"""
Intelligent Knowledge Retrieval System - Minimal Frontend
"""

import streamlit as st
import time
from guards import input as input_guard, output as output_guard
from rag import retriever

# ============================================
# BACKEND INTEGRATION
# ============================================

def mock_llm(query: str, context: str) -> str:
    """
    Placeholder for the LLM generation.
    """
    return f"Based on the context, the answer to '{query}' is likely found within the retrieved documents:\n\n{context}"

def query_rag_pipeline(user_query: str) -> dict:
    start_time = time.time()

    # 1. Input Guardrails
    if not input_guard.is_valid(user_query):
        return {"answer": "Invalid query. Please try again.", "sources": []}

    # 2. Retrieval
    retrieved_docs = retriever.search(user_query)

    # 3. Context Generation
    context = "\n".join([str(doc) for doc in retrieved_docs])

    # 4. LLM Generation (Mock)
    raw_answer = mock_llm(query=user_query, context=context)

    # 5. Output Guardrails
    clean_answer = output_guard.sanitize(raw_answer)

    # 6. Final Response
    end_time = time.time()
    return {
        "answer": clean_answer,
        "sources": [{"title": str(doc), "relevance": 1.0} for doc in retrieved_docs],
        "processing_time": end_time - start_time
    }

# ============================================
# HAUPTANWENDUNG
# ============================================

def main():
    st.set_page_config(
        page_title="Knowledge Retrieval System",
        page_icon="🤖",
        layout="centered"
    )
    
    # Header
    st.title("🤖 Intelligent Knowledge Retrieval System")
    st.markdown("Stelle Fragen in natürlicher Sprache - das System durchsucht die Wissensdatenbank und generiert eine Antwort.")
    st.markdown("---")
    
    # Chat-Historie initialisieren
    if "messages" not in st.session_state:
        st.session_state.messages = []
        # Willkommensnachricht
        st.session_state.messages.append({
            "role": "assistant",
            "content": "Hallo! Ich kann dir bei Fragen zu unserer Wissensdatenbank helfen. Was möchtest du wissen?"
        })
    
    # Chat-Historie anzeigen
    for message in st.session_state.messages:
        with st.chat_message(message["role"]):
            st.write(message["content"])
            
            # Zeige Quellen wenn vorhanden
            if message.get("sources"):
                with st.expander("📚 Verwendete Quellen"):
                    for source in message["sources"]:
                        relevance_percent = source["relevance"] * 100
                        st.write(f"• {source['title']} (Relevanz: {relevance_percent:.0f}%)")
    
    # Chat-Eingabe
    if prompt := st.chat_input("Stelle eine Frage..."):
        # Benutzer-Nachricht hinzufügen und anzeigen
        st.session_state.messages.append({"role": "user", "content": prompt})
        with st.chat_message("user"):
            st.write(prompt)
        
        # RAG-Antwort generieren
        with st.chat_message("assistant"):
            with st.spinner("Durchsuche Datenbank und generiere Antwort..."):
                # RAG Pipeline aufrufen
                response = query_rag_pipeline(prompt)
                
                # Antwort anzeigen
                st.write(response["answer"])
                
                # Antwort in Historie speichern
                st.session_state.messages.append({
                    "role": "assistant",
                    "content": response["answer"],
                    "sources": response.get("sources", [])
                })
                
                # Quellen anzeigen wenn vorhanden
                if response.get("sources"):
                    with st.expander("📚 Verwendete Quellen"):
                        for source in response["sources"]:
                            relevance_percent = source["relevance"] * 100
                            st.write(f"• {source['title']} (Relevanz: {relevance_percent:.0f}%)")
    
    # Footer mit Info
    st.markdown("---")
    with st.expander("ℹ️ Für Entwickler"):
        st.markdown("""
        **Integration der Backend-Komponenten:**
        
        Ersetzt die Funktion `query_rag_pipeline()` mit eurer echten Implementation:
        
        ```python
        def query_rag_pipeline(user_query: str) -> dict:
            # 1. Retrieval
            documents = database.search(user_query, top_k=5)
            
            # 2. Context erstellen
            context = format_documents(documents)
            
            # 3. LLM Generation
            answer = llm.generate(user_query, context)
            
            return {
                "answer": answer,
                "sources": documents
            }
        ```
        
        **Erwartete Response-Struktur:**
        - `answer`: String mit der generierten Antwort
        - `sources`: Liste von Dokumenten (optional)
        - `processing_time`: Verarbeitungszeit in Sekunden (optional)
        """)

if __name__ == "__main__":
    main()