import os
import json
import requests
import pandas as pd
import gradio as gr

APP_NAME = "StayWise AI"

DATA_FILE = "synthetic_airbnb_project_data.csv"
N8N_WEBHOOK_URL = os.getenv("N8N_WEBHOOK_URL", "").strip()


def load_data():
    return pd.read_csv(DATA_FILE)


df = load_data()


def safe_mean(series, default=0):
    series = pd.to_numeric(series, errors="coerce").dropna()
    if len(series) == 0:
        return default
    return float(series.mean())


def get_choices(column):
    return sorted(df[column].dropna().astype(str).unique().tolist())


neighbourhood_groups = get_choices("neighbourhood_group")
neighbourhoods = get_choices("neighbourhood")
room_types = get_choices("room_type")
seasons = get_choices("season") if "season" in df.columns else ["Low Season", "Medium Season", "High Season"]


def call_n8n(payload):
    if not N8N_WEBHOOK_URL:
        return {
            "status": "not_configured",
            "insight": "n8n webhook is not configured yet.",
            "next_step": "Add N8N_WEBHOOK_URL in Hugging Face Space secrets.",
            "log": "Pipeline ran locally only."
        }

    try:
        response = requests.post(N8N_WEBHOOK_URL, json=payload, timeout=15)

        if response.status_code >= 200 and response.status_code < 300:
            try:
                data = response.json()
                return {
                    "status": data.get("status", "success"),
                    "insight": data.get("insight", "n8n processed the recommendation."),
                    "next_step": data.get("next_step", "Review the generated automation output."),
                    "log": data.get("log", "Automation completed.")
                }
            except Exception:
                return {
                    "status": "success",
                    "insight": "n8n received the data.",
                    "next_step": "Check your n8n workflow output.",
                    "log": response.text[:500]
                }

        return {
            "status": "error",
            "insight": f"n8n returned status code {response.status_code}.",
            "next_step": "Check your webhook and Respond to Webhook node.",
            "log": response.text[:500]
        }

    except Exception as e:
        return {
            "status": "error",
            "insight": "The app could not reach n8n.",
            "next_step": "Check that the n8n production webhook is active.",
            "log": str(e)
        }


def run_pipeline(
    neighbourhood_group,
    neighbourhood,
    room_type,
    price,
    availability_365,
    season,
    local_event_score,
    rating,
    sentiment_score,
    send_to_n8n
):
    price = float(price)
    availability_365 = float(availability_365)
    local_event_score = float(local_event_score)
    rating = float(rating)
    sentiment_score = float(sentiment_score)

    comparable = df.copy()

    comparable = comparable[
        (comparable["neighbourhood_group"].astype(str) == str(neighbourhood_group)) &
        (comparable["room_type"].astype(str) == str(room_type))
    ]

    local_comparable = comparable[comparable["neighbourhood"].astype(str) == str(neighbourhood)]

    if len(local_comparable) >= 5:
        comparable = local_comparable

    if len(comparable) < 5:
        comparable = df[df["room_type"].astype(str) == str(room_type)]

    competitor_avg_price = safe_mean(comparable["price"], price)
    avg_occupancy = safe_mean(comparable["occupancy_rate"], 0.5)
    avg_demand = safe_mean(comparable["demand_score"], 50)

    price_gap_pct = ((price - competitor_avg_price) / competitor_avg_price) * 100 if competitor_avg_price else 0

    season_boost = {
        "Low Season": -0.08,
        "Medium Season": 0.00,
        "High Season": 0.08,
        "Low": -0.08,
        "Medium": 0.00,
        "High": 0.08,
        "Peak": 0.12,
        "Spring": 0.03,
        "Summer": 0.08,
        "Autumn": 0.00,
        "Winter": -0.05
    }.get(str(season), 0.00)

    price_penalty = max(min(price_gap_pct / 100, 0.35), -0.35) * 0.30
    event_boost = (local_event_score / 100) * 0.12
    rating_boost = (rating - 4.0) * 0.06
    sentiment_boost = sentiment_score * 0.08

    occupancy = avg_occupancy + season_boost + event_boost + rating_boost + sentiment_boost - price_penalty
    occupancy = max(0.05, min(0.95, occupancy))

    booked_nights = round(occupancy * 30)
    monthly_revenue = round(price * booked_nights, 2)

