global_func/find_name_mismatches.py

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):
    """
    Find and handle name mismatches between portfolio and projections dataframes.
    Returns the updated projections dataframe with matched names.
    """
    # Get all player names from portfolio
    portfolio_players = set()
    for col in portfolio_df.columns:
        if col not in ['salary', 'median', 'Own', 'Finish_percentile', 'Dupes', 'Stack', 'Win%', 'Lineup Edge']:
            portfolio_players.update(portfolio_df[col].unique())
    
    # Get all player names from projections
    projection_players_list = projections_df['player_names'].tolist()
    
    # Find players in portfolio that are missing from projections
    players_missing_from_projections = [player for player in portfolio_players if player not in projection_players_list]
    
    # 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] == 90:  # 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)
    
    # Display results
    if players_missing_from_projections:
        st.warning("Players in portfolio but missing from projections")
        
        # Display remaining players
        if players_to_process:
            st.info(f"Players to process ({len(players_to_process)}):\n" + 
                    "\n".join(f"- {player}" for player in players_to_process))
            
            # Create a form for batch processing
            with st.form("player_matching_form"):
                # Create tabs for each player
                tabs = st.tabs([f"Player {i+1}" for i in range(len(players_to_process))])
                
                # Dictionary to store selections
                selections = {}
                
                # Process each tab
                for idx, (tab, player) in enumerate(zip(tabs, players_to_process)):
                    with tab:
                        st.write(f"**Missing Player {idx + 1} of {len(players_to_process)}:** {player}")
                        
                        # Find the top 3 closest matches
                        closest_matches = process.extract(player, projection_players_list, limit=3)
                        
                        # 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 for {player}:",
                            options,
                            key=f"radio_{player}"
                        )
                        
                        selections[player] = selected_option
                
                # Submit button for the entire form
                submitted = st.form_submit_button("Submit All Changes")
                
                if submitted:
                    # Process all selections
                    for player, selection in selections.items():
                        if selection != "None of these":
                            selected_name = selection.split(" (")[0]
                            projections_df.loc[projections_df['player_names'] == selected_name, 'player_names'] = player
                            st.success(f"Replaced '{selected_name}' with '{player}'")
                    
                    # Update session state
                    st.session_state['projections_df'] = projections_df
                    st.success("All player name changes have been applied!")
    else:
        st.success("All portfolio players found in projections!")
    
    return projections_df