# --- import statements remain unchanged ---
import streamlit as st
import requests
import pandas as pd
from gtts import gTTS
import base64
from io import BytesIO
from PIL import Image
import os
import plotly.express as px

# --- Config ---
st.set_page_config(page_title="NeuroPulse AI", page_icon="🧠", layout="wide")
logo_path = "logo.png"
if os.path.exists(logo_path):
    st.image(logo_path, width=180)

# --- Session State ---
if "review" not in st.session_state:
    st.session_state.review = ""
if "dark_mode" not in st.session_state:
    st.session_state.dark_mode = False
if "intelligence_mode" not in st.session_state:
    st.session_state.intelligence_mode = True
if "trigger_example_analysis" not in st.session_state:
    st.session_state.trigger_example_analysis = False

# --- Dark Mode Styling ---
if st.session_state.dark_mode:
    st.markdown("""
        <style>
        html, body, [class*="st-"] {
            background-color: #121212;
            color: #f5f5f5;
        }
        .stTextInput > div > div > input,
        .stTextArea > div > textarea,
        .stSelectbox div div,
        .stDownloadButton > button,
        .stButton > button {
            background-color: #1e1e1e;
            color: white;
        }
        </style>
    """, unsafe_allow_html=True)

# --- Sidebar ---
with st.sidebar:
    st.header("⚙️ Global Settings")
    st.session_state.dark_mode = st.toggle("🌙 Dark Mode", value=st.session_state.dark_mode)
    st.session_state.intelligence_mode = st.toggle("🧠 Intelligence Mode", value=st.session_state.intelligence_mode)

    DEFAULT_DEMO_TOKEN = "my-secret-key"
    api_token = st.text_input("🔐 API Token", value=DEFAULT_DEMO_TOKEN, type="password")
    if not api_token or api_token.strip() == "my-secret-key":
        st.warning("🧪 Running in demo mode — for full access, enter a valid API key.")

    backend_url = st.text_input("🌐 Backend URL", value="http://localhost:8000")

    sentiment_model = st.selectbox("📊 Sentiment Model", [
        "distilbert-base-uncased-finetuned-sst-2-english",
        "nlptown/bert-base-multilingual-uncased-sentiment"
    ])

    industry = st.selectbox("🏭 Industry", [
        "Auto-detect", "Generic", "E-commerce", "Healthcare", "Education", "Travel", "Banking", "Insurance",
        "Gaming", "Food Delivery", "Real Estate", "Fitness", "Entertainment"
    ])

    product_category = st.selectbox("🧩 Product Category", [
        "Auto-detect", "General", "Mobile Devices", "Laptops", "Healthcare Devices", "Banking App",
        "Travel Service", "Educational Tool", "Insurance Portal", "Streaming App", "Wearables",
        "Home Appliances", "Food Apps"
    ])

    use_aspects = st.checkbox("🔬 Enable Aspect Analysis")
    use_smart_summary = st.checkbox("🧠 Smart Summary (Single)")
    use_smart_summary_bulk = st.checkbox("🧠 Smart Summary for Bulk")
    use_explain_bulk = st.checkbox("🧠 Generate Explanations (Bulk)")
    verbosity = st.radio("🗣️ Response Style", ["Brief", "Detailed"])
    voice_lang = st.selectbox("🔈 Voice Language", ["en", "fr", "es", "de", "hi", "zh"])

# --- TTS ---
def speak(text, lang='en'):
    tts = gTTS(text, lang=lang)
    mp3 = BytesIO()
    tts.write_to_fp(mp3)
    b64 = base64.b64encode(mp3.getvalue()).decode()
    st.markdown(f'<audio controls><source src="data:audio/mp3;base64,{b64}" type="audio/mp3"></audio>', unsafe_allow_html=True)
    mp3.seek(0)
    return mp3

# --- Tabs ---
tab1, tab2 = st.tabs(["🧠 Single Review", "📚 Bulk CSV"])

# -------------------
# SINGLE REVIEW TAB
# -------------------
with tab1:
    st.title("🧠 NeuroPulse AI – Multimodal Review Analyzer")
    st.markdown("<div style='font-size:16px;color:#888;'>Minimum 20–50 words recommended.</div>", unsafe_allow_html=True)
    review = st.text_area("📝 Enter Review", value=st.session_state.review, height=180)

    col1, col2, col3 = st.columns(3)
    with col1:
        analyze = st.button("🔍 Analyze", use_container_width=True, disabled=not api_token)
    with col2:
        if st.button("🎲 Example", use_container_width=True):
            st.session_state.review = (
                "I love this phone! Super fast performance, great battery, and smooth UI. "
                "Camera is awesome too, though the price is a bit high. Overall, very happy."
            )
            st.session_state.trigger_example_analysis = True
            st.rerun()
    with col3:
        if st.button("🧹 Clear", use_container_width=True):
            st.session_state.review = ""
            st.rerun()

    if st.session_state.trigger_example_analysis and st.session_state.review:
        analyze = True
        st.session_state.trigger_example_analysis = False

    if analyze and review:
        if len(review.split()) < 20:
            st.warning("⚠️ Please enter at least 20 words.")
        else:
            with st.spinner("Analyzing..."):
                try:
                    payload = {
                        "text": review,
                        "model": sentiment_model,
                        "industry": industry,
                        "aspects": use_aspects,
                        "follow_up": None,
                        "product_category": product_category,
                        "verbosity": verbosity,
                        "intelligence": st.session_state.intelligence_mode,
                        "explain": True
                    }
                    headers = {"x-api-key": st.session_state.get("api_token", api_token)}
                    params = {"smart": "1"} if use_smart_summary else {}
                    res = requests.post(f"{backend_url}/analyze/", json=payload, headers=headers, params=params)

                    if res.status_code == 200:
                        data = res.json()
                        st.success("✅ Analysis Complete")
                        st.subheader("📌 Summary")
                        st.info(data["summary"])
                        st.caption(f"🧠 Summary Type: {'Smart' if use_smart_summary else 'Standard'} | {verbosity} Response")
                        st.markdown(f"**Context:** `{data['industry']}` | `{data['product_category']}` | `Web`")

