global_func/load_contest_file.py

import streamlit as st
import pandas as pd

def load_contest_file(upload, sport):
    pos_values = ['P', 'C', '1B', '2B', '3B', 'SS', 'OF']
    if upload is not None:
        try:
            try:

                if upload.name.endswith('.csv'):
                    raw_df = pd.read_csv(upload)
                elif upload.name.endswith(('.xls', '.xlsx')):
                    raw_df = pd.read_excel(upload)
                else:
                    st.error('Please upload either a CSV or Excel file')
                    return None
            except:
                raw_df = upload
            
            df = raw_df[['EntryId', 'EntryName', 'TimeRemaining', 'Points', 'Lineup', 'Player', 'Roster Position', '%Drafted', 'FPTS', 'Salary', 'Team']]
            df = df.rename(columns={'Roster Position': 'Pos', '%Drafted': 'Own'})
            
            # Split EntryName into base name and entry count
            df['BaseName'] = df['EntryName'].str.replace(r'\s*\(\d+/\d+\)$', '', regex=True)
            df['EntryCount'] = df['EntryName'].str.extract(r'\((\d+/\d+)\)')
            df['EntryCount'] = df['EntryCount'].fillna('1/1')  # Default to 1/1 if no entry count
            
            # Split the lineup string by replacing position indicators with commas
            # We need to ensure we only replace position indicators that are at the start of a player entry
            # and not those that might appear within player names
            df['Lineup'] = df['Lineup'].str.replace(r'\b(' + '|'.join(pos_values) + r')\b', r'\1,', regex=True)
            
            # Split into individual columns and remove position indicators
            # First, determine the maximum number of players in any lineup
            max_players = int(df['Lineup'].str.split(',').str.len().max())
            
            if max_players <= 0:
                st.error('No valid lineups found in the uploaded file')
                return None
                
            # Create columns for each player
            for i in range(1, max_players):
                df[i] = df['Lineup'].str.split(',').str[i].str.strip()
                # Remove position indicators from the end of each entry
                df[i] = df[i].str.replace(r'\s+(' + '|'.join(pos_values) + r')$', '', regex=True)
                # Replace None with -1
                df[i] = df[i].fillna('-1')
            
            if sport == 'MLB':
                df = df.rename(columns={1: '1B', 2: '2B', 3: '3B', 4: 'C', 5: 'OF1', 6: 'OF2', 7: 'OF3', 8: 'SP1', 9: 'SP2', 10: 'SS'})
            try:
                df['Own'] = df['Own'].str.replace('%', '').astype(float)
            except:
                df['Own'] = df['Own'].astype(float)
            ownership_df = df[['Player', 'Own']]
            fpts_df = df[['Player', 'FPTS']]
            salary_df = df[['Player', 'Salary']]
            team_df = df[['Player', 'Team']]
            pos_df = df[['Player', 'Pos']]
            
            # Create position mapping dictionary
            pos_dict = dict(zip(pos_df['Player'], pos_df['Pos']))
            
            # Debug prints
            print("\nPosition Dictionary:")
            print(pos_dict)
            
            print("\nSample Lineup String:")
            print(df['Lineup'].iloc[0])  # Print first lineup
            
            # Function to check if player is eligible for position
            def is_eligible_for_position(player, target_pos):
                if player not in pos_dict:
                    print(f"Player not found in pos_dict: {player}")
                    return False
                player_positions = pos_dict[player].split('/')
                print(f"Checking {player} for {target_pos}. Player positions: {player_positions}")
                # Handle special cases
                if target_pos.startswith('SP') and 'P' in player_positions:
                    return True
                if target_pos.startswith('OF') and 'OF' in player_positions:
                    return True
                return target_pos in player_positions
            
            # Process each lineup
            for idx, row in df.iterrows():
                # Get all players in the lineup and clean up the strings
                players = row['Lineup'].split(',')
                cleaned_players = []
                current_position = None
                
                for item in players:
                    item = item.strip()
                    # If the item is just a position indicator, store it
                    if item in pos_values:
                        current_position = item
                        continue
                    
                    # If we have a position and a player name
                    if current_position:
                        # Remove any trailing position indicators
                        for pos in pos_values:
                            if item.endswith(pos):
                                item = item[:-len(pos)].strip()
                        cleaned_players.append(item)
                        current_position = None
                    else:
                        # If we somehow got a player without a position, try to clean it
                        for pos in pos_values:
                            if item.endswith(pos):
                                item = item[:-len(pos)].strip()
                                break
                        cleaned_players.append(item)
                
                print(f"\nProcessing lineup {idx}:")
                print(f"Original lineup: {row['Lineup']}")
                print(f"Cleaned players: {cleaned_players}")
                
                # First pass: fill required positions (excluding OF)
                required_positions = ['SP1', 'SP2', 'C', '1B', '2B', '3B', 'SS']
                for pos in required_positions:
                    for player in cleaned_players:
                        if is_eligible_for_position(player, pos):
                            print(f"Assigning {player} to {pos}")
                            df.at[idx, pos] = player
                            cleaned_players.remove(player)
                            break
                    else:
                        print(f"No player found for {pos}")
                
                # Second pass: fill OF positions with remaining players
                of_positions = ['OF1', 'OF2', 'OF3']
                for pos in of_positions:
                    for player in cleaned_players:
                        if 'OF' in pos_dict.get(player, '').split('/'):
                            print(f"Assigning {player} to {pos}")
                            df.at[idx, pos] = player
                            cleaned_players.remove(player)
                            break
                    else:
                        print(f"No player found for {pos}, using -1")
                        df.at[idx, pos] = '-1'
            
            cleaned_df = df.drop(columns=['EntryId', 'EntryName', 'TimeRemaining', 'Points', 'Lineup', 'Player', 'Pos', 'Own', 'FPTS', 'Salary', 'Team'])
            cleaned_df = cleaned_df[['BaseName', 'EntryCount', 'SP1', 'SP2', 'C', '1B', '2B', '3B', 'SS', 'OF1', 'OF2', 'OF3']]
            entry_list = list(set(df['BaseName']))
            entry_list.sort()
            
            return cleaned_df, ownership_df, fpts_df, salary_df, team_df, pos_df, entry_list
        except Exception as e:
            st.error(f'Error loading file: {str(e)}')
            return None
    return None