"""VNEWS - FastAPI backend with livescore + xemlaibongda highlights + YouTube VTV shorts""" import re, time, subprocess, json, os, threading import html as html_lib from datetime import datetime, timezone, timedelta from collections import defaultdict VN_TZ = timezone(timedelta(hours=7)) from concurrent.futures import ThreadPoolExecutor, as_completed from fastapi import FastAPI, Query, Request from fastapi.responses import HTMLResponse, JSONResponse, StreamingResponse, Response from urllib.parse import quote import requests from bs4 import BeautifulSoup app = FastAPI() # ===== RATE LIMITING =====app = FastAPI() # ===== WORLD CUP 2026 SCRAPER ===== from wc2026_scraper import get_wc2026_all, scrape_fixtures, scrape_standings, scrape_stats, scrape_wc_news # ===== RATE LIMITING ===== _rate_limit_data = defaultdict(list) # {ip: [timestamp1, timestamp2, ...]} _rate_limit_lock = threading.Lock() RATE_LIMIT_MAX = 60 # Max requests per minute per IP RATE_LIMIT_WINDOW = 60 # seconds def _check_rate_limit(ip: str) -> bool: """Kiểm tra rate limit, return True nếu OK, False nếu bị limit""" with _rate_limit_lock: now = time.time() # Xóa các request cũ _rate_limit_data[ip] = [t for t in _rate_limit_data[ip] if now - t < RATE_LIMIT_WINDOW] if len(_rate_limit_data[ip]) >= RATE_LIMIT_MAX: return False _rate_limit_data[ip].append(now) return True @app.middleware("http") async def rate_limit_middleware(request: Request, call_next): """Middleware để kiểm tra rate limit""" # Chỉ rate limit API endpoints if request.url.path.startswith("/api/"): ip = request.client.host if not _check_rate_limit(ip): return JSONResponse({"error": "rate limit exceeded"}, status_code=429) response = await call_next(request) return response # ===== VTV CHANNELS API ===== from vtv_api import router as vtv_router app.include_router(vtv_router) HEADERS = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36","Accept-Language":"vi-VN,vi;q=0.9,en;q=0.8"} BONGDA_HEADERS = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36","Accept-Language":"vi-VN,vi;q=0.9","Referer":"https://bongda.com.vn/lich-thi-dau","X-Requested-With":"XMLHttpRequest"} BASE_BDP = "https://bongdaplus.vn" _cache = {} _cache_ttl = 300 _cache_ttl_live = 60 _cache_ttl_yt = 1800 # ===== VTV NAM BO SHORTS FALLBACK ===== SHORTS_FALLBACK = [ {"id":"nqlLH6chLRo","title":"Tin nóng VTV Nam Bộ | #shorts","channel":"vtvnambo"}, {"id":"E7Kq0v3hG6w","title":"VTV Nam Bộ - Tin tức miền Nam | #shorts","channel":"vtvnambo"}, {"id":"Lu_iCQ5YwNM","title":"Công an lập hồ sơ xử lý người phụ nữ chửi bới tát nam tài xế ô tô ở Hà Nội","channel":"baodantri7941"}, {"id":"CwWvijF8BOA","title":"Chú rể Ninh Bình bật khóc nhận món quà bí mật người cha quá cố gửi 26 năm trước","channel":"baodantri7941"}, {"id":"tvPewsc2ph4","title":"Tính năng ẩn trên iPhone giúp giảm mỏi mắt","channel":"baodantri7941"}, {"id":"b1Nxzv9ixlU","title":"Y án 3 năm tù với nữ tài xế uống 8 lon bia lái xe tông chủ tịch xã tử vong","channel":"baodantri7941"}, {"id":"Xp5eTwAZAis","title":"Người đánh hàng xóm tại chung cư ở Hà Nội bị tuyên hơn 4 tháng tù","channel":"baodantri7941"}, {"id":"Htzvwg6iOBM","title":"Xe điện Audi S6 Sportback e-tron có gì đặc biệt?","channel":"baodantri7941"}, {"id":"iMdFmWvYdlo","title":"Cô gái người Nga yêu thời trang và đất nước Việt Nam","channel":"baodantri7941"}, {"id":"IVaRc6moEv8","title":"Người nông dân Trung Quốc đột quỵ bệnh viện giúp bán sạch 4 tấn táo","channel":"baodantri7941"}, {"id":"uVxqPxToItU","title":"Công an vào cuộc vụ người phụ nữ chửi bới hành hung tài xế ô tô ở Hà Nội","channel":"baodantri7941"}, {"id":"VAfgNNgZDRs","title":"Khởi tố 4 đối tượng ném bom xăng vào nhà dân ở Đồng Nai","channel":"baodantri7941"}, {"id":"sBH_-zGh0Xw","title":"Vì sao Times New Roman vẫn nổi tiếng sau hàng chục năm?","channel":"baodantri7941"}, {"id":"woKn5f2bLHM","title":"Quảng Ninh ngập sâu diện rộng sau đợt mưa lớn","channel":"baodantri7941"}, {"id":"bcpgRoxbLPw","title":"Giông