Fix: restore working app_v2_entry.py + add /api/match/{event_id}/detail route

app_v2_entry.py

@@ -71,25 +71,15 @@ def _s_vietnamnet(topic,limit=6):
             if len(items)>=limit:break
     except:pass
     return items
-def _s_bongda(topic,limit=6):
-    """bongda.com.vn via Jina homepage (site is JS-protected); keyword-filtered."""
+def _s_bongda(topic,limit=5):
     items=[]
     try:
-        from main import scrape_bongda_jina
-        for a in scrape_bongda_jina(40):
-            if _has_kw(topic,a.get('title','')):
-                items.append({'title':a['title'],'url':a['link'],'via':'Bóng Đá'})
-            if len(items)>=limit:break
-    except:pass
-    return items
-def _s_bongdaplus(topic,limit=6):
-    """bongdaplus.vn via Jina homepage (JS SPA); keyword-filtered."""
-    items=[]
-    try:
-        from main import scrape_bongdaplus_jina
-        for a in scrape_bongdaplus_jina(40):
-            if _has_kw(topic,a.get('title','')):
-                items.append({'title':a['title'],'url':a['link'],'via':'Bóng Đá+'})
+        r=req.get(f"https://bongda.com.vn/tim-kiem.html?q={quote(topic)}",headers={'User-Agent':'Mozilla/5.0'},timeout=8);soup=BeautifulSoup(r.text,'lxml')
+        for a in soup.select('h3 a[href], .title a[href]')[:limit*2]:
+            t=_clean(a.get_text(strip=True));href=a.get('href','')
+            if t and len(t)>15 and _has_kw(topic,t):
+                if not href.startswith('http'):href='https://bongda.com.vn'+href
+                items.append({'title':t,'url':href,'via':'Bóng Đá'})
             if len(items)>=limit:break
     except:pass
     return items
@@ -102,7 +92,7 @@ def _s_genk(topic,limit=5):
             if t and len(t)>15 and _has_kw(topic,t):
                 if href.startswith('/'):href='https://genk.vn'+href
                 items.append({'title':t,'url':href,'via':'GenK'})
-            if len(items)>=limit)break
+            if len(items)>=limit:break
     except:pass
     return items
 def _s_thanhnien(topic,limit=6):
@@ -114,7 +104,7 @@ def _s_thanhnien(topic,limit=6):
             if t and len(t)>15 and _has_kw(topic,t):
                 if not href.startswith('http'):href='https://thanhnien.vn'+href
                 items.append({'title':t,'url':href,'via':'Thanh Niên'})
-            if len(items)>=limit)break
+            if len(items)>=limit:break
     except:pass
     return items
 def _s_tuoitre(topic,limit=6):
@@ -126,7 +116,7 @@ def _s_tuoitre(topic,limit=6):
             if t and len(t)>15 and _has_kw(topic,t):
                 if not href.startswith('http'):href='https://tuoitre.vn'+href
                 items.append({'title':t,'url':href,'via':'Tuổi Trẻ'})
-            if len(items)>=limit)break
+            if len(items)>=limit:break
     except:pass
     return items
 def _s_thethaovanhoa(topic,limit=5):
@@ -138,14 +128,14 @@ def _s_thethaovanhoa(topic,limit=5):
             if t and len(t)>15 and _has_kw(topic,t):
                 if not href.startswith('http'):href='https://thethaovanhoa.vn'+href
                 items.append({'title':t,'url':href,'via':'TT&VH'})
-            if len(items)>=limit)break
+            if len(items)>=limit:break
     except:pass
     return items
 def _search_all(topic,limit=36):
     results={}
-    with ThreadPoolExecutor(10) as ex:
-        futs={ex.submit(_s_vnexpress,topic,8):'vne',ex.submit(_s_dantri,topic,8):'dt',ex.submit(_s_vietnamnet,topic,6):'vnn',ex.submit(_s_bongda,topic,6):'bd',ex.submit(_s_bongdaplus,topic,6):'bdp',ex.submit(_s_genk,topic,5):'gk',ex.submit(_s_thanhnien,topic,6):'tn',ex.submit(_s_tuoitre,topic,6):'tt',ex.submit(_s_thethaovanhoa,topic,5):'tvh'}
-        for f in as_completed(futs,timeout=16):
+    with ThreadPoolExecutor(8) as ex:
+        futs={ex.submit(_s_vnexpress,topic,8):'vne',ex.submit(_s_dantri,topic,8):'dt',ex.submit(_s_vietnamnet,topic,6):'vnn',ex.submit(_s_bongda,topic,5):'bd',ex.submit(_s_genk,topic,5):'gk',ex.submit(_s_thanhnien,topic,6):'tn',ex.submit(_s_tuoitre,topic,6):'tt',ex.submit(_s_thethaovanhoa,topic,5):'tvh'}
