"""
农业智能决策系统
基于多因子 Cobb-Douglas 产量模型的作物种植决策支持应用
"""

import streamlit as st
import numpy as np
import pandas as pd
import plotly.graph_objects as go
import plotly.express as px
from plotly.subplots import make_subplots

# ─── Page Config ────────────────────────────────────────────────────────────
st.set_page_config(
    page_title="农业智能决策系统",
    page_icon="🌾",
    layout="wide",
    initial_sidebar_state="expanded",
)

# ─── Custom CSS ──────────────────────────────────────────────────────────────
st.markdown("""
<style>
@import url('https://fonts.googleapis.com/css2?family=Noto+Serif+SC:wght@400;600;700&family=JetBrains+Mono:wght@400;500&display=swap');

:root {
    --bg-dark: #0a1628;
    --bg-card: #0f2040;
    --accent-green: #4ade80;
    --accent-gold: #f59e0b;
    --accent-blue: #38bdf8;
    --text-primary: #e2e8f0;
    --text-muted: #64748b;
    --border: rgba(74, 222, 128, 0.2);
}

html, body, [class*="css"] {
    font-family: 'Noto Serif SC', serif;
    background-color: var(--bg-dark);
    color: var(--text-primary);
}

.stApp {
    background: linear-gradient(135deg, #0a1628 0%, #0d1f3c 50%, #091520 100%);
}

/* Sidebar */
[data-testid="stSidebar"] {
    background: linear-gradient(180deg, #0f2040 0%, #0a1628 100%);
    border-right: 1px solid var(--border);
}
[data-testid="stSidebar"] .stSlider label,
[data-testid="stSidebar"] .stNumberInput label,
[data-testid="stSidebar"] .stSelectbox label {
    color: var(--accent-green) !important;
    font-size: 0.82rem;
    font-family: 'JetBrains Mono', monospace;
    letter-spacing: 0.05em;
}

/* Metric cards */
.metric-card {
    background: linear-gradient(135deg, #0f2040, #132b55);
    border: 1px solid var(--border);
    border-radius: 12px;
    padding: 20px 24px;
    text-align: center;
    position: relative;
    overflow: hidden;
}
.metric-card::before {
    content: '';
    position: absolute;
    top: 0; left: 0; right: 0;
    height: 2px;
    background: linear-gradient(90deg, var(--accent-green), var(--accent-blue));
}
.metric-value {
    font-family: 'JetBrains Mono', monospace;
    font-size: 2.2rem;
    font-weight: 700;
    color: var(--accent-green);
    line-height: 1;
}
.metric-unit {
    font-size: 0.85rem;
    color: var(--text-muted);
    margin-top: 4px;
}
.metric-label {
    font-size: 0.9rem;
    color: var(--text-primary);
    margin-top: 8px;
}

/* Section headers */
.section-header {
    font-size: 0.75rem;
    font-family: 'JetBrains Mono', monospace;
    letter-spacing: 0.15em;
    color: var(--accent-gold);
    text-transform: uppercase;
    border-bottom: 1px solid rgba(245,158,11,0.3);
    padding-bottom: 6px;
    margin-bottom: 16px;
    margin-top: 24px;
}

/* Crop badge */
.crop-badge {
    display: inline-block;
    background: linear-gradient(135deg, rgba(74,222,128,0.15), rgba(56,189,248,0.15));
    border: 1px solid var(--accent-green);
    border-radius: 6px;
    padding: 4px 12px;
    font-family: 'JetBrains Mono', monospace;
    font-size: 0.78rem;
    color: var(--accent-green);
    margin: 3px;
}

/* Recommendation card */
.rec-card {
    background: linear-gradient(135deg, rgba(74,222,128,0.08), rgba(56,189,248,0.05));
    border: 1px solid rgba(74,222,128,0.3);
    border-radius: 12px;
    padding: 18px 22px;
    margin: 10px 0;
}
.rec-rank {
    font-family: 'JetBrains Mono', monospace;
    font-size: 1.5rem;
    color: var(--accent-gold);
    font-weight: 700;
}
.rec-crop {
    font-size: 1.1rem;
    color: var(--text-primary);
    font-weight: 600;
}
.rec-score {
    font-family: 'JetBrains Mono', monospace;
    font-size: 0.85rem;
    color: var(--accent-blue);
}

