import pandas as pd
import streamlit as st
import matplotlib.pyplot as plt

# Function to load and process the CSV data
def load_data(file):
    try:
        df = pd.read_csv(file)
        
        # Automatically detect the datetime column or ask the user to specify it
        timestamp_col = None
        for col in df.columns:
            if 'date' in col.lower() or 'time' in col.lower():
                timestamp_col = col
                break
        
        if timestamp_col is None:
            timestamp_col = st.selectbox("Select the column containing the timestamp:", df.columns.tolist())
        
        if timestamp_col:
            df['timestamp'] = pd.to_datetime(df[timestamp_col], errors='coerce')  # Convert to datetime
            if df['timestamp'].isnull().any():
                st.error("There are invalid or missing date/time values in the selected column.")
                return None, None
            
            # Show available columns for traffic data selection
            st.write(f"Available columns: {df.columns.tolist()}")
            traffic_column = st.selectbox("Select the column containing traffic flow data:", df.columns.tolist())
            
            return df, traffic_column
        else:
            st.error("No valid timestamp column found in the dataset.")
            return None, None
    except Exception as e:
        st.error(f"Error loading file: {e}")
        return None, None

# Function to generate peak traffic hour
def peak_traffic_hour(df, traffic_column):
    df['hour'] = df['timestamp'].dt.hour  # Extract hour from timestamp
    traffic_by_hour = df.groupby('hour')[traffic_column].sum()  # Sum the traffic flow per hour
    peak_hour = traffic_by_hour.idxmax()  # Find the hour with the maximum traffic
    peak_traffic = traffic_by_hour.max()  # Find the traffic flow for that peak hour
    return peak_hour, peak_traffic

# Function to generate hourly traffic summary
def hourly_traffic_summary(df, traffic_column):
    df['hour'] = df['timestamp'].dt.hour
    df[traffic_column] = pd.to_numeric(df[traffic_column], errors='coerce')  # Ensure numeric data
    hourly_summary = df.groupby('hour')[traffic_column].sum().reset_index()  # Aggregate by hour
    return hourly_summary

# Initialize Streamlit app
def main():
    st.title("Traffic Flow Analyzer")

    # Upload CSV file
    uploaded_file = st.file_uploader("Upload Traffic Data (CSV)", type=["csv"])
    if uploaded_file is not None:
        df, traffic_column = load_data(uploaded_file)
        
        if df is not None and traffic_column is not None:
            st.write("Data loaded successfully! Now, you can ask your questions.")
            st.write(df.head())  # Display the first few rows of the dataset for user reference
            
            # Ask the user what they want to know
            user_question = st.text_input("Ask a question about the traffic data (e.g., 'What is the peak traffic hour?')")
            
            if user_question:
                user_question = user_question.lower()
                
                if "peak traffic hour" in user_question:
                    peak_hour, peak_traffic = peak_traffic_hour(df, traffic_column)
                    st.write(f"The peak traffic hour is {peak_hour}:00 with a total traffic flow of {peak_traffic} vehicles.")
                    
                elif "hourly traffic summary" in user_question:
                    hourly_summary = hourly_traffic_summary(df, traffic_column)
                    st.write("Total traffic flow per hour:")
                    st.write(hourly_summary)
                    
                else:
                    st.write("Sorry, I couldn't understand your question. Please try asking something else.")

if __name__ == "__main__":
    main()