lốc quật bay mái tôn ở TP.HCM","channel":"baodantri7941"}, {"id":"ZIIC5osy544","title":"Bé trai Trung Quốc rơi từ tầng 11 vẫn sống sót kỳ diệu","channel":"baodantri7941"}, {"id":"uTMJ49NQpyc","title":"Sau lớp mascot 40kg Câu chuyện mưu sinh của người trẻ ở TPHCM","channel":"baodantri7941"}, {"id":"7Pd6vZ2Lz1M","title":"Hành động ấm lòng của người đàn ông tìm kiếm 5 học sinh tử vong ở sông Lô","channel":"baosuckhoedoisongboyte"}, {"id":"SlHLt_ZyPiE","title":"Xử phạt người đàn ông xóa số điện thoại cứu hộ trên cao tốc Bắc Nam","channel":"baosuckhoedoisongboyte"}, {"id":"IUOprcJyYr4","title":"Phụ nữ táo bón có phải do lười ăn rau?","channel":"baosuckhoedoisongboyte"}, {"id":"YY8ojFNE-AU","title":"Quái xế tự quay clip nẹt pô đánh võng đăng TikTok bị xử lý","channel":"baosuckhoedoisongboyte"}, {"id":"OV7_oGdQGII","title":"Bố cô dâu khóc sụt sùi rồi quẩy cực sung gây bão mạng","channel":"baosuckhoedoisongboyte"}, {"id":"FoxhFyz2skY","title":"Người đàn ông nước ngoài đập phá ô tô bẻ cần gạt nước ở Đà Nẵng","channel":"baosuckhoedoisongboyte"}, {"id":"R1oC_I8dFPU","title":"Thanh niên buông tay lái đứng trên xe máy khi đổ đèo ở Đắk Lắk","channel":"baosuckhoedoisongboyte"}, {"id":"U0Ft6ChWAIo","title":"Cô giáo kể phút tháo chạy khỏi xe khách trước khi bị lũ vò nát ở Cao Bằng","channel":"baosuckhoedoisongboyte"}, {"id":"hH0ANeze_4E","title":"Liên tiếp hàng chục con bò bị sét đánh chết trong ngày mưa dông","channel":"baosuckhoedoisongboyte"}, {"id":"pXWt0QbAzRQ","title":"Va chạm giao thông người phụ nữ lăng mạ tài xế ô tô","channel":"baosuckhoedoisongboyte"}, {"id":"UWWLPY1OYt4","title":"CSGT chặn xe khách khống chế đối tượng cướp dây chuyền tại Gia Lai","channel":"baosuckhoedoisongboyte"}, {"id":"AxhVTQutsuo","title":"Xuất tinh sớm và những hiểu lầm thường gặp","channel":"baosuckhoedoisongboyte"}, {"id":"cNy6FgaNxYM","title":"Cô dâu khóc sưng mắt vì 6 chỉ vàng không cánh mày bay trong ngày cưới","channel":"baosuckhoedoisongboyte"}, {"id":"IDt_S6q59Ro","title":"Chở bạn gái không đội mũ bảo hiểm thanh niên đấm CSGT","channel":"baosuckhoedoisongboyte"}, {"id":"LFxJ9Ik6W0A","title":"Mệnh lệnh từ trái tim CSGT Hà Nội mở đường đưa bé 5 tháng tuổi đi cấp cứu","channel":"baosuckhoedoisongboyte"}, ] for _v in SHORTS_FALLBACK: _v.setdefault("link", "https://www.youtube.com/watch?v="+_v["id"]) _v.setdefault("img", "https://i.ytimg.com/vi/"+_v["id"]+"/hqdefault.jpg") _v.setdefault("source", "yt") SHORT_STATS_FILE = "/data/short_stats.json" if os.path.isdir("/data") else "/app/short_stats.json" _short_lock = threading.Lock() def _load_short_db(): try: if os.path.exists(SHORT_STATS_FILE): with open(SHORT_STATS_FILE,"r",encoding="utf-8") as f: return json.load(f) except: pass return {} def _save_short_db(db): try: os.makedirs(os.path.dirname(SHORT_STATS_FILE), exist_ok=True) tmp = SHORT_STATS_FILE + ".tmp" with open(tmp,"w",encoding="utf-8") as f: json.dump(db, f, ensure_ascii=False) os.replace(tmp, SHORT_STATS_FILE) except: pass def _short_default(): return {"views":0,"likes":0,"shares":0,"comments":[]} PRIORITY_LEAGUES = ["Ngoại Hạng Anh","FA Cup","Champions League","LaLiga","Copa del Rey","Serie A","Bundesliga","Ligue 1","V-League"] LEAGUE_IDS = {"nha":27110,"laliga":27233,"seriea":27044,"bundesliga":26891,"ligue1":27212} HL_LEAGUES = { "premier-league":{"path":"anh/premier-league","name":"Premier League","emoji":"🏴󠁧󠁢󠁥󠁮󠁧󠁿"}, "fa-cup":{"path":"anh/fa-cup","name":"FA Cup","emoji":"🏆"}, "bundesliga":{"path":"duc/bundesliga","name":"Bundesliga","emoji":"🇩🇪"}, "serie-a":{"path":"italy/serie-a","name":"Serie A","emoji":"🇮🇹"}, "la-liga":{"path":"tay-ban-nha/la-liga","name":"La Liga","emoji":"🇪🇸"}, "champions-league":{"path":"cup-chau-au/uefa-champions-league","name":"Champions League","emoji":"⭐"}, "europa-league":{"path":"cup-chau-au/uefa-europa-league","name":"Europa League","emoji":"🟠"}, "world-cup":{"path":"the-gioi/world-cup","name":"World Cup 2026","emoji":"🌍"}, } def _cached(key, fn, ttl=None): now=time.time(); t=ttl or _cache_ttl if key in _cache and now-_cache[key]["t"]=3: score=f"{spans[0].get_text(strip=True)} - {spans[2].get_text(strip=True)}" if len(spans)>=4: minute=spans[3].get_text(strip=True) if not score and status_el and status_el.select_one(".vs"): score="VS" league=league_el.get_text(strip=True) if league_el else "" return {"home":home_el.get_text(strip=True),"away":away_el.get_text(strip=True),"score":score or"VS","minute":minute,"league":league,"time":time_el.get_text(strip=True) if time_el else "","event_id":event_id,"home_logo":home_logo.get("src","") if home_logo else "","away_logo":away_logo.get("src","") if away_logo else "","status":status_type} # ===== VIDEO PROXY ===== @app.get("/api/proxy/m3u8") def proxy_m3u8(url: str = Query(...)): try: r = requests.get(url, headers=HEADERS, timeout=15) if r.status_code != 200: return Response(status_code=502, content="upstream error") lines = r.text.strip().split('\n'); rewritten = [] for line in lines: if line.startswith('#') or not line.strip(): rewritten.append(line) else: rewritten.append("/api/proxy/seg?url=" + quote(line.strip(), safe="")) return Response(content='\n'.join(rewritten).encode('utf-8'), media_type="application/vnd.apple.mpegurl", headers={"Access-Control-Allow-Origin":"*","Cache-Control":"public, max-age=300"}) except: return Response(status_code=502, content="proxy error") @app.get("/api/proxy/seg") def proxy_segment(url: str = Query(...)): try: r = requests.get(url, headers=HEADERS, timeout=30) if r.status_code != 200: return Response(status_code=502, content="upstream error") data = r.content if len(data) > 188 and data[0:4] == b'\x89PNG' and data[188] == 0x47: data = data[188:] return Response(content=data, media_type="video/mp2t", headers={"Access-Control-Allow-Origin":"*","Cache-Control":"public, max-age=3600"}) except: return Response(status_code=502, content="proxy error") @app.get("/api/proxy/video") def proxy_video(url: str = Query(...), request: Request = None): try: req_headers = dict(HEADERS) if request and request.headers.get("range"): req_headers["Range"] = request.headers["range"] r = requests.get(url, headers=req_headers, timeout=30, stream=True) resp_headers = {"Access-Control-Allow-Origin":"*","Accept-Ranges":"bytes","Content-Type":r.headers.get("Content-Type","video/mp4")} if "Content-Range" in r.headers: resp_headers["Content-Range"] = r.headers["Content-Range"] if "Content-Length" in r.headers: resp_headers["Content-Length"] = r.headers["Content-Length"] return StreamingResponse(r.iter_content(chunk_size=256*1024), status_code=r.status_code, headers=resp_headers) except: return Response(status_code=502, content="proxy error") @app.get("/api/proxy/img") def proxy_img(url: str = Query(...)): try: r = requests.get(url, headers={**HEADERS, "Referer": "https://dantri.com.vn/"}, timeout=10) if r.status_code != 200: return Response(status_code=502) ct = r.headers.get("Content-Type", "image/jpeg") return Response(content=r.content, media_type=ct, headers={"Cache-Control": "public, max-age=86400", "Access-Control-Allow-Origin": "*"}) except: return Response(status_code=502) # ===== XEMLAIBONGDA HIGHLIGHTS ===== def _scrape_xemlaibongda_page(page_path, limit=20): """ Scrape video từ xemlaibongda.top - Simple & Reliable Dùng logic cũ đã test, không fetch từng trang (tránh timeout) """ try: url = f"https://xemlaibongda.top/{page_path}" if page_path else "https://xemlaibongda.top/" r = requests.get(url, headers=HEADERS, timeout=15) if r.status_code != 200: return [] r.encoding = "utf-8" soup = BeautifulSoup(r.text, "lxml") videos = [] seen = set() for a in soup.find_all("a", href=True): href = a.get("href", "") if "/video/" not in href and "/xem-lai/" not in href: continue if not href.startswith("http"): href = "https://xemlaibongda.top" + href # Bỏ query params clean_href = href.split("?")