    demand_score = (
        0.45 * avg_demand +
        0.25 * (occupancy * 100) +
        0.15 * local_event_score +
        0.10 * ((rating / 5) * 100) +
        0.05 * ((sentiment_score + 1) / 2 * 100)
    )

    demand_score = round(max(0, min(100, demand_score)), 1)

    if demand_score >= 70:
        demand_level = "High"
        demand_badge = "🟢 High"
    elif demand_score >= 45:
        demand_level = "Medium"
        demand_badge = "🟡 Medium"
    else:
        demand_level = "Low"
        demand_badge = "🔴 Low"

    if price_gap_pct > 15 and demand_level != "High":
        pricing_recommendation = "Consider lowering price"
        suggested_price = round(competitor_avg_price * 1.05, 2)
        insight = "The listing appears overpriced compared with similar properties."
        next_step = f"Test a lower price around ${suggested_price} to improve occupancy."
        recommendation_badge = "🔻 Price Reduction Suggested"
    elif price_gap_pct < -10 and demand_level in ["Medium", "High"]:
        pricing_recommendation = "Consider raising price"
        suggested_price = round(min(competitor_avg_price * 0.98, price * 1.12), 2)
        insight = "The listing appears underpriced relative to comparable demand."
        next_step = f"Consider increasing the price toward ${suggested_price}."
        recommendation_badge = "🚀 Revenue Opportunity"
    else:
        pricing_recommendation = "Keep price stable"
        suggested_price = round(price, 2)
        insight = "The current price is aligned with comparable listings."
        next_step = "Keep price stable and focus on visibility, reviews, and conversion."
        recommendation_badge = "✅ Stable Positioning"

    opportunity_score = round(
        demand_score * 0.45 +
        occupancy * 100 * 0.25 +
        rating / 5 * 100 * 0.15 +
        ((sentiment_score + 1) / 2 * 100) * 0.15,
        2
    )

    payload = {
        "app_name": APP_NAME,
        "neighbourhood_group": neighbourhood_group,
        "neighbourhood": neighbourhood,
        "room_type": room_type,
        "current_price": price,
        "suggested_price": suggested_price,
        "competitor_avg_price": round(competitor_avg_price, 2),
        "price_vs_competitor_pct": round(price_gap_pct, 2),
        "occupancy_estimate": round(occupancy, 3),
        "booked_nights_month": booked_nights,
        "monthly_revenue": monthly_revenue,
        "demand_score": demand_score,
        "demand_level": demand_level,
        "opportunity_score": opportunity_score,
        "pricing_recommendation": pricing_recommendation,
        "insight": insight,
        "next_step": next_step
    }

    if send_to_n8n:
        n8n_response = call_n8n(payload)
    else:
        n8n_response = {
            "status": "not_sent",
            "insight": "n8n automation was not triggered.",
            "next_step": "Tick the n8n checkbox to send this result to the workflow.",
            "log": "Local pipeline only."
        }

    result_text = f"""
<div class="result-card">

<div class="section-label">Pipeline Result</div>

<h2>{recommendation_badge}</h2>

<div class="kpi-grid">
    <div class="kpi-card">
        <div class="kpi-title">Current Price</div>
        <div class="kpi-value">${price:,.0f}</div>
    </div>
    <div class="kpi-card">
        <div class="kpi-title">Suggested Price</div>
        <div class="kpi-value">${suggested_price:,.0f}</div>
    </div>
    <div class="kpi-card">
        <div class="kpi-title">Monthly Revenue</div>
        <div class="kpi-value">${monthly_revenue:,.0f}</div>
    </div>
    <div class="kpi-card">
        <div class="kpi-title">Demand</div>
        <div class="kpi-value">{demand_badge}</div>
    </div>
</div>

<h3>Key Metrics</h3>

<table class="metric-table">
<tr><td>Competitor average price</td><td>${competitor_avg_price:,.2f}</td></tr>
<tr><td>Price vs competitors</td><td>{price_gap_pct:.2f}%</td></tr>
<tr><td>Estimated occupancy</td><td>{occupancy * 100:.1f}%</td></tr>
<tr><td>Estimated booked nights / month</td><td>{booked_nights}</td></tr>
<tr><td>Demand score</td><td>{demand_score}/100</td></tr>
<tr><td>Opportunity score</td><td>{opportunity_score}/100</td></tr>
</table>

