import streamlit as st import numpy as np import pandas as pd import time from fuzzywuzzy import process def find_name_mismatches(portfolio_df, projections_df): # Create a copy of the projections dataframe to avoid modifying the original projections_df = projections_df.copy() col_count = len(portfolio_df.columns) portfolio_df.columns = range(col_count) if 'player_names' not in projections_df.columns: st.error("No 'player_names' column found in projections file") return projections_df # Get unique player names from portfolio and projections portfolio_players = set() for col in portfolio_df.columns: portfolio_players.update(portfolio_df[col].unique()) projection_players = set(projections_df['player_names'].unique()) projection_players_list = list(projection_players) # Find players in portfolio that are missing from projections players_missing_from_projections = list(portfolio_players - projection_players) # Automatically handle 100% matches before starting interactive process players_to_process = [] for player in players_missing_from_projections: if not isinstance(player, str): st.warning(f"Skipping non-string value: {player}") continue closest_matches = process.extract(player, projection_players_list, limit=1) if closest_matches[0][1] == 100: # If perfect match found match_name = closest_matches[0][0] projections_df.loc[projections_df['player_names'] == match_name, 'player_names'] = player st.success(f"Automatically matched '{match_name}' with '{player}' (100% match)") else: players_to_process.append(player) # Initialize session state for tracking current player if not exists if 'current_player_index' not in st.session_state: st.session_state.current_player_index = 0 st.session_state.players_to_process = players_to_process # Display results if players_missing_from_projections: st.warning("Players in portfolio but missing from projections") # Display remaining players remaining_players = st.session_state.players_to_process[st.session_state.current_player_index:] st.info(f"Remaining players to process ({len(remaining_players)}):\n" + "\n".join(f"- {player}" for player in remaining_players)) if st.session_state.current_player_index < len(st.session_state.players_to_process): current_player = st.session_state.players_to_process[st.session_state.current_player_index] # Find the top 3 closest matches closest_matches = process.extract(current_player, projection_players_list, limit=3) st.write(f"**Missing Player {st.session_state.current_player_index + 1} of {len(st.session_state.players_to_process)}:** {current_player}") # Create radio buttons for selection options = [f"{match[0]} ({match[1]}%)" for match in closest_matches] options.append("None of these") selected_option = st.radio( f"Select correct match:", options, key=f"radio_{current_player}" ) if st.button("Confirm Selection"): if selected_option != "None of these": selected_name = selected_option.split(" (")[0] projections_df.loc[projections_df['player_names'] == selected_name, 'player_names'] = current_player st.success(f"Replaced '{selected_name}' with '{current_player}'") st.session_state['projections_df'] = projections_df # Move to next player st.session_state.current_player_index += 1 st.rerun() else: st.success("All players have been processed!") # Reset the index for future runs st.session_state.current_player_index = 0 st.session_state.players_to_process = [] else: st.success("All portfolio players found in projections!") return projections_df