import gradio as gr
import pandas as pd
import bm25s
from rapidfuzz import fuzz
import numpy as np
import matplotlib.pyplot as plt


# 1. Load the Data
abt_df = pd.read_csv("Abt.csv", encoding='latin1') #.rename(columns={'id': 'idAbt', 'name': 'name_A'})
buy_df = pd.read_csv("Buy.csv", encoding='latin1') #.rename(columns={'id': 'idBuy', 'name': 'name_B'})

print("Abt Missing Data:")
print(abt_df.isnull().sum())

print("\nBuy Missing Data:")
print(buy_df.isnull().sum())


# Calculate string lengths
abt_df['name_length'] = abt_df['name'].str.len()
buy_df['name_length'] = buy_df['name'].str.len()


def generate_plot():
    fig = plt.figure()
    plt.hist(abt_df['name_length'], alpha=0.5, label='Abt')
    plt.hist(buy_df['name_length'], alpha=0.5, label='Buy')
    plt.legend()
    return fig # Return the figure, DO NOT use plt.show()

# Then in your Gradio layout:
with gr.Blocks() as demo:
    gr.Markdown("# Welcome to the Product Matcher")
    
    with gr.Row():
        my_plot = gr.Plot(value=generate_plot()) # Gradio handles the drawing




print(f"Abt Price Range: ${abt_df['price'].min()} to ${abt_df['price'].max()}")
print(f"Buy Price Range: ${buy_df['price'].min()} to ${buy_df['price'].max()}")