<h3>Business Insight</h3>
<p>{insight}</p>

<h3>Next Step</h3>
<p>{next_step}</p>

</div>
"""

    automation_text = f"""
<div class="automation-card">

<div class="section-label">n8n Automation Output</div>

<h2>Automation Status: {n8n_response.get("status", "unknown")}</h2>

<table class="metric-table">
<tr><td>Insight</td><td>{n8n_response.get("insight", "No insight returned.")}</td></tr>
<tr><td>Next step</td><td>{n8n_response.get("next_step", "No next step returned.")}</td></tr>
<tr><td>Log</td><td>{n8n_response.get("log", "No log returned.")}</td></tr>
</table>

</div>
"""

    cols = [
        "id",
        "name",
        "neighbourhood_group",
        "neighbourhood",
        "room_type",
        "price",
        "occupancy_rate",
        "monthly_revenue",
        "demand_score",
        "demand_level",
        "pricing_recommendation"
    ]

    available_cols = [c for c in cols if c in comparable.columns]
    comparable_table = comparable[available_cols].head(10)

    json_output = json.dumps(payload, indent=2)

    return result_text, automation_text, comparable_table, json_output


custom_css = """
/* Main background */
body {
    background: radial-gradient(circle at top left, #1e3a8a 0%, #0f172a 35%, #020617 100%) !important;
    color: #f8fafc !important;
}

/* Main container */
.gradio-container {
    max-width: 1250px !important;
    margin: auto !important;
    font-family: Inter, system-ui, -apple-system, BlinkMacSystemFont, sans-serif !important;
}

/* Header */
#hero {
    background: linear-gradient(135deg, rgba(37, 99, 235, 0.95), rgba(14, 165, 233, 0.85));
    border-radius: 28px;
    padding: 38px 42px;
    margin-bottom: 26px;
    box-shadow: 0 25px 60px rgba(0, 0, 0, 0.35);
    border: 1px solid rgba(255,255,255,0.16);
}

#hero h1 {
    font-size: 48px;
    margin: 0 0 8px 0;
    color: white;
    letter-spacing: -1px;
}

#hero p {
    font-size: 18px;
    margin: 6px 0 0 0;
    color: #e0f2fe;
}

.hero-badge {
    display: inline-block;
    padding: 6px 12px;
    background: rgba(255,255,255,0.16);
    border: 1px solid rgba(255,255,255,0.22);
    border-radius: 999px;
    font-size: 13px;
    margin-bottom: 14px;
    color: white;
}

/* Panels */
.gr-block, .gr-form, .gr-box, .gr-panel {
    border-radius: 20px !important;
}

.input-panel {
    background: rgba(15, 23, 42, 0.82);
    border: 1px solid rgba(148, 163, 184, 0.25);
    border-radius: 24px;
    padding: 20px;
    box-shadow: 0 15px 45px rgba(0,0,0,0.25);
}

/* Buttons */
button.primary, .gr-button {
    border-radius: 14px !important;
    font-weight: 700 !important;
    background: linear-gradient(90deg, #2563eb, #06b6d4) !important;
    border: none !important;
    color: white !important;
    box-shadow: 0 10px 25px rgba(37, 99, 235, 0.32) !important;
}

button.primary:hover, .gr-button:hover {
    transform: translateY(-1px);
    filter: brightness(1.08);
}

/* Labels */
label, .wrap label {
    color: #dbeafe !important;
    font-weight: 600 !important;
}

/* Inputs */
input, textarea, select {
    border-radius: 12px !important;
}

/* Slider */
input[type="range"] {
    accent-color: #38bdf8 !important;
}

/* Output cards */
.result-card, .automation-card {
    background: rgba(15, 23, 42, 0.88);
    border: 1px solid rgba(148, 163, 184, 0.25);
    border-radius: 26px;
    padding: 26px;
    margin-bottom: 22px;
    box-shadow: 0 18px 50px rgba(0,0,0,0.28);
}

.result-card h2, .automation-card h2 {
    margin-top: 8px;
    color: #f8fafc;
    font-size: 26px;
}

.result-card h3, .automation-card h3 {
    margin-top: 24px;
    color: #bae6fd;
    font-size: 20px;
}

.result-card p, .automation-card p {
    color: #e5e7eb;
    line-height: 1.6;
}

.section-label {
    display: inline-block;
    background: rgba(56, 189, 248, 0.13);
    border: 1px solid rgba(56, 189, 248, 0.35);
    color: #7dd3fc;
    padding: 7px 13px;
    border-radius: 999px;
    font-size: 13px;
    font-weight: 700;
    letter-spacing: 0.4px;
    text-transform: uppercase;
}