/* Hero title */
.hero-title {
    font-size: 2rem;
    font-weight: 700;
    background: linear-gradient(135deg, var(--accent-green), var(--accent-blue));
    -webkit-background-clip: text;
    -webkit-text-fill-color: transparent;
    background-clip: text;
    line-height: 1.2;
}
.hero-sub {
    font-family: 'JetBrains Mono', monospace;
    font-size: 0.8rem;
    color: var(--text-muted);
    letter-spacing: 0.1em;
    margin-top: 4px;
}

/* Alert boxes */
.alert-good {
    background: rgba(74,222,128,0.1);
    border-left: 3px solid var(--accent-green);
    border-radius: 0 8px 8px 0;
    padding: 12px 16px;
    margin: 8px 0;
    font-size: 0.9rem;
}
.alert-warn {
    background: rgba(245,158,11,0.1);
    border-left: 3px solid var(--accent-gold);
    border-radius: 0 8px 8px 0;
    padding: 12px 16px;
    margin: 8px 0;
    font-size: 0.9rem;
}

/* Override streamlit slider colors */
.stSlider [data-baseweb="slider"] [data-testid="stTickBarMin"],
.stSlider [data-baseweb="slider"] [data-testid="stTickBarMax"] {
    color: var(--text-muted);
}
</style>
""", unsafe_allow_html=True)


# ─── Crop Database ───────────────────────────────────────────────────────────
CROPS = {
    "水稻": {
        "emoji": "🌾",
        "optimal": {"ph": (6.0, 7.0), "N": (80, 120), "P": (30, 60), "K": (40, 80),
                    "rainfall": (150, 250), "temp": (22, 30)},
        "base_yield": 7500,  # kg/ha
        "color": "#4ade80"
    },
    "小麦": {
        "emoji": "🌿",
        "optimal": {"ph": (6.0, 7.5), "N": (60, 100), "P": (20, 50), "K": (30, 60),
                    "rainfall": (60, 120), "temp": (15, 22)},
        "base_yield": 6000,
        "color": "#f59e0b"
    },
    "玉米": {
        "emoji": "🌽",
        "optimal": {"ph": (5.8, 7.0), "N": (100, 150), "P": (40, 70), "K": (60, 100),
                    "rainfall": (100, 180), "temp": (20, 28)},
        "base_yield": 8500,
        "color": "#fbbf24"
    },
    "大豆": {
        "emoji": "🫘",
        "optimal": {"ph": (6.0, 7.0), "N": (20, 50), "P": (30, 60), "K": (40, 80),
                    "rainfall": (80, 150), "temp": (18, 26)},
        "base_yield": 3500,
        "color": "#a78bfa"
    },
    "油菜": {
        "emoji": "🌻",
        "optimal": {"ph": (6.0, 7.5), "N": (80, 130), "P": (30, 60), "K": (50, 90),
                    "rainfall": (80, 130), "temp": (15, 20)},
        "base_yield": 3000,
        "color": "#f97316"
    },
    "棉花": {
        "emoji": "☁️",
        "optimal": {"ph": (6.0, 8.0), "N": (60, 100), "P": (20, 45), "K": (40, 70),
                    "rainfall": (70, 120), "temp": (25, 32)},
        "base_yield": 4500,
        "color": "#e2e8f0"
    },
}


# ─── Yield Model ─────────────────────────────────────────────────────────────
def compute_factor(value, optimal_low, optimal_high, penalty=0.5):
    """Score 0-1: 1 if in optimal range, decays outside."""
    mid = (optimal_low + optimal_high) / 2
    width = (optimal_high - optimal_low) / 2 + 1e-9
    if optimal_low <= value <= optimal_high:
        return 1.0
    dist = min(abs(value - optimal_low), abs(value - optimal_high))
    return max(0.0, 1.0 - penalty * (dist / width))


def predict_yield(crop_name, ph, N, P, K, rainfall, temp, pesticide, area):
    crop = CROPS[crop_name]
    opt = crop["optimal"]

    f_ph       = compute_factor(ph, *opt["ph"], penalty=0.6)
    f_N        = compute_factor(N, *opt["N"], penalty=0.4)
    f_P        = compute_factor(P, *opt["P"], penalty=0.4)
    f_K        = compute_factor(K, *opt["K"], penalty=0.4)
    f_rain     = compute_factor(rainfall, *opt["rainfall"], penalty=0.5)
    f_temp     = compute_factor(temp, *opt["temp"], penalty=0.7)
    f_pest     = 0.5 + 0.5 * min(pesticide / 100, 1.0)

