import streamlit as st
import requests
import pandas as pd

from functions import cosing_download
from functions_ui import download_pdf

#API_BASE = "https://api.cosmoguard.it/api/v1"
API_BASE = "http://localhost:8000/api/v1"

st.set_page_config(page_title="Ricerca Ingredienti", layout="wide")
st.title("Ricerca Ingredienti per CAS")

if "ingredient_data" not in st.session_state:
    st.session_state.ingredient_data = None
if "ingredient_cas" not in st.session_state:
    st.session_state.ingredient_cas = ""

cas_input = st.session_state.selected_cas
force_refresh = st.session_state.force_refresh if "force_refresh" in st.session_state else False

if cas_input:
    with st.spinner(f"Ricerca in corso per {cas_input}..."):
        try:
            resp = requests.post(
                f"{API_BASE}/ingredients/search",
                json={"cas": cas_input, "force": force_refresh},
                timeout=120,
            )
            result = resp.json()

            if result.get("success") and result.get("data"):
                st.session_state.ingredient_data = result["data"]
                st.session_state.ingredient_cas = cas_input
                st.success(f"Ingrediente {cas_input} trovato")
            else:
                st.session_state.ingredient_data = None
                st.error(result.get("error", f"Nessun dato per CAS {cas_input}"))

        except requests.ConnectionError:
            st.error("Impossibile connettersi all'API. Verifica che il server sia attivo.")
        except Exception as e:
            st.error(f"Errore: {e}")


data = st.session_state.ingredient_data

if data is None:
    st.info("Effettua una ricerca per visualizzare i risultati.")
else:
    cas = data.get("cas", "")
    st.subheader(f"Ingrediente: {cas}")

    # --- Header con INCI e data ---
    col_h1, col_h2 = st.columns(2)
    with col_h1:
        inci = data.get("inci") or []
        st.markdown(f"**INCI:** {', '.join(inci) if inci else 'N/A'}")
    with col_h2:
        st.markdown(f"**Data creazione:** {data.get('creation_date', 'N/A')}")

    st.markdown("---")

    # --- DAP Info ---
    st.subheader("DAP Info")
    dap = data.get("dap_info")
    if dap:
        col_d1, col_d2, col_d3 = st.columns(3)
        with col_d1:
            st.metric("Peso Molecolare (Da)", dap.get("molecular_weight", "N/A"))
            st.metric("Log Pow", dap.get("log_pow", "N/A"))
        with col_d2:
            st.metric("TPSA", dap.get("tpsa", "N/A"))
            st.metric("Punto Fusione (C)", dap.get("melting_point") or "N/A")
        with col_d3:
            dap_val = dap.get("dap_value", 0.5)
            st.metric("DAP Value", f"{dap_val * 100:.0f}%")
            st.metric("Alta Ionizzazione", dap.get("high_ionization") or "N/A")
    else:
        st.warning("Dati DAP non disponibili")

    st.markdown("---")

    # --- COSING Info ---
    st.subheader("COSING Info")
    cosing_list = data.get("cosing_info")
    if cosing_list:
        for i, cosing in enumerate(cosing_list):
            if len(cosing_list) > 1:
                st.markdown(f"**Voce {i + 1}**")

            col_c1, col_c2 = st.columns(2)
            with col_c1:
                names = cosing.get("common_names", [])
                st.markdown(f"**Nomi comuni:** {', '.join(names) if names else 'N/A'}")
                st.markdown(f"**INCI:** {', '.join(cosing.get('inci', [])) or 'N/A'}")
                st.markdown(f"**CAS:** {', '.join(cosing.get('cas', [])) or 'N/A'}")
                
                funcs = cosing.get("functionName", [])
                if funcs:
                    st.markdown("**Funzioni:**")
                    for f in funcs:
                        st.markdown(f"- {f}")
                
                ref_no = cosing.get("reference", "")
                if ref_no:
                    pdf_bytes = cosing_download(ref_no)

                    if isinstance(pdf_bytes, bytes):
                        st.download_button(
                            label="Download CosIng PDF",
                            data=pdf_bytes,
                            file_name=f"{cas_input}_cosing.pdf",
                            mime="application/pdf"
                        )
                    else:
                        st.error(pdf_bytes) 
                

            with col_c2:
                annex = cosing.get("annex", [])
                if annex:
                    st.markdown("**Annex/Restrizioni:**")
                    for a in annex:
                        st.markdown(f"- {a}")
                else:
                    st.markdown("**Annex:** Nessuna restrizione")

                other = cosing.get("otherRestrictions", [])
                if other:
                    st.markdown("**Altre restrizioni:**")
                    for o in other:
                        st.markdown(f"- {o}")

                restriction = cosing.get("cosmeticRestriction", "")
                if restriction:
                    st.warning(f"Restrizione cosmetica: {restriction}")
                
                link_opinions = cosing.get("sccsOpinionUrls", [])
                if link_opinions:
                    st.markdown("**SCCS Opinions:**")
                    for url in link_opinions:
                        st.link_button("View SCCS Opinion", url)

            if i < len(cosing_list) - 1:
                st.divider()
    else:
        st.warning("Dati COSING non disponibili")

    st.markdown("---")

    # --- Toxicity ---
    st.subheader("Tossicologia")
    tox = data.get("toxicity")
    if tox:
        # Best case highlight
        best = tox.get("best_case")
        if best:
            st.success(
                f"Miglior indicatore: **{best['indicator']}** = {best['value']} {best['unit']} "
                f"({best['route']}) — Fattore: {tox.get('factor', 'N/A')}"
            )
        
            download_pdf(
            casNo=cas_input,
            origin=best.get("source"),
            link=best.get("ref")
            )

        # Indicators table
        indicators = tox.get("indicators", [])
        if indicators:
            df = pd.DataFrame(indicators)
            col_order = ["indicator", "value", "unit", "route", "toxicity_type"]
            existing = [c for c in col_order if c in df.columns]
            df = df[existing]
            df.columns = ["Indicatore", "Valore", "Unita", "Via", "Tipo"][:len(existing)]
            st.dataframe(df, width="stretch", hide_index=True)
    else:
        st.warning("Dati tossicologici non disponibili")    

    st.markdown("---")

    # --- JSON completo ---
    with st.expander("JSON completo"):
        st.json(data)