fix: hash routing works, shorts fallback scraper, highlight img from og, video comments+views

main.py

@@ -1,4 +1,4 @@
-"""VNEWS - FastAPI backend with news scrapers and API."""
+"""VNEWS - FastAPI backend."""
 import hashlib, re, time
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from fastapi import FastAPI, Query
@@ -8,305 +8,274 @@ import requests
 from bs4 import BeautifulSoup
 
 app = FastAPI()
-
-HEADERS = {
-    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
-    "Accept-Language": "vi-VN,vi;q=0.9,en;q=0.8",
-}
+HEADERS = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36","Accept-Language":"vi-VN,vi;q=0.9,en;q=0.8"}
 BASE_BDP = "https://bongdaplus.vn"
 BASE_24H = "https://www.24h.com.vn"
 SPACE_URL = "https://bep40-vnews.hf.space"
-
 _cache = {}
 _cache_ttl = 300
 
 def _cached(key, fn):
-    now = time.time()
-    if key in _cache and now - _cache[key]["t"] < _cache_ttl:
-        return _cache[key]["d"]
-    try: data = fn()
-    except: data = _cache.get(key, {}).get("d", [])
-    _cache[key] = {"d": data, "t": now}
-    return data
+    now=time.time()
+    if key in _cache and now-_cache[key]["t"]<_cache_ttl: return _cache[key]["d"]
+    try: data=fn()
+    except: data=_cache.get(key,{}).get("d",[])
+    _cache[key]={"d":data,"t":now}; return data
 
-def _get(url, headers=None):
-    h = headers or HEADERS
-    r = requests.get(url, headers=h, timeout=15); r.encoding="utf-8"
-    return BeautifulSoup(r.text, "lxml")
+def _get(url,headers=None):
+    h=headers or HEADERS; r=requests.get(url,headers=h,timeout=15); r.encoding="utf-8"
+    return BeautifulSoup(r.text,"lxml")
 
-# ═══ Scrapers ═══
 def scrape_vne(cat_url):
     try:
-        soup = _get(cat_url); arts = []
+        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 ""
+            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]
-            desc = it.select_one("p.description")
+                src=it.find("source")
+                if src:img=src.get("srcset","").split(",")[0].strip().split(" ")[0]
+            desc=it.select_one("p.description")
             arts.append({"title":t,"link":lk,"img":img,"summary":(desc.get_text(strip=True)[:150] if desc else ""),"source":"vne"})
         return arts
-    except: return []
+    except:return[]
 
 def scrape_vne_article(url):
     try:
-        soup = _get(url)
-        h1 = soup.select_one("h1.title-detail")
-        title = h1.get_text(strip=True) if h1 else ""
-        desc = soup.select_one("p.description")
-        summary = desc.get_text(strip=True) if desc else ""
-        cd = soup.select_one("article.fck_detail")
-        body = []
+        soup=_get(url);h1=soup.select_one("h1.title-detail");desc=soup.select_one("p.description")
+        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)
-                    if t: body.append({"type":"p","text":t})
-                elif ch.name == "figure":
-                    im = ch.find("img")
+                if not hasattr(ch,'name') or not ch.name:continue
+                if ch.name=="p":
+                    t=ch.get_text(strip=True)
+                    if t:body.append({"type":"p","text":t})
+                elif ch.name=="figure":
+                    im=ch.find("img")
                     if im:
-                        s = im.get("data-src") or im.get("src","")
-                        cap = ch.find("figcaption")
-                        if s: body.append({"type":"img","src":s,"alt":cap.get_text(strip=True) if cap else ""})
-                elif ch.name in ("h2","h3"):
-                    body.append({"type":"heading","text":ch.get_text(strip=True)})
-        return {"title":title,"summary":summary,"body":body,"source":"vne","url":url}
-    except: return None
+                        s=im.get("data-src") or im.get("src","")
+                        cap=ch.find("figcaption")
+                        if s:body.append({"type":"img","src":s,"alt":cap.get_text(strip=True) if cap else ""})
+                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 "","body":body,"source":"vne","url":url}
+    except:return None
 