/* KPI Cards */
.kpi-grid {
    display: grid;
    grid-template-columns: repeat(4, minmax(120px, 1fr));
    gap: 14px;
    margin: 22px 0;
}

.kpi-card {
    background: linear-gradient(180deg, rgba(30, 41, 59, 0.95), rgba(15, 23, 42, 0.95));
    border: 1px solid rgba(148, 163, 184, 0.25);
    border-radius: 18px;
    padding: 18px;
}

.kpi-title {
    color: #94a3b8;
    font-size: 13px;
    margin-bottom: 8px;
}

.kpi-value {
    color: #f8fafc;
    font-size: 22px;
    font-weight: 800;
}

/* Tables */
.metric-table {
    width: 100%;
    border-collapse: collapse;
    overflow: hidden;
    border-radius: 14px;
    margin-top: 12px;
}

.metric-table td {
    padding: 12px 14px;
    border-bottom: 1px solid rgba(148, 163, 184, 0.18);
    color: #e5e7eb;
}

.metric-table td:first-child {
    color: #93c5fd;
    font-weight: 650;
    width: 45%;
}

.metric-table tr:last-child td {
    border-bottom: none;
}

/* Dataframe and code areas */
.dataframe, .wrap, .contain {
    border-radius: 18px !important;
}

/* Footer text */
.small-note {
    color: #94a3b8;
    font-size: 13px;
    margin-top: -8px;
    margin-bottom: 18px;
}

@media (max-width: 900px) {
    .kpi-grid {
        grid-template-columns: repeat(2, minmax(120px, 1fr));
    }

    #hero h1 {
        font-size: 36px;
    }
}
"""


with gr.Blocks(css=custom_css) as demo:
    gr.HTML(
        """
        <div id="hero">
            <div class="hero-badge">Short-Term Rental Pricing Assistant</div>
            <h1>🏠 StayWise AI</h1>
            <p>AI-powered pricing and performance optimization for short-term rentals.</p>
            <p class="small-note">Run the full pipeline, benchmark a listing against comparable properties, and return automation insights from n8n.</p>
        </div>
        """
    )

    with gr.Row():
        with gr.Column(scale=1):
            gr.Markdown("## 📊 Property Inputs")

            neighbourhood_group = gr.Dropdown(
                choices=neighbourhood_groups,
                label="Neighbourhood Group",
                value=neighbourhood_groups[0]
            )

            neighbourhood = gr.Dropdown(
                choices=neighbourhoods,
                label="Neighbourhood",
                value=neighbourhoods[0]
            )

            room_type = gr.Dropdown(
                choices=room_types,
                label="Room Type",
                value=room_types[0]
            )

            price = gr.Slider(
                minimum=20,
                maximum=1000,
                value=150,
                step=5,
                label="Current Nightly Price ($)"
            )

            availability_365 = gr.Slider(
                minimum=0,
                maximum=365,
                value=180,
                step=1,
                label="Availability per Year"
            )

            season = gr.Dropdown(
                choices=seasons,
                label="Season",
                value=seasons[0]
            )

            local_event_score = gr.Slider(
                minimum=0,
                maximum=100,
                value=50,
                step=1,
                label="Local Event Demand Score"
            )

            rating = gr.Slider(
                minimum=1,
                maximum=5,
                value=4.4,
                step=0.1,
                label="Guest Rating"
            )

            sentiment_score = gr.Slider(
                minimum=-1,
                maximum=1,
                value=0.2,
                step=0.05,
                label="Customer Sentiment Score"
            )

            send_to_n8n = gr.Checkbox(
                label="Send output to n8n",
                value=False
            )

            run_button = gr.Button("🚀 Run Full Pipeline", variant="primary")

        with gr.Column(scale=2):
            result_output = gr.HTML()
            automation_output = gr.HTML()

    gr.Markdown("## 📈 Comparable Listings")
    comparable_output = gr.Dataframe()

    gr.Markdown("## 🧾 Pipeline Data Output")
    json_output = gr.Code(language="json")

    run_button.click(
        fn=run_pipeline,
        inputs=[
            neighbourhood_group,
            neighbourhood,
            room_type,
            price,
            availability_365,
            season,
            local_event_score,
            rating,
            sentiment_score,
            send_to_n8n
        ],
        outputs=[
            result_output,
            automation_output,
            comparable_output,
            json_output
        ]
    )


demo.launch()