import os, json, logging, requests, re, time from datetime import datetime import pytz from flask import Flask, render_template, redirect, url_for from bs4 import BeautifulSoup from apscheduler.schedulers.background import BackgroundScheduler app = Flask(__name__) # Formatage des logs pour inclure l'heure et le niveau logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s') logger = logging.getLogger("CineSynthese") CACHE_FILE = "/data/movies_cache.json" HEADERS = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/121.0.0.0 Safari/537.36'} paris_tz = pytz.timezone("Europe/Paris") def get_today_formats(): now = datetime.now(paris_tz) jours_courts = ["Lun", "Mar", "Mer", "Jeu", "Ven", "Sam", "Dim"] day_short = jours_courts[now.weekday()] return { "cosmo_id": f"seance-{now.strftime('%Y-%m-%d')}", "cratere_date": f"{day_short}. {now.day:02d}/{now.month:02d}" } # --- 1. SCRAPER ABC --- def scrape_abc(): url = "https://abc-toulouse.fr/horaires.html" logger.info("📡 [ABC] Début du scraping...") movies = [] try: r = requests.get(url, headers=HEADERS, timeout=15) soup = BeautifulSoup(r.content, 'html.parser') movie_blocks = soup.find_all('div', class_='table_infos') for block in movie_blocks: title_tag = block.find('h2') if not title_tag: continue title = title_tag.get_text(strip=True) img_tag = block.find('img') img_url = "https://abc-toulouse.fr" + img_tag.get('src') if img_tag else "" next_div = block.find_next_sibling() showtimes_text = next_div.get_text(" ") if next_div else block.get_text(" ") times = re.findall(r'(\d{2}:\d{2})', showtimes_text) if times: found_times = sorted(list(set(times))) movies.append({ "title": title, "img": img_url, "times": found_times, "cinema": "ABC", "color": "#00bcd4" }) logger.info(f" ✅ [ABC] Trouvé: {title} ({', '.join(found_times)})") logger.info(f"📊 [ABC] Total: {len(movies)} films récupérés.") except Exception as e: logger.error(f"❌ [ABC] Erreur lors du scraping: {e}") return movies # --- 2. SCRAPER LE CRATÈRE --- def scrape_cratere(): url_main = "https://www.cinemalecratere.fr/films" target_date = get_today_formats()["cratere_date"] logger.info(f"📡 [Le Cratère] Début du scraping pour la date: {target_date}...") movies_dict = {} session = requests.Session() try: r = session.get(url_main, headers=HEADERS, timeout=15) soup = BeautifulSoup(r.content, 'html.parser') items = soup.find_all('div', attrs={'data-search': True}) for item in items: if target_date in item['data-search']: title_tag = item.find('span', class_='uk-text-bold') title = title_tag.get_text(strip=True) if title_tag else "Inconnu" time_match = re.search(r'(\d{2}h\d{2})', item['data-search']) show_time = time_match.group(0).replace('h', ':') if time_match else "Séance" if title not in movies_dict: img_url = "" link_tag = item.find('a', href=True) if link_tag: r_det = session.get("https://www.cinemalecratere.fr" + link_tag['href'], timeout=10) img_tag = BeautifulSoup(r_det.content, 'html.parser').find('img', class_='el-image') if img_tag: img_url = img_tag.get('src') if not img_url.startswith('http'): img_url = "https://www.cinemalecratere.fr" + img_url movies_dict[title] = {"title": title, "img": img_url, "times": [show_time], "cinema": "Le Cratère", "color": "#e62117"} logger.info(f" ✅ [Le Cratère] Nouveau film: {title}") else: if show_time not in movies_dict[title]["times"]: movies_dict[title]["times"].append(show_time) logger.info(f" ➕ [Le Cratère] Séance ajoutée pour {title}: {show_time}") logger.info(f"📊 [Le Cratère] Total: {len(movies_dict)} films récupérés.") except Exception as e: logger.error(f"❌ [Le Cratère] Erreur lors du scraping: {e}") return list(movies_dict.values()) # --- 3. SCRAPER COSMO --- def