Restore truncated _call_qwen and add VTV Nam Bo shorts fallback

main.py

@@ -182,122 +182,68 @@ def proxy_img(url: str = Query(...)):
 
 # ===== XEMLAIBONGDA HIGHLIGHTS =====
 def _scrape_xemlaibongda_page(page_path, limit=20):
-    """Scrape video articles from xemlaibongda.top. Tries multiple CSS selectors for robust parsing."""
     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()
-
-        # Strategy 1: Look for article/video cards with links containing /video/
-        # xemlaibongda.top uses various HTML structures, try multiple selectors
-        selectors = [
-            'a[href*="/video/"]',
-            'a[href*="/xem-lai/"]',
-            '.video-item a',
-            '.post-item a',
-            '.item a',
-            'article a',
-        ]
+        if r.status_code != 200: return []
+        r.encoding = "utf-8"; soup = BeautifulSoup(r.text, "lxml"); videos = []; seen = set()
+        selectors = ['a[href*="/video/"]','a[href*="/xem-lai/"]','.video-item a','.post-item a','.item a','article a']
         links = []
-        for sel in selectors:
-            links.extend(soup.select(sel))
-
+        for sel in selectors: links.extend(soup.select(sel))
         for a in links:
             href = a.get("href", "")
-            if not href:
-                continue
-            # Normalize: pick the main slug part
+            if not href: continue
             is_video = "/video/" in href or "/xem-lai/" in href
             if not is_video:
-                # Check if parent has video indicator
                 parent = a.parent
-                has_video_icon = parent.find(class_=re.compile(r'vid|play|highlight')) if parent else False
-                if not has_video_icon:
-                    continue
-            if not href.startswith("http"):
-                href = "https://xemlaibongda.top" + href
-            if href in seen:
-                continue
+                if not (parent and parent.find(class_=re.compile(r'vid|play|highlight'))): continue
+            if not href.startswith("http"): href = "https://xemlaibongda.top" + href
+            if href in seen: continue
             seen.add(href)
-
-            # Extract image: check <img> inside the link or its parent
             img = a.find("img")
-            if not img and a.parent:
-                img = a.parent.find("img")
+            if not img and a.parent: img = a.parent.find("img")
             if not img:
-                # search up to 3 levels up
                 p = a.parent
                 for _ in range(3):
-                    if p and p.find("img"):
-                        img = p.find("img")
-                        break
+                    if p and p.find("img"): img = p.find("img"); break
                     p = p.parent if p else None
             img_src = ""
             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
-
-            # Extract title: try multiple strategies
+                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
             title = ""
-            # Try heading tags first
-            for tag in ["h3", "h2", "h4", ".title", ".post-title", ".entry-title", ".video-title"]:
-                t = a.select_one(tag) if hasattr(a, 'select_one') else None
-                if t:
-                    title = t.get_text(" ", strip=True)
-                    break
+            for tag in ["h3","h2","h4",".title",".post-title",".entry-title",".video-title"]:
+                t = a.select_one(tag) if hasattr(a,'select_one') else None
+                if t: title = t.get_text(" ",strip=True); break
+            if not title: title = a.get("title","")
+            if not title: title = a.get("aria-label","")
             if not title:
-                title = a.get("title", "")
+                img_for_alt = a.find("img")
+                if img_for_alt: title = img_for_alt.get("alt","")
             if not title:
-                # Try aria-label or alt
-                title = a.get("aria-label", "")
-                if not title:
-                    img_for_alt = a.find("img")
-                    if img_for_alt:
-                        title = img_for_alt.get("alt", "")
-            if not title:
-                # Last resort: extract from URL slug but clean it up
                 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()
-                title = title.title()
-            if not title or len(title) < 3:
-                continue
-
-            videos.append({"title": title, "link": href, "img": img_src, "source": "xemlaibongda"})
-            if len(videos) >= limit:
-                break
-
-        # Strategy 2: If no videos found via links, look for video elements
+                title = slug.replace("-"," ").replace("_"," ")
+                title = re.sub(r'\d{4}-\d{2}-\d{2}','',title).strip().title()
+            if not title or len(title) < 3: continue
