""" BET.CUTTALO.COM - FIXTURE COLLECTOR Raccoglie partite future con data e ora da API gratuite """ import requests import pandas as pd import json import os from datetime import datetime, timedelta from typing import List, Dict import time DATA_DIR = os.path.join(os.path.dirname(__file__), '..', 'data') os.makedirs(DATA_DIR, exist_ok=True) # Stadium coordinates for major teams (for timezone/location info) TEAM_LOCATIONS = { # Premier League 'Arsenal': {'city': 'London', 'stadium': 'Emirates Stadium', 'country': 'England'}, 'Aston Villa': {'city': 'Birmingham', 'stadium': 'Villa Park', 'country': 'England'}, 'Bournemouth': {'city': 'Bournemouth', 'stadium': 'Vitality Stadium', 'country': 'England'}, 'Brentford': {'city': 'London', 'stadium': 'Gtech Community Stadium', 'country': 'England'}, 'Brighton': {'city': 'Brighton', 'stadium': 'Amex Stadium', 'country': 'England'}, 'Chelsea': {'city': 'London', 'stadium': 'Stamford Bridge', 'country': 'England'}, 'Crystal Palace': {'city': 'London', 'stadium': 'Selhurst Park', 'country': 'England'}, 'Everton': {'city': 'Liverpool', 'stadium': 'Goodison Park', 'country': 'England'}, 'Fulham': {'city': 'London', 'stadium': 'Craven Cottage', 'country': 'England'}, 'Ipswich': {'city': 'Ipswich', 'stadium': 'Portman Road', 'country': 'England'}, 'Leicester': {'city': 'Leicester', 'stadium': 'King Power Stadium', 'country': 'England'}, 'Liverpool': {'city': 'Liverpool', 'stadium': 'Anfield', 'country': 'England'}, 'Man City': {'city': 'Manchester', 'stadium': 'Etihad Stadium', 'country': 'England'}, 'Man United': {'city': 'Manchester', 'stadium': 'Old Trafford', 'country': 'England'}, 'Newcastle': {'city': 'Newcastle', 'stadium': "St James' Park", 'country': 'England'}, "Nott'm Forest": {'city': 'Nottingham', 'stadium': 'City Ground', 'country': 'England'}, 'Southampton': {'city': 'Southampton', 'stadium': "St Mary's Stadium", 'country': 'England'}, 'Tottenham': {'city': 'London', 'stadium': 'Tottenham Hotspur Stadium', 'country': 'England'}, 'West Ham': {'city': 'London', 'stadium': 'London Stadium', 'country': 'England'}, 'Wolves': {'city': 'Wolverhampton', 'stadium': 'Molineux Stadium', 'country': 'England'}, # Serie A 'AC Milan': {'city': 'Milan', 'stadium': 'San Siro', 'country': 'Italy'}, 'Milan': {'city': 'Milan', 'stadium': 'San Siro', 'country': 'Italy'}, 'Inter': {'city': 'Milan', 'stadium': 'San Siro', 'country': 'Italy'}, 'Juventus': {'city': 'Turin', 'stadium': 'Allianz Stadium', 'country': 'Italy'}, 'Napoli': {'city': 'Naples', 'stadium': 'Diego Armando Maradona', 'country': 'Italy'}, 'Roma': {'city': 'Rome', 'stadium': 'Stadio Olimpico', 'country': 'Italy'}, 'Lazio': {'city': 'Rome', 'stadium': 'Stadio Olimpico', 'country': 'Italy'}, 'Atalanta': {'city': 'Bergamo', 'stadium': 'Gewiss Stadium', 'country': 'Italy'}, 'Fiorentina': {'city': 'Florence', 'stadium': 'Stadio Artemio Franchi', 'country': 'Italy'}, 'Bologna': {'city': 'Bologna', 'stadium': 'Stadio Renato Dall\'Ara', 'country': 'Italy'}, 'Torino': {'city': 'Turin', 'stadium': 'Stadio Olimpico Grande Torino', 'country': 'Italy'}, # La Liga 'Real Madrid': {'city': 'Madrid', 'stadium': 'Santiago Bernabeu', 'country': 'Spain'}, 'Barcelona': {'city': 'Barcelona', 'stadium': 'Spotify Camp Nou', 'country': 'Spain'}, 'Ath Madrid': {'city': 'Madrid', 'stadium': 'Civitas