Add match detail API (commentaries + stats) + featured match endpoint

main.py

@@ -17,7 +17,7 @@ BASE_24H = "https://www.24h.com.vn"
 SPACE_URL = "https://bep40-vnews.hf.space"
 _cache = {}
 _cache_ttl = 300
-_cache_ttl_live = 60  # 1 min for live data
+_cache_ttl_live = 60
 def _cached(key, fn, ttl=None):
     now=time.time();t=ttl or _cache_ttl
     if key in _cache and now-_cache[key]["t"]<t:return _cache[key]["d"]
@@ -32,7 +32,6 @@ def _get(url,headers=None):
 LEAGUE_IDS = {"nha":27110,"laliga":27233,"seriea":27044,"bundesliga":26891,"ligue1":27212}
 
 def fetch_bongda_api(endpoint):
-    """Fetch bongda.com.vn API and return HTML content."""
     try:
         r=requests.get(f"https://bongda.com.vn{endpoint}",headers=BONGDA_HEADERS,timeout=10)
         if r.status_code==200:
@@ -43,43 +42,107 @@ def fetch_bongda_api(endpoint):
 
 @app.get("/api/livescore/live")
 def api_livescore_live():
-    """Trận đang diễn ra."""
     html=_cached("ls_live",lambda:fetch_bongda_api("/api/fixtures/live"),ttl=_cache_ttl_live)
     return JSONResponse({"html":html})
 
 @app.get("/api/livescore/incoming")
 def api_livescore_incoming():
-    """Trận sắp diễn ra."""
     html=_cached("ls_incoming",lambda:fetch_bongda_api("/api/fixtures/incoming"),ttl=_cache_ttl_live)
     return JSONResponse({"html":html})
 
 @app.get("/api/livescore/today")
 def api_livescore_today():
-    """Tất cả trận hôm nay."""
     today=datetime.now().strftime("%Y-%m-%d")
     html=_cached("ls_today",lambda:fetch_bongda_api(f"/api/fixtures/get-by-date?date={today}"),ttl=_cache_ttl)
     return JSONResponse({"html":html})
 
 @app.get("/api/livescore/results")
 def api_livescore_results():
-    """Kết quả hôm nay."""
     today=datetime.now().strftime("%Y-%m-%d")
     html=_cached("ls_results",lambda:fetch_bongda_api(f"/api/fixtures/get-by-date?date={today}&status=finished"),ttl=_cache_ttl)
     return JSONResponse({"html":html})
 
 @app.get("/api/livescore/standings/{league}")
 def api_livescore_standings(league:str):
-    """Bảng xếp hạng theo giải."""
     tid=LEAGUE_IDS.get(league,27110)
     html=_cached(f"ls_bxh_{league}",lambda:fetch_bongda_api(f"/api/league-table/home?tournament_id={tid}&is_detail=True"),ttl=_cache_ttl)
     return JSONResponse({"html":html})
 
 @app.get("/api/livescore/date/{date}")
 def api_livescore_date(date:str):
-    """Lịch thi đấu theo ngày (YYYY-MM-DD)."""
     html=fetch_bongda_api(f"/api/fixtures/get-by-date?date={date}")
     return JSONResponse({"html":html})
 