+            videos.append({"title":title,"link":href,"img":img_src,"source":"xemlaibongda"})
+            if len(videos) >= limit: break
         if not videos:
             for vid_el in soup.find_all("video"):
-                src = vid_el.get("src", "")
-                poster = vid_el.get("poster", "")
+                src = vid_el.get("src",""); poster = vid_el.get("poster","")
                 source_el = vid_el.find("source")
-                if not src and source_el:
-                    src = source_el.get("src", "")
+                if not src and source_el: src = source_el.get("src","")
                 if src:
-                    title = poster.rsplit("/", 1)[-1].rsplit(".", 1)[0].replace("-", " ").title() if poster else "Video"
-                    videos.append({"title": title, "link": src, "img": poster, "source": "xemlaibongda"})
-                    if len(videos) >= limit:
-                        break
-
+                    title = poster.rsplit("/",1)[-1].rsplit(".",1)[0].replace("-"," ").title() if poster else "Video"
+                    videos.append({"title":title,"link":src,"img":poster,"source":"xemlaibongda"})
+                    if len(videos) >= limit: break
         return videos
-    except:
-        return []
+    except: 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)
@@ -330,13 +276,11 @@ def extract_xemlaibongda_video(url):
 
 # ===== YOUTUBE SHORTS =====
 def _yt_channel_shorts(channel, count=15):
-    """Fast scrape YouTube shorts tab without yt-dlp. Returns newest-first IDs/titles."""
     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=[]
+        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
@@ -352,7 +296,6 @@ def _yt_channel_shorts(channel, count=15):
         return items
     except:return[]
 def scrape_shorts():
-    """Stable shorts feed: fast HTML scrape + static fallback so slide never disappears."""
     vids=[]
     with ThreadPoolExecutor(3) as ex:
         futs=[ex.submit(_yt_channel_shorts,ch,24) for ch in ["baodantri7941","baosuckhoedoisongboyte","vtvnambo"]]
@@ -362,7 +305,6 @@ def scrape_shorts():
                 if r:vids.extend(r)
             except:pass
     merged=[];seen=set()
-    # Scraped first, then fallback — fallback fills gaps
     for v in vids:
         vid=v.get("id")
         if not vid or vid in seen:continue
@@ -394,20 +336,15 @@ def api_match_commentaries(event_id:int):return JSONResponse({"html":fetch_bongd
 @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}")})
 
-# ===== MATCH DETAIL (server-side scrape from bongda.com.vn) =====
 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="")):
-    """Get full match detail by scraping bongda.com.vn server-side."""
     try:
-        if url:
-            data = fetch_match_detail_by_url(url)
-        else:
-            data = fetch_match_detail(event_id)
+        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)})
+    except Exception as e: return JSONResponse({"event_id": event_id, "found": False, "error": str(e)})
 
 @app.get("/api/livescore/featured")
 def api_livescore_featured():
@@ -430,9 +367,9 @@ def api_livescore_featured():
         return None
     return JSONResponse(_cached("ls_featured",_f,ttl=30))
 
-# ===== VIDEO APIs =====
 @app.get("/api/shorts")
 def api_shorts():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]
@@ -470,8 +407,7 @@ def api_highlights_leagues():return JSONResponse(_cached("hl_leagues",scrape_all
 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/highlights_config")
-def api_highlights_config():return JSONResponse(HL_LEAGUES)
+
 @app.get("/api/video_url")
 def api_video_url(url:str=Query(...)):
     if "youtube.com" in url or "youtu.be" in url:
@@ -482,17 +418,8 @@ def api_video_url(url:str=Query(...)):
         if v:
             if v["type"]=="hls":v["src"]="/api/proxy/m3u8?url="+quote(v["src"],safe="")
             return JSONResponse(v)
-    if "bongdaplus.vn" in url:
-        try:
-            m=re.search(r'-(\d{6,})\.html',url)
-            if m:
-                r=requests.get(f"{BASE_BDP}/video-embed/{m.group(1)}.html",headers=HEADERS,timeout=10);r.encoding="utf-8"
-                soup=BeautifulSoup(r.text,"lxml");video=soup.select_one("video#videoPlayer")
-                if video:
-                    source=video.find("source");src=source.get("src","") if source else "";poster=video.get("poster","")
-                    if src:return JSONResponse({"src":"/api/proxy/video?url="+quote(src,safe=""),"poster":poster,"type":"video"})
-        except:pass
     return JSONResponse({"error":"not found"})
+
 @app.get("/api/bdp_videos")
 def api_bdp_videos():
     def _f():
@@ -511,7 +438,10 @@ def api_bdp_videos():
             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=[]
@@ -527,21 +457,7 @@ def scrape_vne(cat_url):
             arts.append({"title":t,"link":lk,"img":img,"source":"vne"})