[0].split("#")[0] if clean_href in seen: continue seen.add(clean_href) # ===== Lấy THUMBNAIL ===== img_src = "" img = a.find("img") if not img and a.parent: img = a.parent.find("img") if not img: p = a.parent for _ in range(4): if p and p.find("img"): img = p.find("img") break p = p.parent if p else None if img: img_src = (img.get("data-src", "") or img.get("src", "") or img.get("data-lazy", "") or img.get("data-original", "")) if img_src.startswith("//"): img_src = "https:" + img_src elif img_src.startswith("/"): img_src = "https://xemlaibongda.top" + img_src # ===== Lấy TITLE ===== title = "" # Thử attribute for attr in ["title", "aria-label"]: val = a.get(attr, "") if val and len(val) >= 5: title = val break # Thử các selector if not title: for selector in ["h3", "h2", "h4", ".title", ".video-title", "strong"]: try: el = a.select_one(selector) if el: t = el.get_text(strip=True) if len(t) >= 5: title = t break except: pass # Thử text content if not title: text = a.get_text(strip=True) if text and len(text) >= 5: title = text[:100] # Fallback: tạo title từ slug if not title or len(title) < 3: slug = clean_href.split("/video/")[-1].rstrip("/").split("/xem-lai/")[-1].rstrip("/") title = slug.replace("-", " ").replace("_", " ").title() title = re.sub(r'\d{4}-\d{2}-\d{2}', '', title).strip() if not title or len(title) < 3: continue # Fallback thumbnail từ slug if not img_src: slug = clean_href.split("/video/")[-1].rstrip("/").split("/xem-lai/")[-1].rstrip("/") img_src = f"https://xemlaibongda.top/uploads/thumb/{slug}.jpg" videos.append({ "title": title[:100], "link": clean_href, "img": img_src, "source": "xemlaibongda" }) if len(videos) >= limit: break return videos except Exception as e: print(f"[xemlaibongda] Error: {e}") return [] def _extract_img_src(a_tag): """Extract image URL từ thẻ và parent elements""" img = a_tag.find("img") if not img and a_tag.parent: img = a_tag.parent.find("img") if not img: p = a_tag.parent for _ in range(5): # Tìm sâu hơn if p and p.find("img"): img = p.find("img") break p = p.parent if p else None if not img: return "" # Thử tất cả các attribute có thể chứa img URL attrs = ["data-src", "src", "data-lazy", "data-original", "data-srcset", "data-thumb", "data-image"] for attr in attrs: val = img.get(attr, "") if val: if attr == "data-srcset": val = val.split(",")[0].strip().split(" ")[0] break else: val = "" # Thử background-image từ style if not val: style = img.get("style", "") or img.get("data-bg", "") bg_match = re.search(r'url\(["\']?(.*?)["\']?\)', style) if bg_match: val = bg_match.group(1) # Normalize URL if val.startswith("//"): val = "https:" + val elif val.startswith("/"): val = "https://xemlaibongda.top" + val return val def _extract_title(a_tag, href): """Extract title từ thẻ và child/parent elements""" title = "" # 1. Thử các selector phổ biến cho title title_selectors = [ "h3", "h2", "h4", "h5", ".title", ".post-title", ".entry-title", ".video-title", ".card-title", ".item-title", ".news-title", "span.title", "strong", "b", ".name", ".caption" ] for tag in title_selectors: try: t = a_tag.select_one(tag) if hasattr(a_tag, 'select_one') else None if t: title = t.get_text(" ", strip=True) if len(title) >= 3: return title except: pass # 2. Thử attribute của for attr in ["title", "aria-label", "data-title"]: val = a_tag.get(attr, "") if val and len(val) >= 3: return val # 3. Thử alt text của img img = a_tag.find("img") if img: alt = img.get("alt", "") if alt and len(alt) >= 3: return alt # 4. Thử text content của (loại bỏ quá dài) text = a_tag.get_text(" ", strip=True) if text and len(text) >= 3: # Lấy dòng đầu tiên nếu có nhiều dòng lines = [l.strip() for l in text.split("\n") if l.strip()] if lines: first_line = lines[0] if len(first_line) >= 3: return first_line[:100] # 5. Fallback: tạo title từ slug slug = href.split("/video/")[-1].rstrip("/").split("/xem-lai/")[-1].rstrip("/") title = slug.replace("-", " ").replace("_", " ") title = re.sub(r'\d{4}-\d{2}-\d{2}', '', title).strip() if