+@app.get("/api/match/{event_id}/commentaries")
+def api_match_commentaries(event_id:int):
+    """Diễn biến trận đấu."""
+    html=fetch_bongda_api(f"/api/fixtures/commentaries?event_id={event_id}")
+    return JSONResponse({"html":html})
+
+@app.get("/api/match/{event_id}/stats")
+def api_match_stats(event_id:int):
+    """Thống kê cầu thủ trận đấu."""
+    html=fetch_bongda_api(f"/api/event-standing/player-performance?event_id={event_id}")
+    return JSONResponse({"html":html})
+
+@app.get("/api/livescore/featured")
+def api_livescore_featured():
+    """Trận tâm điểm: lấy trận live đầu tiên từ giải lớn, hoặc trận sắp tới."""
+    def _f():
+        # Try live first
+        html=fetch_bongda_api("/api/fixtures/live")
+        if html and len(html)>100:
+            soup=BeautifulSoup(html,"lxml")
+            # Find first match from top leagues
+            for li in soup.select("li.match-detail"):
+                match_div=li.select_one("div.match")
+                if not match_div:continue
+                home_el=match_div.select_one(".home-team .name")
+                away_el=match_div.select_one(".away-team .name")
+                status_el=match_div.select_one(".status a")
+                league_el=li.find_previous("strong")
+                time_el=match_div.select_one(".match-time")
+                home_logo=match_div.select_one(".home-team .logo img")
+                away_logo=match_div.select_one(".away-team .logo img")
+                if home_el and away_el:
+                    # Get event_id from link
+                    event_id=""
+                    if status_el:
+                        href=status_el.get("href","")
+                        m=re.search(r'/tran-dau/(\d+)/',href)
+                        if m:event_id=m.group(1)
+                    spans=status_el.find_all("span") if status_el else []
+                    score=""
+                    minute=""
+                    if len(spans)>=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)
+                    return{"home":home_el.get_text(strip=True),"away":away_el.get_text(strip=True),"score":score,"minute":minute,"league":league_el.get_text(strip=True) if league_el else "","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":"live"}
+        # Fallback: incoming
+        html2=fetch_bongda_api("/api/fixtures/incoming")
+        if html2 and len(html2)>100:
+            soup2=BeautifulSoup(html2,"lxml")
+            for li in soup2.select("li.match-detail"):
+                match_div=li.select_one("div.match")
+                if not match_div:continue
+                home_el=match_div.select_one(".home-team .name")
+                away_el=match_div.select_one(".away-team .name")
+                time_el=match_div.select_one(".match-time")
+                league_el=li.find_previous("strong")
+                home_logo=match_div.select_one(".home-team .logo img")
+                away_logo=match_div.select_one(".away-team .logo img")
+                status_el=match_div.select_one(".status a")
+                event_id=""
+                if status_el:
+                    href=status_el.get("href","")
+                    m=re.search(r'/tran-dau/(\d+)/',href)
+                    if m:event_id=m.group(1)
+                if home_el and away_el:
+                    return{"home":home_el.get_text(strip=True),"away":away_el.get_text(strip=True),"score":"VS","minute":"","league":league_el.get_text(strip=True) if league_el else "","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":"upcoming"}
+        return None
+    return JSONResponse(_cached("ls_featured",_f,ttl=_cache_ttl_live))
+
 # ===== NEWS SCRAPERS =====
 def scrape_vne(cat_url):
     try:
@@ -112,6 +175,41 @@ def scrape_vne_article(url):
                 elif ch.name in("h2","h3"):body.append({"type":"heading","text":ch.get_text(strip=True)})
         return{"title":h1.get_text(strip=True) if h1 else "","summary":desc.get_text(strip=True) if desc else "","og_image":og_img,"body":body,"source":"vne","url":url}
     except:return None
+def scrape_dantri_hot():
+    try:
+        soup=_get("https://dantri.com.vn/tin-nong.htm");arts=[];seen=set()
+        for a in soup.find_all("a",href=True):
