global_func/grab_contest_data.py

import pandas as pd
import requests

def grab_contest_data(sport, contest_name, contest_id_map, contest_date, contest_date2):

    if sport == 'GOLF':
        sport_var = 'pga'
    else:
        sport_var = sport.lower()
        
    contest_id = contest_id_map[contest_name]
    raw_url = f'https://dh5nxc6yx3kwy.cloudfront.net/contests/{sport_var}/{contest_date}/{contest_id}/'
    raw_url2 = f'https://dh5nxc6yx3kwy.cloudfront.net/contests/{sport_var}/{contest_date2}/{contest_id}/'

    def format_lineup_string(lineup_hash, positions):
        """Replaces colons in a lineup hash with sequential positions."""
        # Remove the leading colon and split by the remaining colons
        # Only split on colons that are between numbers
        player_ids = [pid for pid in lineup_hash.lstrip(':').split(':') if pid.isdigit() or pid == '-1']
        
        # Get the actual positions needed for this lineup
        actual_positions = positions[:len(player_ids)]
        
        # Check if the number of IDs matches the number of positions
        if len(player_ids) != len(actual_positions):
            print(f"Warning: Mismatch for hash {lineup_hash}. IDs: {len(player_ids)}, Positions: {len(actual_positions)}")
            return lineup_hash

        # Combine positions and player IDs, only for valid player IDs
        combined_parts = []
        for pos, pid in zip(actual_positions, player_ids):
            if pid == '-1':  # Handle empty slots
                combined_parts.append(pos + '')
            else:
                combined_parts.append(pos + pid_map.get(pid, pid))
        
        # # Just combine positions and PIDs without name mapping
        # combined_parts = [pos + pid for pos, pid in zip(actual_positions, player_ids)]
        
        # Join them into a single string
        return "".join(combined_parts)

    try:
        lineups_json = requests.get(raw_url + 'lineups/').json()
        data_json = requests.get(raw_url + 'data/').json()
    except:
        lineups_json = requests.get(raw_url2 + 'lineups/').json()
        data_json = requests.get(raw_url2 + 'data/').json()

    lineup_data = []
    player_data = []
    position_inserts = ['1B ', ' 2B ', ' 3B ', ' C ', ' OF ', ' OF ', ' OF ', ' P ', ' P ', ' SS ']

    for players, player_info in data_json['players'].items():
        player_data.append({
            'fullName': player_info['fullName'],
            'playerId': player_info['playerId'],
            'salary': player_info['salary'],
            'currentTeam': player_info['currentTeam'],
            'rosterPosition': player_info['rosterPosition'],
            'ownership': player_info['ownership'],
            'actualPoints': player_info['actualPoints']
        })

    players_df = pd.DataFrame(player_data)
    players_df = players_df.sort_values(by='ownership', ascending=False).reset_index(drop=True)
    players_df = players_df.rename(columns={'fullName': 'Player', 'rosterPosition': 'Roster Position', 'ownership': '%Drafted', 'actualPoints': 'FPTS', 'salary': 'Salary', 'currentTeam': 'Team'})
    pid_map = dict(zip(players_df['playerId'].astype(str), players_df['Player']))

    for lineup_hash, lineup_info in lineups_json['lineups'].items():
        lineup_data.append({
            'lineupHash': lineup_hash,
            'points': lineup_info['points'],
            'entryNameList': lineup_info['entryNameList'][0]
        })

    lineups_df = pd.DataFrame(lineup_data)
    lineups_df = lineups_df.sort_values(by='points', ascending=False)
    lineups_df = lineups_df.reset_index()
    lineups_df['index'] = lineups_df.index + 1
    lineups_df['TimeRemaining'] = str(0)
    lineups_df['EntryId'] = lineups_df['index'].astype(str) + lineups_df['entryNameList'].astype(str)
    lineups_df['lineupHash'] = ':' + lineups_df['lineupHash']
    lineups_df = lineups_df.rename(columns={'index': 'Rank', 'points': 'Points', 'entryNameList': 'EntryName', 'lineupHash': 'Lineup'})
    lineups_df['EntryName'] = lineups_df['EntryName'] + ' (1/1)'
    lineups_df['Lineup'] = lineups_df['Lineup'].apply(lambda x: format_lineup_string(x, position_inserts))
    lineups_df = lineups_df[['Rank', 'EntryId', 'EntryName', 'TimeRemaining', 'Points', 'Lineup']]

    total_data = lineups_df.merge(players_df, how='left', left_index=True, right_index=True)
    
    return total_data