title: return title.title() return "" def scrape_xemlaibongda(): return _scrape_xemlaibongda_page("", 20) def scrape_highlights_by_league(league_key): if league_key not in HL_LEAGUES: return [] return _scrape_xemlaibongda_page(HL_LEAGUES[league_key]["path"], 20) def scrape_all_league_highlights(): results = {} def _fetch(key): return key, scrape_highlights_by_league(key) with ThreadPoolExecutor(8) as ex: futs = [ex.submit(_fetch, k) for k in HL_LEAGUES] for f in as_completed(futs, timeout=25): try: key, vids = f.result() if vids: results[key] = vids except: pass return results def extract_xemlaibongda_video(url): try: r=requests.get(url, headers=HEADERS, timeout=15) if r.status_code!=200: return None r.encoding="utf-8"; soup=BeautifulSoup(r.text,"lxml"); video=soup.find("video") if video: src=video.get("src",""); poster=video.get("poster","") if not src: source=video.find("source") if source: src=source.get("src","") if src: return{"src":src,"poster":poster,"type":"hls" if".m3u8" in src else"video"} m3u8s=re.findall(r'(https?://[^\s"\'<>]+\.m3u8)',r.text) if m3u8s: og=soup.find("meta",property="og:image"); poster=og.get("content","") if og else "" return{"src":m3u8s[0],"poster":poster,"type":"hls"} return None except: return None # ===== YOUTUBE SHORTS SCRAPING ===== def _yt_channel_shorts_requests(channel, count=15): try: url=f"https://www.youtube.com/@{channel}/shorts" r=requests.get(url, headers={**HEADERS,"Accept-Language":"vi,en;q=0.8"}, timeout=15) if r.status_code!=200: return [] html=r.text; ids=[]; items=[] for m in re.finditer(r'"videoId":"([A-Za-z0-9_-]{11})"',html): vid=m.group(1) if vid in ids: continue ids.append(vid) snip=html[max(0,m.start()-900):m.start()+1600] title="" mt=re.search(r'"title":\{"runs":\[\{"text":"([^"]+)"',snip) if not mt: mt=re.search(r'"accessibilityText":"([^"]+)"',snip) if mt: title=html_lib.unescape(mt.group(1)).replace('\n',' ').strip() if not title: title="YouTube Short" items.append({"title":title,"link":f"https://www.youtube.com/watch?v={vid}","img":f"https://i.ytimg.com/vi/{vid}/hqdefault.jpg","source":"yt","id":vid,"channel":channel}) if len(items)>=count: break return items except: return [] def scrape_shorts(): vids=[] with ThreadPoolExecutor(3) as ex: futs=[ex.submit(_yt_channel_shorts_requests,ch,24) for ch in ["baodantri7941","baosuckhoedoisongboyte","vtvnambo"]] for f in as_completed(futs): try: r=f.result() if r: vids.extend(r) except: pass merged=[]; seen=set() for v in vids: vid=v.get("id") if not vid or vid in seen: continue seen.add(vid); merged.append(v) for v in SHORTS_FALLBACK: vid=v.get("id") if not vid or vid in seen: continue seen.add(vid); merged.append(v) return merged[:60] # ===== VTV NAM BO & WC SHORTS - using yt-dlp ===== from yt_scraper import get_vtvnambo_shorts, get_wc_related_shorts @app.get("/api/shorts/vtvnamo") def api_shorts_vtvnamo(count: int = Query(default=50, le=100)): items = get_vtvnambo_shorts(count) if not items: items = [v for v in SHORTS_FALLBACK if v.get("channel") == "vtvnambo"] nql = [v for v in items if v.get("id") == "nqlLH6chLRo"] rest = [v for v in items if v.get("id") != "nqlLH6chLRo"] items = nql + rest return JSONResponse(items) @app.get("/api/shorts/wc") def api_shorts_wc(count: int = Query(default=50, le=100)): items = get_wc_related_shorts(count) if not items: items = [v for v in SHORTS_FALLBACK if v.get("channel") == "vtvnambo"] nql = [v for v in items if v.get("id") == "nqlLH6chLRo"] rest = [v for v in items if v.get("id") != "nqlLH6chLRo"] items = nql + rest return JSONResponse(items) # ===== LIVESCORE ===== @app.get("/api/livescore/live") def api_livescore_live(): return JSONResponse({"html":_cached("ls_live",lambda:fetch_bongda_api("/api/fixtures/live"),ttl=_cache_ttl_live)}) @app.get("/api/livescore/incoming") def api_livescore_incoming(): return JSONResponse({"html":_cached("ls_incoming",lambda:fetch_bongda_api("/api/fixtures/incoming"),ttl=_cache_ttl_live)}) @app.get("/api/livescore/today") def