Metropolitano', 'country': 'Spain'}, 'Sevilla': {'city': 'Seville', 'stadium': 'Ramon Sanchez Pizjuan', 'country': 'Spain'}, 'Valencia': {'city': 'Valencia', 'stadium': 'Mestalla', 'country': 'Spain'}, 'Villarreal': {'city': 'Villarreal', 'stadium': 'Estadio de la Ceramica', 'country': 'Spain'}, 'Real Betis': {'city': 'Seville', 'stadium': 'Benito Villamarin', 'country': 'Spain'}, 'Ath Bilbao': {'city': 'Bilbao', 'stadium': 'San Mames', 'country': 'Spain'}, # Bundesliga 'Bayern Munich': {'city': 'Munich', 'stadium': 'Allianz Arena', 'country': 'Germany'}, 'Dortmund': {'city': 'Dortmund', 'stadium': 'Signal Iduna Park', 'country': 'Germany'}, 'RB Leipzig': {'city': 'Leipzig', 'stadium': 'Red Bull Arena', 'country': 'Germany'}, 'Leverkusen': {'city': 'Leverkusen', 'stadium': 'BayArena', 'country': 'Germany'}, 'Ein Frankfurt': {'city': 'Frankfurt', 'stadium': 'Deutsche Bank Park', 'country': 'Germany'}, 'Wolfsburg': {'city': 'Wolfsburg', 'stadium': 'Volkswagen Arena', 'country': 'Germany'}, "M'gladbach": {'city': 'Monchengladbach', 'stadium': 'Borussia-Park', 'country': 'Germany'}, 'Stuttgart': {'city': 'Stuttgart', 'stadium': 'MHPArena', 'country': 'Germany'}, # Ligue 1 'Paris SG': {'city': 'Paris', 'stadium': 'Parc des Princes', 'country': 'France'}, 'Marseille': {'city': 'Marseille', 'stadium': 'Stade Velodrome', 'country': 'France'}, 'Lyon': {'city': 'Lyon', 'stadium': 'Groupama Stadium', 'country': 'France'}, 'Monaco': {'city': 'Monaco', 'stadium': 'Stade Louis II', 'country': 'France'}, 'Lille': {'city': 'Lille', 'stadium': 'Stade Pierre-Mauroy', 'country': 'France'}, 'Nice': {'city': 'Nice', 'stadium': 'Allianz Riviera', 'country': 'France'}, } class FixtureCollector: """ Raccoglie fixture future da API gratuite """ def __init__(self): self.session = requests.Session() self.session.headers.update({ 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36' }) def get_fixtures_openligadb(self, league: str = 'bl1', season: int = 2024) -> List[Dict]: """ Ottieni fixture da OpenLigaDB (gratuito, no key) Supporta: bl1 (Bundesliga), bl2, bl3, dfb (DFB Pokal) """ fixtures = [] try: url = f"https://api.openligadb.de/getmatchdata/{league}/{season}" response = self.session.get(url, timeout=30) data = response.json() now = datetime.now() for match in data: match_date = datetime.fromisoformat(match['matchDateTime'].replace('Z', '+00:00')) # Solo partite future non ancora giocate if match_date.replace(tzinfo=None) > now and not match.get('matchIsFinished', True): home_team = match['team1']['teamName'] away_team = match['team2']['teamName'] fixtures.append({ 'id': f"openliga_{match['matchID']}", 'source': 'OpenLigaDB', 'league_code': 'D1', 'league_name': 'Bundesliga', 'country': 'Germany', 'matchday': match.get('group', {}).get('groupOrderID', 0), 'date': match_date.strftime('%Y-%m-%d'), 'time': match_date.strftime('%H:%M'), 'datetime': match_date.isoformat(), 'timestamp': int(match_date.timestamp()), 'home_team': home_team, 'away_team': away_team, 'home_team_logo': match['team1'].get('teamIconUrl', ''), 'away_team_logo': match['team2'].get('teamIconUrl', ''), 'venue': TEAM_LOCATIONS.get(home_team, {}).get('stadium', ''), 'city': TEAM_LOCATIONS.get(home_team, {}).get('city', ''), 'status': 'SCHEDULED' }) print(f"OpenLigaDB: {len(fixtures)} fixture Bundesliga") except Exception as e: print(f"Errore OpenLigaDB: {e}") return fixtures def get_fixtures_from_footballdata(self) -> List[Dict]: """ Scarica fixture REALI da football-data.co.uk Include molte più leghe con quote vere """ fixtures = [] # Tutte le leghe disponibili su football-data.co.uk LEAGUES_INFO = { # Inghilterra 'E0': {'name': 'Premier League', 'country': 'England', 'tier': 1, 'flag': '🏴󠁧󠁢󠁥󠁮󠁧󠁿', 'color': '#38003c'}, 'E1': {'name': 'Championship', 'country': 'England', 'tier': 2, 'flag': '🏴󠁧󠁢󠁥󠁮󠁧󠁿', 'color': '#ff6b00'}, 'E2': {'name': 'League One', 'country': 'England', 'tier': 3, 'flag': '🏴󠁧󠁢󠁥󠁮󠁧󠁿', 'color': '#e62333'}, 'E3': {'name': 'League Two', 'country': 'England', 'tier': 4, 'flag': '🏴󠁧󠁢󠁥󠁮󠁧󠁿', 'color': '#0057b8'}, 'EC': {'name': 'Conference', 'country': 'England', 'tier': 5, 'flag': '🏴󠁧󠁢󠁥󠁮󠁧󠁿', 'color': '#1e3a5f'}, # Scozia 'SC0': {'name': 'Scottish Premiership', 'country': 'Scotland', 'tier': 1, 'flag': '🏴󠁧󠁢󠁳󠁣󠁴󠁿', 'color': '#004b87'}, 'SC1': {'name': 'Scottish Championship', 'country': 'Scotland', 'tier': 2, 'flag': '🏴󠁧󠁢󠁳󠁣󠁴󠁿', 'color': '#4a7c59'}, 'SC2': {'name': 'Scottish League 1', 'country': 'Scotland', 'tier': 3, 'flag': '🏴󠁧󠁢󠁳󠁣󠁴󠁿', 'color': '#6b4423'}, 'SC3': {'name': 'Scottish League 2', 'country': 'Scotland', 'tier': 4, 'flag': '🏴󠁧󠁢󠁳󠁣󠁴󠁿', 'color': '#8b4513'}, # Germania 'D1': {'name': 'Bundesliga', 'country': 'Germany', 'tier': 1, 'flag': '🇩🇪', 'color': '#d20515'}, 'D2': {'name': '2. Bundesliga', 'country': 'Germany', 'tier': 2, 'flag': '🇩🇪', 'color': '#004d40'}, # Italia 'I1': {'name': 'Serie A', 'country': 'Italy', 'tier': 1, 'flag': '🇮🇹', 'color': '#024494'}, 'I2': {'name': 'Serie B', 'country': 'Italy', 'tier': 2, 'flag': '🇮🇹', 'color': '#005c3a'}, # Spagna 'SP1': {'name': 'La Liga', 'country': 'Spain', 'tier': 1, 'flag': '🇪🇸', 'color': '#ee324e'}, 'SP2': {'name': 'La Liga 2', 'country': 'Spain', 'tier': 2, 'flag': '🇪🇸', 'color': '#1a472a'}, # Francia 'F1': {'name': 'Ligue 1', 'country': 'France', 'tier': 1, 'flag': '🇫🇷', 'color': '#091c3e'}, 'F2': {'name': 'Ligue 2', 'country': 'France', 'tier': 2, 'flag': '🇫🇷', 'color': '#2d572c'}, # Olanda 'N1': {'name': 'Eredivisie', 'country': 'Netherlands', 'tier': 1, 'flag': '🇳🇱', 'color': '#ff6600'}, # Belgio 'B1': {'name': 'Jupiler Pro League', 'country': 'Belgium', 'tier': 1, 'flag': '🇧🇪', 'color': '#0a3d62'}, # Portogallo 'P1': {'name': 'Primeira Liga', 'country': 'Portugal', 'tier': 1, 'flag': '🇵🇹', 'color': '#006847'}, # Turchia 'T1': {'name': 'Super Lig', 'country': 'Turkey', 'tier': 1, 'flag': '🇹🇷', 'color': '#e30a17'}, # Grecia 'G1': {'name': 'Super League', 'country': 'Greece', 'tier': 1, 'flag': '🇬🇷', 'color': '#0d5eaf'}, } try: print("\n📥 Scaricando fixture reali da football-data.co.uk...") url = "https://www.football-data.co.uk/fixtures.csv" response = self.session.get(url, timeout=30) response.raise_for_status() from io import StringIO # Remove BOM from text text = response.text.replace('\ufeff', '') if text.startswith(''): text = text[3:] # Remove BOM bytes df = pd.read_csv(StringIO(text)) # Clean column names df.columns = [col.strip() for col in df.columns] now = datetime.now() for idx, row in df.iterrows(): league_code = str(row.get('Div', '')).strip() if not league_code or league_code == 'nan' or league_code not in LEAGUES_INFO: continue league = LEAGUES_INFO[league_code] home_team = str(row.get('HomeTeam', '')).strip() away_team = str(row.get('AwayTeam', '')).strip() if not home_team or not away_team: continue # Parse date date_str = str(row.get('Date', '')) time_str = str(row.get('Time', '15:00')) try: # Try different date formats for fmt in ['%d/%m/%Y', '%Y-%m-%d', '%d/%m/%y']: try: match_date = datetime.strptime(date_str, fmt) break except: continue else: continue # Add time if time_str and time_str != 'nan': try: time_parts = time_str.split(':') match_date = match_date.replace(hour=int(time_parts[0]), minute=int(time_parts[1]) if len(time_parts) > 1 else 0) except: match_date = match_date.replace(hour=15, minute=0) else: match_date = match_date.replace(hour=15, minute=0) # Only include future matches (not started yet) # Skip matches that have already started (kickoff + 15 min buffer) if match_date < now - timedelta(minutes=15): continue except Exception as e: continue # Get odds from the CSV (Bet365 primary, then average) try: odds_home = float(row.get('B365H') or row.get('AvgH') or 2.0) odds_draw = float(row.get('B365D') or row.get('AvgD') or 3.5) odds_away = float(row.get('B365A') or row.get('AvgA') or 3.0) except: odds_home, odds_draw, odds_away = 2.0, 3.5, 3.0 # All matches here are future/upcoming is_past = False status = 'SCHEDULED' fixtures.append({ 'id': f"fd_{league_code}_{idx}_{int(match_date.timestamp())}", 'source': 'football-data.co.uk', 'league_code': league_code, 'league_name': league['name'], 'country': league['country'], 'tier': league['tier'], 'flag': league['flag'], 'league_color': league['color'], 'matchday': 0, 'date': match_date.strftime('%Y-%m-%d'), 'time': match_date.strftime('%H:%M'), 'datetime': match_date.isoformat(), 'timestamp': int(match_date.timestamp()), 'home_team': home_team, 'away_team': away_team, 'home_team_logo': '', 'away_team_logo': '', 'venue': TEAM_LOCATIONS.get(home_team, {}).get('stadium', ''), 'city': TEAM_LOCATIONS.get(home_team, {}).get('city', ''), 'status': status, 'is_past': is_past, 'odds_home': odds_home, 'odds_draw': odds_draw, 'odds_away': odds_away, 'odds_source': 'Bet365' }) # Count by league by_league = {} for f in fixtures: lc = f['league_code'] by_league[lc] = by_league.get(lc, 0) + 1 print(f" ✓ {len(fixtures)} fixture reali scaricate") for lc, count in sorted(by_league.items()): if lc in LEAGUES_INFO: print(f" {LEAGUES_INFO[lc]['flag']} {LEAGUES_INFO[lc]['name']}: {count}") except Exception as e: print(f" ✗ Errore scaricamento: {e}") import traceback traceback.print_exc() return fixtures def get_fixtures_from_csv(self) -> List[Dict]: """ Genera fixture simulate basate sui dati storici per i campionati senza fixture reali """ fixtures = [] try: # Carica dati storici csv_path = os.path.join(DATA_DIR, 'historical_matches.csv') if not os.path.exists(csv_path): return fixtures df = pd.read_csv(csv_path) df['Date'] = pd.to_datetime(df['Date'], errors='coerce') # Principali leghe europee per cui generare fixture stimate leagues = { 'E0': {'name': 'Premier League', 'country': 'England', 'teams': 20, 'flag': '🏴󠁧󠁢󠁥󠁮󠁧󠁿', 'tier': 1}, 'I1': {'name': 'Serie A', 'country': 'Italy', 'teams': 20, 'flag': '🇮🇹', 'tier': 1}, 'SP1': {'name': 'La Liga', 'country': 'Spain', 'teams': 20, 'flag': '🇪🇸', 'tier': 1}, 'D1': {'name': 'Bundesliga', 'country': 'Germany', 'teams': 18, 'flag': '🇩🇪', 'tier': 1}, 'F1': {'name': 'Ligue 1', 'country': 'France', 