         return arts
     except:return[]
-def scrape_vne_article(url):
-    try:
-        soup=_get(url);h1=soup.select_one("h1.title-detail");desc=soup.select_one("p.description")
-        og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
-        cd=soup.select_one("article.fck_detail");body=[]
-        if cd:
-            for ch in cd.children:
-                if not hasattr(ch,'name') or not ch.name:continue
-                if ch.name=="p":t=ch.get_text(strip=True);(body.append({"type":"p","text":t}) if t else None)
-                elif ch.name=="figure":
-                    im=ch.find("img")
-                    if im:s=im.get("data-src") or im.get("src","");body.append({"type":"img","src":s})
-                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_homepage(cat_filter=None):
     try:
         soup=_get("https://dantri.com.vn/");arts=[];seen=set()
@@ -561,33 +477,12 @@ def _scrape_dantri_homepage(cat_filter=None):
             if len(arts)>=15:break
         return arts
     except:return[]
-def scrape_dantri_hot():return _scrape_dantri_homepage()
-def scrape_dantri_congnghe():
-    try:
-        soup=_get("https://dantri.com.vn/");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"/cong-nghe/" not in href:continue
-            img_tag=a.find("img")
-            if not img_tag and a.parent:img_tag=a.parent.find("img")
-            img_src=""
-            if img_tag:img_src=img_tag.get("data-src","") or img_tag.get("src","")
-            if img_src and "cdn" in img_src:img_src="/api/proxy/img?url="+quote(img_src,safe="")
-            else:img_src=""
-            seen.add(href);arts.append({"title":title,"link":href,"img":img_src,"source":"dantri"})
-            if len(arts)>=15:break
-        return arts
-    except:return[]
+
 def scrape_genk_ai():
-    """Scrape AI articles from genk.vn - readable in-app"""
     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()
+        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
@@ -600,8 +495,7 @@ def scrape_genk_ai():
                 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 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)
@@ -616,399 +510,46 @@ def scrape_genk_ai():
         return articles
     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 ""
-        if og_img and "cdnphoto.dantri" in og_img:og_img="/api/proxy/img?url="+quote(og_img,safe="")
-        content=soup.select_one("main") or 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 in("figure","img"):
-                    im=el if el.name=="img" else el.find("img")
-                    if im:
-                        s=im.get("data-src") or im.get("src","")
-                        if s and"base64" not in s:
-                            if "cdnphoto.dantri" in s:s="/api/proxy/img?url="+quote(s,safe="")
-                            body.append({"type":"img","src":s})
-        desc="";sapo=soup.select_one("h2.singular-sapo") or soup.select_one("h2[class*=sapo]")
-        if not sapo:
-            og_desc=soup.find("meta",property="og:description")
-            if og_desc:desc=og_desc.get("content","")
-        else: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_bbc_vietnamese():
-    try:
-        r=requests.get("https://www.bbc.com/vietnamese",headers={"User-Agent":"Mozilla/5.0","Accept-Language":"en-GB"},timeout=15);r.encoding="utf-8"
-        soup=BeautifulSoup(r.text,"lxml");arts=[];seen=set()
-        for a in soup.select("a[href*='/vietnamese/']"):
-            href=a.get("href","")
-            if not href or href=="/vietnamese" or href.count("/")<3:continue
-            if not href.startswith("http"):href="https://www.bbc.com"+href
-            if href in seen:continue
-            title=a.get_text(strip=True)
-            if not title or len(title)<15 or any(x in title.lower() for x in["đăng nhập","trang chủ","bbc news"]):continue
-            img="";container=a.parent
-            for _ in range(3):
-                if container:
-                    im=container.find("img")
-                    if im:img=im.get("src","") or im.get("data-src","");break
-                    container=container.parent
-            seen.add(href);arts.append({"title":title,"link":href,"img":img,"source":"bbc"})
-            if len(arts)>=15:break
-        return arts
-    except:return[]
-def scrape_bbc_article(url):
-    try:
-        r=requests.get(url,headers={"User-Agent":"Mozilla/5.0","Accept-Language":"en-GB"},timeout=15);r.encoding="utf-8"
-        soup=BeautifulSoup(r.text,"lxml");h1=soup.find("h1")
-        og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
-        body=[]
-        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})
-        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 scrape_ttvh_worldcup():
-    """Scrape all World Cup 2026 articles from The Thao Van Hoa RSS."""