api_livescore_today(): today=datetime.now(VN_TZ).strftime("%Y-%m-%d");return JSONResponse({"html":_cached("ls_today",lambda:fetch_bongda_api(f"/api/fixtures/get-by-date?date={today}"),ttl=_cache_ttl)}) @app.get("/api/livescore/results") def api_livescore_results(): today=datetime.now(VN_TZ).strftime("%Y-%m-%d");return JSONResponse({"html":_cached("ls_results",lambda:fetch_bongda_api(f"/api/fixtures/get-by-date?date={today}&status=finished"),ttl=_cache_ttl)}) @app.get("/api/livescore/standings/{league}") def api_livescore_standings(league:str): tid=LEAGUE_IDS.get(league,27110);return JSONResponse({"html":_cached(f"ls_bxh_{league}",lambda:fetch_bongda_api(f"/api/league-table/home?tournament_id={tid}&is_detail=True"),ttl=_cache_ttl)}) @app.get("/api/livescore/date/{date}") def api_livescore_date(date:str):return JSONResponse({"html":fetch_bongda_api(f"/api/fixtures/get-by-date?date={date}")}) @app.get("/api/match/{event_id}/commentaries") def api_match_commentaries(event_id:int):return JSONResponse({"html":fetch_bongda_api(f"/api/fixtures/commentaries?event_id={event_id}")}) @app.get("/api/match/{event_id}/stats") def api_match_stats(event_id:int):return JSONResponse({"html":fetch_bongda_api(f"/api/event-standing/player-performance?event_id={event_id}")}) from match_detail_v2 import fetch_match_detail, fetch_match_detail_by_url @app.get("/api/match/{event_id}/detail") def api_match_detail(event_id: int, url: str = Query(default="")): try: if url: data = fetch_match_detail_by_url(url) else: data = fetch_match_detail(event_id) return JSONResponse(data) except Exception as e: return JSONResponse({"event_id": event_id, "found": False, "error": str(e)}) @app.get("/api/livescore/featured") def api_livescore_featured(): def _f(): sources=[("/api/fixtures/live","live"),("/api/fixtures/get-by-date?date="+datetime.now(VN_TZ).strftime("%Y-%m-%d"),"today"),("/api/fixtures/incoming","upcoming")] for endpoint, stype in sources: html=fetch_bongda_api(endpoint) if not html or len(html)<100:continue soup=BeautifulSoup(html,"lxml");all_matches=[] for li in soup.select("li.match-detail"): match=_parse_match_from_li(li, stype) if not match or not match["event_id"]:continue if stype=="today" and "KT" in match.get("minute",""):continue all_matches.append(match) if not all_matches:continue for pl in PRIORITY_LEAGUES: for match in all_matches: if pl in match["league"]:return match return all_matches[0] return None return JSONResponse(_cached("ls_featured",_f,ttl=30)) @app.get("/api/shorts") def api_shorts(channel: str = Query(default="")): if channel == "vtvnambo": return api_shorts_vtvnamo() if channel == "wc": return api_shorts_wc() return JSONResponse(_cached("yt_shorts_v3",scrape_shorts,ttl=_cache_ttl_yt)) @app.get("/api/short-stats") def api_short_stats(ids:str=Query(default="")): arr=[x for x in ids.split(",") if x] with _short_lock: db=_load_short_db();out={} for vid in arr: st=db.get(vid) or _short_default() out[vid]={"views":int(st.get("views",0)),"likes":int(st.get("likes",0)),"shares":int(st.get("shares",0)),"comments":st.get("comments",[])[:80]} return JSONResponse({"stats":out}) @app.post("/api/short-action") async def api_short_action(request:Request): try: body=await request.json() except: body={} vid=str(body.get("id","")).strip(); action=str(body.get("action","")).strip(); txt=str(body.get("text","")).strip() if not vid: return JSONResponse({"error":"missing id"},status_code=400) with _short_lock: db=_load_short_db(); st=db.get(vid) or _short_default() if action=="view": st["views"]=int(st.get("views",0))+1 elif action=="like": st["likes"]=int(st.get("likes",0))+1 elif action=="share": st["shares"]=int(st.get("shares",0))+1 elif action=="comment" and txt: comments=st.get("comments",[]) comments.insert(0,{"text":txt[:180],"ts":int(time.time())}) st["comments"]=comments[:80] st["updated"]=int(time.time()); db[vid]=st; _save_short_db(db) out={"views":int(st.get("views",0)),"likes":int(st.get("likes",0)),"shares":int(st.get("shares",0)),"comments":st.get("comments",[])[:80]} return