+        for f in as_completed(futs,timeout=14):
             try:results[futs[f]]=f.result()
             except:results[futs[f]]=[]
     srcs=list(results.values());out=[];seen=set();mx=max((len(s) for s in srcs),default=0)
@@ -153,132 +143,11 @@ def _search_all(topic,limit=36):
         for s in srcs:
             if i<len(s) and s[i].get('url') and s[i]['url'] not in seen:seen.add(s[i]['url']);out.append(s[i])
     return out[:limit]
-
-from main import (_yt_jina_page, _yt_rss_shorts, SHORTS_FALLBACK,
-                  _YT_CHANNEL_IDS, _cached, _cache_ttl_yt)
-
-def _yt_jina_shorts_classified(channel, count=24):
-    items=[];seen=set()
-    for path, tab_short in (("videos", False), ("shorts", True)):
-        md=_yt_jina_page(channel, path)
-        if not md:continue
-        for m in re.finditer(r'\[([^\]\[]{8,160})\]\(https?://(?:www\.)?youtube\.com/(shorts/|watch\?v=)([A-Za-z0-9_-]{11})[^)]*\)', md):
-            title=html_lib.unescape(m.group(1)).strip().strip('"')
-            form=m.group(2); vid=m.group(3)
-            if vid in seen:continue
-            tl=title.lower()
-            if tl in ("home home","shorts shorts","image") or title.startswith("!") or title.startswith("Image"):continue
-            is_short = True if form.startswith("shorts/") else tab_short
-            seen.add(vid)
-            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,"is_short":is_short})
-            if len(items)>=count:break
-    return items
-
-def _scrape_shorts_v2():
-    vids=[]
-    with ThreadPoolExecutor(2) as ex:
-        futs=[ex.submit(_yt_jina_shorts_classified,ch,24) for ch in ["baodantri7941","baosuckhoedoisongboyte"]]
-        for f in as_completed(futs):
-            try:
-                r=f.result()
-                if r:vids.extend(r)
-            except:pass
-    have=set(v.get("channel") for v in vids)
-    for ch in ["baodantri7941","baosuckhoedoisongboyte"]:
-        if ch not in have:
-            for v in _yt_rss_shorts(ch,24):
-                v.setdefault("is_short",True);vids.append(v)
-    merged=[];seen=set()
-    for v in vids+[dict(x) for x in SHORTS_FALLBACK]:
-        vid=v.get("id")
-        if not vid or vid in seen:continue
-        if "is_short" not in v:
-            t=(v.get("title","") or "").lower()
-            v["is_short"]= ("#shorts" in t) or ("short" in t)
-        seen.add(vid);merged.append(v)
-    return merged[:40]
-
-app.router.routes=[r for r in app.router.routes if not(getattr(r,'path',None)=='/api/shorts' and 'GET' in getattr(r,'methods',set()))]
-@app.get('/api/shorts')
-def api_shorts_v2():
-    return JSONResponse(_cached("yt_shorts_v4",_scrape_shorts_v2,ttl=_cache_ttl_yt))
-
+# Override article endpoint
 app.router.routes=[r for r in app.router.routes if not(getattr(r,'path',None)=='/api/article' and 'GET' in getattr(r,'methods',set()))]
-_JINA_BOILER_STOP=['facebooktelegram','copy link','follow us','bài viết liên quan','tin liên quan',
-    'cùng chuyên mục','bình luận','trả lời','bản quyền','địa chỉ:','lưu ý:','đăng ký nhận tin',
-    'chia sẻ bài viết','tags:','© ','bongdaplus.vn.','vật vờ studio để cập nhật','mời các bạn theo dõi',
-    'cập nhật tin tức, thủ thuật']
-_JINA_NAV=['đăng nhập','tạo tài khoản','reading:','sponsored by']
-_JINA_IMG_DENY=['logo','avatar','sponsor','vvaba','ductrinh','longnguyen','hoangtrinh','vvs_logo',
-    'cropped-','login.png','/icon','placeholder','/ads/','banner','giaminhmedia','-330x220','-420x280',
-    '-150x','-100x','thumb']
-_JINA_AD_DENY=['giaminhmedia','delivery/cl.php','[hot]','[video','săn ngay','giảm giá','khuyến mãi',
-    'mua ngay','đặt mua','shopee','lazada','tiki.vn']
-def _jina_strip(s):
-    s=re.sub(r'!\[[^\]]*\]\([^)]*\)','',s)
-    s=re.sub(r'\[([^\]]*)\]\([^)]*\)',r'\1',s)
-    s=re.sub(r'</?[a-zA-Z][^>]*>','',s)
-    s=re.sub(r'[#>*`_]+','',s)
-    return re.sub(r'\s+',' ',s).strip()
-def _jina_good_img(u,alt=''):
