import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from datetime import datetime, timedelta
import random
import gradio as gr
from sklearn.linear_model import LinearRegression
import tempfile

def generate_synthetic_data(n_days=90):
    dates = pd.date_range(end=datetime.today(), periods=n_days)
    data = []
    store_names = [f"Store_{i}" for i in range(1, 21)]
    categories = ['Fashion', 'Electronics', 'Food', 'Entertainment', 'Home Decor']
    zones = ['Ground Floor', 'First Floor', 'Second Floor']

    for date in dates:
        for store in store_names:
            visitors = np.random.poisson(lam=150)
            sales = np.random.normal(loc=visitors * 10, scale=visitors * 2)
            dwell_time = np.random.normal(loc=30, scale=5)
            category = random.choice(categories)
            zone = random.choice(zones)
            data.append({
                'date': date,
                'store': store,
                'category': category,
                'zone': zone,
                'visitors': max(0, visitors),
                'sales': max(0, sales),
                'dwell_time': max(5, dwell_time)
            })
    return pd.DataFrame(data)

def forecast_sales(df, days_ahead=7):
    daily_sales = df.groupby('date')['sales'].sum().reset_index()
    daily_sales['day_num'] = (daily_sales['date'] - daily_sales['date'].min()).dt.days
    X = daily_sales[['day_num']]
    y = daily_sales['sales']
    model = LinearRegression().fit(X, y)

    last_day = daily_sales['day_num'].max()
    future_days = np.arange(last_day + 1, last_day + 1 + days_ahead).reshape(-1, 1)
    future_dates = [daily_sales['date'].max() + timedelta(days=i) for i in range(1, days_ahead + 1)]
    predictions = model.predict(future_days)
    forecast_df = pd.DataFrame({'date': future_dates, 'sales': predictions})
    return forecast_df

def generate_chart(chart_type):
    df = generate_synthetic_data()
    forecast_df = forecast_sales(df)
    
    daily_visits = df.groupby('date')['visitors'].sum().reset_index()
    category_sales = df.groupby('category')['sales'].sum().reset_index()
    zone_traffic = df.groupby('zone')['visitors'].sum().reset_index()
    top_stores = df.groupby('store')['sales'].sum().nlargest(10).reset_index()
    dwell_time = df.groupby('category')['dwell_time'].mean().reset_index()

    fig, ax = plt.subplots(figsize=(8, 5))
    plt.tight_layout()

    if chart_type == "Daily Footfall":
        ax.plot(daily_visits['date'], daily_visits['visitors'], color='blue')
        ax.set_title('Daily Footfall')
        ax.set_xlabel('Date')
        ax.set_ylabel('Visitors')
    elif chart_type == "Sales by Category":
        ax.bar(category_sales['category'], category_sales['sales'], color='green')
        ax.set_title('Sales by Category')
        ax.set_ylabel('Sales')
    elif chart_type == "Footfall by Zone":
        ax.pie(zone_traffic['visitors'], labels=zone_traffic['zone'], autopct='%1.1f%%')
        ax.set_title('Footfall by Zone')
    elif chart_type == "Top 10 Stores by Sales":
        ax.bar(top_stores['store'], top_stores['sales'], color='purple')
        ax.set_title('Top 10 Stores by Sales')
        ax.set_ylabel('Sales')
        plt.xticks(rotation=45)
    elif chart_type == "Avg Dwell Time by Category":
        ax.bar(dwell_time['category'], dwell_time['dwell_time'], color='orange')
        ax.set_title('Avg Dwell Time by Category')
        ax.set_ylabel('Dwell Time (mins)')
    elif chart_type == "Forecasted Sales (Next 7 Days)":
        ax.plot(forecast_df['date'], forecast_df['sales'], color='red')
        ax.set_title('Forecasted Sales (Next 7 Days)')
        ax.set_xlabel('Date')
        ax.set_ylabel('Predicted Sales')
    else:
        return None

    tmp_file = tempfile.NamedTemporaryFile(suffix=".png", delete=False)
    plt.savefig(tmp_file.name, bbox_inches='tight')
    plt.close(fig)
    return tmp_file.name

with gr.Blocks(title="📊 Mall Dashboard with Matplotlib") as demo:
    gr.Markdown("## 🏬 Mall Analytics Dashboard (Matplotlib Version)")
    chart_dropdown = gr.Dropdown(
        choices=[
            "Daily Footfall",
            "Sales by Category",
            "Footfall by Zone",
            "Top 10 Stores by Sales",
            "Avg Dwell Time by Category",
            "Forecasted Sales (Next 7 Days)"
        ],
        value="Daily Footfall",
        label="Choose a Chart"
    )
    generate_button = gr.Button("Generate Chart")
    image_output = gr.Image(label="Chart Output")

    generate_button.click(fn=generate_chart, inputs=chart_dropdown, outputs=image_output)

demo.launch()