    # Cobb-Douglas style yield function
    nutrient_idx = (f_N * f_P * f_K) ** (1/3)
    soil_idx     = f_ph
    climate_idx  = (f_rain * f_temp) ** 0.5

    total_factor = soil_idx ** 0.2 * nutrient_idx ** 0.4 * climate_idx ** 0.3 * f_pest ** 0.1

    yield_per_ha = crop["base_yield"] * total_factor
    total_yield  = yield_per_ha * area

    factors = {
        "土壤pH": f_ph, "氮(N)": f_N, "磷(P)": f_P,
        "钾(K)": f_K, "降雨量": f_rain, "温度": f_temp, "农药": f_pest
    }
    return yield_per_ha, total_yield, factors


def rank_crops(ph, N, P, K, rainfall, temp, pesticide, area):
    results = []
    for name in CROPS:
        yph, ytotal, factors = predict_yield(name, ph, N, P, K, rainfall, temp, pesticide, area)
        score = np.mean(list(factors.values()))
        results.append({
            "crop": name,
            "emoji": CROPS[name]["emoji"],
            "yield_ha": yph,
            "total_yield": ytotal,
            "score": score,
            "color": CROPS[name]["color"],
            "factors": factors
        })
    results.sort(key=lambda x: x["score"], reverse=True)
    return results


# ─── Sidebar Inputs ──────────────────────────────────────────────────────────
with st.sidebar:
    st.markdown('<div class="hero-title">🌾 农业决策</div>', unsafe_allow_html=True)
    st.markdown('<div class="hero-sub">SMART FARMING SYSTEM v2.0</div>', unsafe_allow_html=True)
    st.markdown("---")

    st.markdown('<div class="section-header">🧪 土壤参数</div>', unsafe_allow_html=True)
    col1, col2 = st.columns(2)
    with col1:
        ph = st.slider("pH 值", 4.0, 9.0, 6.5, 0.1)
        N  = st.slider("氮 N (mg/kg)", 0, 200, 90, 5)
    with col2:
        P  = st.slider("磷 P (mg/kg)", 0, 100, 45, 5)
        K  = st.slider("钾 K (mg/kg)", 0, 150, 60, 5)

    st.markdown('<div class="section-header">🌦 气象数据</div>', unsafe_allow_html=True)
    col3, col4 = st.columns(2)
    with col3:
        rainfall = st.slider("降雨量 (mm/月)", 0, 400, 120, 10)
    with col4:
        temp = st.slider("温度 (°C)", 0, 45, 22, 1)

    st.markdown('<div class="section-header">🌱 种植参数</div>', unsafe_allow_html=True)
    area      = st.number_input("种植面积 (公顷)", 0.1, 10000.0, 100.0, 10.0)
    pesticide = st.slider("农药用量 (kg/ha)", 0, 200, 50, 5)

    st.markdown('<div class="section-header">🎯 目标作物</div>', unsafe_allow_html=True)
    selected_crop = st.selectbox(
        "选择分析作物",
        list(CROPS.keys()),
        format_func=lambda x: f"{CROPS[x]['emoji']} {x}"
    )


# ─── Compute ──────────────────────────────────────────────────────────────────
yph, ytotal, factors = predict_yield(selected_crop, ph, N, P, K, rainfall, temp, pesticide, area)
rankings = rank_crops(ph, N, P, K, rainfall, temp, pesticide, area)
best_crop = rankings[0]


# ─── Main Layout ─────────────────────────────────────────────────────────────
st.markdown(f"""
<div style="display:flex; align-items:baseline; gap:16px; margin-bottom:4px;">
  <div class="hero-title">农业智能决策系统</div>
</div>
<div class="hero-sub">YIELD = f(SOIL · WEATHER · PESTICIDE) &nbsp;|&nbsp; 基于多因子 Cobb-Douglas 产量模型</div>
""", unsafe_allow_html=True)

st.markdown("<br>", unsafe_allow_html=True)

