import streamlit as st
import pandas as pd
from app_config import AppConfig  # Import the configurations class
from data_processor import DataProcessor  # Import the data analysis class
from visualization import Visualization  # Import the data viz class
from ai_analysis import AIAnalysis  # Import the ai analysis class
from sidebar import Sidebar  # Import the Sidebar class


def main():
    # Initialize the app configuration
    app_config = AppConfig()

    # Initialize the sidebar
    sidebar = Sidebar()
    sidebar.display()

    # Initialize the data processor
    data_processor = DataProcessor()

    # Initialize the visualization handler
    visualization = Visualization()

    # Initialize the AI analysis handler
    ai_analysis = AIAnalysis(data_processor.client)

    st.title("Intervention Program Analysis")

    # Date selection option
    date_option = st.radio(
        "Select data range:",
        ("All Data", "Date Range")
    )

    # Initialize start and end date variables
    start_date = None
    end_date = None

    if date_option == "Date Range":
        # Prompt user to enter start and end dates
        start_date = st.date_input("Start Date")
        end_date = st.date_input("End Date")

        # Ensure start date is before end date
        if start_date > end_date:
            st.error("Start date must be before end date.")
            return

    # File uploader
    uploaded_file = st.file_uploader("Upload your Excel file", type=["xlsx"])

    if uploaded_file is not None:
        try:
            # Read the Excel file into a DataFrame
            df = data_processor.read_excel(uploaded_file)

            # Format the session data
            df = data_processor.format_session_data(df)

            # Replace student names with initials
            df = data_processor.replace_student_names_with_initials(df)

            # Filter data if date range is selected
            if date_option == "Date Range":
                # Convert start_date and end_date to datetime
                start_date = pd.to_datetime(start_date).date()
                end_date = pd.to_datetime(end_date).date()

                # Filter the DataFrame based on the selected date range
                df = df[(df['Date of Session'] >= start_date) & (df['Date of Session'] <= end_date)]


            st.subheader("Uploaded Data")
            st.write(df)

            # Ensure expected column is available
            if DataProcessor.INTERVENTION_COLUMN not in df.columns:
                st.error(f"Expected column '{DataProcessor.INTERVENTION_COLUMN}' not found.")
                return

            # Compute Intervention Session Statistics
            intervention_stats = data_processor.compute_intervention_statistics(df)
            st.subheader("Intervention Session Statistics")
            st.write(intervention_stats)

            # Plot and download intervention statistics
            # intervention_fig = visualization.plot_intervention_statistics(intervention_stats)
            # visualization.download_chart(intervention_fig, "intervention_statistics_chart.png")

            # Plot and download intervention statistics: Two-column layout for the visualization and intervention frequency
            col1, col2 = st.columns([3, 1])  # Set the column width ratio
            
            with col1:
                intervention_fig = visualization.plot_intervention_statistics(intervention_stats)
            
            with col2:
                intervention_frequency = intervention_stats['Intervention Frequency (%)'].values[0]
                # Display the "Intervention Frequency (%)" text
                st.markdown("<h3 style='color: #358E66;'>Intervention Frequency</h3>", unsafe_allow_html=True)
                # Display the frequency value below it
                st.markdown(f"<h1 style='color: #358E66;'>{intervention_frequency}%</h1>", unsafe_allow_html=True)

            visualization.download_chart(intervention_fig, "intervention_statistics_chart.png")
            
            # Compute Student Metrics
            student_metrics_df = data_processor.compute_student_metrics(df)
            st.subheader("Student Metrics")
            st.write(student_metrics_df)

            # Compute Student Metric Averages
            attendance_avg_stats, engagement_avg_stats = data_processor.compute_average_metrics(student_metrics_df)

            # Plot and download student metrics
            student_metrics_fig = visualization.plot_student_metrics(student_metrics_df, attendance_avg_stats, engagement_avg_stats)
            visualization.download_chart(student_metrics_fig, "student_metrics_chart.png")

            # Evaluate each student and build decision tree diagrams
            student_metrics_df['Evaluation'] = student_metrics_df.apply(
                lambda row: data_processor.evaluate_student(row), axis=1
            )
            st.subheader("Student Evaluations")
            st.write(student_metrics_df[['Student', 'Evaluation']])
            
            # # Build and display decision tree diagrams for each student
            # for index, row in student_metrics_df.iterrows():
            #     tree_diagram = visualization.build_tree_diagram(row)
            #     st.graphviz_chart(tree_diagram.source)

            # Build and display decision tree diagrams for each student
            for index, row in student_metrics_df.iterrows():
                tree_diagram = visualization.build_tree_diagram(row)
                
                # Get the student's name from the DataFrame
                student_name = row['Student']
                
                # Use st.expander to wrap the graphviz chart with the student's name
                with st.expander(f"{student_name} Decision Tree", expanded=False):
                    st.graphviz_chart(tree_diagram.source)
            
            # Prepare input for the language model
            llm_input = ai_analysis.prepare_llm_input(student_metrics_df)

            # Generate Notes and Recommendations using Hugging Face LLM
            with st.spinner("Generating AI analysis..."):
                recommendations = ai_analysis.prompt_response_from_hf_llm(llm_input)

            st.subheader("AI Analysis")
            st.markdown(recommendations)

            # Download AI output
            ai_analysis.download_llm_output(recommendations, "llm_output.txt")

        except Exception as e:
            st.error(f"Error processing the file: {str(e)}")

if __name__ == '__main__':
    main()