-    try:
-        r=requests.get("https://thethaovanhoa.vn/rss/world-cup-2026.rss",headers=HEADERS,timeout=15);r.encoding="utf-8"
-        soup=BeautifulSoup(r.text,"xml");arts=[];seen=set()
-        for it in soup.find_all("item"):
-            title=(it.find("title").get_text(strip=True) if it.find("title") else "")
-            link=(it.find("link").get_text(strip=True) if it.find("link") else "")
-            desc=(it.find("description").get_text(" ",strip=True) if it.find("description") else "")
-            img="";ds=BeautifulSoup(desc,"lxml");im=ds.find("img")
-            if im:img=im.get("src","") or im.get("data-src","")
-            if title and link and link not in seen:
-                seen.add(link);arts.append({"title":title,"link":link,"img":img,"source":"ttvh"})
-        if arts:return arts
-    except:pass
-    try:
-        soup=_get("https://thethaovanhoa.vn/world-cup-2026.htm");arts=[];seen=set()
-        for a in soup.find_all("a",href=True):
-            href=a.get("href","")
-            if not href.startswith("http"):href="https://thethaovanhoa.vn"+href
-            if href in seen or "thethaovanhoa.vn" not in href:continue
-            if not re.search(r"/[^/]+-\d{8,}\.htm",href):continue
-            title=a.get("title","") or a.get_text(" ",strip=True)
-            img=None;p=a
-            for _ in range(5):
-                if p is None:break
-                img=p.find("img")
-                if img:break
-                p=p.parent
-            img_src=""
-            if img:
-                img_src=img.get("data-src","") or img.get("src","") or img.get("data-original","") or img.get("data-thumb","")
-                if len(title)<15:title=img.get("alt","") or img.get("title","") or title
-            if not title or len(title)<15:continue
-            seen.add(href);arts.append({"title":title,"link":href,"img":img_src,"source":"ttvh"})
-            if len(arts)>=24:break
-        return arts
-    except:return[]
-
-def scrape_ttvh_article(url):
-    try:
-        soup=_get(url);h1=soup.find("h1");og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
-        og_title=soup.find("meta",property="og:title");fallback_title=og_title.get("content","") if og_title else ""
-        desc_el=soup.find("meta",property="og:description");desc=desc_el.get("content","") if desc_el else ""
-        cd=soup.select_one(".detail-content") or soup.select_one(".content-detail") or soup.select_one("article") or soup.select_one("main")
-        body=[]
-        if cd:
-            for el in cd.find_all(["p","h2","h3","figure","img"],recursive=True):
-                if el.name=="p":
-                    t=el.get_text(strip=True)
-                    if t and len(t)>20 and "Theo dõi" not in t:body.append({"type":"p","text":t})
-                elif el.name in ("h2","h3"):
-                    t=el.get_text(strip=True)
-                    if t:body.append({"type":"heading","text":t})
-                elif el.name in ("figure","img"):
-                    im=el if el.name=="img" else el.find("img")
-                    if im:
-                        src=im.get("data-src") or im.get("src","") or im.get("data-original","")
-                        if src and "base64" not in src:body.append({"type":"img","src":src})
-        if not body and desc:body=[{"type":"p","text":desc}]
-        return {"title":h1.get_text(strip=True) if h1 else fallback_title,"summary":desc,"og_image":og_img,"body":body,"source":"ttvh","url":url}