JSONResponse({"stats":out}) @app.get("/api/highlights") def api_highlights(): return JSONResponse(_cached("xemlaibongda_hl",scrape_xemlaibongda,ttl=_cache_ttl)) @app.get("/api/highlights/leagues") def api_highlights_leagues(): return JSONResponse(_cached("hl_leagues",scrape_all_league_highlights,ttl=_cache_ttl)) @app.get("/api/highlights/{league}") def api_highlights_league(league:str): if league not in HL_LEAGUES: return JSONResponse({"error":"league not found"}) return JSONResponse(_cached(f"hl_{league}",lambda:scrape_highlights_by_league(league),ttl=_cache_ttl)) @app.get("/api/video_url") def api_video_url(url:str=Query(...)): if "youtube.com" in url or "youtu.be" in url: m=re.search(r'(?:v=|shorts/|youtu\.be/)([a-zA-Z0-9_-]{11})',url) if m: vid=m.group(1); return JSONResponse({"src":f"https://www.youtube.com/embed/{vid}?autoplay=1&rel=0&enablejsapi=1","poster":f"https://i.ytimg.com/vi/{vid}/hqdefault.jpg","type":"youtube"}) if "xemlaibongda.top" in url: v=extract_xemlaibongda_video(url) if v: if v["type"]=="hls": v["src"]="/api/proxy/m3u8?url="+quote(v["src"],safe="") return JSONResponse(v) return JSONResponse({"error":"not found"}) # ===== WORLD CUP 2026 API ===== # Rate limiting cho WC API _wc_request_times = [] _wc_rate_limit_lock = threading.Lock() _WC_RATE_LIMIT = 10 # Max 10 requests per minute def _wc_rate_limit(): """Kiểm tra rate limit cho WC API""" global _wc_request_times with _wc_rate_limit_lock: now = time.time() # Xóa các request cũ hơn 60 giây _wc_request_times = [t for t in _wc_request_times if now - t < 60] if len(_wc_request_times) >= _WC_RATE_LIMIT: return False _wc_request_times.append(now) return True @app.get("/api/wc2026") def api_wc2026(): """Trả về tất cả dữ liệu World Cup 2026""" return JSONResponse(_cached("wc2026", get_wc2026_all, ttl=_cache_ttl)) @app.get("/api/wc2026/{tab}") def api_wc2026_tab(tab: str): """Trả về từng tab của World Cup""" valid_tabs = ["news", "fixtures", "standings", "stats", "highlights"] if tab not in valid_tabs: return JSONResponse({"error": "invalid tab"}, status_code=400) def _fetch_tab(): if tab == "highlights": return scrape_highlights_by_league("world-cup") elif tab == "news": return scrape_wc_news() elif tab == "fixtures": return scrape_fixtures() elif tab == "standings": return scrape_standings() elif tab == "stats": return scrape_stats() return [] return JSONResponse(_cached(f"wc2026_{tab}", _fetch_tab, ttl=_cache_ttl)) # Note: WC functions (scrape_wc_news, scrape_fixtures, scrape_stats, scrape_standings) # are imported from wc2026_scraper.py at the top of this file @app.get("/api/video_url") def api_video_url(url:str=Query(...)): if "youtube.com" in url or "youtu.be" in url: m=re.search(r'(?:v=|shorts/|youtu\.be/)([a-zA-Z0-9_-]{11})',url) if m: vid=m.group(1); return JSONResponse({"src":f"https://www.youtube.com/embed/{vid}?autoplay=1&rel=0&enablejsapi=1","poster":f"https://i.ytimg.com/vi/{vid}/hqdefault.jpg","type":"youtube"}) if "xemlaibongda.top" in url: v=extract_xemlaibongda_video(url) if v: if v["type"]=="hls": v["src"]="/api/proxy/m3u8?url="+quote(v["src"],safe="") return JSONResponse(v) return JSONResponse({"error":"not found"}) @app.get("/api/bdp_videos") def api_bdp_videos(): def _f(): try: soup=_get(f"{BASE_BDP}/video"); arts=[]; seen=set() for a in soup.find_all("a",href=True): href=a.get("href","") if"/video/" not in href or href in("/video/","/video/ban-thang-dep","/video/highlight"):continue if not href.startswith("http"): href=BASE_BDP+href if href in seen: continue title=re.sub(r'^\d{2}:\d{2}','',a.get_text(strip=True)).strip() if not title or len(title)<5: continue img_tag=a.find("img") or(a.parent.find("img") if a.parent else None) img=(img_tag.get("data-src") or img_tag.get("src","")) if img_tag else "" seen.add(href); arts.append({"title":title,"link":href,"img":img,"source":"bdp"}) return arts[:20] except: return [] return JSONResponse(_cached("bdp_videos",_f)) # ===== NEWS ===== VNE_CATS={"thoi-su":("https://vnexpress.net/thoi-su","Thời Sự"),"the-gioi":("https://vnexpress.net/the-gioi","Thế Giới"),"kinh-doanh":("https://vnexpress.net/kinh-doanh","Kinh Doanh"),"the-thao":("https://vnexpress.net/the-thao","Thể Thao"),"giai-tri":("https://vnexpress.net/giai-tri","Giải Trí"),"suc-khoe":("https://vnexpress.net/suc-khoe","Sức Khỏe"),"phap-luat":("https://vnexpress.net/phap-luat","Pháp Luật"),"giao-duc":("https://vnexpress.net/giao-duc","Giáo Dục"),"du-lich":("https://vnexpress.net/du-lich","Du Lịch"),"doi-song":("https://vnexpress.net/doi-song","Đời Sống")} def scrape_vne(cat_url): try: soup=_get(cat_url); arts=[] for it in soup.select("article.item-news")[:15]: a=it.select_one("h2.title-news a") or it.select_one("h3.title-news a") if not a: continue t=a.get("title","") or a.get_text(strip=True); lk=a.get("href","") if not t or not lk: continue im=it.find("img"); img=(im.get("data-src") or im.get("src","")) if im else "" if img and 'blank' in img: src=it.find("source") if src: img=src.get("srcset","").split(",")[0].strip().split(" ")[0] arts.append({"title":t,"link":lk,"img":img,"source":"vne"}) return arts except: return [] def scrape_genk_ai(): try: r=requests.get("https://genk.vn/ai.chn",headers=HEADERS,timeout=15) if r.status_code!=200: return [] r.encoding="utf-8"; soup=BeautifulSoup(r.text,"lxml"); articles=[]; seen=set() for a in soup.find_all("a",href=True): href=a.get("href","") if not href.endswith(".chn") or href=="/ai.chn": continue if href.startswith("/"): href="https://genk.vn"+href if href in seen or "genk.vn" not in href: continue title=a.get("title","") or a.get_text(strip=True) if not title or len(title)<20: continue container=a.parent; img_src="" for _ in range(6): if container is None: break for img in container.find_all("img"): s=img.get("data-src","") or img.get("src","") if s and "mediacdn" in s and "avatar" not in s and "logo" not in s: img_src=s; break if img_src: break container=container.parent seen.add(href) if not img_src: try: og_r=requests.get(href,headers=HEADERS,timeout=8); og_r.encoding="utf-8" og_soup=BeautifulSoup(og_r.text,"lxml"); og_tag=og_soup.find("meta",property="og:image") if og_tag: img_src=og_tag.get("content","") except: pass articles.append({"title":title,"link":href,"img":img_src,"source":"genk"}) if len(articles)>=30: break return articles except: return [] @app.get("/api/homepage") def api_homepage(): def _f(): articles=[] with ThreadPoolExecutor(12) as ex: futs={ex.submit(scrape_vne,VNE_CATS[k][0]):VNE_CATS[k][1] for k in["thoi-su","the-gioi","kinh-doanh","the-thao","giai-tri","phap-luat","giao-duc","du-lich","doi-song"]} for f in as_completed(futs): try: for a in f.result(): a["group"]=futs[f]; articles.append(a) except: pass return articles return JSONResponse(_cached("homepage",_f)) @app.get("/api/category/{cat_id}") def api_category(cat_id:str): def _f(): if cat_id=="cong-nghe": return scrape_genk_ai() if cat_id in VNE_CATS: arts=scrape_vne(VNE_CATS[cat_id][0]) [a.update({"group":VNE_CATS[cat_id][1]}) for a in arts] return arts return [] return JSONResponse(_cached(f"cat_{cat_id}",_f)) @app.get("/api/categories") def api_categories(): cats=[{"id":"cong-nghe","name":"Công Nghệ","source":"genk"}] for k,(u,n) in VNE_CATS.items(): cats.append({"id":k,"name":n,"source":"vne"}) return JSONResponse(cats) @app.get("/api/article") def api_article(url:str=Query(...)): return JSONResponse({"error":"not supported"}) @app.get("/api/storage_status") def api_storage_status(): return JSONResponse({"persistent":os.path.isdir("/data")}) @app.get("/api/hot_topics") def api_hot_topics(): return JSONResponse({"topics":[]}) @app.get("/", response_class=HTMLResponse) async def root(): return HTMLResponse("

VNEWS

Running

") # v14 rebuild 2026-06-18T12:21:41.503230