# KPI row
k1, k2, k3, k4 = st.columns(4)
with k1:
    st.markdown(f"""
    <div class="metric-card">
        <div class="metric-value">{yph:,.0f}</div>
        <div class="metric-unit">kg / 公顷</div>
        <div class="metric-label">{CROPS[selected_crop]['emoji']} {selected_crop} 单产</div>
    </div>""", unsafe_allow_html=True)
with k2:
    st.markdown(f"""
    <div class="metric-card">
        <div class="metric-value">{ytotal/1000:,.1f}</div>
        <div class="metric-unit">吨 / 总产量</div>
        <div class="metric-label">📦 {area:.0f} 公顷总产</div>
    </div>""", unsafe_allow_html=True)
with k3:
    overall = np.mean(list(factors.values()))
    st.markdown(f"""
    <div class="metric-card">
        <div class="metric-value">{overall*100:.1f}%</div>
        <div class="metric-unit">综合适宜度</div>
        <div class="metric-label">🎯 环境匹配指数</div>
    </div>""", unsafe_allow_html=True)
with k4:
    st.markdown(f"""
    <div class="metric-card">
        <div class="metric-value">{best_crop['emoji']}</div>
        <div class="metric-unit">{best_crop['crop']} ({best_crop['score']*100:.0f}%)</div>
        <div class="metric-label">🏆 最优推荐作物</div>
    </div>""", unsafe_allow_html=True)

st.markdown("<br>", unsafe_allow_html=True)

# ─── Charts Row ──────────────────────────────────────────────────────────────
col_left, col_right = st.columns([3, 2])

with col_left:
    st.markdown('<div class="section-header">📊 影响因子雷达图</div>', unsafe_allow_html=True)

    factor_names = list(factors.keys())
    factor_vals  = [round(v * 100, 1) for v in factors.values()]
    factor_names_closed = factor_names + [factor_names[0]]
    factor_vals_closed  = factor_vals + [factor_vals[0]]

    fig_radar = go.Figure()
    fig_radar.add_trace(go.Scatterpolar(
        r=factor_vals_closed,
        theta=factor_names_closed,
        fill='toself',
        fillcolor='rgba(74,222,128,0.15)',
        line=dict(color='#4ade80', width=2),
        name=selected_crop,
    ))
    fig_radar.add_trace(go.Scatterpolar(
        r=[100]*len(factor_names_closed),
        theta=factor_names_closed,
        line=dict(color='rgba(255,255,255,0.1)', width=1, dash='dot'),
        mode='lines',
        name='理想值',
    ))
    fig_radar.update_layout(
        polar=dict(
            bgcolor='rgba(0,0,0,0)',
            radialaxis=dict(range=[0, 100], showticklabels=True,
                            tickfont=dict(color='#64748b', size=9),
                            gridcolor='rgba(255,255,255,0.06)'),
            angularaxis=dict(tickfont=dict(color='#e2e8f0', size=11),
                             gridcolor='rgba(255,255,255,0.08)'),
        ),
        paper_bgcolor='rgba(0,0,0,0)',
        plot_bgcolor='rgba(0,0,0,0)',
        font=dict(color='#e2e8f0'),
        legend=dict(orientation='h', y=-0.12, font=dict(size=10)),
        margin=dict(t=20, b=40, l=40, r=40),
        height=320,
    )
    st.plotly_chart(fig_radar, use_container_width=True)

with col_right:
    st.markdown('<div class="section-header">🏅 作物推荐排行</div>', unsafe_allow_html=True)
    for i, r in enumerate(rankings[:4]):
        rank_icons = ["🥇", "🥈", "🥉", "4️⃣"]
        bar_width  = int(r['score'] * 100)
        bar_color  = r['color']
        st.markdown(f"""
        <div class="rec-card" style="margin-bottom:8px;">
          <div style="display:flex; justify-content:space-between; align-items:center;">
            <div>
              <span class="rec-rank">{rank_icons[i]}</span>
              <span class="rec-crop" style="margin-left:8px;">{r['emoji']} {r['crop']}</span>
            </div>
            <span class="rec-score">{r['score']*100:.1f}%</span>
          </div>
          <div style="background:rgba(255,255,255,0.06); border-radius:4px; height:4px; margin-top:8px; overflow:hidden;">
            <div style="width:{bar_width}%; height:100%; background:{bar_color}; border-radius:4px;"></div>
          </div>
          <div style="font-size:0.78rem; color:#64748b; margin-top:4px; font-family:'JetBrains Mono',monospace;">
            {r['yield_ha']:,.0f} kg/ha &nbsp;·&nbsp; 总产 {r['total_yield']/1000:,.1f} 吨
          </div>
        </div>""", unsafe_allow_html=True)