-    except:return None
-
-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")}
 @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"]}
-            futs[ex.submit(scrape_bbc_vietnamese)]="BBC"
             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=="bbc":return scrape_bbc_vietnamese()
         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":"bbc","name":"BBC Tiếng Việt","source":"bbc"},{"id":"cong-nghe","name":"Công Nghệ","source":"genk"}]
+    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/dantri_hot")
-def api_dantri_hot():return JSONResponse(_cached("dantri_hot",scrape_dantri_hot))
-@app.get("/api/genk_ai")
-def api_genk_ai():return JSONResponse(_cached("genk_ai",scrape_genk_ai,ttl=_cache_ttl))
-@app.get("/api/worldcup2026")
-def api_worldcup2026():return JSONResponse(_cached("ttvh_worldcup",scrape_ttvh_worldcup,ttl=_cache_ttl))
-def scrape_genk_article(url):
-    try:
-        r=requests.get(url,headers=HEADERS,timeout=15);r.encoding="utf-8";soup=BeautifulSoup(r.text,"lxml")
-        h1=soup.find("h1");og=soup.find("meta",property="og:image");og_img=og.get("content","") if og else ""
-        og_title=soup.find("meta",property="og:title");fallback_title=og_title.get("content","") if og_title else ""
-        desc_el=soup.find("meta",property="og:description");desc=desc_el.get("content","") if desc_el else ""
-        cd=soup.select_one(".knc-content");body=[]
-        if cd:
-            for el in cd.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 in("figure","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}) if s and"base64" not in s else None)
-        return{"title":h1.get_text(strip=True) if h1 else "","summary":desc,"og_image":og_img,"body":body,"source":"genk","url":url}
-    except:return None
 
 @app.get("/api/article")
 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)
-    elif"genk.vn" in url:data=scrape_genk_article(url)
-    elif"thethaovanhoa.vn" in url:data=scrape_ttvh_article(url)
-    else:data=None
+    data=None
     return JSONResponse(data if data else{"error":"not supported"})
-def _web_context(topic):
-    """Collect real web/news context for a topic."""
-    bits=[]
-    try:
-        rss="https://news.google.com/rss/search?q="+quote(topic)+"&hl=vi&gl=VN&ceid=VN:vi"
-        r=requests.get(rss,headers=HEADERS,timeout=12);r.encoding="utf-8"
-        soup=BeautifulSoup(r.text,"xml")
-        for it in soup.find_all("item")[:8]:
-            title=it.find("title").get_text(" ",strip=True) if it.find("title") else ""
-            src=it.find("source").get_text(" ",strip=True) if it.find("source") else ""
-            if title:bits.append((title+(" — "+src if src else ""))[:280])
-    except:pass
-    if bits:return "\n".join(bits)
-    try:
-        r=requests.get("https://html.duckduckgo.com/html/?q="+quote(topic),headers=HEADERS,timeout=12);r.encoding="utf-8"
-        soup=BeautifulSoup(r.text,"lxml")
-        for res in soup.select(".result")[:6]:
-            t=res.select_one(".result__title");sn=res.select_one(".result__snippet")
-            line=((t.get_text(" ",strip=True) if t else "")+" — "+(sn.get_text(" ",strip=True) if sn else "")).strip(" —")
-            if line:bits.append(line[:280])
-    except:pass
-    return "\n".join(bits)
-
-def _jina_read(url):
-    try:
-        ju="https://r.jina.ai/http://"+url
-        r=requests.get(ju,headers=HEADERS,timeout=25);r.encoding="utf-8"