-    ul=u.lower();al=(alt or '').lower()
-    if any(k in ul for k in _JINA_IMG_DENY):return False
-    if 'vật vờ studio' in al or 'sponsored' in al:return False
-    if not re.search(r'\.(jpg|jpeg|png|webp|gif)',ul) and 'imgthumbnail' not in ul and '/media/' not in ul.lower():return False
-    return True
-def _jina_article(url):
-    try:
-        r=req.get("https://r.jina.ai/"+url,headers={'Accept':'text/markdown','X-Return-Format':'markdown','User-Agent':'Mozilla/5.0'},timeout=18)
-        if r.status_code!=200 or not r.text:return None
-        lines=r.text.splitlines()
-        title=''
-        for ln in lines[:30]:
-            if ln.startswith('Title:'):title=ln.replace('Title:','',1).strip();break
-        tw=set(w for w in re.findall(r'[A-Za-zÀ-ỹ0-9]+',title.lower()) if len(w)>3)
-        start=0
-        for i,ln in enumerate(lines):
-            if ln.startswith('# '):
-                hw=set(re.findall(r'[A-Za-zÀ-ỹ0-9]+',_jina_strip(ln).lower()))
-                if tw and len(tw&hw)>=max(2,len(tw)//2):start=i+1
-        body=[];og_img=''
-        for ln in lines[start:]:
-            low=_jina_strip(ln).lower()
-            if low and any(b in low for b in _JINA_BOILER_STOP):break
-            if any(a in ln.lower() for a in _JINA_AD_DENY):continue
-            mi=re.search(r'!\[([^\]]*)\]\((https?://[^)\s]+)\)',ln)
-            if mi and _jina_good_img(mi.group(2),mi.group(1)):
-                src=mi.group(2)
-                if not og_img:og_img=src
-                body.append({'type':'img','src':src});continue
-            s=ln.strip();is_h=bool(re.match(r'^#{2,4}\s',s));txt=_jina_strip(ln)
-            if not txt or len(txt)<40:continue
-            if any(p in txt.lower() for p in _JINA_NAV):continue
-            raw=re.sub(r'!\[[^\]]*\]\([^)]*\)','',ln);lt=''.join(re.findall(r'\[([^\]]*)\]\([^)]*\)',raw))
-            if len(lt)>len(txt)*0.5:continue
-            body.append({'type':'heading' if is_h else 'p','text':txt})
-            if len([b for b in body if b['type']!='img'])>=40:break
-        if not og_img:
-            for ln in lines:
-                mi=re.search(r'!\[([^\]]*)\]\((https?://[^)\s]+)\)',ln)
-                if mi and _jina_good_img(mi.group(2),mi.group(1)):og_img=mi.group(2);break
-        out=[]
-        for b in body:
-            if out and b==out[-1]:continue
-            out.append(b)
-        if not title and out:title=next((b['text'] for b in out if b['type']=='p'),'')[:90]
-        if not out:return None
-        first_p=next((b['text'] for b in out if b['type']=='p'),'')
-        return {'title':_clean(title),'summary':_clean(first_p[:200]),'og_image':og_img,'image':og_img,'body':out,'source':'jina','url':url}
-    except:return None
 def _scrape_generic(url):
     try:
-        r=req.get(url,headers={'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36','Accept-Language':'vi-VN,vi;q=0.9'},timeout=15,allow_redirects=True);r.encoding='utf-8'
-        if r.status_code>=400 or 'bongdaplus.vn' in url or 'vatvostudio.vn' in url or 'Just a moment' in r.text[:600]:
-            jr=_jina_article(url)
-            if jr and jr.get('body'):return jr
-        soup=BeautifulSoup(r.text,'lxml')
+        r=req.get(url,headers={'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36','Accept-Language':'vi-VN,vi;q=0.9'},timeout=15,allow_redirects=True);r.encoding='utf-8';soup=BeautifulSoup(r.text,'lxml')
         for tag in soup.find_all(['script','style','nav','footer','aside','form','noscript']):tag.decompose()
         h1=soup.find('h1');ogt=soup.find('meta',property='og:title');title=(h1.get_text(strip=True) if h1 else '')or(ogt.get('content','') if ogt else '')
         ogd=soup.find('meta',property='og:description');summary=ogd.get('content','') if ogd else ''
@@ -297,39 +166,19 @@ def _scrape_generic(url):
                 im=el if el.name=='img' else el.find('img')
                 if im:src=im.get('data-src') or im.get('src') or'';(body.append({'type':'img','src':'https:'+src if src.startswith('//') else src}) if src and'base64' not in src else None)