# ─── Sensitivity Analysis ─────────────────────────────────────────────────────
st.markdown('<div class="section-header">📈 产量敏感性分析</div>', unsafe_allow_html=True)

sa_col1, sa_col2 = st.columns(2)

with sa_col1:
    N_range = np.linspace(0, 200, 60)
    y_N = [predict_yield(selected_crop, ph, n, P, K, rainfall, temp, pesticide, 1)[0] for n in N_range]

    fig_N = go.Figure()
    fig_N.add_trace(go.Scatter(
        x=N_range, y=y_N,
        mode='lines', line=dict(color='#4ade80', width=2.5),
        fill='tozeroy', fillcolor='rgba(74,222,128,0.08)',
        name='产量'
    ))
    fig_N.add_vline(x=N, line=dict(color='#f59e0b', width=1.5, dash='dot'),
                    annotation_text=f"当前 {N}", annotation_font_color='#f59e0b')
    fig_N.update_layout(
        title=dict(text="氮肥用量 vs 产量", font=dict(color='#94a3b8', size=12)),
        xaxis=dict(title="氮 N (mg/kg)", color='#64748b', gridcolor='rgba(255,255,255,0.05)'),
        yaxis=dict(title="产量 (kg/ha)", color='#64748b', gridcolor='rgba(255,255,255,0.05)'),
        paper_bgcolor='rgba(0,0,0,0)', plot_bgcolor='rgba(0,0,0,0)',
        font=dict(color='#e2e8f0', size=10),
        margin=dict(t=36, b=36, l=50, r=20), height=220,
        showlegend=False,
    )
    st.plotly_chart(fig_N, use_container_width=True)

with sa_col2:
    rain_range = np.linspace(0, 400, 60)
    y_rain = [predict_yield(selected_crop, ph, N, P, K, r, temp, pesticide, 1)[0] for r in rain_range]

    fig_R = go.Figure()
    fig_R.add_trace(go.Scatter(
        x=rain_range, y=y_rain,
        mode='lines', line=dict(color='#38bdf8', width=2.5),
        fill='tozeroy', fillcolor='rgba(56,189,248,0.08)',
        name='产量'
    ))
    fig_R.add_vline(x=rainfall, line=dict(color='#f59e0b', width=1.5, dash='dot'),
                    annotation_text=f"当前 {rainfall}mm", annotation_font_color='#f59e0b')
    fig_R.update_layout(
        title=dict(text="月降雨量 vs 产量", font=dict(color='#94a3b8', size=12)),
        xaxis=dict(title="降雨量 (mm/月)", color='#64748b', gridcolor='rgba(255,255,255,0.05)'),
        yaxis=dict(title="产量 (kg/ha)", color='#64748b', gridcolor='rgba(255,255,255,0.05)'),
        paper_bgcolor='rgba(0,0,0,0)', plot_bgcolor='rgba(0,0,0,0)',
        font=dict(color='#e2e8f0', size=10),
        margin=dict(t=36, b=36, l=50, r=20), height=220,
        showlegend=False,
    )
    st.plotly_chart(fig_R, use_container_width=True)

# ─── All Crops Comparison Bar Chart ───────────────────────────────────────────
st.markdown('<div class="section-header">🌐 全作物产量对比</div>', unsafe_allow_html=True)

crop_names   = [f"{r['emoji']} {r['crop']}" for r in rankings]
crop_yields  = [r['yield_ha'] for r in rankings]
crop_colors  = [r['color'] for r in rankings]

fig_bar = go.Figure()
fig_bar.add_trace(go.Bar(
    x=crop_names, y=crop_yields,
    marker=dict(color=crop_colors, opacity=0.85,
                line=dict(color='rgba(255,255,255,0.2)', width=1)),
    text=[f"{y:,.0f}" for y in crop_yields],
    textposition='outside',
    textfont=dict(color='#94a3b8', size=10, family='JetBrains Mono'),
))
fig_bar.update_layout(
    xaxis=dict(color='#64748b', gridcolor='rgba(255,255,255,0.04)'),
    yaxis=dict(title="预期产量 (kg/ha)", color='#64748b', gridcolor='rgba(255,255,255,0.05)'),
    paper_bgcolor='rgba(0,0,0,0)', plot_bgcolor='rgba(0,0,0,0)',
    font=dict(color='#e2e8f0', size=11),
    margin=dict(t=20, b=30, l=60, r=20), height=240,
    showlegend=False,
)
st.plotly_chart(fig_bar, use_container_width=True)