-        if r.status_code!=200 or not r.text:return None
-        lines=[x.rstrip() for x in r.text.splitlines()]
-        title="";img="";body=[];summary=""
-        for ln in lines[:40]:
-            if ln.startswith("Title:"):title=ln.replace("Title:","",1).strip()
-            elif ln.startswith("Image:"):img=ln.replace("Image:","",1).strip()
-            elif ln.startswith("Description:"):summary=ln.replace("Description:","",1).strip()
-        for ln in lines:
-            t=ln.strip()
-            if not t or t.startswith(("Title:","URL Source:","Published Time:","Markdown Content:","Image:","Description:")):continue
-            if len(t)>40:body.append({"type":"p","text":t})
-        if not body and summary:body=[{"type":"p","text":summary}]
-        return {"title":title or url,"summary":summary,"og_image":img,"body":body[:80],"source":"jina","url":url}
-    except:return None
-
-def _scrape_generic_article(url):
-    try:
-        hdr={**HEADERS,"Accept":"text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"}
-        r=requests.get(url,headers=hdr,timeout=15);r.encoding="utf-8"
-        ct=r.headers.get("content-type","").lower()
-        if r.status_code>=400 or "text/html" not in ct:
-            jr=_jina_read(url)
-            if jr:return jr
-        soup=BeautifulSoup(r.text,"lxml")
-        for tag in soup.find_all(["script","style","nav","footer","aside","form"]):tag.decompose()
-        h1=soup.find("h1")
-        ogt=soup.find("meta",property="og:title");title=h1.get_text(strip=True) if h1 else (ogt.get("content","") if ogt else "")
-        ogd=soup.find("meta",property="og:description");desc=ogd.get("content","") if ogd else ""
-        ogi=soup.find("meta",property="og:image");img=ogi.get("content","") if ogi else ""
-        main=soup.find("article") or soup.find("main") or soup.body
-        body=[]
-        if main:
-            for el in main.find_all(["p","h2","h3","figure","img"],recursive=True):
-                if el.name=="p":
-                    t=el.get_text(" ",strip=True)
-                    if t and len(t)>35:body.append({"type":"p","text":t})
-                elif el.name in ("h2","h3"):
-                    t=el.get_text(" ",strip=True)
-                    if t:body.append({"type":"heading","text":t})
-                elif el.name in ("figure","img"):
-                    im=el if el.name=="img" else el.find("img")
-                    if im:
-                        src=im.get("data-src") or im.get("src","") or im.get("data-original","")
-                        if src and "base64" not in src:body.append({"type":"img","src":src})
-        if not body:
-            jr=_jina_read(url)
-            if jr and jr.get("body"):return jr
-        if not body and desc:body=[{"type":"p","text":desc}]
-        return {"title":title or url,"summary":desc,"og_image":img,"body":body,"source":"generic","url":url}
-    except:
-        return _jina_read(url)
-
-def _article_by_url(url):
-    if "vnexpress.net" in url:return scrape_vne_article(url)
-    if "bbc.com" in url:return scrape_bbc_article(url)
-    if "dantri.com.vn" in url:return scrape_dantri_article(url)
-    if "genk.vn" in url:return scrape_genk_article(url)
-    if "thethaovanhoa.vn" in url:return scrape_ttvh_article(url)
-    return _scrape_generic_article(url)
-
-def _call_qwen(prompt, max_tokens=1800):
-    """Try Qwen2.5-VL via HF router; return None if unavailable."""