'teams': 18, 'flag': '🇫🇷', 'tier': 1}, } now = datetime.now() for league_code, league_info in leagues.items(): league_df = df[df['league_code'] == league_code] current_season = league_df['season'].iloc[-1] if len(league_df) > 0 else '2024/2025' # Ottieni squadre correnti season_df = league_df[league_df['season'] == current_season] teams = list(set(season_df['home_team'].unique()) | set(season_df['away_team'].unique())) if len(teams) < 2: continue # Calcola ultima giornata giocata last_match_date = season_df['Date'].max() # Genera prossime partite (stima) matchday = len(season_df) // (len(teams) // 2) + 1 # Genera fixture per i prossimi 14 giorni # Tipicamente: Sabato/Domenica + partite infrasettimanali time_slots_weekend = ['15:00', '17:30', '18:00', '20:45', '21:00'] time_slots_midweek = ['18:30', '19:00', '20:45', '21:00'] import random # Genera per i prossimi 14 giorni for day_offset in range(1, 15): # Prossimi 14 giorni match_date = now + timedelta(days=day_offset) weekday = match_date.weekday() # 0=Mon, 5=Sat, 6=Sun # Determina quante partite per questo giorno # Weekend: più partite, Infrasettimanale: meno if weekday in [5, 6]: # Weekend max_matches = min(len(teams) // 2, 10) time_slots = time_slots_weekend elif weekday in [1, 2]: # Martedì/Mercoledì - turni infrasettimanali max_matches = min(len(teams) // 2, 6) time_slots = time_slots_midweek else: max_matches = min(len(teams) // 2, 3) # Altri giorni meno partite time_slots = time_slots_midweek # Seed per riproducibilità random.seed(day_offset * 1000 + hash(league_code) + int(now.timestamp() // 86400)) available_teams = teams.copy() random.shuffle(available_teams) for i in range(max_matches): if len(available_teams) < 2: break home = available_teams.pop() away = available_teams.pop() match_time = time_slots[i % len(time_slots)] full_datetime = datetime.combine( match_date.date(), datetime.strptime(match_time, '%H:%M').time() ) # Only add if match is in the future if full_datetime < now - timedelta(minutes=15): continue fixtures.append({ 'id': f"{league_code}_{day_offset}_{i}_{int(now.timestamp())}", 'source': 'Estimated', 'league_code': league_code, 'league_name': league_info['name'], 'country': league_info['country'], 'tier': league_info.get('tier', 1), 'flag': league_info.get('flag', '⚽'), 'league_color': '#666666', 'matchday': matchday + (day_offset // 7), 'date': full_datetime.strftime('%Y-%m-%d'), 'time': match_time, 'datetime': full_datetime.isoformat(), 'timestamp': int(full_datetime.timestamp()), 'home_team': home, 'away_team': away, 'home_team_logo': '', 'away_team_logo': '', 'venue': TEAM_LOCATIONS.get(home, {}).get('stadium', ''), 'city': TEAM_LOCATIONS.get(home, {}).get('city', ''), 'status': 'SCHEDULED', 'is_past': False, 'odds_home': 2.0, 'odds_draw': 3.5, 'odds_away': 3.0, 'odds_source': 'Estimated' }) print(f"Estimated: {len([f for f in fixtures if f['league_code'] == league_code])} fixture {league_info['name']}") except Exception as e: print(f"Errore generazione fixture: {e}") import traceback traceback.print_exc() return fixtures def collect_all_fixtures(self) -> List[Dict]: """ Raccoglie tutte le fixture disponibili """ all_fixtures = [] print("\n" + "="*60) print("RACCOLTA FIXTURE FUTURE - SISTEMA PROFESSIONALE") print("="*60) # 1. Fixture REALI da football-data.co.uk (priorità) real_fixtures = self.get_fixtures_from_footballdata() all_fixtures.extend(real_fixtures) # 2. Bundesliga extra da OpenLigaDB (dettagli aggiuntivi) # Skip se già abbiamo fixture dalla fonte principale if not any(f['league_code'] == 'D1' for f in all_fixtures): bundesliga = self.get_fixtures_openligadb('bl1', 2024) all_fixtures.extend(bundesliga) # 3. Fixture stimate per date future (se non ci sono fixture reali recenti) # Aggiungi solo se abbiamo poche fixture future future_count = sum(1 for f in all_fixtures if not f.get('is_past', False)) if future_count < 20: print(f"\n📅 Solo {future_count} fixture future - generando stimate...") estimated = self.get_fixtures_from_csv() # Prendi solo le fixture future dalle stime estimated = [f for f in estimated if not f.get('is_past', False)] all_fixtures.extend(estimated) print(f" ✓ Aggiunte {len(estimated)} fixture stimate") # Ordina per data all_fixtures.sort(key=lambda x: x['timestamp']) # Salva output_path = os.path.join(DATA_DIR, 'upcoming_fixtures.json') with open(output_path, 'w') as f: json.dump(all_fixtures, f, indent=2, ensure_ascii=False) print(f"\n{'='*60}") print(f"TOTALE: {len(all_fixtures)} fixture salvate") print("="*60) return all_fixtures def fetch_real_odds(): """ Scarica quote reali da football-data.co.uk fixtures.csv Returns dict: {(home_team, away_team): {'home': x.xx, 'draw': x.xx, 'away': x.xx}} """ print("\n📊 Scaricando quote reali da football-data.co.uk...") # Mapping nomi squadre (football-data -> nostro sistema) TEAM_NAME_MAP = { # Bundesliga 'stuttgart': 'stuttgart', 'ein frankfurt': 'frankfurt', 'eintracht frankfurt': 'frankfurt', 'dortmund': 'dortmund', 'werder bremen': 'bremen', 'hamburg': 'hamburg', 'hamburger sv': 'hamburg', 'leverkusen': 'leverkusen', 'bayer leverkusen': 'leverkusen', 'mainz': 'mainz', 'heidenheim': 'heidenheim', 'wolfsburg': 'wolfsburg', 'st pauli': 'st pauli', 'fc koln': 'koln', 'bayern munich': 'bayern munich', 'hoffenheim': 'hoffenheim', "m'gladbach": 'gladbach', 'monchengladbach': 'gladbach', 'rb leipzig': 'rb leipzig', 'leipzig': 'rb leipzig', 'freiburg': 'freiburg', 'augsburg': 'augsburg', 'union berlin': 'union berlin', 'hertha': 'hertha', 'schalke': 'schalke', 'bochum': 'bochum', # Serie A 'inter': 'inter', 'ac milan': 'milan', 'milan': 'milan', 'juventus': 'juventus', 'napoli': 'napoli', 'roma': 'roma', 'lazio': 'lazio', 'atalanta': 'atalanta', 'fiorentina': 'fiorentina', 'bologna': 'bologna', 'torino': 'torino', 'verona': 'verona', 'udinese': 'udinese', 'sassuolo': 'sassuolo', 'empoli': 'empoli', 'cagliari': 'cagliari', 'genoa': 'genoa', 'sampdoria': 'sampdoria', 'lecce': 'lecce', 'monza': 'monza', 'parma': 'parma', 'como': 'como', 'venezia': 'venezia', # Premier League 'man city': 'man city', 'manchester city': 'man city', 'man united': 'man united', 'manchester united': 'man united', 'liverpool': 'liverpool', 'chelsea': 'chelsea', 'arsenal': 'arsenal', 'tottenham': 'tottenham', 'west ham': 'west ham', 'newcastle': 'newcastle', 'aston villa': 'aston villa', 'brighton': 'brighton', 'wolves': 'wolves', 'wolverhampton': 'wolves', 'everton': 'everton', 'leicester': 'leicester', 'leeds': 'leeds', 'southampton': 'southampton', 'bournemouth': 'bournemouth', 'fulham': 'fulham', 'crystal palace': 'crystal palace', 'brentford': 'brentford', 'ipswich': 'ipswich', "nott'm forest": "nott'm forest", 