 def scrape_bdp(url):
     try:
-        soup = _get(url); arts = []; seen = set()
-        for sel in ["div.news.fst","div.sld-itm.news","li.news"]:
+        soup=_get(url);arts=[];seen=set()
+        for sel in["div.news.fst","div.sld-itm.news","li.news"]:
             for it in soup.select(sel):
-                tag = it.find("a",class_="title") or it.find("a",href=True)
-                if not tag: continue
-                t=tag.get_text(strip=True); lk=tag.get("href","")
-                if not t or len(t)<5: continue
-                if lk and not lk.startswith("http"): lk=BASE_BDP+lk
-                if lk in seen: continue
-                im=it.find("img"); img=(im.get("data-src") or im.get("src","")) if im else ""
-                seen.add(lk); arts.append({"title":t,"link":lk,"img":img,"source":"bdp","summary":""})
+                tag=it.find("a",class_="title") or it.find("a",href=True)
+                if not tag:continue
+                t=tag.get_text(strip=True);lk=tag.get("href","")
+                if not t or len(t)<5:continue
+                if lk and not lk.startswith("http"):lk=BASE_BDP+lk
+                if lk in seen:continue
+                im=it.find("img");img=(im.get("data-src") or im.get("src","")) if im else ""
+                seen.add(lk);arts.append({"title":t,"link":lk,"img":img,"source":"bdp","summary":""})
         return arts[:15]
-    except: return []
+    except:return[]
 
 def scrape_bdp_videos():
     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"})
+        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 []
+    except:return[]
 
 def scrape_24h_highlights():
+    """Scrape highlights + fetch og:image for items missing img."""
     try:
-        soup = _get(f"{BASE_24H}/video-highlight-c953.html"); 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 = BASE_24H + href
-            if href in seen or "video" not in href.lower() or href.endswith("-c953.html"): continue
+        soup=_get(f"{BASE_24H}/video-highlight-c953.html");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=BASE_24H+href
+            if href in seen or "video" not in href.lower() or href.endswith("-c953.html"):continue
             seen.add(href)
-            img = a.find("img")
-            img_src = ""
-            if img:
-                img_src = img.get("data-original") or img.get("data-src") or img.get("src","")
-                if "base64" in img_src: img_src = ""
+            img=a.find("img")
+            img_src=""
+            if img:img_src=img.get("data-original") or img.get("data-src") or img.get("src","")
+            if img_src and "base64" in img_src:img_src=""
             arts.append({"title":title,"link":href,"img":img_src,"source":"24h"})
+        # Fetch og:image for items without img (parallel, max 10)
+        def _fetch_og(art):
+            if art["img"]:return art
+            try:
+                r=requests.get(art["link"],headers=HEADERS,timeout=8);r.encoding="utf-8"
+                s=BeautifulSoup(r.text,"lxml")
+                og=s.find("meta",property="og:image")
+                if og:art["img"]=og.get("content","")
+            except:pass
+            return art
+        need_img=[a for a in arts if not a["img"]][:10]
+        if need_img:
+            with ThreadPoolExecutor(5) as ex:
+                list(ex.map(_fetch_og,need_img))
         return arts[:30]
-    except: return []
+    except:return[]
 
 def scrape_24h_shorts():
+    """Scrape shorts - try <article> first, fallback to <a> with images."""
     try:
-        soup = _get(f"{BASE_24H}/video/video-tin-tuc-cvd769.html"); arts = []; seen = set()
+        r=requests.get(f"{BASE_24H}/video/video-tin-tuc-cvd769.html",headers=HEADERS,timeout=15)
+        r.encoding="utf-8";soup=BeautifulSoup(r.text,"lxml")
+        arts=[];seen=set()
+        # Method 1: <article> tags
         for art in soup.find_all("article"):
-            a = art.find("a", href=True)
-            if not a: continue
-            href = a.get("href","")
-            if not href.startswith("http"): href = BASE_24H + href
-            if href in seen: continue
-            img_tag = art.find("img")
-            title = (img_tag.get("alt","") if img_tag else "") or a.get("title","") or a.get_text(strip=True)
-            if not title or len(title)<10: continue
-            img_src = ""
-            if img_tag:
-                img_src = img_tag.get("data-original") or img_tag.get("data-src") or img_tag.get("src","")
-                if "base64" in img_src: img_src = ""
-            seen.add(href); arts.append({"title":title,"link":href,"img":img_src,"source":"24h-shorts"})
+            a=art.find("a",href=True)
+            if not a:continue
+            href=a.get("href","")
+            if not href.startswith("http"):href=BASE_24H+href
+            if href in seen:continue
+            img_tag=art.find("img")
+            title=(img_tag.get("alt","") if img_tag else "") or a.get("title","") or a.get_text(strip=True)
+            if not title or len(title)<10:continue
+            img_src=""
+            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"})
+        # Method 2: fallback - all <a> with <img> inside
+        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")
+                if not img:
+                    p=a.parent
+                    if p:img=p.find("img")
+                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 []
+    except:return[]
 