         if not body and summary:body=[{'type':'p','text':summary}]
-        if not body:
-            jr=_jina_article(url)
-            if jr and jr.get('body'):return jr
         return{'title':_clean(title),'summary':_clean(summary),'og_image':og_img,'body':body[:50],'source':'generic','url':url}
-    except:
-        return _jina_article(url)
-_ARTICLE_CACHE={};_ARTICLE_TTL=900
-def _article_cache_get(url):
-    e=_ARTICLE_CACHE.get(url)
-    if e and time.time()-e['t']<_ARTICLE_TTL:return e['d']
-    return None
-def _article_cache_put(url,data):
-    if data and data.get('body'):
-        _ARTICLE_CACHE[url]={'d':data,'t':time.time()}
-        if len(_ARTICLE_CACHE)>300:
-            for k in sorted(_ARTICLE_CACHE,key=lambda k:_ARTICLE_CACHE[k]['t'])[:100]:_ARTICLE_CACHE.pop(k,None)
+    except:return None
 @app.get('/api/article')
 def api_article_v2(url:str=Query(...)):
-    cached=_article_cache_get(url)
-    if cached:return JSONResponse(cached)
     from main import scrape_vne_article,scrape_bbc_article,scrape_dantri_article,scrape_genk_article,scrape_ttvh_article
-    used_generic=False
     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=_scrape_generic(url);used_generic=True
-    if not(data and data.get('body')) and not used_generic:
-        data=_scrape_generic(url)
-    if data and data.get('body'):
-        _article_cache_put(url,data);return JSONResponse(data)
-    return JSONResponse(data if data else{'error':'Không đọc được','url':url})
+    else:data=_scrape_generic(url)
+    if data and data.get('body'):return JSONResponse(data)
+    data=_scrape_generic(url);return JSONResponse(data if data else{'error':'Không đọc được','url':url})
 _hot_cache={'t':0,'d':[]}
 def _get_hot_topics():
     now=time.time()
@@ -349,22 +198,6 @@ def _get_hot_topics():
                         phrase=' '.join(words[i:i+n])
                         if 8<=len(phrase)<=45:key=phrase.lower();freq[key]=freq.get(key,0)+1;display[key]=phrase
         except:continue
-    def _ingest_titles(titles):
-        for title in titles:
-            title=_clean(title)
-            if not title:continue
-            title=re.sub(r'\s*[-|].*$','',title)
-            words=[w for w in re.findall(r'[A-Za-zÀ-ỹ0-9]+',title) if len(w)>2 and w.lower() not in _STOP]
-            if len(words)<2:continue
-            for n in(3,4,2):
-                for i in range(max(0,len(words)-n+1)):
-                    phrase=' '.join(words[i:i+n])
-                    if 8<=len(phrase)<=45:key=phrase.lower();freq[key]=freq.get(key,0)+1;display[key]=phrase
-    try:
-        from main import scrape_bongda_jina,scrape_bongdaplus_jina
-        _ingest_titles([a.get('title','') for a in scrape_bongda_jina(20)])
-        _ingest_titles([a.get('title','') for a in scrape_bongdaplus_jina(20)])
-    except:pass
     ranked=sorted(freq.items(),key=lambda x:x[1],reverse=True);topics=[];seen=set()
     for key,count in ranked:
         is_dup=any(len(set(e.split())&set(key.split()))/max(len(set(e.split())),len(set(key.split())),1)>0.6 for e in seen)
@@ -390,6 +223,8 @@ def _ht(topic:str=Query(...),page:int=Query(default=0)):
 def _cat():return JSONResponse([])
 @app.get('/api/storage_status')
 def _st():return JSONResponse({'persistent':os.path.isdir('/data') and os.access('/data',os.W_OK)})
+@app.get('/s')
+async def _sh(url:str='',title:str='',img:str=''):return HTMLResponse(f'<!DOCTYPE html><html><head><meta property="og:title" content="{_clean(title)}"><meta property="og:image" content="{_clean(img)}"><meta http-equiv="refresh" content="0;url={_clean(url) or "/"}"></head><body></body></html>')
 DATA_DIR='/data' if os.path.isdir('/data') else os.path.join(os.path.dirname(os.path.abspath(__file__)),'data')
 os.makedirs(DATA_DIR,exist_ok=True);IF=os.path.join(DATA_DIR,'interactions_v2.json');CF=os.path.join(DATA_DIR,'comments_v2.json')
 _il=threading.Lock();_cl=threading.Lock()