+            href=a.get("href","");title=a.get("title","") or a.get_text(strip=True)
+            if not title or len(title)<15 or"javascript:" in href:continue
+            if not href.startswith("http"):href="https://dantri.com.vn"+href
+            if href in seen or not href.endswith(".htm"):continue
+            if href=="https://dantri.com.vn/tin-nong.htm":continue
+            seen.add(href);arts.append({"title":title,"link":href,"img":"","source":"dantri"})
+            if len(arts)>=12:break
+        def _og(art):
+            try:r2=requests.get(art["link"],headers=HEADERS,timeout=8);r2.encoding="utf-8";og=BeautifulSoup(r2.text,"lxml").find("meta",property="og:image");art["img"]=og.get("content","") if og else ""
+            except:pass
+        with ThreadPoolExecutor(4) as ex:list(ex.map(_og,arts[:10]))
+        return[a for a in arts if a["img"]][:10]
+    except:return[]
+def scrape_dantri_article(url):
+    try:
+        r=requests.get(url,headers=HEADERS,timeout=15);r.encoding="utf-8";soup=BeautifulSoup(r.text,"lxml")
+        for tag in soup.find_all(["script","style","nav","footer","aside"]):tag.decompose()
+        h1=soup.find("h1");og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
+        content=soup.select_one("div.singular-content") or soup.select_one("article")
+        body=[]
+        if content:
+            for el in content.find_all(["p","h2","h3","figure","img"],recursive=True):
+                if el.name=="p":t=el.get_text(strip=True);(body.append({"type":"p","text":t}) if t and len(t)>15 else None)
+                elif el.name in("h2","h3"):t=el.get_text(strip=True);(body.append({"type":"heading","text":t}) if t else None)
+                elif el.name=="figure" or el.name=="img":
+                    im=el if el.name=="img" else el.find("img")
+                    if im:s=im.get("data-src") or im.get("src","");(body.append({"type":"img","src":s,"alt":im.get("alt","")}) if s and"base64" not in s else None)
+        desc="";sapo=soup.select_one("h2.singular-sapo");
+        if sapo:desc=sapo.get_text(strip=True)
+        return{"title":h1.get_text(strip=True) if h1 else "","summary":desc,"og_image":og_img,"body":body,"source":"dantri","url":url}
+    except:return None
 def scrape_bdp_videos():
     try:
         soup=_get(f"{BASE_BDP}/video");arts=[];seen=set()
@@ -165,19 +263,6 @@ def scrape_24h_shorts():
             if img_tag:img_src=img_tag.get("data-original") or img_tag.get("data-src") or img_tag.get("src","")
             if img_src and"base64" in img_src:img_src=""
             seen.add(href);arts.append({"title":title,"link":href,"img":img_src,"source":"24h-shorts"})
-        if len(arts)<3:
-            for a in soup.find_all("a",href=True):
-                href=a.get("href","")
-                if not href.endswith(".html") or"javascript:" in href or"-cvd" in href:continue
-                if not href.startswith("http"):href=BASE_24H+href
-                if href in seen:continue
-                img=a.find("img") or(a.parent.find("img") if a.parent else None)
-                if not img:continue
-                img_src=img.get("data-original") or img.get("data-src") or img.get("src","")
-                if not img_src or"base64" in img_src:continue
-                title=img.get("alt","") or a.get("title","") or a.get_text(strip=True)
-                if not title or len(title)<8:continue
-                seen.add(href);arts.append({"title":title,"link":href,"img":img_src,"source":"24h-shorts"})
         return arts[:20]
     except:return[]
 def scrape_bbc_vietnamese():
@@ -212,9 +297,6 @@ def scrape_bbc_article(url):
         for p in soup.select("[data-component='text-block'] p, article p, main p"):
             t=p.get_text(strip=True)
             if t and len(t)>20:body.append({"type":"p","text":t})
-        for img in soup.select("main img, article img"):
-            src=img.get("src","")
-            if src and("ichef" in src or"bbci" in src):body.append({"type":"img","src":src,"alt":img.get("alt","")})
         return{"title":h1.get_text(strip=True) if h1 else "","summary":"","og_image":og_img,"body":body,"source":"bbc","url":url}
     except:return None
 def extract_video_url(article_url):