 def scrape_bbc_vietnamese():
-    """Scrape BBC Vietnamese - uses UK-based server headers."""
     try:
-        bbc_headers = {
-            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
-            "Accept-Language": "en-GB,en;q=0.9,vi;q=0.8",
-            "Accept": "text/html,application/xhtml+xml",
-        }
-        r = requests.get("https://www.bbc.com/vietnamese", headers=bbc_headers, timeout=15)
-        r.encoding = "utf-8"
-        soup = BeautifulSoup(r.text, "lxml")
-        arts = []; seen = set()
-        # BBC uses data-testid or specific class patterns
+        bbc_h={"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36","Accept-Language":"en-GB,en;q=0.9"}
+        r=requests.get("https://www.bbc.com/vietnamese",headers=bbc_h,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 == "/vietnamese/": continue
-            if 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: continue
-            # Skip navigation/footer links
-            if any(x in title.lower() for x in ["đăng nhập","trang chủ","bbc news"]): continue
-            img = ""
-            # Look for image in parent/sibling
-            container = a.parent
+            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:continue
+            if 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","summary":""})
-            if len(arts) >= 15: break
+                    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","summary":""})
+            if len(arts)>=15:break
         return arts
-    except: return []
+    except:return[]
 
 def scrape_bbc_article(url):
     try:
-        bbc_headers = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36","Accept-Language":"en-GB,en;q=0.9"}
-        r = requests.get(url, headers=bbc_headers, timeout=15); r.encoding="utf-8"
-        soup = BeautifulSoup(r.text,"lxml")
-        h1 = soup.find("h1")
-        title = h1.get_text(strip=True) if h1 else ""
-        body = []
-        # BBC article paragraphs
+        bbc_h={"User-Agent":"Mozilla/5.0","Accept-Language":"en-GB,en;q=0.9"}
+        r=requests.get(url,headers=bbc_h,timeout=15);r.encoding="utf-8"
+        soup=BeautifulSoup(r.text,"lxml");h1=soup.find("h1")
+        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 and t != title:
-                body.append({"type":"p","text":t})
-        # Images
+            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":title,"summary":"","body":body,"source":"bbc","url":url}
-    except: return None
+            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":"","body":body,"source":"bbc","url":url}
+    except:return None
 
