st.chat_input("Type your question here..."):

    # Display the customer's message immediately
    st.session_state.messages.append({"role": "user", "content": prompt})
    with st.chat_message("user"):
        st.markdown(prompt)

    # Get the Chat Agent's response and display it
    with st.chat_message("assistant"):
        with st.spinner("Looking into your query..."):
            response = run_chat_agent_query(prompt)
        st.markdown(response)

    # Save assistant response to history
    st.session_state.messages.append({"role": "assistant", "content": response})