import pandas as pd
import streamlit as st
from difflib import SequenceMatcher

def read_csv_or_excel(file):
    # Read CSV or Excel file
    if file.name.endswith('.csv'):
        return pd.read_csv(file)
    elif file.name.endswith('.xlsx') or file.name.endswith('.xls'):
        return pd.read_excel(file)
    else:
        raise ValueError("Unsupported file format. Only CSV and Excel files are supported.")

def find_exact_matches(df1, df2, column_name):
    # Find rows with exact matches in the specified column
    matches = pd.merge(df1, df2, on=column_name, how='inner')
    return matches

def find_similar_texts(df1, df2, column_name, threshold=0.8):
    # Find rows with similar texts in the specified column
    similar_texts = []
    for index1, row1 in df1.iterrows():
        for index2, row2 in df2.iterrows():
            similarity = SequenceMatcher(None, str(row1[column_name]), str(row2[column_name])).ratio()
            if similarity >= threshold:
                similar_texts.append((index1, index2, row1[column_name], row2[column_name]))
    return similar_texts

def main():
    st.title("Item Comparison App")

    # Upload files
    st.header("Upload Files")
    warehouse_file = st.file_uploader("Upload Warehouse Item Stocks (CSV or Excel)")
    industry_file = st.file_uploader("Upload Industry Item Stocks (CSV or Excel)")

    if warehouse_file is not None and industry_file is not None:
        # Read files
        warehouse_df = read_csv_or_excel(warehouse_file)
        industry_df = read_csv_or_excel(industry_file)

        # Get column names
        warehouse_columns = warehouse_df.columns.tolist()
        industry_columns = industry_df.columns.tolist()

        # Select columns using dropdowns
        st.header("Select Columns")
        warehouse_column = st.selectbox("Choose column from warehouse item stocks:", warehouse_columns)
        industry_column = st.selectbox("Choose column from industry item stocks:", industry_columns)

        # Find exact matches
        exact_matches = find_exact_matches(warehouse_df, industry_df, warehouse_column)

        # Find similar texts
        similar_texts = find_similar_texts(warehouse_df, industry_df, warehouse_column)

        # Display results
        st.header("Exact Matches")
        st.write(exact_matches)

        st.header("Similar Texts")
        for text_pair in similar_texts:
            st.write(f"Row {text_pair[0]} in warehouse item stocks is similar to Row {text_pair[1]} in industry item stocks:")
            st.write(f"Warehouse: {text_pair[2]}")
            st.write(f"Industry: {text_pair[3]}")
            st.write("")

if __name__ == "__main__":
    main()