 def extract_video_url(article_url):
-    """Extract m3u8 video URL + probe for multi-part."""
     try:
-        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
-        if not primary: return None
-        soup = BeautifulSoup(r.text,"lxml")
-        og = soup.find("meta",property="og:image")
-        poster = og.get("content","") if og else ""
-        
-        # Probe for multi-part
-        base_url = primary[0]
-        parts = [base_url]
-        is_720p = '_720p.m3u8' in base_url
-        name = base_url.replace('_720p.m3u8','').replace('.m3u8','')
-        m = re.search(r'(\D)(\d{1,2})$', name)
+        r=requests.get(article_url,headers={**HEADERS,"Referer":"https://www.24h.com.vn/"},timeout=10);r.encoding="utf-8"
+        m3u8s=re.findall(r'(https?://cdn\.24h\.com\.vn/[^\s"\'\\<>]+\.m3u8)',r.text)
+        esc=[u.replace('\\//','/').replace('\\/','/') for u in re.findall(r'(https?:\\/\\/cdn\.24h\.com\.vn\\/[^\s"\'<>]+\.m3u8)',r.text)]
+        all_urls=list(dict.fromkeys(m3u8s+esc))
+        full=[u for u in all_urls if'_720p' not in u];p720=[u for u in all_urls if'_720p' in u]
+        primary=full or p720
+        if not primary:return None
+        soup=BeautifulSoup(r.text,"lxml");og=soup.find("meta",property="og:image")
+        poster=og.get("content","") if og else ""
+        # Multi-part probe
+        base_url=primary[0];parts=[base_url]
+        is_720p='_720p.m3u8' in base_url
+        name=base_url.replace('_720p.m3u8','').replace('.m3u8','')
+        m=re.search(r'(\D)(\d{1,2})$',name)
         if m:
-            cur = int(m.group(2)); w = len(m.group(2)); base_name = name[:m.start(2)]
-            for i in range(cur+1, cur+5):
-                nn = f"{base_name}{i:02d}" if w==2 else f"{base_name}{i}"
-                for sfx in [('_720p.m3u8' if is_720p else '.m3u8'),'_720p.m3u8','.m3u8']:
-                    pu = nn+sfx
-                    if pu in parts: continue
+            cur=int(m.group(2));w=len(m.group(2));bn=name[:m.start(2)]
+            for i in range(cur+1,cur+5):
+                nn=f"{bn}{i:02d}" if w==2 else f"{bn}{i}"
+                for sfx in[('_720p.m3u8' if is_720p else '.m3u8'),'_720p.m3u8','.m3u8']:
+                    pu=nn+sfx
+                    if pu in parts:continue
                     try:
-                        if requests.head(pu, headers=HEADERS, timeout=3, allow_redirects=True).status_code==200:
-                            parts.append(pu); break
-                    except: continue
-                else: break
-        # Pattern B: CDN id increment
+                        if requests.head(pu,headers=HEADERS,timeout=3,allow_redirects=True).status_code==200:parts.append(pu);break
+                    except:continue
+                else:break
         if len(parts)==1:
-            m2 = re.match(r'^(.+-)(\d{7,})(-.+?)(\d{1,2})((?:_720p)?\.m3u8)$', base_url)
+            m2=re.match(r'^(.+-)(\d{7,})(-.+?)(\d{1,2})((?:_720p)?\.m3u8)$',base_url)
             if m2:
-                pfx,cid,mid,pn,sfx = m2.group(1),int(m2.group(2)),m2.group(3),int(m2.group(4)),m2.group(5)
+                pfx,cid,mid,pn,sfx=m2.group(1),int(m2.group(2)),m2.group(3),int(m2.group(4)),m2.group(5)
                 for i in range(1,5):
-                    pu = f"{pfx}{cid+i}{mid}{pn+i}{sfx}"
+                    pu=f"{pfx}{cid+i}{mid}{pn+i}{sfx}"
                     try:
-                        if requests.head(pu, headers=HEADERS, timeout=3, allow_redirects=True).status_code==200:
-                            parts.append(pu)
-                        else: break
-                    except: break
-        
-        if len(parts)>1:
-            return {"src":parts[0],"poster":poster,"all_parts":parts}
-        return {"src":primary[0],"poster":poster}
-    except: return None
+                        if requests.head(pu,headers=HEADERS,timeout=3,allow_redirects=True).status_code==200:parts.append(pu)
+                        else:break
+                    except:break
+        if len(parts)>1:return{"src":parts[0],"poster":poster,"all_parts":parts}
+        return{"src":primary[0],"poster":poster}
+    except:return None
 
 def extract_bdp_video(url):
     try:
-        m = re.search(r'-(\d{6,})\.html', url)
-        if not m: return None
-        vid_id = m.group(1)
-        r = requests.get(f"{BASE_BDP}/video-embed/{vid_id}.html", headers=HEADERS, timeout=10)
-        r.encoding="utf-8"
-        soup = BeautifulSoup(r.text,"lxml")
-        video = soup.select_one("video#videoPlayer")
-        if not video: return None
-        source = video.find("source")
-        return {"src":source.get("src","") if source else "","poster":video.get("poster","")}
-    except: return None
+        m=re.search(r'-(\d{6,})\.html',url)
+        if not m:return None
+        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 not video:return None
+        source=video.find("source")
+        return{"src":source.get("src","") if source else "","poster":video.get("poster","")}
+    except:return None
 
 # ═══ API ═══
-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"),
-    "cong-nghe":("https://vnexpress.net/so-hoa","Công Nghệ"),
-    "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"),
-}
-BDP_CATS = {
-    "ngoai-hang-anh":("https://bongdaplus.vn/ngoai-hang-anh","Ngoại Hạng Anh"),
-    "la-liga":("https://bongdaplus.vn/la-liga","La Liga"),
-    "champions-league":("https://bongdaplus.vn/champions-league-cup-c1","Champions League"),
-    "bong-da-vn":("https://bongdaplus.vn/bong-da-viet-nam","Bóng Đá VN"),
-}
+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"),"cong-nghe":("https://vnexpress.net/so-hoa","Công Nghệ"),"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")}
+BDP_CATS={"ngoai-hang-anh":("https://bongdaplus.vn/ngoai-hang-anh","Ngoại Hạng Anh"),"la-liga":("https://bongdaplus.vn/la-liga","La Liga"),"champions-league":("https://bongdaplus.vn/champions-league-cup-c1","Champions League"),"bong-da-vn":("https://bongdaplus.vn/bong-da-viet-nam","Bóng Đá VN")}
 