# ─── Advisory Panel ───────────────────────────────────────────────────────────
st.markdown('<div class="section-header">💡 智能建议</div>', unsafe_allow_html=True)
adv1, adv2 = st.columns(2)

with adv1:
    crop_opt = CROPS[selected_crop]["optimal"]
    advisories = []

    if not (crop_opt["ph"][0] <= ph <= crop_opt["ph"][1]):
        advisories.append(("warn", f"pH {ph} 偏离 {selected_crop} 适宜范围 {crop_opt['ph']}，建议{'施石灰' if ph < crop_opt['ph'][0] else '施硫磺'}调节"))
    else:
        advisories.append(("good", f"土壤 pH {ph} 处于 {selected_crop} 适宜范围内 ✓"))

    if N < crop_opt["N"][0]:
        advisories.append(("warn", f"氮肥不足（{N} vs 建议 {crop_opt['N'][0]}-{crop_opt['N'][1]} mg/kg），建议追施尿素"))
    elif N > crop_opt["N"][1]:
        advisories.append(("warn", f"氮肥过量（{N} mg/kg），可能造成徒长，建议减施"))
    else:
        advisories.append(("good", f"氮肥水平 {N} mg/kg 适宜 ✓"))

    if rainfall < crop_opt["rainfall"][0]:
        advisories.append(("warn", f"降雨量不足，建议增加灌溉（缺水 {crop_opt['rainfall'][0]-rainfall} mm）"))
    elif rainfall > crop_opt["rainfall"][1]:
        advisories.append(("warn", f"降雨量偏多，注意防涝排水"))
    else:
        advisories.append(("good", f"降雨量 {rainfall}mm 适合 {selected_crop} 生长 ✓"))

    for typ, msg in advisories:
        css_class = "alert-good" if typ == "good" else "alert-warn"
        st.markdown(f'<div class="{css_class}">{msg}</div>', unsafe_allow_html=True)

with adv2:
    st.markdown(f"""
    <div class="rec-card">
      <div style="font-size:0.82rem; color:#64748b; font-family:'JetBrains Mono',monospace; margin-bottom:12px;">
        当前环境参数下适宜种植：
      </div>
    """, unsafe_allow_html=True)
    badges = "".join([
        f'<span class="crop-badge">{r["emoji"]} {r["crop"]} {r["score"]*100:.0f}%</span>'
        for r in rankings if r['score'] > 0.6
    ])
    st.markdown(f'{badges}</div>', unsafe_allow_html=True)

    st.markdown(f"""
    <div style="margin-top:16px; font-size:0.85rem; color:#94a3b8; line-height:1.7;">
    <b style="color:#4ade80;">最优方案：</b>{best_crop['emoji']} {best_crop['crop']}<br>
    预期单产：<span style="font-family:'JetBrains Mono',monospace; color:#38bdf8;">{best_crop['yield_ha']:,.0f} kg/ha</span><br>
    {area:.0f}公顷总产：<span style="font-family:'JetBrains Mono',monospace; color:#38bdf8;">{best_crop['total_yield']/1000:,.1f} 吨</span>
    </div>
    """, unsafe_allow_html=True)

# ─── Footer ───────────────────────────────────────────────────────────────────
st.markdown("<br>", unsafe_allow_html=True)
st.markdown("""
<div style="text-align:center; font-family:'JetBrains Mono',monospace; font-size:0.72rem;
            color:#334155; padding:16px; border-top:1px solid rgba(74,222,128,0.1);">
  YIELD = f(Soil, Weather, Pesticide) &nbsp;|&nbsp; Cobb-Douglas Multi-Factor Model &nbsp;|&nbsp; 农业智能决策系统
</div>
""", unsafe_allow_html=True)