'nottingham forest': "nott'm forest", # La Liga 'barcelona': 'barcelona', 'real madrid': 'real madrid', 'ath madrid': 'ath madrid', 'atletico madrid': 'ath madrid', 'sevilla': 'sevilla', 'valencia': 'valencia', 'villarreal': 'villarreal', 'real betis': 'betis', 'betis': 'betis', 'real sociedad': 'sociedad', 'ath bilbao': 'ath bilbao', 'athletic bilbao': 'ath bilbao', 'athletic club': 'ath bilbao', 'celta': 'celta', 'celta vigo': 'celta', 'getafe': 'getafe', 'alaves': 'alaves', 'osasuna': 'osasuna', 'mallorca': 'mallorca', 'granada': 'granada', 'cadiz': 'cadiz', 'girona': 'girona', 'almeria': 'almeria', 'las palmas': 'las palmas', 'rayo vallecano': 'vallecano', 'espanol': 'espanol', 'espanyol': 'espanol', 'leganes': 'leganes', 'valladolid': 'valladolid', # Ligue 1 'paris sg': 'paris sg', 'paris saint germain': 'paris sg', 'psg': 'paris sg', 'marseille': 'marseille', 'lyon': 'lyon', 'monaco': 'monaco', 'lille': 'lille', 'nice': 'nice', 'rennes': 'rennes', 'lens': 'lens', 'nantes': 'nantes', 'strasbourg': 'strasbourg', 'montpellier': 'montpellier', 'toulouse': 'toulouse', 'reims': 'reims', 'brest': 'brest', 'lorient': 'lorient', 'le havre': 'le havre', 'clermont': 'clermont', 'metz': 'metz', 'auxerre': 'auxerre', 'angers': 'angers', 'st etienne': 'st etienne', # Scottish 'celtic': 'celtic', 'rangers': 'rangers', 'hearts': 'hearts', 'aberdeen': 'aberdeen', 'hibernian': 'hibernian', 'dundee': 'dundee', 'motherwell': 'motherwell', 'st mirren': 'st mirren', 'st johnstone': 'st johnstone', 'kilmarnock': 'kilmarnock', 'ross county': 'ross county', 'dundee utd': 'dundee utd', } def normalize_team(name): """Normalizza nome squadra""" name = name.lower().strip() return TEAM_NAME_MAP.get(name, name) try: url = "https://www.football-data.co.uk/fixtures.csv" response = requests.get(url, timeout=30) response.raise_for_status() # Parse CSV from io import StringIO df = pd.read_csv(StringIO(response.text)) odds_map = {} for _, row in df.iterrows(): home = str(row.get('HomeTeam', '')).strip() away = str(row.get('AwayTeam', '')).strip() if not home or not away: continue # Normalizza nomi squadre home_norm = normalize_team(home) away_norm = normalize_team(away) # Get Bet365 odds (most popular), fallback to average home_odds = row.get('B365H') or row.get('AvgH') or 2.0 draw_odds = row.get('B365D') or row.get('AvgD') or 3.5 away_odds = row.get('B365A') or row.get('AvgA') or 3.0 try: odds_map[(home_norm, away_norm)] = { 'home': float(home_odds), 'draw': float(draw_odds), 'away': float(away_odds), 'league': row.get('Div', ''), 'date': row.get('Date', ''), 'time': row.get('Time', ''), 'original_home': home, 'original_away': away } except (ValueError, TypeError): pass print(f" ✓ {len(odds_map)} partite con quote reali") return odds_map except Exception as e: print(f" ✗ Errore scaricamento quote: {e}") return {} def generate_predictions_for_fixtures(): """ Genera predizioni per tutte le fixture usando l'API (V4 Ensemble) """ # Carica fixture fixtures_path = os.path.join(DATA_DIR, 'upcoming_fixtures.json') if not os.path.exists(fixtures_path): print("Nessuna fixture trovata") return [] with open(fixtures_path) as f: fixtures = json.load(f) # Scarica quote reali real_odds = fetch_real_odds() API_URL = 'http://localhost:3098/api/predict' RESULT_MAP = {0: 'AWAY', 1: 'DRAW', 2: 'HOME'} RESULT_LABELS = {0: 