@@ -222,8 +304,7 @@ def extract_video_url(article_url):
         r=requests.get(article_url,headers={**HEADERS,"Referer":"https://www.24h.com.vn/"},timeout=10);r.encoding="utf-8"
         m3u8s=re.findall(r'(https?://cdn\.24h\.com\.vn/[^\s"\'\\<>]+\.m3u8)',r.text)
         esc=[u.replace('\\//','/').replace('\\/','/') for u in re.findall(r'(https?:\\/\\/cdn\.24h\.com\.vn\\/[^\s"\'<>]+\.m3u8)',r.text)]
-        all_urls=list(dict.fromkeys(m3u8s+esc));full=[u for u in all_urls if'_720p' not in u];p720=[u for u in all_urls if'_720p' in u]
-        primary=full or p720
+        all_urls=list(dict.fromkeys(m3u8s+esc));primary=[u for u in all_urls if'_720p' not in u] or [u for u in all_urls if'_720p' in u]
         if not primary:return None
         soup=BeautifulSoup(r.text,"lxml");og=soup.find("meta",property="og:image");poster=og.get("content","") if og else ""
         return{"src":primary[0],"poster":poster}
@@ -270,6 +351,8 @@ def api_categories():
 def api_highlights():return JSONResponse(_cached("highlights",scrape_24h_highlights))
 @app.get("/api/shorts")
 def api_shorts():return JSONResponse(_cached("shorts",scrape_24h_shorts))
+@app.get("/api/dantri_hot")
+def api_dantri_hot():return JSONResponse(_cached("dantri_hot",scrape_dantri_hot))
 @app.get("/api/bdp_videos")
 def api_bdp_videos():return JSONResponse(_cached("bdp_videos",scrape_bdp_videos))
 @app.get("/api/video_url")
@@ -282,6 +365,7 @@ def api_video_url(url:str=Query(...)):
 def api_article(url:str=Query(...)):
     if"vnexpress.net" in url:data=scrape_vne_article(url)
     elif"bbc.com" in url:data=scrape_bbc_article(url)
+    elif"dantri.com.vn" in url:data=scrape_dantri_article(url)
     else:data=None
     return JSONResponse(data if data else{"error":"not supported"})
 @app.get("/v")
@@ -289,13 +373,13 @@ async def video_share(url:str=Query(default=""),title:str=Query(default="VNEWS V
     og_image=unquote(img) if img else "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
     decoded_url=unquote(url);decoded_title=unquote(title)
     redirect_script=f'<script>localStorage.setItem("pending_video",JSON.stringify({{"url":"{decoded_url}","type":"{type}"}}));location.href="{SPACE_URL}";</script>' if decoded_url else f'<script>location.href="{SPACE_URL}";</script>'
-    return HTMLResponse(f'''<!DOCTYPE html><html><head><meta charset="utf-8"><title>{decoded_title}</title><meta property="og:title" content="{decoded_title}"><meta property="og:image" content="{og_image}"><meta property="og:type" content="video.other"><meta property="og:site_name" content="VNEWS"></head><body style="background:#111;color:#fff;text-align:center;padding:40px"><p>⏳</p>{redirect_script}</body></html>''')
+    return HTMLResponse(f'''<!DOCTYPE html><html><head><meta charset="utf-8"><title>{decoded_title}</title><meta property="og:title" content="{decoded_title}"><meta property="og:image" content="{og_image}"></head><body style="background:#111;color:#fff;text-align:center;padding:40px"><p>⏳</p>{redirect_script}</body></html>''')
 @app.get("/s")
 async def share_redirect(url:str=Query(default=""),title:str=Query(default="VNEWS"),img:str=Query(default="")):
     og_image=unquote(img) if img else "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
     decoded_url=unquote(url)
     redirect_script=f'<script>localStorage.setItem("pending_article","{decoded_url}");location.href="{SPACE_URL}";</script>' if decoded_url else f'<script>location.href="{SPACE_URL}";</script>'
-    return HTMLResponse(f'''<!DOCTYPE html><html><head><meta charset="utf-8"><title>{unquote(title)}</title><meta property="og:title" content="{unquote(title)}"><meta property="og:image" content="{og_image}"><meta property="og:type" content="article"><meta property="og:site_name" content="VNEWS"></head><body>{redirect_script}</body></html>''')
+    return HTMLResponse(f'''<!DOCTYPE html><html><head><meta charset="utf-8"><title>{unquote(title)}</title><meta property="og:title" content="{unquote(title)}"><meta property="og:image" content="{og_image}"></head><body>{redirect_script}</body></html>''')
 @app.get("/")
 async def index():
     with open("/app/static/index.html","r",encoding="utf-8") as f:return HTMLResponse(content=f.read())