                        st.subheader("🔊 Audio")
                        audio = speak(data["summary"], lang=voice_lang)
                        st.download_button("⬇️ Download Summary Audio", audio.read(), "summary.mp3", mime="audio/mp3")

                        st.metric("📊 Sentiment", data["sentiment"]["label"], delta=f"{data['sentiment']['score']:.2%}")
                        st.info(f"💢 Emotion: {data['emotion']}")

                        if data.get("explanation"):
                            st.subheader("🧮 Explanation")
                            st.markdown(data["explanation"])

                        st.markdown("### 🔁 Got questions?")
                        st.info("💬 Ask a follow-up question about this review.")
                        sample_questions = [
                            "What did the user like most?",
                            "Any complaints mentioned?",
                            "Is it positive overall?",
                            "What are the improvement areas?"
                        ]
                        selected_q = st.selectbox("💡 Sample Questions", ["Type your own..."] + sample_questions)
                        custom_q = st.text_input("🔍 Ask a follow-up", value="" if selected_q == "Type your own..." else selected_q)

                        if custom_q:
                            with st.spinner("Thinking..."):
                                payload["follow_up"] = custom_q
                                res = requests.post(f"{backend_url}/analyze/", json=payload, headers=headers, params=params)
                                if res.status_code == 200:
                                    follow = res.json().get("follow_up")
                                    if follow:
                                        st.subheader("🔁 Follow-Up Answer")
                                        if isinstance(follow, list):
                                            for q in follow:
                                                st.write("➡️", q)
                                        else:
                                            st.warning(follow)
                                else:
                                    st.error(f"❌ Follow-up failed: {res.json().get('detail')}")
                    else:
                        st.error(f"❌ API Error {res.status_code}: {res.json().get('detail', 'Unknown error')}")
                except Exception as e:
                    st.error(f"🚫 Exception occurred: {e}")

# -------------------
# BULK CSV TAB
# -------------------
with tab2:
    st.title("📚 Bulk CSV Upload")
    st.markdown("""
        Upload a CSV with the following columns:<br>
        <code>review</code> (required), <code>industry</code>, <code>product_category</code>, <code>device</code>, <code>follow_up</code> (optional)
    """, unsafe_allow_html=True)

    with st.expander("📄 Sample CSV"):
        with open("sample_reviews.csv", "rb") as f:
            st.download_button("⬇️ Download sample CSV", f, file_name="sample_reviews.csv")

    uploaded_file = st.file_uploader("📁 Upload your CSV", type="csv")

    if uploaded_file:
        if not api_token:
            st.error("🔐 Please enter your API token in the sidebar.")
        else:
            try:
                df = pd.read_csv(uploaded_file)
                if "review" not in df.columns:
                    st.error("CSV must contain a `review` column.")
                else:
                    st.success(f"✅ Loaded {len(df)} reviews")
                    for col in ["industry", "product_category", "device", "follow_up"]:
                        if col not in df.columns:
                            df[col] = ["Auto-detect"] * len(df)
                        df[col] = df[col].fillna("Auto-detect").astype(str)

                    df["industry"] = df["industry"].apply(lambda x: "Generic" if x.lower() == "auto-detect" else x)
                    df["product_category"] = df["product_category"].apply(lambda x: "General" if x.lower() == "auto-detect" else x)
                    df["device"] = df["device"].apply(lambda x: "Web" if x.lower() == "auto-detect" else x)

                    if st.button("📊 Analyze Bulk Reviews", use_container_width=True):
                        with st.spinner("Processing..."):
                            try:
                                payload = {
                                    "reviews": df["review"].tolist(),
                                    "model": sentiment_model,
                                    "aspects": use_aspects,
                                    "industry": df["industry"].tolist(),
                                    "product_category": df["product_category"].tolist(),
                                    "device": df["device"].tolist(),
                                    "follow_up": df["follow_up"].tolist(),
                                    "explain": use_explain_bulk,
                                    "intelligence": st.session_state.intelligence_mode,
                                }
                                res = requests.post(
                                    f"{backend_url}/bulk/?token={st.session_state.get('api_token', api_token)}",
                                    json=payload
                                )
                                if res.status_code == 200:
                                    results = pd.DataFrame(res.json()["results"])
                                    st.dataframe(results)
                                    if "sentiment" in results.columns:
                                        fig = px.pie(results, names="sentiment", title="Sentiment Distribution")
                                        st.plotly_chart(fig)
                                    st.download_button("⬇️ Download Results CSV", results.to_csv(index=False), "results.csv", mime="text/csv")
                                else:
                                    st.error(f"❌ Bulk Error {res.status_code}: {res.json().get('detail', 'Unknown error')}")
                            except Exception as e:
                                st.error(f"🚨 Processing Error: {e}")
            except Exception as e:
                st.error(f"❌ File Read Error: {e}")