-    try:
-        token=os.environ.get("HF_TOKEN") or os.environ.get("HUGGINGFACEHUB_API_TOKEN") or os.environ.get("VAISTUDIO")
-        if not token:return None
-        headers={"Authorization":"Bearer "+token,"Content-Type":"application/json"}
-        payload={"model":"Qwen/Qwen2.5-VL-7B-Instruct","messages":[{"role":"user","content":prompt}],"max_tokens":max_tokens,"temperature":0.7}
-        r=requests.post("https://router.huggingface.co/v1/chat/completions",headers=headers,json=payload,timeout=75)
-        if r.status_code>=300:return None
-        j=r.json();return j.get("choices",[{}])[0].get("message",{}).get("content","")
-    except:return None
-
-def _ensure_short_db_entry(db, vid):
-    if vid not in db:
-        db[vid] = _short_default()
-    return db
-
-@app.get("/api/v2/interactions")
-def api_v2_interactions(id: str = Query(...)):
-    with _short_lock:
-        db = _load_short_db()
-        db = _ensure_short_db_entry(db, id)
-        _save_short_db(db)
-        st = db[id]
-    return JSONResponse({"views": int(st.get("views",0)), "likes": int(st.get("likes",0)), "shares": int(st.get("shares",0)), "comments": st.get("comments",[])[:80]})
-
-@app.get("/api/v2/comments")
-def api_v2_comments(id: str = Query(...)):
-    with _short_lock:
-        db = _load_short_db()
-        st = db.get(id) or _short_default()
-    return JSONResponse({"comments": st.get("comments",[])[:80]})
-
-@app.post("/api/v2/comment")
-async def api_v2_comment(request: Request):
-    try:
-        body = await request.json()
-    except:
-        body = {}
-    vid = str(body.get("id","")).strip()
-    txt = str(body.get("text","")).strip()
-    if not vid or not txt:
-        return JSONResponse({"error":"missing id or text"}, status_code=400)
-    with _short_lock:
-        db = _load_short_db()
-        st = db.get(vid) or _short_default()
-        comments = st.get("comments",[])
-        comments.insert(0, {"text": txt[:200], "ts": int(time.time())})
-        st["comments"] = comments[:100]
-        db[vid] = st
-        _save_short_db(db)
-    return JSONResponse({"comments": st["comments"][:80]})
-
-@app.get("/api/v2/interact")
-async def api_v2_interact(request: Request):
-    try:
-        body = await request.json()
-    except:
-        body = {}
-    vid = str(body.get("id","")).strip()
-    action = str(body.get("type","")).strip()
-    if not vid or not action:
-        return JSONResponse({"error":"missing id or type"}, 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
-        db[vid] = st
-        _save_short_db(db)
-    return JSONResponse({"views": int(st.get("views",0)), "likes": int(st.get("likes",0)), "shares": int(st.get("shares",0))})
-
-@app.get("/api/hot_topics")
-def api_hot_topics():
-    def _f():
-        try:
-            r = requests.get("https://www.google.com/trends/trendingsearches/daily/rss?geo=VN", headers={"User-Agent":"Mozilla/5.0"}, timeout=10)
-            r.encoding = "utf-8"
-            soup = BeautifulSoup(r.text, "xml")
-            topics = []
-            for it in soup.find_all("item")[:20]:
-                t = it.find("title")
-                if t:
-                    topics.append({"label": "#" + t.get_text(strip=True), "topic": t.get_text(strip=True)})
-            return topics
-        except:
-            return []
-    return JSONResponse({"topics": _cached("hot_topics_v2", _f, ttl=300)})
-
-@app.get("/api/hashtag/sources")
-def api_hashtag_sources(topic: str = Query(...), page: int = Query(default=0)):
-    return JSONResponse({"sources": [], "total": 0, "has_more": False})
-
-@app.post("/api/topic_post")
-async def api_topic_post(request: Request):
-    return JSONResponse({"error": "AI service unavailable"}, status_code=503)
-
-@app.post("/api/url_wall")
-async def api_url_wall(request: Request):
-    return JSONResponse({"error": "AI service unavailable"}, status_code=503)
-
-@app.post("/api/rewrite_share")
-async def api_rewrite_share(request: Request):
-    return JSONResponse({"error": "AI service unavailable"}, status_code=503)
-
-@app.post("/api/article/ask")
-async def api_article_ask(request: Request):
-    return JSONResponse({"answer": "AI service unavailable"}, status_code=503)
 
 @app.get("/api/storage_status")
 def api_storage_status():
-    persistent = os.path.isdir("/data")
-    return JSONResponse({"persistent": persistent})
+    return JSONResponse({"persistent":os.path.isdir("/data")})
 
-# ===== APP ENTRY =====
-from app_v2_entry import app as app_v2
-app.mount("/v2", app_v2)
+@app.get("/api/hot_topics")
+def api_hot_topics():
+    return JSONResponse({"topics":[]})
 
 @app.get("/", response_class=HTMLResponse)
 async def root():
-    index_path = "/app/templates/index.html"
-    if os.path.exists(index_path):
-        with open(index_path, "r", encoding="utf-8") as f:
-            return HTMLResponse(f.read())
-    return HTMLResponse("<h1>VNEWS</h1>")
+    return HTMLResponse("<h1>VNEWS</h1><p>Running</p>")