import streamlit as st
import pandas as pd
import requests
import json

# Configure page
st.set_page_config(
    page_title="SuperKart Sales Forecasting",
    page_icon="🛒",
    layout="wide"
)

# Backend API URL
BACKEND_URL = "https://deepakdm411-shoppingcartbackend.hf.space"

st.title("🛒 SuperKart Sales Forecasting System")
st.write("Predict sales revenue for SuperKart products using our advanced ML model")

# API connection status
def check_backend_connection():
    try:
        response = requests.get(f"{BACKEND_URL}/", timeout=10)
        return response.status_code == 200
    except Exception as e:
        st.error(f"Connection error: {str(e)}")
        return False

# Check backend status
with st.spinner("Checking API connection..."):
    backend_online = check_backend_connection()

if backend_online:
    st.success("✅ Connected to backend API")
else:
    st.error("❌ Backend API not available. Please check the backend URL.")
    st.info(f"Current backend URL: {BACKEND_URL}")

st.subheader("Enter Product and Store Details:")

# Create input form
with st.form("prediction_form"):
    col1, col2 = st.columns(2)
    
    with col1:
        st.markdown("**🏪 Store Information**")
        store_type = st.selectbox(
            "Store Type", 
            ["Supermarket Type1", "Supermarket Type2", "Supermarket Type3", "Grocery Store"]
        )
        store_size = st.selectbox("Store Size", ["Small", "Medium", "High"])
        store_location = st.selectbox("Store Location City Type", ["Tier 1", "Tier 2", "Tier 3"])
        store_year = st.number_input("Store Establishment Year", min_value=1980, max_value=2024, value=2010)
    
    with col2:
        st.markdown("**📦 Product Information**")
        product_type = st.selectbox(
            "Product Type", 
            ["Dairy", "Soft Drinks", "Meat", "Fruits and Vegetables", "Household", 
             "Baking Goods", "Snack Foods", "Frozen Foods", "Breakfast", 
             "Health and Hygiene", "Hard Drinks", "Canned", "Bread", 
             "Starchy Foods", "Others", "Seafood"]
        )
        product_sugar = st.selectbox("Product Sugar Content", ["Low Fat", "Regular"])
        product_weight = st.number_input("Product Weight", min_value=0.01, value=1.0)
        product_mrp = st.number_input("Product MRP", min_value=1.0, value=100.0)
        product_area = st.number_input("Product Allocated Area", min_value=0.001, value=0.1)
    
    # Submit button
    submitted = st.form_submit_button("🎯 Predict Sales Revenue", type="primary")

# Handle form submission
if submitted and backend_online:
    # Prepare data for API
    prediction_data = {
        "Product_Weight": product_weight,
        "Product_Sugar_Content": product_sugar,
        "Product_Allocated_Area": product_area,
        "Product_Type": product_type,
        "Product_MRP": product_mrp,
        "Store_Establishment_Year": store_year,
        "Store_Size": store_size,
        "Store_Location_City_Type": store_location,
        "Store_Type": store_type
    }
    
    # Make API call
    with st.spinner("Making prediction..."):
        try:
            response = requests.post(
                f"{BACKEND_URL}/predict",
                json=prediction_data,
                headers={"Content-Type": "application/json"},
                timeout=30
            )
            
            if response.status_code == 200:
                result = response.json()
                
                # Display results
                st.success("✅ Prediction Complete!")
                
                # Main prediction display
                col1, col2, col3 = st.columns([1, 2, 1])
                with col2:
                    st.metric(
                        label="Predicted Sales Revenue",
                        value=result["formatted_prediction"]
                    )
                
                # Additional details
                st.markdown("---")
                st.markdown("### 📊 Prediction Details")
                
                detail_col1, detail_col2 = st.columns(2)
                
                with detail_col1:
                    st.info(f"""
                    **Store Profile:**
                    - Type: {store_type}
                    - Size: {store_size}
                    - Location: {store_location}
                    - Established: {store_year}
                    """)
                
                with detail_col2:
                    st.info(f"""
                    **Product Profile:**
                    - Category: {product_type}
                    - Weight: {product_weight} kg
                    - MRP: ₹{product_mrp}
                    - Sugar Content: {product_sugar}
                    """)
                
                # Business insights
                prediction_value = result["prediction"]
                if prediction_value > product_mrp * 10:
                    st.success("🚀 Excellent Revenue Potential!")
                elif prediction_value > product_mrp * 5:
                    st.info("📈 Good Revenue Potential")
                else:
                    st.warning("📊 Moderate Revenue Potential")
                    
            else:
                error_data = response.json()
                st.error(f"Prediction failed: {error_data.get('error', 'Unknown error')}")
                
        except requests.exceptions.Timeout:
            st.error("⏰ Request timed out. Please try again.")
        except requests.exceptions.ConnectionError:
            st.error("🔌 Connection error. Please check if backend is running.")
        except Exception as e:
            st.error(f"An error occurred: {str(e)}")

elif submitted and not backend_online:
    st.error("Cannot make prediction - backend API is not available.")

# Sidebar with API information
st.sidebar.markdown("### 🔧 API Information")
if backend_online:
    try:
        model_info = requests.get(f"{BACKEND_URL}/model-info", timeout=10).json()
        st.sidebar.success("✅ API Online")
        st.sidebar.json(model_info)
    except:
        st.sidebar.warning("⚠️ Could not fetch model info")
else:
    st.sidebar.error("❌ API Offline")

st.sidebar.markdown(f"**Backend URL:** {BACKEND_URL}")

# Footer
st.markdown("---")
st.markdown("""
<div style='text-align: center; color: #666;'>
    <p>SuperKart Sales Forecasting System | Built with Streamlit & Flask</p>
</div>
""", unsafe_allow_html=True)