scrape_cosmo(): BASE_URL = "https://www.american-cosmograph.fr" url_horaires = f"{BASE_URL}/les-horaires.html" target_id = get_today_formats()["cosmo_id"] logger.info(f"📡 [Cosmograph] Début du scraping pour le bloc: {target_id}...") movies_today = [] session = requests.Session() try: r = session.get(url_horaires, headers=HEADERS, timeout=15) soup = BeautifulSoup(r.content, 'html.parser') day_container = soup.find('div', id=target_id) if not day_container: logger.warning(f" ⚠️ [Cosmograph] Bloc {target_id} non trouvé sur le site.") return [] items = day_container.find_all('li', class_='heureReady') for item in items: title_tag = item.select_one('.filmTitle a') if not title_tag: continue title = title_tag.get_text(strip=True) detail_path = title_tag.get('href') time_tag = item.select_one('.horaire_debut') time_raw = time_tag.get_text(strip=True) if time_tag else "" st_match = re.search(r'\d{2}h\d{2}', time_raw) show_time = st_match.group(0).replace('h', ':') if st_match else "Séance" img_url = "" if detail_path: try: r_detail = session.get(f"{BASE_URL}{detail_path}", headers=HEADERS, timeout=10) soup_detail = BeautifulSoup(r_detail.content, 'html.parser') img_tag = soup_detail.find('img', class_='fc_field_image') if img_tag: img_url = img_tag.get('src') if img_url and not img_url.startswith('http'): img_url = f"{BASE_URL}{img_url}" except: pass movies_today.append({"title": title, "img": img_url, "times": [show_time], "cinema": "American Cosmograph", "color": "#ffc107"}) logger.info(f" ✅ [Cosmograph] Séance trouvée: {title} à {show_time}") time.sleep(0.1) except Exception as e: logger.error(f"❌ [Cosmograph] Erreur lors du scraping: {e}") unique_movies = {} for m in movies_today: if m['title'] not in unique_movies: unique_movies[m['title']] = m else: if m['times'][0] not in unique_movies[m['title']]['times']: unique_movies[m['title']]['times'].append(m['times'][0]) unique_movies[m['title']]['times'].sort() logger.info(f"📊 [Cosmograph] Total: {len(unique_movies)} films uniques récupérés.") return list(unique_movies.values()) # --- LOGIQUE MISE À JOUR --- def update_all_data(): logger.info("🚀 --- LANCEMENT DE LA MISE À JOUR GLOBALE ---") start_time = time.time() try: abc = scrape_abc() cratere = scrape_cratere() cosmo = scrape_cosmo() data = abc + cratere + cosmo data = sorted(data, key=lambda x: x['times'][0] if x['times'] else "99:99") now_paris = datetime.now(paris_tz) update_str = now_paris.strftime("%d/%m/%Y à %H:%M") payload = {"last_update": update_str, "movies": data} os.makedirs(os.path.dirname(CACHE_FILE), exist_ok=True) with open(CACHE_FILE, 'w') as f: json.dump(payload, f) duration = round(time.time() - start_time, 2) logger.info(f"✨ --- MISE À JOUR TERMINÉE EN {duration}s ---") logger.info(f"📈 Total final: {len(data)} films à l'affiche aujourd'hui.") return payload except Exception as e: logger.error(f"💥 Erreur critique lors de la mise à jour: {e}") return None scheduler = BackgroundScheduler(timezone=paris_tz) scheduler.add_job(func=update_all_data, trigger="cron", hour=0, minute=5) scheduler.start() @app.route('/') def index(): if not os.path.exists(CACHE_FILE): logger.info("ℹ️ Cache inexistant, lancement du premier scraping...") cache = update_all_data() else: with open(CACHE_FILE, 'r') as f: cache = json.load(f) if isinstance(cache, list): logger.info("ℹ️ Cache ancien format détecté, conversion...") cache = update_all_data() return render_template('index.html', films=cache["movies"], updated=cache["last_update"]) @app.route('/refresh') def refresh(): logger.info("🔄 Rafraîchissement manuel demandé via /refresh") update_all_data() return redirect(url_for('index')) if __name__ == '__main__': app.run(host='0.0.0.0', port=5000)