'Away Win', 1: 'Draw', 2: 'Home Win'} predictions = [] odds_found = 0 print(f"\nGenerando predizioni per {len(fixtures)} fixture usando API (V4 Ensemble)...") for fixture in fixtures: league_code = fixture['league_code'] home_team = fixture['home_team'] away_team = fixture['away_team'] try: # Chiama l'API per la predizione V4 response = requests.post(API_URL, json={ 'home_team': home_team, 'away_team': away_team, 'league': league_code }, timeout=10) if response.status_code == 200: pred_data = response.json() prediction = { **fixture, 'prediction': pred_data.get('prediction', 'HOME'), 'prediction_label': pred_data.get('prediction_label', 'Home Win'), 'prediction_code': pred_data.get('prediction_code', 2), 'confidence': pred_data.get('confidence', 40), 'home_prob': pred_data.get('home_prob', 33), 'draw_prob': pred_data.get('draw_prob', 33), 'away_prob': pred_data.get('away_prob', 33), 'home_form': pred_data.get('home_form', {}), 'away_form': pred_data.get('away_form', {}), 'generated_at': datetime.now().isoformat() } else: # Fallback se API non risponde prediction = { **fixture, 'prediction': 'HOME', 'prediction_label': 'Home Win', 'prediction_code': 2, 'confidence': 40, 'home_prob': 40, 'draw_prob': 30, 'away_prob': 30, 'home_form': {}, 'away_form': {}, 'generated_at': datetime.now().isoformat() } except Exception as e: print(f" Errore predizione {home_team} vs {away_team}: {e}") prediction = { **fixture, 'prediction': 'HOME', 'prediction_label': 'Home Win', 'prediction_code': 2, 'confidence': 40, 'home_prob': 40, 'draw_prob': 30, 'away_prob': 30, 'home_form': {}, 'away_form': {}, 'generated_at': datetime.now().isoformat() } # Aggiungi quote reali se disponibili # Normalizza i nomi per il matching home_norm = home_team.lower().strip() away_norm = away_team.lower().strip() odds_key = (home_norm, away_norm) if odds_key in real_odds: odds = real_odds[odds_key] prediction['odds_home'] = odds['home'] prediction['odds_draw'] = odds['draw'] prediction['odds_away'] = odds['away'] prediction['odds_source'] = 'Bet365' odds_found += 1 else: # Calcola quote implicite dalla probabilità del modello # EV sarà mostrato come N/A nel frontend per queste partite prediction['odds_home'] = round(100 / max(prediction['home_prob'], 1), 2) prediction['odds_draw'] = round(100 / max(prediction['draw_prob'], 1), 2) prediction['odds_away'] = round(100 / max(prediction['away_prob'], 1), 2) prediction['odds_source'] = 'Implied' predictions.append(prediction) # Ordina per data predictions.sort(key=lambda x: x['timestamp']) # Salva output_path = os.path.join(DATA_DIR, 'predictions_upcoming.json') with open(output_path, 'w') as f: json.dump(predictions, f, indent=2, ensure_ascii=False) print(f"Salvate {len(predictions)} predizioni in predictions_upcoming.json") print(f" ✓ {odds_found} con quote REALI (Bet365)") print(f" ✓ {len(predictions) - odds_found} con quote implicite") return predictions if __name__ == "__main__": # Raccogli fixture collector = FixtureCollector() fixtures = collector.collect_all_fixtures() # Genera predizioni predictions = generate_predictions_for_fixtures() print(f"\nPrime 5 predizioni:") for p in predictions[:5]: print(f" {p['date']} {p['time']} | {p['league_name']}") print(f" {p['home_team']} vs {p['away_team']}") print(f" Pred: {p['prediction']} ({p['confidence']:.0f}%)") print()