 @app.get("/api/homepage")
 def api_homepage():
     def _f():
         articles=[]
         with ThreadPoolExecutor(6) 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"]}
+            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"]}
             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
+                    for a in f.result():a["group"]=futs[f];articles.append(a)
+                except:pass
         try:
-            for a in scrape_bdp(f"{BASE_BDP}/tin-moi")[:6]: a["group"]="Bóng Đá"; articles.append(a)
-        except: pass
+            for a in scrape_bdp(f"{BASE_BDP}/tin-moi")[:6]:a["group"]="Bóng Đá";articles.append(a)
+        except:pass
         return articles
     return JSONResponse(_cached("homepage",_f))
 
@@ -314,14 +283,8 @@ def api_homepage():
 def api_category(cat_id:str):
     def _f():
         if cat_id=="bbc":return scrape_bbc_vietnamese()
-        if cat_id in VNE_CATS:
-            arts=scrape_vne(VNE_CATS[cat_id][0])
-            for a in arts:a["group"]=VNE_CATS[cat_id][1]
-            return arts
-        if cat_id in BDP_CATS:
-            arts=scrape_bdp(BDP_CATS[cat_id][0])
-            for a in arts:a["group"]=BDP_CATS[cat_id][1]
-            return arts
+        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
+        if cat_id in BDP_CATS:arts=scrape_bdp(BDP_CATS[cat_id][0]);[a.update({"group":BDP_CATS[cat_id][1]}) for a in arts];return arts
         return[]
     return JSONResponse(_cached(f"cat_{cat_id}",_f))
 
@@ -346,28 +309,25 @@ def api_bdp_videos():
 
 @app.get("/api/video_url")
 def api_video_url(url:str=Query(...)):
-    """Extract video URL. Returns {src, poster, all_parts?}."""
-    if "24h.com.vn" in url: v=extract_video_url(url)
-    elif "bongdaplus.vn" in url: v=extract_bdp_video(url)
-    else: v=None
-    return JSONResponse(v if v else {"error":"not found"})
+    if "24h.com.vn" in url:v=extract_video_url(url)
+    elif "bongdaplus.vn" in url:v=extract_bdp_video(url)
+    else:v=None
+    return JSONResponse(v if v else{"error":"not found"})
 
 @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)
-    else: data=None
-    return JSONResponse(data if data else {"error":"not supported"})
+    if "vnexpress.net" in url:data=scrape_vne_article(url)
+    elif "bbc.com" in url:data=scrape_bbc_article(url)
+    else:data=None
+    return JSONResponse(data if data else{"error":"not supported"})
 
 @app.get("/share/{path:path}")
 async def share_page(path:str,img:str=Query(default=""),title:str=Query(default="VNEWS")):
     og_image=img or "https://s1.vnecdn.net/vnexpress/restruct/i/v9505/logo_default.jpg"
-    html=f'<!DOCTYPE html><html><head><meta charset="utf-8"><title>{title}</title><meta property="og:title" content="{title}"><meta property="og:image" content="{og_image}"><meta property="og:type" content="article"><meta name="twitter:card" content="summary_large_image"><meta name="twitter:image" content="{og_image}"><meta http-equiv="refresh" content="0;url={SPACE_URL}"></head><body>Redirecting...</body></html>'
-    return HTMLResponse(content=html)
+    return HTMLResponse(f'<!DOCTYPE html><html><head><meta charset="utf-8"><title>{title}</title><meta property="og:title" content="{title}"><meta property="og:image" content="{og_image}"><meta property="og:type" content="article"><meta name="twitter:card" content="summary_large_image"><meta name="twitter:image" content="{og_image}"><meta http-equiv="refresh" content="0;url={SPACE_URL}/#/{path}"></head><body>Redirecting...</body></html>')
 
 @app.get("/")
 async def index():
-    with open("/app/static/index.html","r",encoding="utf-8") as f:
-        return HTMLResponse(content=f.read())
+    with open("/app/static/index.html","r",encoding="utf-8") as f:return HTMLResponse(content=f.read())
 
 app.mount("/static",StaticFiles(directory="/app/static"),name="static")