import gradio as gr
import pandas as pd
import matplotlib.pyplot as plt
from io import StringIO
from termcolor import colored

# Load the dataset with debug prints
def load_dataset():
    try:
        print(colored("Loading dataset...", "blue"))
        df = pd.read_csv("data/source/all_networks_developer_classification.csv")
        # Ensure the month_year column is in the correct datetime format
        df['month_year'] = pd.to_datetime(df['month_year'], format='%B_%Y')  # Adjust format if necessary
        return df
    except Exception as e:
        print(colored(f"Error loading dataset: {e}", "red"))
        raise

# Process input and generate plot and classification with debug prints
def process_input(input_text, uploaded_file):
    try:
        print(colored("Processing input...", "blue"))

        # Read GitHub handles from input text or uploaded file
        if uploaded_file is not None:
            print(colored("Reading from uploaded file...", "blue"))
            input_text = uploaded_file.read().decode("utf-8")
        github_handles = [handle.strip() for handle in input_text.split(",")]
        print(colored(f"GitHub handles: {github_handles}", "blue"))

        # Load dataset
        df = load_dataset()

        # Filter dataset for the provided GitHub handles
        print(colored("Filtering dataset...", "blue"))
        filtered_df = df[df['developer'].isin(github_handles)]

        # Generate plot
        print(colored("Generating plot...", "blue"))
        fig, ax = plt.subplots()
        for handle in github_handles:
            dev_df = filtered_df[filtered_df['developer'] == handle]
            dev_df = dev_df.sort_values('month_year')
            ax.plot(dev_df['month_year'], dev_df['total_commits'], label=handle)

        ax.set_xlabel("Month")
        ax.set_ylabel("Number of Commits")
        ax.legend()
        plt.xticks(rotation=45)
        plt.tight_layout()

        # Generate classification table
        print(colored("Classifying developers...", "blue"))
        classification = []
        for handle in github_handles:
            dev_df = filtered_df[filtered_df['developer'] == handle]
            last_3_months = pd.Timestamp.now() - pd.DateOffset(months=3)
            recent_activity = dev_df[dev_df['month_year'] >= last_3_months]
            total_recent_commits = recent_activity['total_commits'].sum()
            
            if dev_df.empty:
                status = "Always been inactive"
            elif recent_activity.empty:
                status = "Previously active but no longer"
            elif total_recent_commits < 20:
                status = "Low-level active"
            else:
                status = "Highly involved"
            
            classification.append((handle, status))
        
        classification_df = pd.DataFrame(classification, columns=["Developer", "Classification"]).sort_values("Classification", ascending=False)
        print(colored("Classification completed.", "blue"))

        # Return plot and classification table
        return fig, classification_df
    except Exception as e:
        print(colored(f"Error processing input: {e}", "red"))
        raise

# Gradio interface with descriptions and debug prints
with gr.Blocks() as app:
    gr.Markdown("## GitHub Starknet Developer Insights")
    gr.Markdown("""
    This tool allows you to analyze the GitHub activity of developers within the Starknet ecosystem. 
    Enter GitHub handles separated by commas or upload a CSV file with GitHub handles in a single column 
    to see their monthly commit activity and their current involvement classification.
    """)
    
    with gr.Row():
        text_input = gr.Textbox(label="Enter GitHub handles separated by commas", placeholder="e.g., user1,user2,user3")
        file_input = gr.File(label="Or upload a CSV file with GitHub handles in a single column", type="binary")
    gr.Markdown("""
    *Note:* When uploading a CSV, ensure it contains a single column of GitHub handles without a header row.
    """)
    btn = gr.Button("Analyze")
    plot_output = gr.Plot(label="Commits per Month")
    table_output = gr.Dataframe(label="Developer Classification")

    btn.click(process_input, inputs=[text_input, file_input], outputs=[plot_output, table_output])

    print(colored("Gradio app initialized.", "blue"))

if __name__ == "__main__":
    print(colored("Launching app...", "blue"))
    app.launch(share=True)